diff options
323 files changed, 5589 insertions, 3407 deletions
diff --git a/Documentation/DocBook/kgdb.tmpl b/Documentation/DocBook/kgdb.tmpl index 97618bed4d6..028a8444d95 100644 --- a/Documentation/DocBook/kgdb.tmpl +++ b/Documentation/DocBook/kgdb.tmpl @@ -72,7 +72,7 @@ kgdb is a source level debugger for linux kernel. It is used along with gdb to debug a linux kernel. The expectation is that gdb can be used to "break in" to the kernel to inspect memory, variables - and look through a cal stack information similar to what an + and look through call stack information similar to what an application developer would use gdb for. It is possible to place breakpoints in kernel code and perform some limited execution stepping. @@ -93,8 +93,10 @@ <chapter id="CompilingAKernel"> <title>Compiling a kernel</title> <para> - To enable <symbol>CONFIG_KGDB</symbol>, look under the "Kernel debugging" - and then select "KGDB: kernel debugging with remote gdb". + To enable <symbol>CONFIG_KGDB</symbol> you should first turn on + "Prompt for development and/or incomplete code/drivers" + (CONFIG_EXPERIMENTAL) in "General setup", then under the + "Kernel debugging" select "KGDB: kernel debugging with remote gdb". </para> <para> Next you should choose one of more I/O drivers to interconnect debugging diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking index c2992bc54f2..8b22d7d8b99 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -92,7 +92,6 @@ prototypes: void (*destroy_inode)(struct inode *); void (*dirty_inode) (struct inode *); int (*write_inode) (struct inode *, int); - void (*put_inode) (struct inode *); void (*drop_inode) (struct inode *); void (*delete_inode) (struct inode *); void (*put_super) (struct super_block *); @@ -115,7 +114,6 @@ alloc_inode: no no no destroy_inode: no dirty_inode: no (must not sleep) write_inode: no -put_inode: no drop_inode: no !!!inode_lock!!! delete_inode: no put_super: yes yes no diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt index 81e5be6e6e3..b7522c6cbae 100644 --- a/Documentation/filesystems/vfs.txt +++ b/Documentation/filesystems/vfs.txt @@ -205,7 +205,6 @@ struct super_operations { void (*dirty_inode) (struct inode *); int (*write_inode) (struct inode *, int); - void (*put_inode) (struct inode *); void (*drop_inode) (struct inode *); void (*delete_inode) (struct inode *); void (*put_super) (struct super_block *); @@ -246,9 +245,6 @@ or bottom half). inode to disc. The second parameter indicates whether the write should be synchronous or not, not all filesystems check this flag. - put_inode: called when the VFS inode is removed from the inode - cache. - drop_inode: called when the last access to the inode is dropped, with the inode_lock spinlock held. diff --git a/Documentation/kbuild/kconfig-language.txt b/Documentation/kbuild/kconfig-language.txt index 00b950d1c19..c412c245848 100644 --- a/Documentation/kbuild/kconfig-language.txt +++ b/Documentation/kbuild/kconfig-language.txt @@ -377,27 +377,3 @@ config FOO limits FOO to module (=m) or disabled (=n). - -Build limited by a third config symbol which may be =y or =m -~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -A common idiom that we see (and sometimes have problems with) is this: - -When option C in B (module or subsystem) uses interfaces from A (module -or subsystem), and both A and B are tristate (could be =y or =m if they -were independent of each other, but they aren't), then we need to limit -C such that it cannot be built statically if A is built as a loadable -module. (C already depends on B, so there is no dependency issue to -take care of here.) - -If A is linked statically into the kernel image, C can be built -statically or as loadable module(s). However, if A is built as loadable -module(s), then C must be restricted to loadable module(s) also. This -can be expressed in kconfig language as: - -config C - depends on A = y || A = B - -or for real examples, use this command in a kernel tree: - -$ find . -name Kconfig\* | xargs grep -ns "depends on.*=.*||.*=" | grep -v orig - diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index a3c35446e75..cdd5b934f43 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1094,9 +1094,6 @@ and is between 256 and 4096 characters. It is defined in the file mac5380= [HW,SCSI] Format: <can_queue>,<cmd_per_lun>,<sg_tablesize>,<hostid>,<use_tags> - mac53c9x= [HW,SCSI] Format: - <num_esps>,<disconnect>,<nosync>,<can_queue>,<cmd_per_lun>,<sg_tablesize>,<hostid>,<use_tags> - machvec= [IA64] Force the use of a particular machine-vector (machvec) in a generic kernel. Example: machvec=hpzx1_swiotlb @@ -1525,6 +1522,8 @@ and is between 256 and 4096 characters. It is defined in the file This is normally done in pci_enable_device(), so this option is a temporary workaround for broken drivers that don't call it. + skip_isa_align [X86] do not align io start addr, so can + handle more pci cards firmware [ARM] Do not re-enumerate the bus but instead just use the configuration from the bootloader. This is currently used on diff --git a/Documentation/s390/CommonIO b/Documentation/s390/CommonIO index 8fbc0a85287..bf0baa19ec2 100644 --- a/Documentation/s390/CommonIO +++ b/Documentation/s390/CommonIO @@ -8,17 +8,6 @@ Command line parameters Enable logging of debug information in case of ccw device timeouts. - -* cio_msg = yes | no - - Determines whether information on found devices and sensed device - characteristics should be shown during startup or when new devices are - found, i. e. messages of the types "Detected device 0.0.4711 on subchannel - 0.0.0042" and "SenseID: Device 0.0.4711 reports: ...". - - Default is off. - - * cio_ignore = {all} | {<device> | <range of devices>} | {!<device> | !<range of devices>} diff --git a/Documentation/scheduler/sched-design.txt b/Documentation/scheduler/sched-design.txt deleted file mode 100644 index 1605bf0cba8..00000000000 --- a/Documentation/scheduler/sched-design.txt +++ /dev/null @@ -1,165 +0,0 @@ - Goals, Design and Implementation of the - new ultra-scalable O(1) scheduler - - - This is an edited version of an email Ingo Molnar sent to - lkml on 4 Jan 2002. It describes the goals, design, and - implementation of Ingo's new ultra-scalable O(1) scheduler. - Last Updated: 18 April 2002. - - -Goal -==== - -The main goal of the new scheduler is to keep all the good things we know -and love about the current Linux scheduler: - - - good interactive performance even during high load: if the user - types or clicks then the system must react instantly and must execute - the user tasks smoothly, even during considerable background load. - - - good scheduling/wakeup performance with 1-2 runnable processes. - - - fairness: no process should stay without any timeslice for any - unreasonable amount of time. No process should get an unjustly high - amount of CPU time. - - - priorities: less important tasks can be started with lower priority, - more important tasks with higher priority. - - - SMP efficiency: no CPU should stay idle if there is work to do. - - - SMP affinity: processes which run on one CPU should stay affine to - that CPU. Processes should not bounce between CPUs too frequently. - - - plus additional scheduler features: RT scheduling, CPU binding. - -and the goal is also to add a few new things: - - - fully O(1) scheduling. Are you tired of the recalculation loop - blowing the L1 cache away every now and then? Do you think the goodness - loop is taking a bit too long to finish if there are lots of runnable - processes? This new scheduler takes no prisoners: wakeup(), schedule(), - the timer interrupt are all O(1) algorithms. There is no recalculation - loop. There is no goodness loop either. - - - 'perfect' SMP scalability. With the new scheduler there is no 'big' - runqueue_lock anymore - it's all per-CPU runqueues and locks - two - tasks on two separate CPUs can wake up, schedule and context-switch - completely in parallel, without any interlocking. All - scheduling-relevant data is structured for maximum scalability. - - - better SMP affinity. The old scheduler has a particular weakness that - causes the random bouncing of tasks between CPUs if/when higher - priority/interactive tasks, this was observed and reported by many - people. The reason is that the timeslice recalculation loop first needs - every currently running task to consume its timeslice. But when this - happens on eg. an 8-way system, then this property starves an - increasing number of CPUs from executing any process. Once the last - task that has a timeslice left has finished using up that timeslice, - the recalculation loop is triggered and other CPUs can start executing - tasks again - after having idled around for a number of timer ticks. - The more CPUs, the worse this effect. - - Furthermore, this same effect causes the bouncing effect as well: - whenever there is such a 'timeslice squeeze' of the global runqueue, - idle processors start executing tasks which are not affine to that CPU. - (because the affine tasks have finished off their timeslices already.) - - The new scheduler solves this problem by distributing timeslices on a - per-CPU basis, without having any global synchronization or - recalculation. - - - batch scheduling. A significant proportion of computing-intensive tasks - benefit from batch-scheduling, where timeslices are long and processes - are roundrobin scheduled. The new scheduler does such batch-scheduling - of the lowest priority tasks - so nice +19 jobs will get - 'batch-scheduled' automatically. With this scheduler, nice +19 jobs are - in essence SCHED_IDLE, from an interactiveness point of view. - - - handle extreme loads more smoothly, without breakdown and scheduling - storms. - - - O(1) RT scheduling. For those RT folks who are paranoid about the - O(nr_running) property of the goodness loop and the recalculation loop. - - - run fork()ed children before the parent. Andrea has pointed out the - advantages of this a few months ago, but patches for this feature - do not work with the old scheduler as well as they should, - because idle processes often steal the new child before the fork()ing - CPU gets to execute it. - - -Design -====== - -The core of the new scheduler contains the following mechanisms: - - - *two* priority-ordered 'priority arrays' per CPU. There is an 'active' - array and an 'expired' array. The active array contains all tasks that - are affine to this CPU and have timeslices left. The expired array - contains all tasks which have used up their timeslices - but this array - is kept sorted as well. The active and expired array is not accessed - directly, it's accessed through two pointers in the per-CPU runqueue - structure. If all active tasks are used up then we 'switch' the two - pointers and from now on the ready-to-go (former-) expired array is the - active array - and the empty active array serves as the new collector - for expired tasks. - - - there is a 64-bit bitmap cache for array indices. Finding the highest - priority task is thus a matter of two x86 BSFL bit-search instructions. - -the split-array solution enables us to have an arbitrary number of active -and expired tasks, and the recalculation of timeslices can be done -immediately when the timeslice expires. Because the arrays are always -access through the pointers in the runqueue, switching the two arrays can -be done very quickly. - -this is a hybride priority-list approach coupled with roundrobin -scheduling and the array-switch method of distributing timeslices. - - - there is a per-task 'load estimator'. - -one of the toughest things to get right is good interactive feel during -heavy system load. While playing with various scheduler variants i found -that the best interactive feel is achieved not by 'boosting' interactive -tasks, but by 'punishing' tasks that want to use more CPU time than there -is available. This method is also much easier to do in an O(1) fashion. - -to establish the actual 'load' the task contributes to the system, a -complex-looking but pretty accurate method is used: there is a 4-entry -'history' ringbuffer of the task's activities during the last 4 seconds. -This ringbuffer is operated without much overhead. The entries tell the -scheduler a pretty accurate load-history of the task: has it used up more -CPU time or less during the past N seconds. [the size '4' and the interval -of 4x 1 seconds was found by lots of experimentation - this part is -flexible and can be changed in both directions.] - -the penalty a task gets for generating more load than the CPU can handle -is a priority decrease - there is a maximum amount to this penalty -relative to their static priority, so even fully CPU-bound tasks will -observe each other's priorities, and will share the CPU accordingly. - -the SMP load-balancer can be extended/switched with additional parallel -computing and cache hierarchy concepts: NUMA scheduling, multi-core CPUs -can be supported easily by changing the load-balancer. Right now it's -tuned for my SMP systems. - -i skipped the prev->mm == next->mm advantage - no workload i know of shows -any sensitivity to this. It can be added back by sacrificing O(1) -schedule() [the current and one-lower priority list can be searched for a -that->mm == current->mm condition], but costs a fair number of cycles -during a number of important workloads, so i wanted to avoid this as much -as possible. - -- the SMP idle-task startup code was still racy and the new scheduler -triggered this. So i streamlined the idle-setup code a bit. We do not call -into schedule() before all processors have started up fully and all idle -threads are in place. - -- the patch also cleans up a number of aspects of sched.c - moves code -into other areas of the kernel where it's appropriate, and simplifies -certain code paths and data constructs. As a result, the new scheduler's -code is smaller than the old one. - - Ingo diff --git a/MAINTAINERS b/MAINTAINERS index abe27871ad6..f5583dc7ea3 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2112,12 +2112,10 @@ L: netdev@vger.kernel.org S: Maintained INTEL ETHERNET DRIVERS (e100/e1000/e1000e/igb/ixgb/ixgbe) -P: Auke Kok -M: auke-jan.h.kok@intel.com -P: Jesse Brandeburg -M: jesse.brandeburg@intel.com P: Jeff Kirsher M: jeffrey.t.kirsher@intel.com +P: Jesse Brandeburg +M: jesse.brandeburg@intel.com P: Bruce Allan M: bruce.w.allan@intel.com P: John Ronciak diff --git a/arch/arm/kernel/sys_arm.c b/arch/arm/kernel/sys_arm.c index 9bd1870d980..0128687ba0f 100644 --- a/arch/arm/kernel/sys_arm.c +++ b/arch/arm/kernel/sys_arm.c @@ -34,23 +34,6 @@ extern unsigned long do_mremap(unsigned long addr, unsigned long old_len, unsigned long new_len, unsigned long flags, unsigned long new_addr); -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - /* common code for old and new mmaps */ inline long do_mmap2( unsigned long addr, unsigned long len, diff --git a/arch/arm/mach-orion5x/addr-map.c b/arch/arm/mach-orion5x/addr-map.c index 9608503d67f..e63fb05dc89 100644 --- a/arch/arm/mach-orion5x/addr-map.c +++ b/arch/arm/mach-orion5x/addr-map.c @@ -34,11 +34,7 @@ * Non-CPU Masters address decoding -- * Unlike the CPU, we setup the access from Orion's master interfaces to DDR * banks only (the typical use case). - * Setup access for each master to DDR is issued by common.c. - * - * Note: although orion_setbits() and orion_clrbits() are not atomic - * no locking is necessary here since code in this file is only called - * at boot time when there is no concurrency issues. + * Setup access for each master to DDR is issued by platform device setup. */ /* @@ -48,10 +44,6 @@ #define TARGET_DEV_BUS 1 #define TARGET_PCI 3 #define TARGET_PCIE 4 -#define ATTR_DDR_CS(n) (((n) ==0) ? 0xe : \ - ((n) == 1) ? 0xd : \ - ((n) == 2) ? 0xb : \ - ((n) == 3) ? 0x7 : 0xf) #define ATTR_PCIE_MEM 0x59 #define ATTR_PCIE_IO 0x51 #define ATTR_PCIE_WA 0x79 @@ -61,17 +53,12 @@ #define ATTR_DEV_CS1 0x1d #define ATTR_DEV_CS2 0x1b #define ATTR_DEV_BOOT 0xf -#define WIN_EN 1 /* * Helpers to get DDR bank info */ -#define DDR_BASE_CS(n) ORION5X_DDR_REG(0x1500 + ((n) * 8)) -#define DDR_SIZE_CS(n) ORION5X_DDR_REG(0x1504 + ((n) * 8)) -#define DDR_MAX_CS 4 -#define DDR_REG_TO_SIZE(reg) (((reg) | 0xffffff) + 1) -#define DDR_REG_TO_BASE(reg) ((reg) & 0xff000000) -#define DDR_BANK_EN 1 +#define DDR_BASE_CS(n) ORION5X_DDR_REG(0x1500 + ((n) << 3)) +#define DDR_SIZE_CS(n) ORION5X_DDR_REG(0x1504 + ((n) << 3)) /* * CPU Address Decode Windows registers @@ -81,17 +68,6 @@ #define CPU_WIN_REMAP_LO(n) ORION5X_BRIDGE_REG(0x008 | ((n) << 4)) #define CPU_WIN_REMAP_HI(n) ORION5X_BRIDGE_REG(0x00c | ((n) << 4)) -/* - * Gigabit Ethernet Address Decode Windows registers - */ -#define ETH_WIN_BASE(win) ORION5X_ETH_REG(0x200 + ((win) * 8)) -#define ETH_WIN_SIZE(win) ORION5X_ETH_REG(0x204 + ((win) * 8)) -#define ETH_WIN_REMAP(win) ORION5X_ETH_REG(0x280 + ((win) * 4)) -#define ETH_WIN_EN ORION5X_ETH_REG(0x290) -#define ETH_WIN_PROT ORION5X_ETH_REG(0x294) -#define ETH_MAX_WIN 6 -#define ETH_MAX_REMAP_WIN 4 - struct mbus_dram_target_info orion5x_mbus_dram_info; @@ -202,39 +178,3 @@ void __init orion5x_setup_pcie_wa_win(u32 base, u32 size) { setup_cpu_win(7, base, size, TARGET_PCIE, ATTR_PCIE_WA, -1); } - -void __init orion5x_setup_eth_wins(void) -{ - int i; - - /* - * First, disable and clear windows - */ - for (i = 0; i < ETH_MAX_WIN; i++) { - orion5x_write(ETH_WIN_BASE(i), 0); - orion5x_write(ETH_WIN_SIZE(i), 0); - orion5x_setbits(ETH_WIN_EN, 1 << i); - orion5x_clrbits(ETH_WIN_PROT, 0x3 << (i * 2)); - if (i < ETH_MAX_REMAP_WIN) - orion5x_write(ETH_WIN_REMAP(i), 0); - } - - /* - * Setup windows for DDR banks. - */ - for (i = 0; i < DDR_MAX_CS; i++) { - u32 base, size; - size = orion5x_read(DDR_SIZE_CS(i)); - base = orion5x_read(DDR_BASE_CS(i)); - if (size & DDR_BANK_EN) { - base = DDR_REG_TO_BASE(base); - size = DDR_REG_TO_SIZE(size); - orion5x_write(ETH_WIN_SIZE(i), (size-1) & 0xffff0000); - orion5x_write(ETH_WIN_BASE(i), (base & 0xffff0000) | - (ATTR_DDR_CS(i) << 8) | - TARGET_DDR); - orion5x_clrbits(ETH_WIN_EN, 1 << i); - orion5x_setbits(ETH_WIN_PROT, 0x3 << (i * 2)); - } - } -} diff --git a/arch/arm/mach-orion5x/common.c b/arch/arm/mach-orion5x/common.c index 968deb58be0..4f13fd037f0 100644 --- a/arch/arm/mach-orion5x/common.c +++ b/arch/arm/mach-orion5x/common.c @@ -190,6 +190,11 @@ static struct platform_device orion5x_ehci1 = { * (The Orion and Discovery (MV643xx) families use the same Ethernet driver) ****************************************************************************/ +struct mv643xx_eth_shared_platform_data orion5x_eth_shared_data = { + .dram = &orion5x_mbus_dram_info, + .t_clk = ORION5X_TCLK, +}; + static struct resource orion5x_eth_shared_resources[] = { { .start = ORION5X_ETH_PHYS_BASE + 0x2000, @@ -201,6 +206,9 @@ static struct resource orion5x_eth_shared_resources[] = { static struct platform_device orion5x_eth_shared = { .name = MV643XX_ETH_SHARED_NAME, .id = 0, + .dev = { + .platform_data = &orion5x_eth_shared_data, + }, .num_resources = 1, .resource = orion5x_eth_shared_resources, }; @@ -223,7 +231,9 @@ static struct platform_device orion5x_eth = { void __init orion5x_eth_init(struct mv643xx_eth_platform_data *eth_data) { + eth_data->shared = &orion5x_eth_shared; orion5x_eth.dev.platform_data = eth_data; + platform_device_register(&orion5x_eth_shared); platform_device_register(&orion5x_eth); } @@ -360,7 +370,6 @@ void __init orion5x_init(void) * Setup Orion address map */ orion5x_setup_cpu_mbus_bridge(); - orion5x_setup_eth_wins(); /* * Register devices. diff --git a/arch/arm/mach-orion5x/common.h b/arch/arm/mach-orion5x/common.h index 14adf8d1a54..bd0f05de6e1 100644 --- a/arch/arm/mach-orion5x/common.h +++ b/arch/arm/mach-orion5x/common.h @@ -22,7 +22,6 @@ void orion5x_setup_dev0_win(u32 base, u32 size); void orion5x_setup_dev1_win(u32 base, u32 size); void orion5x_setup_dev2_win(u32 base, u32 size); void orion5x_setup_pcie_wa_win(u32 base, u32 size); -void orion5x_setup_eth_wins(void); /* * Shared code used internally by other Orion core functions. diff --git a/arch/avr32/kernel/sys_avr32.c b/arch/avr32/kernel/sys_avr32.c index 8deb6003ee6..8e8911e55c8 100644 --- a/arch/avr32/kernel/sys_avr32.c +++ b/arch/avr32/kernel/sys_avr32.c @@ -14,19 +14,6 @@ #include <asm/mman.h> #include <asm/uaccess.h> -asmlinkage int sys_pipe(unsigned long __user *filedes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(filedes, fd, sizeof(fd))) - error = -EFAULT; - } - return error; -} - asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, off_t offset) diff --git a/arch/blackfin/kernel/sys_bfin.c b/arch/blackfin/kernel/sys_bfin.c index efb7b25a263..fce49d7cf00 100644 --- a/arch/blackfin/kernel/sys_bfin.c +++ b/arch/blackfin/kernel/sys_bfin.c @@ -45,23 +45,6 @@ #include <asm/cacheflush.h> #include <asm/dma.h> -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2 * sizeof(int))) - error = -EFAULT; - } - return error; -} - /* common code for old and new mmaps */ static inline long do_mmap2(unsigned long addr, unsigned long len, diff --git a/arch/cris/kernel/sys_cris.c b/arch/cris/kernel/sys_cris.c index 8b9984197ed..d124066e172 100644 --- a/arch/cris/kernel/sys_cris.c +++ b/arch/cris/kernel/sys_cris.c @@ -40,8 +40,11 @@ asmlinkage int sys_pipe(unsigned long __user * fildes) error = do_pipe(fd); unlock_kernel(); if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) + if (copy_to_user(fildes, fd, 2*sizeof(int))) { + sys_close(fd[0]); + sys_close(fd[1]); error = -EFAULT; + } } return error; } diff --git a/arch/frv/kernel/sys_frv.c b/arch/frv/kernel/sys_frv.c index 04c6b1677cc..49b2cf2c38f 100644 --- a/arch/frv/kernel/sys_frv.c +++ b/arch/frv/kernel/sys_frv.c @@ -28,23 +28,6 @@ #include <asm/setup.h> #include <asm/uaccess.h> -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -asmlinkage long sys_pipe(unsigned long __user * fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff) diff --git a/arch/h8300/kernel/sys_h8300.c b/arch/h8300/kernel/sys_h8300.c index 00608be6d56..2745656dcc5 100644 --- a/arch/h8300/kernel/sys_h8300.c +++ b/arch/h8300/kernel/sys_h8300.c @@ -27,23 +27,6 @@ #include <asm/traps.h> #include <asm/unistd.h> -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long * fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - /* common code for old and new mmaps */ static inline long do_mmap2( unsigned long addr, unsigned long len, diff --git a/arch/m32r/kernel/sys_m32r.c b/arch/m32r/kernel/sys_m32r.c index 6d7a80fdad4..319c79720b8 100644 --- a/arch/m32r/kernel/sys_m32r.c +++ b/arch/m32r/kernel/sys_m32r.c @@ -90,8 +90,11 @@ sys_pipe(unsigned long r0, unsigned long r1, unsigned long r2, error = do_pipe(fd); if (!error) { - if (copy_to_user((void __user *)r0, fd, 2*sizeof(int))) + if (copy_to_user((void __user *)r0, fd, 2*sizeof(int))) { + sys_close(fd[0]); + sys_close(fd[1]); error = -EFAULT; + } } return error; } diff --git a/arch/m68k/kernel/sys_m68k.c b/arch/m68k/kernel/sys_m68k.c index e892f17ba3f..7f54efaf60b 100644 --- a/arch/m68k/kernel/sys_m68k.c +++ b/arch/m68k/kernel/sys_m68k.c @@ -30,23 +30,6 @@ #include <asm/page.h> #include <asm/unistd.h> -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long __user * fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - /* common code for old and new mmaps */ static inline long do_mmap2( unsigned long addr, unsigned long len, diff --git a/arch/m68k/kernel/traps.c b/arch/m68k/kernel/traps.c index fd4858e2dd6..75b8340b254 100644 --- a/arch/m68k/kernel/traps.c +++ b/arch/m68k/kernel/traps.c @@ -468,15 +468,26 @@ static inline void access_error040(struct frame *fp) * (if do_page_fault didn't fix the mapping, * the writeback won't do good) */ +disable_wb: #ifdef DEBUG printk(".. disabling wb2\n"); #endif if (fp->un.fmt7.wb2a == fp->un.fmt7.faddr) fp->un.fmt7.wb2s &= ~WBV_040; + if (fp->un.fmt7.wb3a == fp->un.fmt7.faddr) + fp->un.fmt7.wb3s &= ~WBV_040; } - } else if (send_fault_sig(&fp->ptregs) > 0) { - printk("68040 access error, ssw=%x\n", ssw); - trap_c(fp); + } else { + /* In case of a bus error we either kill the process or expect + * the kernel to catch the fault, which then is also responsible + * for cleaning up the mess. + */ + current->thread.signo = SIGBUS; + current->thread.faddr = fp->un.fmt7.faddr; + if (send_fault_sig(&fp->ptregs) >= 0) + printk("68040 bus error (ssw=%x, faddr=%lx)\n", ssw, + fp->un.fmt7.faddr); + goto disable_wb; } do_040writebacks(fp); diff --git a/arch/m68k/mac/config.c b/arch/m68k/mac/config.c index 735a49b4b93..ad3e3bacae3 100644 --- a/arch/m68k/mac/config.c +++ b/arch/m68k/mac/config.c @@ -48,9 +48,6 @@ struct mac_booter_data mac_bi_data; int mac_bisize = sizeof mac_bi_data; -struct mac_hw_present mac_hw_present; -EXPORT_SYMBOL(mac_hw_present); - /* New m68k bootinfo stuff and videobase */ extern int m68k_num_memory; @@ -817,27 +814,6 @@ void __init mac_identify(void) m68k_ramdisk.addr, m68k_ramdisk.size); #endif - /* - * TODO: set the various fields in macintosh_config->hw_present here! - */ - switch (macintosh_config->scsi_type) { - case MAC_SCSI_OLD: - MACHW_SET(MAC_SCSI_80); - break; - case MAC_SCSI_QUADRA: - case MAC_SCSI_QUADRA2: - case MAC_SCSI_QUADRA3: - MACHW_SET(MAC_SCSI_96); - if ((macintosh_config->ident == MAC_MODEL_Q900) || - (macintosh_config->ident == MAC_MODEL_Q950)) - MACHW_SET(MAC_SCSI_96_2); - break; - default: - printk(KERN_WARNING "config.c: wtf: unknown scsi, using 53c80\n"); - MACHW_SET(MAC_SCSI_80); - break; - } - iop_init(); via_init(); oss_init(); diff --git a/arch/m68knommu/kernel/sys_m68k.c b/arch/m68knommu/kernel/sys_m68k.c index 65f7a95f056..70028163862 100644 --- a/arch/m68knommu/kernel/sys_m68k.c +++ b/arch/m68knommu/kernel/sys_m68k.c @@ -28,23 +28,6 @@ #include <asm/cacheflush.h> #include <asm/unistd.h> -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long * fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - /* common code for old and new mmaps */ static inline long do_mmap2( unsigned long addr, unsigned long len, diff --git a/arch/mn10300/kernel/sys_mn10300.c b/arch/mn10300/kernel/sys_mn10300.c index 5f17a1ebc82..bca5a84dc72 100644 --- a/arch/mn10300/kernel/sys_mn10300.c +++ b/arch/mn10300/kernel/sys_mn10300.c @@ -29,23 +29,6 @@ #define MIN_MAP_ADDR PAGE_SIZE /* minimum fixed mmap address */ /* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way Unix traditionally does this, though. - */ -asmlinkage long sys_pipe(unsigned long __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2 * sizeof(int))) - error = -EFAULT; - } - return error; -} - -/* * memory mapping syscall */ asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, diff --git a/arch/parisc/kernel/sys_parisc.c b/arch/parisc/kernel/sys_parisc.c index 4f589216b39..71b31957c8f 100644 --- a/arch/parisc/kernel/sys_parisc.c +++ b/arch/parisc/kernel/sys_parisc.c @@ -33,19 +33,6 @@ #include <linux/utsname.h> #include <linux/personality.h> -int sys_pipe(int __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - static unsigned long get_unshared_area(unsigned long addr, unsigned long len) { struct vm_area_struct *vma; diff --git a/arch/powerpc/kernel/syscalls.c b/arch/powerpc/kernel/syscalls.c index e722a4eeb5d..4fe69ca2448 100644 --- a/arch/powerpc/kernel/syscalls.c +++ b/arch/powerpc/kernel/syscalls.c @@ -136,23 +136,6 @@ int sys_ipc(uint call, int first, unsigned long second, long third, return ret; } -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -int sys_pipe(int __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - static inline unsigned long do_mmap2(unsigned long addr, size_t len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long off, int shift) diff --git a/arch/powerpc/kvm/booke_guest.c b/arch/powerpc/kvm/booke_guest.c index 6d9884a6884..712d89a28c4 100644 --- a/arch/powerpc/kvm/booke_guest.c +++ b/arch/powerpc/kvm/booke_guest.c @@ -49,6 +49,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "inst_emu", VCPU_STAT(emulated_inst_exits) }, { "dec", VCPU_STAT(dec_exits) }, { "ext_intr", VCPU_STAT(ext_intr_exits) }, + { "halt_wakeup", VCPU_STAT(halt_wakeup) }, { NULL } }; @@ -338,6 +339,11 @@ int kvmppc_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu, } break; + case BOOKE_INTERRUPT_FP_UNAVAIL: + kvmppc_queue_exception(vcpu, exit_nr); + r = RESUME_GUEST; + break; + case BOOKE_INTERRUPT_DATA_STORAGE: vcpu->arch.dear = vcpu->arch.fault_dear; vcpu->arch.esr = vcpu->arch.fault_esr; diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index bad40bd2d3a..777e0f34e0e 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -36,13 +36,12 @@ gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn) int kvm_cpu_has_interrupt(struct kvm_vcpu *v) { - /* XXX implement me */ - return 0; + return !!(v->arch.pending_exceptions); } int kvm_arch_vcpu_runnable(struct kvm_vcpu *v) { - return 1; + return !(v->arch.msr & MSR_WE); } @@ -214,6 +213,11 @@ static void kvmppc_decrementer_func(unsigned long data) struct kvm_vcpu *vcpu = (struct kvm_vcpu *)data; kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_DECREMENTER); + + if (waitqueue_active(&vcpu->wq)) { + wake_up_interruptible(&vcpu->wq); + vcpu->stat.halt_wakeup++; + } } int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) @@ -339,6 +343,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) int r; sigset_t sigsaved; + vcpu_load(vcpu); + if (vcpu->sigset_active) sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved); @@ -363,12 +369,20 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) if (vcpu->sigset_active) sigprocmask(SIG_SETMASK, &sigsaved, NULL); + vcpu_put(vcpu); + return r; } int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq) { kvmppc_queue_exception(vcpu, BOOKE_INTERRUPT_EXTERNAL); + + if (waitqueue_active(&vcpu->wq)) { + wake_up_interruptible(&vcpu->wq); + vcpu->stat.halt_wakeup++; + } + return 0; } diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile index 4bb023f4c86..f1d2cdc5331 100644 --- a/arch/powerpc/lib/Makefile +++ b/arch/powerpc/lib/Makefile @@ -23,3 +23,4 @@ obj-$(CONFIG_SMP) += locks.o endif obj-$(CONFIG_PPC_LIB_RHEAP) += rheap.o +obj-$(CONFIG_HAS_IOMEM) += devres.o diff --git a/arch/powerpc/lib/devres.c b/arch/powerpc/lib/devres.c new file mode 100644 index 00000000000..292115d98ea --- /dev/null +++ b/arch/powerpc/lib/devres.c @@ -0,0 +1,42 @@ +/* + * Copyright (C) 2008 Freescale Semiconductor, Inc. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include <linux/device.h> /* devres_*(), devm_ioremap_release() */ +#include <linux/io.h> /* ioremap_flags() */ +#include <linux/module.h> /* EXPORT_SYMBOL() */ + +/** + * devm_ioremap_prot - Managed ioremap_flags() + * @dev: Generic device to remap IO address for + * @offset: BUS offset to map + * @size: Size of map + * @flags: Page flags + * + * Managed ioremap_prot(). Map is automatically unmapped on driver + * detach. + */ +void __iomem *devm_ioremap_prot(struct device *dev, resource_size_t offset, + size_t size, unsigned long flags) +{ + void __iomem **ptr, *addr; + + ptr = devres_alloc(devm_ioremap_release, sizeof(*ptr), GFP_KERNEL); + if (!ptr) + return NULL; + + addr = ioremap_flags(offset, size, flags); + if (addr) { + *ptr = addr; + devres_add(dev, ptr); + } else + devres_free(ptr); + + return addr; +} +EXPORT_SYMBOL(devm_ioremap_prot); diff --git a/arch/powerpc/platforms/chrp/pegasos_eth.c b/arch/powerpc/platforms/chrp/pegasos_eth.c index 5bcc58d9a4d..130ff72d99d 100644 --- a/arch/powerpc/platforms/chrp/pegasos_eth.c +++ b/arch/powerpc/platforms/chrp/pegasos_eth.c @@ -58,7 +58,9 @@ static struct resource mv643xx_eth0_resources[] = { static struct mv643xx_eth_platform_data eth0_pd = { + .shared = &mv643xx_eth_shared_device, .port_number = 0, + .tx_sram_addr = PEGASOS2_SRAM_BASE_ETH0, .tx_sram_size = PEGASOS2_SRAM_TXRING_SIZE, .tx_queue_size = PEGASOS2_SRAM_TXRING_SIZE/16, @@ -88,7 +90,9 @@ static struct resource mv643xx_eth1_resources[] = { }; static struct mv643xx_eth_platform_data eth1_pd = { + .shared = &mv643xx_eth_shared_device, .port_number = 1, + .tx_sram_addr = PEGASOS2_SRAM_BASE_ETH1, .tx_sram_size = PEGASOS2_SRAM_TXRING_SIZE, .tx_queue_size = PEGASOS2_SRAM_TXRING_SIZE/16, diff --git a/arch/powerpc/platforms/pseries/scanlog.c b/arch/powerpc/platforms/pseries/scanlog.c index bec3803f061..417eca79df6 100644 --- a/arch/powerpc/platforms/pseries/scanlog.c +++ b/arch/powerpc/platforms/pseries/scanlog.c @@ -55,11 +55,6 @@ static ssize_t scanlog_read(struct file *file, char __user *buf, dp = PDE(inode); data = (unsigned int *)dp->data; - if (!data) { - printk(KERN_ERR "scanlog: read failed no data\n"); - return -EIO; - } - if (count > RTAS_DATA_BUF_SIZE) count = RTAS_DATA_BUF_SIZE; @@ -146,11 +141,6 @@ static int scanlog_open(struct inode * inode, struct file * file) struct proc_dir_entry *dp = PDE(inode); unsigned int *data = (unsigned int *)dp->data; - if (!data) { - printk(KERN_ERR "scanlog: open failed no data\n"); - return -EIO; - } - if (data[0] != 0) { /* This imperfect test stops a second copy of the * data (or a reset while data is being copied) @@ -168,10 +158,6 @@ static int scanlog_release(struct inode * inode, struct file * file) struct proc_dir_entry *dp = PDE(inode); unsigned int *data = (unsigned int *)dp->data; - if (!data) { - printk(KERN_ERR "scanlog: release failed no data\n"); - return -EIO; - } data[0] = 0; return 0; @@ -200,12 +186,11 @@ static int __init scanlog_init(void) if (!data) goto err; - ent = proc_create("ppc64/rtas/scan-log-dump", S_IRUSR, NULL, - &scanlog_fops); + ent = proc_create_data("ppc64/rtas/scan-log-dump", S_IRUSR, NULL, + &scanlog_fops, data); if (!ent) goto err; - ent->data = data; proc_ppc64_scan_log_dump = ent; return 0; diff --git a/arch/powerpc/sysdev/mv64x60_dev.c b/arch/powerpc/sysdev/mv64x60_dev.c index 41af1223e2a..a132e0de8ca 100644 --- a/arch/powerpc/sysdev/mv64x60_dev.c +++ b/arch/powerpc/sysdev/mv64x60_dev.c @@ -239,6 +239,8 @@ static int __init mv64x60_eth_device_setup(struct device_node *np, int id, memset(&pdata, 0, sizeof(pdata)); + pdata.shared = shared_pdev; + prop = of_get_property(np, "reg", NULL); if (!prop) return -ENODEV; diff --git a/arch/ppc/syslib/mv64x60.c b/arch/ppc/syslib/mv64x60.c index 90fe904d361..418f3053de5 100644 --- a/arch/ppc/syslib/mv64x60.c +++ b/arch/ppc/syslib/mv64x60.c @@ -341,6 +341,7 @@ static struct resource mv64x60_eth0_resources[] = { }; static struct mv643xx_eth_platform_data eth0_pd = { + .shared = &mv64x60_eth_shared_device; .port_number = 0, }; @@ -366,6 +367,7 @@ static struct resource mv64x60_eth1_resources[] = { }; static struct mv643xx_eth_platform_data eth1_pd = { + .shared = &mv64x60_eth_shared_device; .port_number = 1, }; @@ -391,6 +393,7 @@ static struct resource mv64x60_eth2_resources[] = { }; static struct mv643xx_eth_platform_data eth2_pd = { + .shared = &mv64x60_eth_shared_device; .port_number = 2, }; diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 29a7940f284..1d035082e78 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -430,6 +430,13 @@ config CMM_IUCV Select this option to enable the special message interface to the cooperative memory management. +config PAGE_STATES + bool "Unused page notification" + help + This enables the notification of unused pages to the + hypervisor. The ESSA instruction is used to do the states + changes between a page that has content and the unused state. + config VIRT_TIMER bool "Virtual CPU timer support" help diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S index 743d54f0b8d..d003a6e16af 100644 --- a/arch/s390/kernel/compat_wrapper.S +++ b/arch/s390/kernel/compat_wrapper.S @@ -121,7 +121,7 @@ sys32_ptrace_wrapper: lgfr %r3,%r3 # long llgtr %r4,%r4 # long llgfr %r5,%r5 # long - jg sys_ptrace # branch to system call + jg compat_sys_ptrace # branch to system call .globl sys32_alarm_wrapper sys32_alarm_wrapper: diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index bdbb3bcd78a..708cf9cf9a3 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -279,8 +279,6 @@ sysc_do_restart: st %r2,SP_R2(%r15) # store return value (change R2 on stack) sysc_return: - tm SP_PSW+1(%r15),0x01 # returning to user ? - bno BASED(sysc_restore) tm __TI_flags+3(%r9),_TIF_WORK_SVC bnz BASED(sysc_work) # there is work to do (signals etc.) sysc_restore: @@ -312,6 +310,8 @@ sysc_work_loop: # One of the work bits is on. Find out which one. # sysc_work: + tm SP_PSW+1(%r15),0x01 # returning to user ? + bno BASED(sysc_restore) tm __TI_flags+3(%r9),_TIF_MCCK_PENDING bo BASED(sysc_mcck_pending) tm __TI_flags+3(%r9),_TIF_NEED_RESCHED @@ -602,12 +602,6 @@ io_no_vtime: la %r2,SP_PTREGS(%r15) # address of register-save area basr %r14,%r1 # branch to standard irq handler io_return: - tm SP_PSW+1(%r15),0x01 # returning to user ? -#ifdef CONFIG_PREEMPT - bno BASED(io_preempt) # no -> check for preemptive scheduling -#else - bno BASED(io_restore) # no-> skip resched & signal -#endif tm __TI_flags+3(%r9),_TIF_WORK_INT bnz BASED(io_work) # there is work to do (signals etc.) io_restore: @@ -629,10 +623,18 @@ io_restore_trace_psw: .long 0, io_restore_trace + 0x80000000 #endif -#ifdef CONFIG_PREEMPT -io_preempt: +# +# switch to kernel stack, then check the TIF bits +# +io_work: + tm SP_PSW+1(%r15),0x01 # returning to user ? +#ifndef CONFIG_PREEMPT + bno BASED(io_restore) # no-> skip resched & signal +#else + bnz BASED(io_work_user) # no -> check for preemptive scheduling + # check for preemptive scheduling icm %r0,15,__TI_precount(%r9) - bnz BASED(io_restore) + bnz BASED(io_restore) # preemption disabled l %r1,SP_R15(%r15) s %r1,BASED(.Lc_spsize) mvc SP_PTREGS(__PT_SIZE,%r1),SP_PTREGS(%r15) @@ -646,10 +648,7 @@ io_resume_loop: br %r1 # call schedule #endif -# -# switch to kernel stack, then check the TIF bits -# -io_work: +io_work_user: l %r1,__LC_KERNEL_STACK s %r1,BASED(.Lc_spsize) mvc SP_PTREGS(__PT_SIZE,%r1),SP_PTREGS(%r15) diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S index 5a4a7bcd2bb..fee10177dbf 100644 --- a/arch/s390/kernel/entry64.S +++ b/arch/s390/kernel/entry64.S @@ -271,8 +271,6 @@ sysc_noemu: stg %r2,SP_R2(%r15) # store return value (change R2 on stack) sysc_return: - tm SP_PSW+1(%r15),0x01 # returning to user ? - jno sysc_restore tm __TI_flags+7(%r9),_TIF_WORK_SVC jnz sysc_work # there is work to do (signals etc.) sysc_restore: @@ -304,6 +302,8 @@ sysc_work_loop: # One of the work bits is on. Find out which one. # sysc_work: + tm SP_PSW+1(%r15),0x01 # returning to user ? + jno sysc_restore tm __TI_flags+7(%r9),_TIF_MCCK_PENDING jo sysc_mcck_pending tm __TI_flags+7(%r9),_TIF_NEED_RESCHED @@ -585,12 +585,6 @@ io_no_vtime: la %r2,SP_PTREGS(%r15) # address of register-save area brasl %r14,do_IRQ # call standard irq handler io_return: - tm SP_PSW+1(%r15),0x01 # returning to user ? -#ifdef CONFIG_PREEMPT - jno io_preempt # no -> check for preemptive scheduling -#else - jno io_restore # no-> skip resched & signal -#endif tm __TI_flags+7(%r9),_TIF_WORK_INT jnz io_work # there is work to do (signals etc.) io_restore: @@ -612,10 +606,41 @@ io_restore_trace_psw: .quad 0, io_restore_trace #endif -#ifdef CONFIG_PREEMPT -io_preempt: +# +# There is work todo, we need to check if we return to userspace, then +# check, if we are in SIE, if yes leave it +# +io_work: + tm SP_PSW+1(%r15),0x01 # returning to user ? +#ifndef CONFIG_PREEMPT +#if defined(CONFIG_KVM) || defined(CONFIG_KVM_MODULE) + jnz io_work_user # yes -> no need to check for SIE + la %r1, BASED(sie_opcode) # we return to kernel here + lg %r2, SP_PSW+8(%r15) + clc 0(2,%r1), 0(%r2) # is current instruction = SIE? + jne io_restore # no-> return to kernel + lg %r1, SP_PSW+8(%r15) # yes-> add 4 bytes to leave SIE + aghi %r1, 4 + stg %r1, SP_PSW+8(%r15) + j io_restore # return to kernel +#else + jno io_restore # no-> skip resched & signal +#endif +#else + jnz io_work_user # yes -> do resched & signal +#if defined(CONFIG_KVM) || defined(CONFIG_KVM_MODULE) + la %r1, BASED(sie_opcode) + lg %r2, SP_PSW+8(%r15) + clc 0(2,%r1), 0(%r2) # is current instruction = SIE? + jne 0f # no -> leave PSW alone + lg %r1, SP_PSW+8(%r15) # yes-> add 4 bytes to leave SIE + aghi %r1, 4 + stg %r1, SP_PSW+8(%r15) +0: +#endif + # check for preemptive scheduling icm %r0,15,__TI_precount(%r9) - jnz io_restore + jnz io_restore # preemption is disabled # switch to kernel stack lg %r1,SP_R15(%r15) aghi %r1,-SP_SIZE @@ -629,10 +654,7 @@ io_resume_loop: jg preempt_schedule_irq #endif -# -# switch to kernel stack, then check TIF bits -# -io_work: +io_work_user: lg %r1,__LC_KERNEL_STACK aghi %r1,-SP_SIZE mvc SP_PTREGS(__PT_SIZE,%r1),SP_PTREGS(%r15) @@ -653,6 +675,11 @@ io_work_loop: j io_restore io_work_done: +#if defined(CONFIG_KVM) || defined(CONFIG_KVM_MODULE) +sie_opcode: + .long 0xb2140000 +#endif + # # _TIF_MCCK_PENDING is set, call handler # diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index 7f427016374..35827b9bd4d 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -292,8 +292,7 @@ poke_user(struct task_struct *child, addr_t addr, addr_t data) return 0; } -static int -do_ptrace_normal(struct task_struct *child, long request, long addr, long data) +long arch_ptrace(struct task_struct *child, long request, long addr, long data) { ptrace_area parea; int copied, ret; @@ -529,35 +528,19 @@ poke_user_emu31(struct task_struct *child, addr_t addr, addr_t data) return 0; } -static int -do_ptrace_emu31(struct task_struct *child, long request, long addr, long data) +long compat_arch_ptrace(struct task_struct *child, compat_long_t request, + compat_ulong_t caddr, compat_ulong_t cdata) { - unsigned int tmp; /* 4 bytes !! */ + unsigned long addr = caddr; + unsigned long data = cdata; ptrace_area_emu31 parea; int copied, ret; switch (request) { - case PTRACE_PEEKTEXT: - case PTRACE_PEEKDATA: - /* read word at location addr. */ - copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 0); - if (copied != sizeof(tmp)) - return -EIO; - return put_user(tmp, (unsigned int __force __user *) data); - case PTRACE_PEEKUSR: /* read the word at location addr in the USER area. */ return peek_user_emu31(child, addr, data); - case PTRACE_POKETEXT: - case PTRACE_POKEDATA: - /* write the word at location addr. */ - tmp = data; - copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 1); - if (copied != sizeof(tmp)) - return -EIO; - return 0; - case PTRACE_POKEUSR: /* write the word at location addr in the USER area */ return poke_user_emu31(child, addr, data); @@ -587,82 +570,11 @@ do_ptrace_emu31(struct task_struct *child, long request, long addr, long data) copied += sizeof(unsigned int); } return 0; - case PTRACE_GETEVENTMSG: - return put_user((__u32) child->ptrace_message, - (unsigned int __force __user *) data); - case PTRACE_GETSIGINFO: - if (child->last_siginfo == NULL) - return -EINVAL; - return copy_siginfo_to_user32((compat_siginfo_t - __force __user *) data, - child->last_siginfo); - case PTRACE_SETSIGINFO: - if (child->last_siginfo == NULL) - return -EINVAL; - return copy_siginfo_from_user32(child->last_siginfo, - (compat_siginfo_t - __force __user *) data); } - return ptrace_request(child, request, addr, data); + return compat_ptrace_request(child, request, addr, data); } #endif -long arch_ptrace(struct task_struct *child, long request, long addr, long data) -{ - switch (request) { - case PTRACE_SYSCALL: - /* continue and stop at next (return from) syscall */ - case PTRACE_CONT: - /* restart after signal. */ - if (!valid_signal(data)) - return -EIO; - if (request == PTRACE_SYSCALL) - set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - else - clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - child->exit_code = data; - /* make sure the single step bit is not set. */ - user_disable_single_step(child); - wake_up_process(child); - return 0; - - case PTRACE_KILL: - /* - * make the child exit. Best I can do is send it a sigkill. - * perhaps it should be put in the status that it wants to - * exit. - */ - if (child->exit_state == EXIT_ZOMBIE) /* already dead */ - return 0; - child->exit_code = SIGKILL; - /* make sure the single step bit is not set. */ - user_disable_single_step(child); - wake_up_process(child); - return 0; - - case PTRACE_SINGLESTEP: - /* set the trap flag. */ - if (!valid_signal(data)) - return -EIO; - clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - child->exit_code = data; - user_enable_single_step(child); - /* give it a chance to run. */ - wake_up_process(child); - return 0; - - /* Do requests that differ for 31/64 bit */ - default: -#ifdef CONFIG_COMPAT - if (test_thread_flag(TIF_31BIT)) - return do_ptrace_emu31(child, request, addr, data); -#endif - return do_ptrace_normal(child, request, addr, data); - } - /* Not reached. */ - return -EIO; -} - asmlinkage void syscall_trace(struct pt_regs *regs, int entryexit) { diff --git a/arch/s390/kernel/sys_s390.c b/arch/s390/kernel/sys_s390.c index 988d0d64c2c..5fdb799062b 100644 --- a/arch/s390/kernel/sys_s390.c +++ b/arch/s390/kernel/sys_s390.c @@ -32,23 +32,6 @@ #include <asm/uaccess.h> #include "entry.h" -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way Unix traditionally does this, though. - */ -asmlinkage long sys_pipe(unsigned long __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - /* common code for old and new mmaps */ static inline long do_mmap2( unsigned long addr, unsigned long len, diff --git a/arch/s390/kvm/Kconfig b/arch/s390/kvm/Kconfig index 1761b74d639..e051cad1f1e 100644 --- a/arch/s390/kvm/Kconfig +++ b/arch/s390/kvm/Kconfig @@ -22,7 +22,6 @@ config KVM select PREEMPT_NOTIFIERS select ANON_INODES select S390_SWITCH_AMODE - select PREEMPT ---help--- Support hosting paravirtualized guest machines using the SIE virtualization capability on the mainframe. This should work diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c index 349581a2610..47a0b642174 100644 --- a/arch/s390/kvm/intercept.c +++ b/arch/s390/kvm/intercept.c @@ -105,6 +105,9 @@ static intercept_handler_t instruction_handlers[256] = { static int handle_noop(struct kvm_vcpu *vcpu) { switch (vcpu->arch.sie_block->icptcode) { + case 0x0: + vcpu->stat.exit_null++; + break; case 0x10: vcpu->stat.exit_external_request++; break; diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 98d1e73e01f..0ac36a649eb 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -31,6 +31,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "userspace_handled", VCPU_STAT(exit_userspace) }, + { "exit_null", VCPU_STAT(exit_null) }, { "exit_validity", VCPU_STAT(exit_validity) }, { "exit_stop_request", VCPU_STAT(exit_stop_request) }, { "exit_external_request", VCPU_STAT(exit_external_request) }, @@ -221,10 +222,6 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) vcpu->arch.guest_fpregs.fpc &= FPC_VALID_MASK; restore_fp_regs(&vcpu->arch.guest_fpregs); restore_access_regs(vcpu->arch.guest_acrs); - - if (signal_pending(current)) - atomic_set_mask(CPUSTAT_STOP_INT, - &vcpu->arch.sie_block->cpuflags); } void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) diff --git a/arch/s390/mm/Makefile b/arch/s390/mm/Makefile index fb988a48a75..2a745813454 100644 --- a/arch/s390/mm/Makefile +++ b/arch/s390/mm/Makefile @@ -5,3 +5,4 @@ obj-y := init.o fault.o extmem.o mmap.o vmem.o pgtable.o obj-$(CONFIG_CMM) += cmm.o obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o +obj-$(CONFIG_PAGE_STATES) += page-states.o diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c index fa31de6ae97..29f3a63806b 100644 --- a/arch/s390/mm/init.c +++ b/arch/s390/mm/init.c @@ -126,6 +126,9 @@ void __init mem_init(void) /* clear the zero-page */ memset(empty_zero_page, 0, PAGE_SIZE); + /* Setup guest page hinting */ + cmma_init(); + /* this will put all low memory onto the freelists */ totalram_pages += free_all_bootmem(); diff --git a/arch/s390/mm/page-states.c b/arch/s390/mm/page-states.c new file mode 100644 index 00000000000..fc0ad73ffd9 --- /dev/null +++ b/arch/s390/mm/page-states.c @@ -0,0 +1,79 @@ +/* + * arch/s390/mm/page-states.c + * + * Copyright IBM Corp. 2008 + * + * Guest page hinting for unused pages. + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/types.h> +#include <linux/mm.h> +#include <linux/init.h> + +#define ESSA_SET_STABLE 1 +#define ESSA_SET_UNUSED 2 + +static int cmma_flag; + +static int __init cmma(char *str) +{ + char *parm; + parm = strstrip(str); + if (strcmp(parm, "yes") == 0 || strcmp(parm, "on") == 0) { + cmma_flag = 1; + return 1; + } + cmma_flag = 0; + if (strcmp(parm, "no") == 0 || strcmp(parm, "off") == 0) + return 1; + return 0; +} + +__setup("cmma=", cmma); + +void __init cmma_init(void) +{ + register unsigned long tmp asm("0") = 0; + register int rc asm("1") = -EOPNOTSUPP; + + if (!cmma_flag) + return; + asm volatile( + " .insn rrf,0xb9ab0000,%1,%1,0,0\n" + "0: la %0,0\n" + "1:\n" + EX_TABLE(0b,1b) + : "+&d" (rc), "+&d" (tmp)); + if (rc) + cmma_flag = 0; +} + +void arch_free_page(struct page *page, int order) +{ + int i, rc; + + if (!cmma_flag) + return; + for (i = 0; i < (1 << order); i++) + asm volatile(".insn rrf,0xb9ab0000,%0,%1,%2,0" + : "=&d" (rc) + : "a" ((page_to_pfn(page) + i) << PAGE_SHIFT), + "i" (ESSA_SET_UNUSED)); +} + +void arch_alloc_page(struct page *page, int order) +{ + int i, rc; + + if (!cmma_flag) + return; + for (i = 0; i < (1 << order); i++) + asm volatile(".insn rrf,0xb9ab0000,%0,%1,%2,0" + : "=&d" (rc) + : "a" ((page_to_pfn(page) + i) << PAGE_SHIFT), + "i" (ESSA_SET_STABLE)); +} diff --git a/arch/sh/kernel/sys_sh64.c b/arch/sh/kernel/sys_sh64.c index 578004d71e0..91fb8445a5a 100644 --- a/arch/sh/kernel/sys_sh64.c +++ b/arch/sh/kernel/sys_sh64.c @@ -31,23 +31,6 @@ #include <asm/unistd.h> /* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way Unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long * fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - -/* * Do a system call from kernel instead of calling sys_execve so we * end up with proper pt_regs. */ diff --git a/arch/sparc/kernel/process.c b/arch/sparc/kernel/process.c index e7f35198ae3..36431f377de 100644 --- a/arch/sparc/kernel/process.c +++ b/arch/sparc/kernel/process.c @@ -419,14 +419,26 @@ asmlinkage int sparc_do_fork(unsigned long clone_flags, unsigned long stack_size) { unsigned long parent_tid_ptr, child_tid_ptr; + unsigned long orig_i1 = regs->u_regs[UREG_I1]; + long ret; parent_tid_ptr = regs->u_regs[UREG_I2]; child_tid_ptr = regs->u_regs[UREG_I4]; - return do_fork(clone_flags, stack_start, - regs, stack_size, - (int __user *) parent_tid_ptr, - (int __user *) child_tid_ptr); + ret = do_fork(clone_flags, stack_start, + regs, stack_size, + (int __user *) parent_tid_ptr, + (int __user *) child_tid_ptr); + + /* If we get an error and potentially restart the system + * call, we're screwed because copy_thread() clobbered + * the parent's %o1. So detect that case and restore it + * here. + */ + if ((unsigned long)ret >= -ERESTART_RESTARTBLOCK) + regs->u_regs[UREG_I1] = orig_i1; + + return ret; } /* Copy a Sparc thread. The fork() return value conventions diff --git a/arch/sparc/kernel/signal.c b/arch/sparc/kernel/signal.c index 3c312290c3c..368157926d2 100644 --- a/arch/sparc/kernel/signal.c +++ b/arch/sparc/kernel/signal.c @@ -245,15 +245,29 @@ static inline int invalid_frame_pointer(void __user *fp, int fplen) static inline void __user *get_sigframe(struct sigaction *sa, struct pt_regs *regs, unsigned long framesize) { - unsigned long sp; + unsigned long sp = regs->u_regs[UREG_FP]; - sp = regs->u_regs[UREG_FP]; + /* + * If we are on the alternate signal stack and would overflow it, don't. + * Return an always-bogus address instead so we will die with SIGSEGV. + */ + if (on_sig_stack(sp) && !likely(on_sig_stack(sp - framesize))) + return (void __user *) -1L; /* This is the X/Open sanctioned signal stack switching. */ if (sa->sa_flags & SA_ONSTACK) { - if (!on_sig_stack(sp) && !((current->sas_ss_sp + current->sas_ss_size) & 7)) + if (sas_ss_flags(sp) == 0) sp = current->sas_ss_sp + current->sas_ss_size; } + + /* Always align the stack frame. This handles two cases. First, + * sigaltstack need not be mindful of platform specific stack + * alignment. Second, if we took this signal because the stack + * is not aligned properly, we'd like to take the signal cleanly + * and report that. + */ + sp &= ~7UL; + return (void __user *)(sp - framesize); } diff --git a/arch/sparc/kernel/sys_sparc.c b/arch/sparc/kernel/sys_sparc.c index f188b5dc9fd..e995491c443 100644 --- a/arch/sparc/kernel/sys_sparc.c +++ b/arch/sparc/kernel/sys_sparc.c @@ -223,8 +223,7 @@ int sparc_mmap_check(unsigned long addr, unsigned long len, unsigned long flags) { if (ARCH_SUN4C_SUN4 && (len > 0x20000000 || - ((flags & MAP_FIXED) && - addr < 0xe0000000 && addr + len > 0x20000000))) + (addr < 0xe0000000 && addr + len > 0x20000000))) return -EINVAL; /* See asm-sparc/uaccess.h */ diff --git a/arch/sparc64/kernel/process.c b/arch/sparc64/kernel/process.c index 500ac6d483a..4129c044985 100644 --- a/arch/sparc64/kernel/process.c +++ b/arch/sparc64/kernel/process.c @@ -503,6 +503,8 @@ asmlinkage long sparc_do_fork(unsigned long clone_flags, unsigned long stack_size) { int __user *parent_tid_ptr, *child_tid_ptr; + unsigned long orig_i1 = regs->u_regs[UREG_I1]; + long ret; #ifdef CONFIG_COMPAT if (test_thread_flag(TIF_32BIT)) { @@ -515,9 +517,19 @@ asmlinkage long sparc_do_fork(unsigned long clone_flags, child_tid_ptr = (int __user *) regs->u_regs[UREG_I4]; } - return do_fork(clone_flags, stack_start, - regs, stack_size, - parent_tid_ptr, child_tid_ptr); + ret = do_fork(clone_flags, stack_start, + regs, stack_size, + parent_tid_ptr, child_tid_ptr); + + /* If we get an error and potentially restart the system + * call, we're screwed because copy_thread() clobbered + * the parent's %o1. So detect that case and restore it + * here. + */ + if ((unsigned long)ret >= -ERESTART_RESTARTBLOCK) + regs->u_regs[UREG_I1] = orig_i1; + + return ret; } /* Copy a Sparc thread. The fork() return value conventions diff --git a/arch/sparc64/kernel/signal.c b/arch/sparc64/kernel/signal.c index 45d6bf632da..07c0443ea3f 100644 --- a/arch/sparc64/kernel/signal.c +++ b/arch/sparc64/kernel/signal.c @@ -376,16 +376,29 @@ save_fpu_state(struct pt_regs *regs, __siginfo_fpu_t __user *fpu) static inline void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, unsigned long framesize) { - unsigned long sp; + unsigned long sp = regs->u_regs[UREG_FP] + STACK_BIAS; - sp = regs->u_regs[UREG_FP] + STACK_BIAS; + /* + * If we are on the alternate signal stack and would overflow it, don't. + * Return an always-bogus address instead so we will die with SIGSEGV. + */ + if (on_sig_stack(sp) && !likely(on_sig_stack(sp - framesize))) + return (void __user *) -1L; /* This is the X/Open sanctioned signal stack switching. */ if (ka->sa.sa_flags & SA_ONSTACK) { - if (!on_sig_stack(sp) && - !((current->sas_ss_sp + current->sas_ss_size) & 7)) + if (sas_ss_flags(sp) == 0) sp = current->sas_ss_sp + current->sas_ss_size; } + + /* Always align the stack frame. This handles two cases. First, + * sigaltstack need not be mindful of platform specific stack + * alignment. Second, if we took this signal because the stack + * is not aligned properly, we'd like to take the signal cleanly + * and report that. + */ + sp &= ~7UL; + return (void __user *)(sp - framesize); } diff --git a/arch/sparc64/kernel/signal32.c b/arch/sparc64/kernel/signal32.c index 9415d2c918c..0f6b7b156ef 100644 --- a/arch/sparc64/kernel/signal32.c +++ b/arch/sparc64/kernel/signal32.c @@ -406,11 +406,27 @@ static void __user *get_sigframe(struct sigaction *sa, struct pt_regs *regs, uns regs->u_regs[UREG_FP] &= 0x00000000ffffffffUL; sp = regs->u_regs[UREG_FP]; + /* + * If we are on the alternate signal stack and would overflow it, don't. + * Return an always-bogus address instead so we will die with SIGSEGV. + */ + if (on_sig_stack(sp) && !likely(on_sig_stack(sp - framesize))) + return (void __user *) -1L; + /* This is the X/Open sanctioned signal stack switching. */ if (sa->sa_flags & SA_ONSTACK) { - if (!on_sig_stack(sp) && !((current->sas_ss_sp + current->sas_ss_size) & 7)) + if (sas_ss_flags(sp) == 0) sp = current->sas_ss_sp + current->sas_ss_size; } + + /* Always align the stack frame. This handles two cases. First, + * sigaltstack need not be mindful of platform specific stack + * alignment. Second, if we took this signal because the stack + * is not aligned properly, we'd like to take the signal cleanly + * and report that. + */ + sp &= ~7UL; + return (void __user *)(sp - framesize); } diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c index 3aba47624df..0d6403a630a 100644 --- a/arch/sparc64/kernel/smp.c +++ b/arch/sparc64/kernel/smp.c @@ -865,21 +865,14 @@ void smp_call_function_client(int irq, struct pt_regs *regs) void *info = call_data->info; clear_softint(1 << irq); - - irq_enter(); - - if (!call_data->wait) { - /* let initiator proceed after getting data */ - atomic_inc(&call_data->finished); - } - - func(info); - - irq_exit(); - if (call_data->wait) { /* let initiator proceed only after completion */ + func(info); atomic_inc(&call_data->finished); + } else { + /* let initiator proceed after getting data */ + atomic_inc(&call_data->finished); + func(info); } } @@ -1041,9 +1034,7 @@ void smp_receive_signal(int cpu) void smp_receive_signal_client(int irq, struct pt_regs *regs) { - irq_enter(); clear_softint(1 << irq); - irq_exit(); } void smp_new_mmu_context_version_client(int irq, struct pt_regs *regs) @@ -1051,8 +1042,6 @@ void smp_new_mmu_context_version_client(int irq, struct pt_regs *regs) struct mm_struct *mm; unsigned long flags; - irq_enter(); - clear_softint(1 << irq); /* See if we need to allocate a new TLB context because @@ -1072,8 +1061,6 @@ void smp_new_mmu_context_version_client(int irq, struct pt_regs *regs) load_secondary_context(mm); __flush_tlb_mm(CTX_HWBITS(mm->context), SECONDARY_CONTEXT); - - irq_exit(); } void smp_new_mmu_context_version(void) @@ -1239,8 +1226,6 @@ void smp_penguin_jailcell(int irq, struct pt_regs *regs) { clear_softint(1 << irq); - irq_enter(); - preempt_disable(); __asm__ __volatile__("flushw"); @@ -1253,8 +1238,6 @@ void smp_penguin_jailcell(int irq, struct pt_regs *regs) prom_world(0); preempt_enable(); - - irq_exit(); } /* /proc/profile writes can call this, don't __init it please. */ diff --git a/arch/sparc64/kernel/sys_sparc.c b/arch/sparc64/kernel/sys_sparc.c index 8d4761f15fa..0dbc941f130 100644 --- a/arch/sparc64/kernel/sys_sparc.c +++ b/arch/sparc64/kernel/sys_sparc.c @@ -549,13 +549,13 @@ int sparc64_mmap_check(unsigned long addr, unsigned long len, if (len >= STACK_TOP32) return -EINVAL; - if ((flags & MAP_FIXED) && addr > STACK_TOP32 - len) + if (addr > STACK_TOP32 - len) return -EINVAL; } else { if (len >= VA_EXCLUDE_START) return -EINVAL; - if ((flags & MAP_FIXED) && invalid_64bit_range(addr, len)) + if (invalid_64bit_range(addr, len)) return -EINVAL; } diff --git a/arch/sparc64/kernel/sys_sparc32.c b/arch/sparc64/kernel/sys_sparc32.c index 161ce4710fe..1aa4288125f 100644 --- a/arch/sparc64/kernel/sys_sparc32.c +++ b/arch/sparc64/kernel/sys_sparc32.c @@ -236,13 +236,6 @@ asmlinkage long sys32_getegid16(void) /* 32-bit timeval and related flotsam. */ -static long get_tv32(struct timeval *o, struct compat_timeval __user *i) -{ - return (!access_ok(VERIFY_READ, i, sizeof(*i)) || - (__get_user(o->tv_sec, &i->tv_sec) | - __get_user(o->tv_usec, &i->tv_usec))); -} - static inline long put_tv32(struct compat_timeval __user *o, struct timeval *i) { return (!access_ok(VERIFY_WRITE, o, sizeof(*o)) || @@ -757,30 +750,6 @@ asmlinkage long sys32_settimeofday(struct compat_timeval __user *tv, return do_sys_settimeofday(tv ? &kts : NULL, tz ? &ktz : NULL); } -asmlinkage long sys32_utimes(char __user *filename, - struct compat_timeval __user *tvs) -{ - struct timespec tv[2]; - - if (tvs) { - struct timeval ktvs[2]; - if (get_tv32(&ktvs[0], tvs) || - get_tv32(&ktvs[1], 1+tvs)) - return -EFAULT; - - if (ktvs[0].tv_usec < 0 || ktvs[0].tv_usec >= 1000000 || - ktvs[1].tv_usec < 0 || ktvs[1].tv_usec >= 1000000) - return -EINVAL; - - tv[0].tv_sec = ktvs[0].tv_sec; - tv[0].tv_nsec = 1000 * ktvs[0].tv_usec; - tv[1].tv_sec = ktvs[1].tv_sec; - tv[1].tv_nsec = 1000 * ktvs[1].tv_usec; - } - - return do_utimes(AT_FDCWD, filename, tvs ? tv : NULL, 0); -} - /* These are here just in case some old sparc32 binary calls it. */ asmlinkage long sys32_pause(void) { diff --git a/arch/sparc64/kernel/systbls.S b/arch/sparc64/kernel/systbls.S index a4fef2ba1ae..8b5282d433c 100644 --- a/arch/sparc64/kernel/systbls.S +++ b/arch/sparc64/kernel/systbls.S @@ -45,7 +45,7 @@ sys_call_table32: /*120*/ .word compat_sys_readv, compat_sys_writev, sys32_settimeofday, sys32_fchown16, sys_fchmod .word sys_nis_syscall, sys32_setreuid16, sys32_setregid16, sys_rename, sys_truncate /*130*/ .word sys_ftruncate, sys_flock, compat_sys_lstat64, sys_nis_syscall, sys_nis_syscall - .word sys_nis_syscall, sys32_mkdir, sys_rmdir, sys32_utimes, compat_sys_stat64 + .word sys_nis_syscall, sys32_mkdir, sys_rmdir, compat_sys_utimes, compat_sys_stat64 /*140*/ .word sys32_sendfile64, sys_nis_syscall, sys32_futex, sys_gettid, compat_sys_getrlimit .word compat_sys_setrlimit, sys_pivot_root, sys32_prctl, sys_pciconfig_read, sys_pciconfig_write /*150*/ .word sys_nis_syscall, sys_inotify_init, sys_inotify_add_watch, sys_poll, sys_getdents64 diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c index 4cad0b32b0a..ec3e2c72302 100644 --- a/arch/sparc64/mm/init.c +++ b/arch/sparc64/mm/init.c @@ -771,6 +771,9 @@ static void __init find_ramdisk(unsigned long phys_base) initrd_end = ramdisk_image + sparc_ramdisk_size; lmb_reserve(initrd_start, initrd_end); + + initrd_start += PAGE_OFFSET; + initrd_end += PAGE_OFFSET; } #endif } @@ -2362,16 +2365,3 @@ void __flush_tlb_all(void) __asm__ __volatile__("wrpr %0, 0, %%pstate" : : "r" (pstate)); } - -#ifdef CONFIG_MEMORY_HOTPLUG - -void online_page(struct page *page) -{ - ClearPageReserved(page); - init_page_count(page); - __free_page(page); - totalram_pages++; - num_physpages++; -} - -#endif /* CONFIG_MEMORY_HOTPLUG */ diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c index 10b86e1cc65..5047490fc29 100644 --- a/arch/um/drivers/line.c +++ b/arch/um/drivers/line.c @@ -191,9 +191,9 @@ void line_flush_chars(struct tty_struct *tty) line_flush_buffer(tty); } -void line_put_char(struct tty_struct *tty, unsigned char ch) +int line_put_char(struct tty_struct *tty, unsigned char ch) { - line_write(tty, &ch, sizeof(ch)); + return line_write(tty, &ch, sizeof(ch)); } int line_write(struct tty_struct *tty, const unsigned char *buf, int len) diff --git a/arch/um/include/line.h b/arch/um/include/line.h index 1223f2c844b..979b73e6352 100644 --- a/arch/um/include/line.h +++ b/arch/um/include/line.h @@ -71,7 +71,7 @@ extern int line_setup(struct line *lines, unsigned int sizeof_lines, char *init, char **error_out); extern int line_write(struct tty_struct *tty, const unsigned char *buf, int len); -extern void line_put_char(struct tty_struct *tty, unsigned char ch); +extern int line_put_char(struct tty_struct *tty, unsigned char ch); extern void line_set_termios(struct tty_struct *tty, struct ktermios * old); extern int line_chars_in_buffer(struct tty_struct *tty); extern void line_flush_buffer(struct tty_struct *tty); diff --git a/arch/um/kernel/syscall.c b/arch/um/kernel/syscall.c index 9cffc628a37..128ee85bc8d 100644 --- a/arch/um/kernel/syscall.c +++ b/arch/um/kernel/syscall.c @@ -73,23 +73,6 @@ long old_mmap(unsigned long addr, unsigned long len, out: return err; } -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -long sys_pipe(unsigned long __user * fildes) -{ - int fd[2]; - long error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, sizeof(fd))) - error = -EFAULT; - } - return error; -} - long sys_uname(struct old_utsname __user * name) { diff --git a/arch/v850/kernel/syscalls.c b/arch/v850/kernel/syscalls.c index 003db9c8c44..1a83daf8e24 100644 --- a/arch/v850/kernel/syscalls.c +++ b/arch/v850/kernel/syscalls.c @@ -132,23 +132,6 @@ sys_ipc (uint call, int first, int second, int third, void *ptr, long fifth) return ret; } -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way unix traditionally does this, though. - */ -int sys_pipe (int *fildes) -{ - int fd[2]; - int error; - - error = do_pipe (fd); - if (!error) { - if (copy_to_user (fildes, fd, 2*sizeof (int))) - error = -EFAULT; - } - return error; -} - static inline unsigned long do_mmap2 (unsigned long addr, size_t len, unsigned long prot, unsigned long flags, diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index c3f880902d6..bbcafaa160c 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -18,6 +18,7 @@ config X86_64 ### Arch settings config X86 def_bool y + select HAVE_UNSTABLE_SCHED_CLOCK select HAVE_IDE select HAVE_OPROFILE select HAVE_KPROBES @@ -1661,6 +1662,7 @@ config GEODE_MFGPT_TIMER config OLPC bool "One Laptop Per Child support" + depends on MGEODE_LX default n help Add support for detecting the unique features of the OLPC diff --git a/arch/x86/boot/compressed/relocs.c b/arch/x86/boot/compressed/relocs.c index d01ea42187e..edaadea90aa 100644 --- a/arch/x86/boot/compressed/relocs.c +++ b/arch/x86/boot/compressed/relocs.c @@ -191,7 +191,7 @@ static void read_ehdr(FILE *fp) die("Cannot read ELF header: %s\n", strerror(errno)); } - if (memcmp(ehdr.e_ident, ELFMAG, 4) != 0) { + if (memcmp(ehdr.e_ident, ELFMAG, SELFMAG) != 0) { die("No ELF magic\n"); } if (ehdr.e_ident[EI_CLASS] != ELFCLASS32) { diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index bbdacb398d4..5e618c3b472 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -83,9 +83,7 @@ obj-$(CONFIG_KVM_GUEST) += kvm.o obj-$(CONFIG_KVM_CLOCK) += kvmclock.o obj-$(CONFIG_PARAVIRT) += paravirt.o paravirt_patch_$(BITS).o -ifdef CONFIG_INPUT_PCSPKR -obj-y += pcspeaker.o -endif +obj-$(CONFIG_PCSPKR_PLATFORM) += pcspeaker.o obj-$(CONFIG_SCx200) += scx200.o scx200-y += scx200_32.o diff --git a/arch/x86/kernel/acpi/Makefile b/arch/x86/kernel/acpi/Makefile index 7335959b6af..fd5ca97a2ad 100644 --- a/arch/x86/kernel/acpi/Makefile +++ b/arch/x86/kernel/acpi/Makefile @@ -10,5 +10,5 @@ endif $(obj)/wakeup_rm.o: $(obj)/realmode/wakeup.bin $(obj)/realmode/wakeup.bin: FORCE - $(Q)$(MAKE) $(build)=$(obj)/realmode $@ + $(Q)$(MAKE) $(build)=$(obj)/realmode diff --git a/arch/x86/kernel/acpi/realmode/Makefile b/arch/x86/kernel/acpi/realmode/Makefile index 092900854ac..1c31cc0e9de 100644 --- a/arch/x86/kernel/acpi/realmode/Makefile +++ b/arch/x86/kernel/acpi/realmode/Makefile @@ -6,7 +6,8 @@ # for more details. # -targets := wakeup.bin wakeup.elf +always := wakeup.bin +targets := wakeup.elf wakeup.lds wakeup-y += wakeup.o wakemain.o video-mode.o copy.o @@ -48,7 +49,7 @@ LDFLAGS_wakeup.elf := -T CPPFLAGS_wakeup.lds += -P -C -$(obj)/wakeup.elf: $(src)/wakeup.lds $(WAKEUP_OBJS) FORCE +$(obj)/wakeup.elf: $(obj)/wakeup.lds $(WAKEUP_OBJS) FORCE $(call if_changed,ld) OBJCOPYFLAGS_wakeup.bin := -O binary diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c index ddee04043ae..4bc1be5d547 100644 --- a/arch/x86/kernel/kvmclock.c +++ b/arch/x86/kernel/kvmclock.c @@ -133,6 +133,7 @@ static int kvm_register_clock(void) return native_write_msr_safe(MSR_KVM_SYSTEM_TIME, low, high); } +#ifdef CONFIG_X86_LOCAL_APIC static void kvm_setup_secondary_clock(void) { /* @@ -143,6 +144,7 @@ static void kvm_setup_secondary_clock(void) /* ok, done with our trickery, call native */ setup_secondary_APIC_clock(); } +#endif /* * After the clock is registered, the host will keep writing to the @@ -177,7 +179,9 @@ void __init kvmclock_init(void) pv_time_ops.get_wallclock = kvm_get_wallclock; pv_time_ops.set_wallclock = kvm_set_wallclock; pv_time_ops.sched_clock = kvm_clock_read; +#ifdef CONFIG_X86_LOCAL_APIC pv_apic_ops.setup_secondary_clock = kvm_setup_secondary_clock; +#endif machine_ops.shutdown = kvm_shutdown; #ifdef CONFIG_KEXEC machine_ops.crash_shutdown = kvm_crash_shutdown; diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index 3e2c54dc8b2..404683b94e7 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -794,6 +794,11 @@ void __init find_smp_config(void) ACPI-based MP Configuration -------------------------------------------------------------------------- */ +/* + * Keep this outside and initialized to 0, for !CONFIG_ACPI builds: + */ +int es7000_plat; + #ifdef CONFIG_ACPI #ifdef CONFIG_X86_IO_APIC @@ -909,8 +914,6 @@ void __init mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi) MP_intsrc_info(&intsrc); } -int es7000_plat; - void __init mp_config_acpi_legacy_irqs(void) { struct mpc_config_intsrc intsrc; diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 07c6d42ab5f..f6be7d5f82f 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -149,7 +149,6 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { .matches = { DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 745"), - DMI_MATCH(DMI_BOARD_NAME, "0WF810"), }, }, { /* Handle problems with rebooting on Dell Optiplex 745's DFF*/ diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index c0c68c18a78..cc6f5eb20b2 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -95,7 +95,7 @@ void __init setup_per_cpu_areas(void) /* Copy section for each CPU (we discard the original) */ size = PERCPU_ENOUGH_ROOM; - printk(KERN_INFO "PERCPU: Allocating %lu bytes of per cpu data\n", + printk(KERN_INFO "PERCPU: Allocating %zd bytes of per cpu data\n", size); for_each_possible_cpu(i) { diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 84241a256dc..6b087ab6cd8 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -299,7 +299,7 @@ static void __cpuinit smp_callin(void) /* * Activate a secondary processor. */ -void __cpuinit start_secondary(void *unused) +static void __cpuinit start_secondary(void *unused) { /* * Don't put *anything* before cpu_init(), SMP booting is too @@ -1306,7 +1306,7 @@ static void remove_siblinginfo(int cpu) cpu_clear(cpu, cpu_sibling_setup_map); } -int additional_cpus __initdata = -1; +static int additional_cpus __initdata = -1; static __init int setup_additional_cpus(char *s) { diff --git a/arch/x86/kernel/sys_i386_32.c b/arch/x86/kernel/sys_i386_32.c index a86d26f036e..d2ab52cc1d6 100644 --- a/arch/x86/kernel/sys_i386_32.c +++ b/arch/x86/kernel/sys_i386_32.c @@ -22,23 +22,6 @@ #include <asm/uaccess.h> #include <asm/unistd.h> -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way Unix traditionally does this, though. - */ -asmlinkage int sys_pipe(unsigned long __user * fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff) diff --git a/arch/x86/kernel/sys_x86_64.c b/arch/x86/kernel/sys_x86_64.c index bd802a5e1aa..3b360ef3381 100644 --- a/arch/x86/kernel/sys_x86_64.c +++ b/arch/x86/kernel/sys_x86_64.c @@ -17,23 +17,6 @@ #include <asm/uaccess.h> #include <asm/ia32.h> -/* - * sys_pipe() is the normal C calling standard for creating - * a pipe. It's not the way Unix traditionally does this, though. - */ -asmlinkage long sys_pipe(int __user *fildes) -{ - int fd[2]; - int error; - - error = do_pipe(fd); - if (!error) { - if (copy_to_user(fildes, fd, 2*sizeof(int))) - error = -EFAULT; - } - return error; -} - asmlinkage long sys_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long off) { diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c index 4c943eabacc..3324d90038e 100644 --- a/arch/x86/kvm/i8254.c +++ b/arch/x86/kvm/i8254.c @@ -288,6 +288,8 @@ static void pit_load_count(struct kvm *kvm, int channel, u32 val) * mode 1 is one shot, mode 2 is period, otherwise del timer */ switch (ps->channels[0].mode) { case 1: + /* FIXME: enhance mode 4 precision */ + case 4: create_pit_timer(&ps->pit_timer, val, 0); break; case 2: diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 2ad6f548167..36c5406b181 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -79,36 +79,6 @@ static int dbg = 1; } #endif -#define PT64_PT_BITS 9 -#define PT64_ENT_PER_PAGE (1 << PT64_PT_BITS) -#define PT32_PT_BITS 10 -#define PT32_ENT_PER_PAGE (1 << PT32_PT_BITS) - -#define PT_WRITABLE_SHIFT 1 - -#define PT_PRESENT_MASK (1ULL << 0) -#define PT_WRITABLE_MASK (1ULL << PT_WRITABLE_SHIFT) -#define PT_USER_MASK (1ULL << 2) -#define PT_PWT_MASK (1ULL << 3) -#define PT_PCD_MASK (1ULL << 4) -#define PT_ACCESSED_MASK (1ULL << 5) -#define PT_DIRTY_MASK (1ULL << 6) -#define PT_PAGE_SIZE_MASK (1ULL << 7) -#define PT_PAT_MASK (1ULL << 7) -#define PT_GLOBAL_MASK (1ULL << 8) -#define PT64_NX_SHIFT 63 -#define PT64_NX_MASK (1ULL << PT64_NX_SHIFT) - -#define PT_PAT_SHIFT 7 -#define PT_DIR_PAT_SHIFT 12 -#define PT_DIR_PAT_MASK (1ULL << PT_DIR_PAT_SHIFT) - -#define PT32_DIR_PSE36_SIZE 4 -#define PT32_DIR_PSE36_SHIFT 13 -#define PT32_DIR_PSE36_MASK \ - (((1ULL << PT32_DIR_PSE36_SIZE) - 1) << PT32_DIR_PSE36_SHIFT) - - #define PT_FIRST_AVAIL_BITS_SHIFT 9 #define PT64_SECOND_AVAIL_BITS_SHIFT 52 @@ -154,10 +124,6 @@ static int dbg = 1; #define PFERR_USER_MASK (1U << 2) #define PFERR_FETCH_MASK (1U << 4) -#define PT64_ROOT_LEVEL 4 -#define PT32_ROOT_LEVEL 2 -#define PT32E_ROOT_LEVEL 3 - #define PT_DIRECTORY_LEVEL 2 #define PT_PAGE_TABLE_LEVEL 1 @@ -186,6 +152,12 @@ static struct kmem_cache *mmu_page_header_cache; static u64 __read_mostly shadow_trap_nonpresent_pte; static u64 __read_mostly shadow_notrap_nonpresent_pte; +static u64 __read_mostly shadow_base_present_pte; +static u64 __read_mostly shadow_nx_mask; +static u64 __read_mostly shadow_x_mask; /* mutual exclusive with nx_mask */ +static u64 __read_mostly shadow_user_mask; +static u64 __read_mostly shadow_accessed_mask; +static u64 __read_mostly shadow_dirty_mask; void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte) { @@ -194,6 +166,23 @@ void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte) } EXPORT_SYMBOL_GPL(kvm_mmu_set_nonpresent_ptes); +void kvm_mmu_set_base_ptes(u64 base_pte) +{ + shadow_base_present_pte = base_pte; +} +EXPORT_SYMBOL_GPL(kvm_mmu_set_base_ptes); + +void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask, + u64 dirty_mask, u64 nx_mask, u64 x_mask) +{ + shadow_user_mask = user_mask; + shadow_accessed_mask = accessed_mask; + shadow_dirty_mask = dirty_mask; + shadow_nx_mask = nx_mask; + shadow_x_mask = x_mask; +} +EXPORT_SYMBOL_GPL(kvm_mmu_set_mask_ptes); + static int is_write_protection(struct kvm_vcpu *vcpu) { return vcpu->arch.cr0 & X86_CR0_WP; @@ -232,7 +221,7 @@ static int is_writeble_pte(unsigned long pte) static int is_dirty_pte(unsigned long pte) { - return pte & PT_DIRTY_MASK; + return pte & shadow_dirty_mask; } static int is_rmap_pte(u64 pte) @@ -387,7 +376,6 @@ static void account_shadowed(struct kvm *kvm, gfn_t gfn) write_count = slot_largepage_idx(gfn, gfn_to_memslot(kvm, gfn)); *write_count += 1; - WARN_ON(*write_count > KVM_PAGES_PER_HPAGE); } static void unaccount_shadowed(struct kvm *kvm, gfn_t gfn) @@ -547,7 +535,7 @@ static void rmap_remove(struct kvm *kvm, u64 *spte) return; sp = page_header(__pa(spte)); pfn = spte_to_pfn(*spte); - if (*spte & PT_ACCESSED_MASK) + if (*spte & shadow_accessed_mask) kvm_set_pfn_accessed(pfn); if (is_writeble_pte(*spte)) kvm_release_pfn_dirty(pfn); @@ -1073,17 +1061,17 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, * whether the guest actually used the pte (in order to detect * demand paging). */ - spte = PT_PRESENT_MASK | PT_DIRTY_MASK; + spte = shadow_base_present_pte | shadow_dirty_mask; if (!speculative) pte_access |= PT_ACCESSED_MASK; if (!dirty) pte_access &= ~ACC_WRITE_MASK; - if (!(pte_access & ACC_EXEC_MASK)) - spte |= PT64_NX_MASK; - - spte |= PT_PRESENT_MASK; + if (pte_access & ACC_EXEC_MASK) + spte |= shadow_x_mask; + else + spte |= shadow_nx_mask; if (pte_access & ACC_USER_MASK) - spte |= PT_USER_MASK; + spte |= shadow_user_mask; if (largepage) spte |= PT_PAGE_SIZE_MASK; @@ -1188,8 +1176,9 @@ static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write, return -ENOMEM; } - table[index] = __pa(new_table->spt) | PT_PRESENT_MASK - | PT_WRITABLE_MASK | PT_USER_MASK; + table[index] = __pa(new_table->spt) + | PT_PRESENT_MASK | PT_WRITABLE_MASK + | shadow_user_mask | shadow_x_mask; } table_addr = table[index] & PT64_BASE_ADDR_MASK; } @@ -1244,7 +1233,6 @@ static void mmu_free_roots(struct kvm_vcpu *vcpu) if (!VALID_PAGE(vcpu->arch.mmu.root_hpa)) return; spin_lock(&vcpu->kvm->mmu_lock); -#ifdef CONFIG_X86_64 if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) { hpa_t root = vcpu->arch.mmu.root_hpa; @@ -1256,7 +1244,6 @@ static void mmu_free_roots(struct kvm_vcpu *vcpu) spin_unlock(&vcpu->kvm->mmu_lock); return; } -#endif for (i = 0; i < 4; ++i) { hpa_t root = vcpu->arch.mmu.pae_root[i]; @@ -1282,7 +1269,6 @@ static void mmu_alloc_roots(struct kvm_vcpu *vcpu) root_gfn = vcpu->arch.cr3 >> PAGE_SHIFT; -#ifdef CONFIG_X86_64 if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) { hpa_t root = vcpu->arch.mmu.root_hpa; @@ -1297,7 +1283,6 @@ static void mmu_alloc_roots(struct kvm_vcpu *vcpu) vcpu->arch.mmu.root_hpa = root; return; } -#endif metaphysical = !is_paging(vcpu); if (tdp_enabled) metaphysical = 1; @@ -1377,7 +1362,7 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, spin_lock(&vcpu->kvm->mmu_lock); kvm_mmu_free_some_pages(vcpu); r = __direct_map(vcpu, gpa, error_code & PFERR_WRITE_MASK, - largepage, gfn, pfn, TDP_ROOT_LEVEL); + largepage, gfn, pfn, kvm_x86_ops->get_tdp_level()); spin_unlock(&vcpu->kvm->mmu_lock); return r; @@ -1484,7 +1469,7 @@ static int init_kvm_tdp_mmu(struct kvm_vcpu *vcpu) context->page_fault = tdp_page_fault; context->free = nonpaging_free; context->prefetch_page = nonpaging_prefetch_page; - context->shadow_root_level = TDP_ROOT_LEVEL; + context->shadow_root_level = kvm_x86_ops->get_tdp_level(); context->root_hpa = INVALID_PAGE; if (!is_paging(vcpu)) { @@ -1633,7 +1618,7 @@ static bool last_updated_pte_accessed(struct kvm_vcpu *vcpu) { u64 *spte = vcpu->arch.last_pte_updated; - return !!(spte && (*spte & PT_ACCESSED_MASK)); + return !!(spte && (*spte & shadow_accessed_mask)); } static void mmu_guess_page_from_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h index e64e9f56a65..1730757bbc7 100644 --- a/arch/x86/kvm/mmu.h +++ b/arch/x86/kvm/mmu.h @@ -3,11 +3,38 @@ #include <linux/kvm_host.h> -#ifdef CONFIG_X86_64 -#define TDP_ROOT_LEVEL PT64_ROOT_LEVEL -#else -#define TDP_ROOT_LEVEL PT32E_ROOT_LEVEL -#endif +#define PT64_PT_BITS 9 +#define PT64_ENT_PER_PAGE (1 << PT64_PT_BITS) +#define PT32_PT_BITS 10 +#define PT32_ENT_PER_PAGE (1 << PT32_PT_BITS) + +#define PT_WRITABLE_SHIFT 1 + +#define PT_PRESENT_MASK (1ULL << 0) +#define PT_WRITABLE_MASK (1ULL << PT_WRITABLE_SHIFT) +#define PT_USER_MASK (1ULL << 2) +#define PT_PWT_MASK (1ULL << 3) +#define PT_PCD_MASK (1ULL << 4) +#define PT_ACCESSED_MASK (1ULL << 5) +#define PT_DIRTY_MASK (1ULL << 6) +#define PT_PAGE_SIZE_MASK (1ULL << 7) +#define PT_PAT_MASK (1ULL << 7) +#define PT_GLOBAL_MASK (1ULL << 8) +#define PT64_NX_SHIFT 63 +#define PT64_NX_MASK (1ULL << PT64_NX_SHIFT) + +#define PT_PAT_SHIFT 7 +#define PT_DIR_PAT_SHIFT 12 +#define PT_DIR_PAT_MASK (1ULL << PT_DIR_PAT_SHIFT) + +#define PT32_DIR_PSE36_SIZE 4 +#define PT32_DIR_PSE36_SHIFT 13 +#define PT32_DIR_PSE36_MASK \ + (((1ULL << PT32_DIR_PSE36_SIZE) - 1) << PT32_DIR_PSE36_SHIFT) + +#define PT64_ROOT_LEVEL 4 +#define PT32_ROOT_LEVEL 2 +#define PT32E_ROOT_LEVEL 3 static inline void kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu) { diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 89e0be2c10d..ab22615eee8 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1863,6 +1863,15 @@ static bool svm_cpu_has_accelerated_tpr(void) return false; } +static int get_npt_level(void) +{ +#ifdef CONFIG_X86_64 + return PT64_ROOT_LEVEL; +#else + return PT32E_ROOT_LEVEL; +#endif +} + static struct kvm_x86_ops svm_x86_ops = { .cpu_has_kvm_support = has_svm, .disabled_by_bios = is_disabled, @@ -1920,6 +1929,7 @@ static struct kvm_x86_ops svm_x86_ops = { .inject_pending_vectors = do_interrupt_requests, .set_tss_addr = svm_set_tss_addr, + .get_tdp_level = get_npt_level, }; static int __init svm_init(void) diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 8e5d6645b90..bfe4db11989 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -42,6 +42,9 @@ module_param(enable_vpid, bool, 0); static int flexpriority_enabled = 1; module_param(flexpriority_enabled, bool, 0); +static int enable_ept = 1; +module_param(enable_ept, bool, 0); + struct vmcs { u32 revision_id; u32 abort; @@ -84,7 +87,7 @@ static inline struct vcpu_vmx *to_vmx(struct kvm_vcpu *vcpu) return container_of(vcpu, struct vcpu_vmx, vcpu); } -static int init_rmode_tss(struct kvm *kvm); +static int init_rmode(struct kvm *kvm); static DEFINE_PER_CPU(struct vmcs *, vmxarea); static DEFINE_PER_CPU(struct vmcs *, current_vmcs); @@ -107,6 +110,11 @@ static struct vmcs_config { u32 vmentry_ctrl; } vmcs_config; +struct vmx_capability { + u32 ept; + u32 vpid; +} vmx_capability; + #define VMX_SEGMENT_FIELD(seg) \ [VCPU_SREG_##seg] = { \ .selector = GUEST_##seg##_SELECTOR, \ @@ -214,6 +222,32 @@ static inline bool cpu_has_vmx_virtualize_apic_accesses(void) SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES); } +static inline int cpu_has_vmx_invept_individual_addr(void) +{ + return (!!(vmx_capability.ept & VMX_EPT_EXTENT_INDIVIDUAL_BIT)); +} + +static inline int cpu_has_vmx_invept_context(void) +{ + return (!!(vmx_capability.ept & VMX_EPT_EXTENT_CONTEXT_BIT)); +} + +static inline int cpu_has_vmx_invept_global(void) +{ + return (!!(vmx_capability.ept & VMX_EPT_EXTENT_GLOBAL_BIT)); +} + +static inline int cpu_has_vmx_ept(void) +{ + return (vmcs_config.cpu_based_2nd_exec_ctrl & + SECONDARY_EXEC_ENABLE_EPT); +} + +static inline int vm_need_ept(void) +{ + return (cpu_has_vmx_ept() && enable_ept); +} + static inline int vm_need_virtualize_apic_accesses(struct kvm *kvm) { return ((cpu_has_vmx_virtualize_apic_accesses()) && @@ -250,6 +284,18 @@ static inline void __invvpid(int ext, u16 vpid, gva_t gva) : : "a"(&operand), "c"(ext) : "cc", "memory"); } +static inline void __invept(int ext, u64 eptp, gpa_t gpa) +{ + struct { + u64 eptp, gpa; + } operand = {eptp, gpa}; + + asm volatile (ASM_VMX_INVEPT + /* CF==1 or ZF==1 --> rc = -1 */ + "; ja 1f ; ud2 ; 1:\n" + : : "a" (&operand), "c" (ext) : "cc", "memory"); +} + static struct kvm_msr_entry *find_msr_entry(struct vcpu_vmx *vmx, u32 msr) { int i; @@ -301,6 +347,33 @@ static inline void vpid_sync_vcpu_all(struct vcpu_vmx *vmx) __invvpid(VMX_VPID_EXTENT_SINGLE_CONTEXT, vmx->vpid, 0); } +static inline void ept_sync_global(void) +{ + if (cpu_has_vmx_invept_global()) + __invept(VMX_EPT_EXTENT_GLOBAL, 0, 0); +} + +static inline void ept_sync_context(u64 eptp) +{ + if (vm_need_ept()) { + if (cpu_has_vmx_invept_context()) + __invept(VMX_EPT_EXTENT_CONTEXT, eptp, 0); + else + ept_sync_global(); + } +} + +static inline void ept_sync_individual_addr(u64 eptp, gpa_t gpa) +{ + if (vm_need_ept()) { + if (cpu_has_vmx_invept_individual_addr()) + __invept(VMX_EPT_EXTENT_INDIVIDUAL_ADDR, + eptp, gpa); + else + ept_sync_context(eptp); + } +} + static unsigned long vmcs_readl(unsigned long field) { unsigned long value; @@ -388,6 +461,8 @@ static void update_exception_bitmap(struct kvm_vcpu *vcpu) eb |= 1u << 1; if (vcpu->arch.rmode.active) eb = ~0; + if (vm_need_ept()) + eb &= ~(1u << PF_VECTOR); /* bypass_guest_pf = 0 */ vmcs_write32(EXCEPTION_BITMAP, eb); } @@ -985,7 +1060,7 @@ static __init int adjust_vmx_controls(u32 ctl_min, u32 ctl_opt, static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf) { u32 vmx_msr_low, vmx_msr_high; - u32 min, opt; + u32 min, opt, min2, opt2; u32 _pin_based_exec_control = 0; u32 _cpu_based_exec_control = 0; u32 _cpu_based_2nd_exec_control = 0; @@ -1003,6 +1078,8 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf) CPU_BASED_CR8_LOAD_EXITING | CPU_BASED_CR8_STORE_EXITING | #endif + CPU_BASED_CR3_LOAD_EXITING | + CPU_BASED_CR3_STORE_EXITING | CPU_BASED_USE_IO_BITMAPS | CPU_BASED_MOV_DR_EXITING | CPU_BASED_USE_TSC_OFFSETING; @@ -1018,11 +1095,13 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf) ~CPU_BASED_CR8_STORE_EXITING; #endif if (_cpu_based_exec_control & CPU_BASED_ACTIVATE_SECONDARY_CONTROLS) { - min = 0; - opt = SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES | + min2 = 0; + opt2 = SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES | SECONDARY_EXEC_WBINVD_EXITING | - SECONDARY_EXEC_ENABLE_VPID; - if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_PROCBASED_CTLS2, + SECONDARY_EXEC_ENABLE_VPID | + SECONDARY_EXEC_ENABLE_EPT; + if (adjust_vmx_controls(min2, opt2, + MSR_IA32_VMX_PROCBASED_CTLS2, &_cpu_based_2nd_exec_control) < 0) return -EIO; } @@ -1031,6 +1110,16 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf) SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES)) _cpu_based_exec_control &= ~CPU_BASED_TPR_SHADOW; #endif + if (_cpu_based_2nd_exec_control & SECONDARY_EXEC_ENABLE_EPT) { + /* CR3 accesses don't need to cause VM Exits when EPT enabled */ + min &= ~(CPU_BASED_CR3_LOAD_EXITING | + CPU_BASED_CR3_STORE_EXITING); + if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_PROCBASED_CTLS, + &_cpu_based_exec_control) < 0) + return -EIO; + rdmsr(MSR_IA32_VMX_EPT_VPID_CAP, + vmx_capability.ept, vmx_capability.vpid); + } min = 0; #ifdef CONFIG_X86_64 @@ -1256,7 +1345,7 @@ static void enter_rmode(struct kvm_vcpu *vcpu) fix_rmode_seg(VCPU_SREG_FS, &vcpu->arch.rmode.fs); kvm_mmu_reset_context(vcpu); - init_rmode_tss(vcpu->kvm); + init_rmode(vcpu->kvm); } #ifdef CONFIG_X86_64 @@ -1304,8 +1393,64 @@ static void vmx_decache_cr4_guest_bits(struct kvm_vcpu *vcpu) vcpu->arch.cr4 |= vmcs_readl(GUEST_CR4) & ~KVM_GUEST_CR4_MASK; } +static void ept_load_pdptrs(struct kvm_vcpu *vcpu) +{ + if (is_paging(vcpu) && is_pae(vcpu) && !is_long_mode(vcpu)) { + if (!load_pdptrs(vcpu, vcpu->arch.cr3)) { + printk(KERN_ERR "EPT: Fail to load pdptrs!\n"); + return; + } + vmcs_write64(GUEST_PDPTR0, vcpu->arch.pdptrs[0]); + vmcs_write64(GUEST_PDPTR1, vcpu->arch.pdptrs[1]); + vmcs_write64(GUEST_PDPTR2, vcpu->arch.pdptrs[2]); + vmcs_write64(GUEST_PDPTR3, vcpu->arch.pdptrs[3]); + } +} + +static void vmx_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4); + +static void ept_update_paging_mode_cr0(unsigned long *hw_cr0, + unsigned long cr0, + struct kvm_vcpu *vcpu) +{ + if (!(cr0 & X86_CR0_PG)) { + /* From paging/starting to nonpaging */ + vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, + vmcs_config.cpu_based_exec_ctrl | + (CPU_BASED_CR3_LOAD_EXITING | + CPU_BASED_CR3_STORE_EXITING)); + vcpu->arch.cr0 = cr0; + vmx_set_cr4(vcpu, vcpu->arch.cr4); + *hw_cr0 |= X86_CR0_PE | X86_CR0_PG; + *hw_cr0 &= ~X86_CR0_WP; + } else if (!is_paging(vcpu)) { + /* From nonpaging to paging */ + vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, + vmcs_config.cpu_based_exec_ctrl & + ~(CPU_BASED_CR3_LOAD_EXITING | + CPU_BASED_CR3_STORE_EXITING)); + vcpu->arch.cr0 = cr0; + vmx_set_cr4(vcpu, vcpu->arch.cr4); + if (!(vcpu->arch.cr0 & X86_CR0_WP)) + *hw_cr0 &= ~X86_CR0_WP; + } +} + +static void ept_update_paging_mode_cr4(unsigned long *hw_cr4, + struct kvm_vcpu *vcpu) +{ + if (!is_paging(vcpu)) { + *hw_cr4 &= ~X86_CR4_PAE; + *hw_cr4 |= X86_CR4_PSE; + } else if (!(vcpu->arch.cr4 & X86_CR4_PAE)) + *hw_cr4 &= ~X86_CR4_PAE; +} + static void vmx_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) { + unsigned long hw_cr0 = (cr0 & ~KVM_GUEST_CR0_MASK) | + KVM_VM_CR0_ALWAYS_ON; + vmx_fpu_deactivate(vcpu); if (vcpu->arch.rmode.active && (cr0 & X86_CR0_PE)) @@ -1323,29 +1468,61 @@ static void vmx_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) } #endif + if (vm_need_ept()) + ept_update_paging_mode_cr0(&hw_cr0, cr0, vcpu); + vmcs_writel(CR0_READ_SHADOW, cr0); - vmcs_writel(GUEST_CR0, - (cr0 & ~KVM_GUEST_CR0_MASK) | KVM_VM_CR0_ALWAYS_ON); + vmcs_writel(GUEST_CR0, hw_cr0); vcpu->arch.cr0 = cr0; if (!(cr0 & X86_CR0_TS) || !(cr0 & X86_CR0_PE)) vmx_fpu_activate(vcpu); } +static u64 construct_eptp(unsigned long root_hpa) +{ + u64 eptp; + + /* TODO write the value reading from MSR */ + eptp = VMX_EPT_DEFAULT_MT | + VMX_EPT_DEFAULT_GAW << VMX_EPT_GAW_EPTP_SHIFT; + eptp |= (root_hpa & PAGE_MASK); + + return eptp; +} + static void vmx_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) { + unsigned long guest_cr3; + u64 eptp; + + guest_cr3 = cr3; + if (vm_need_ept()) { + eptp = construct_eptp(cr3); + vmcs_write64(EPT_POINTER, eptp); + ept_sync_context(eptp); + ept_load_pdptrs(vcpu); + guest_cr3 = is_paging(vcpu) ? vcpu->arch.cr3 : + VMX_EPT_IDENTITY_PAGETABLE_ADDR; + } + vmx_flush_tlb(vcpu); - vmcs_writel(GUEST_CR3, cr3); + vmcs_writel(GUEST_CR3, guest_cr3); if (vcpu->arch.cr0 & X86_CR0_PE) vmx_fpu_deactivate(vcpu); } static void vmx_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) { - vmcs_writel(CR4_READ_SHADOW, cr4); - vmcs_writel(GUEST_CR4, cr4 | (vcpu->arch.rmode.active ? - KVM_RMODE_VM_CR4_ALWAYS_ON : KVM_PMODE_VM_CR4_ALWAYS_ON)); + unsigned long hw_cr4 = cr4 | (vcpu->arch.rmode.active ? + KVM_RMODE_VM_CR4_ALWAYS_ON : KVM_PMODE_VM_CR4_ALWAYS_ON); + vcpu->arch.cr4 = cr4; + if (vm_need_ept()) + ept_update_paging_mode_cr4(&hw_cr4, vcpu); + + vmcs_writel(CR4_READ_SHADOW, cr4); + vmcs_writel(GUEST_CR4, hw_cr4); } static void vmx_set_efer(struct kvm_vcpu *vcpu, u64 efer) @@ -1530,6 +1707,41 @@ out: return ret; } +static int init_rmode_identity_map(struct kvm *kvm) +{ + int i, r, ret; + pfn_t identity_map_pfn; + u32 tmp; + + if (!vm_need_ept()) + return 1; + if (unlikely(!kvm->arch.ept_identity_pagetable)) { + printk(KERN_ERR "EPT: identity-mapping pagetable " + "haven't been allocated!\n"); + return 0; + } + if (likely(kvm->arch.ept_identity_pagetable_done)) + return 1; + ret = 0; + identity_map_pfn = VMX_EPT_IDENTITY_PAGETABLE_ADDR >> PAGE_SHIFT; + r = kvm_clear_guest_page(kvm, identity_map_pfn, 0, PAGE_SIZE); + if (r < 0) + goto out; + /* Set up identity-mapping pagetable for EPT in real mode */ + for (i = 0; i < PT32_ENT_PER_PAGE; i++) { + tmp = (i << 22) + (_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | + _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_PSE); + r = kvm_write_guest_page(kvm, identity_map_pfn, + &tmp, i * sizeof(tmp), sizeof(tmp)); + if (r < 0) + goto out; + } + kvm->arch.ept_identity_pagetable_done = true; + ret = 1; +out: + return ret; +} + static void seg_setup(int seg) { struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg]; @@ -1564,6 +1776,31 @@ out: return r; } +static int alloc_identity_pagetable(struct kvm *kvm) +{ + struct kvm_userspace_memory_region kvm_userspace_mem; + int r = 0; + + down_write(&kvm->slots_lock); + if (kvm->arch.ept_identity_pagetable) + goto out; + kvm_userspace_mem.slot = IDENTITY_PAGETABLE_PRIVATE_MEMSLOT; + kvm_userspace_mem.flags = 0; + kvm_userspace_mem.guest_phys_addr = VMX_EPT_IDENTITY_PAGETABLE_ADDR; + kvm_userspace_mem.memory_size = PAGE_SIZE; + r = __kvm_set_memory_region(kvm, &kvm_userspace_mem, 0); + if (r) + goto out; + + down_read(¤t->mm->mmap_sem); + kvm->arch.ept_identity_pagetable = gfn_to_page(kvm, + VMX_EPT_IDENTITY_PAGETABLE_ADDR >> PAGE_SHIFT); + up_read(¤t->mm->mmap_sem); +out: + up_write(&kvm->slots_lock); + return r; +} + static void allocate_vpid(struct vcpu_vmx *vmx) { int vpid; @@ -1638,6 +1875,9 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx) CPU_BASED_CR8_LOAD_EXITING; #endif } + if (!vm_need_ept()) + exec_control |= CPU_BASED_CR3_STORE_EXITING | + CPU_BASED_CR3_LOAD_EXITING; vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, exec_control); if (cpu_has_secondary_exec_ctrls()) { @@ -1647,6 +1887,8 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx) ~SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES; if (vmx->vpid == 0) exec_control &= ~SECONDARY_EXEC_ENABLE_VPID; + if (!vm_need_ept()) + exec_control &= ~SECONDARY_EXEC_ENABLE_EPT; vmcs_write32(SECONDARY_VM_EXEC_CONTROL, exec_control); } @@ -1722,6 +1964,15 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx) return 0; } +static int init_rmode(struct kvm *kvm) +{ + if (!init_rmode_tss(kvm)) + return 0; + if (!init_rmode_identity_map(kvm)) + return 0; + return 1; +} + static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) { struct vcpu_vmx *vmx = to_vmx(vcpu); @@ -1729,7 +1980,7 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) int ret; down_read(&vcpu->kvm->slots_lock); - if (!init_rmode_tss(vmx->vcpu.kvm)) { + if (!init_rmode(vmx->vcpu.kvm)) { ret = -ENOMEM; goto out; } @@ -1994,6 +2245,9 @@ static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) if (intr_info & INTR_INFO_DELIVER_CODE_MASK) error_code = vmcs_read32(VM_EXIT_INTR_ERROR_CODE); if (is_page_fault(intr_info)) { + /* EPT won't cause page fault directly */ + if (vm_need_ept()) + BUG(); cr2 = vmcs_readl(EXIT_QUALIFICATION); KVMTRACE_3D(PAGE_FAULT, vcpu, error_code, (u32)cr2, (u32)((u64)cr2 >> 32), handler); @@ -2323,6 +2577,64 @@ static int handle_task_switch(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) return kvm_task_switch(vcpu, tss_selector, reason); } +static int handle_ept_violation(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u64 exit_qualification; + enum emulation_result er; + gpa_t gpa; + unsigned long hva; + int gla_validity; + int r; + + exit_qualification = vmcs_read64(EXIT_QUALIFICATION); + + if (exit_qualification & (1 << 6)) { + printk(KERN_ERR "EPT: GPA exceeds GAW!\n"); + return -ENOTSUPP; + } + + gla_validity = (exit_qualification >> 7) & 0x3; + if (gla_validity != 0x3 && gla_validity != 0x1 && gla_validity != 0) { + printk(KERN_ERR "EPT: Handling EPT violation failed!\n"); + printk(KERN_ERR "EPT: GPA: 0x%lx, GVA: 0x%lx\n", + (long unsigned int)vmcs_read64(GUEST_PHYSICAL_ADDRESS), + (long unsigned int)vmcs_read64(GUEST_LINEAR_ADDRESS)); + printk(KERN_ERR "EPT: Exit qualification is 0x%lx\n", + (long unsigned int)exit_qualification); + kvm_run->exit_reason = KVM_EXIT_UNKNOWN; + kvm_run->hw.hardware_exit_reason = 0; + return -ENOTSUPP; + } + + gpa = vmcs_read64(GUEST_PHYSICAL_ADDRESS); + hva = gfn_to_hva(vcpu->kvm, gpa >> PAGE_SHIFT); + if (!kvm_is_error_hva(hva)) { + r = kvm_mmu_page_fault(vcpu, gpa & PAGE_MASK, 0); + if (r < 0) { + printk(KERN_ERR "EPT: Not enough memory!\n"); + return -ENOMEM; + } + return 1; + } else { + /* must be MMIO */ + er = emulate_instruction(vcpu, kvm_run, 0, 0, 0); + + if (er == EMULATE_FAIL) { + printk(KERN_ERR + "EPT: Fail to handle EPT violation vmexit!er is %d\n", + er); + printk(KERN_ERR "EPT: GPA: 0x%lx, GVA: 0x%lx\n", + (long unsigned int)vmcs_read64(GUEST_PHYSICAL_ADDRESS), + (long unsigned int)vmcs_read64(GUEST_LINEAR_ADDRESS)); + printk(KERN_ERR "EPT: Exit qualification is 0x%lx\n", + (long unsigned int)exit_qualification); + return -ENOTSUPP; + } else if (er == EMULATE_DO_MMIO) + return 0; + } + return 1; +} + /* * The exit handlers return 1 if the exit was handled fully and guest execution * may resume. Otherwise they set the kvm_run parameter to indicate what needs @@ -2346,6 +2658,7 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu, [EXIT_REASON_APIC_ACCESS] = handle_apic_access, [EXIT_REASON_WBINVD] = handle_wbinvd, [EXIT_REASON_TASK_SWITCH] = handle_task_switch, + [EXIT_REASON_EPT_VIOLATION] = handle_ept_violation, }; static const int kvm_vmx_max_exit_handlers = @@ -2364,6 +2677,13 @@ static int kvm_handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) KVMTRACE_3D(VMEXIT, vcpu, exit_reason, (u32)vmcs_readl(GUEST_RIP), (u32)((u64)vmcs_readl(GUEST_RIP) >> 32), entryexit); + /* Access CR3 don't cause VMExit in paging mode, so we need + * to sync with guest real CR3. */ + if (vm_need_ept() && is_paging(vcpu)) { + vcpu->arch.cr3 = vmcs_readl(GUEST_CR3); + ept_load_pdptrs(vcpu); + } + if (unlikely(vmx->fail)) { kvm_run->exit_reason = KVM_EXIT_FAIL_ENTRY; kvm_run->fail_entry.hardware_entry_failure_reason @@ -2372,7 +2692,8 @@ static int kvm_handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) } if ((vectoring_info & VECTORING_INFO_VALID_MASK) && - exit_reason != EXIT_REASON_EXCEPTION_NMI) + (exit_reason != EXIT_REASON_EXCEPTION_NMI && + exit_reason != EXIT_REASON_EPT_VIOLATION)) printk(KERN_WARNING "%s: unexpected, valid vectoring info and " "exit reason is 0x%x\n", __func__, exit_reason); if (exit_reason < kvm_vmx_max_exit_handlers @@ -2674,6 +2995,15 @@ static struct kvm_vcpu *vmx_create_vcpu(struct kvm *kvm, unsigned int id) return ERR_PTR(-ENOMEM); allocate_vpid(vmx); + if (id == 0 && vm_need_ept()) { + kvm_mmu_set_base_ptes(VMX_EPT_READABLE_MASK | + VMX_EPT_WRITABLE_MASK | + VMX_EPT_DEFAULT_MT << VMX_EPT_MT_EPTE_SHIFT); + kvm_mmu_set_mask_ptes(0ull, VMX_EPT_FAKE_ACCESSED_MASK, + VMX_EPT_FAKE_DIRTY_MASK, 0ull, + VMX_EPT_EXECUTABLE_MASK); + kvm_enable_tdp(); + } err = kvm_vcpu_init(&vmx->vcpu, kvm, id); if (err) @@ -2706,6 +3036,10 @@ static struct kvm_vcpu *vmx_create_vcpu(struct kvm *kvm, unsigned int id) if (alloc_apic_access_page(kvm) != 0) goto free_vmcs; + if (vm_need_ept()) + if (alloc_identity_pagetable(kvm) != 0) + goto free_vmcs; + return &vmx->vcpu; free_vmcs: @@ -2735,6 +3069,11 @@ static void __init vmx_check_processor_compat(void *rtn) } } +static int get_ept_level(void) +{ + return VMX_EPT_DEFAULT_GAW + 1; +} + static struct kvm_x86_ops vmx_x86_ops = { .cpu_has_kvm_support = cpu_has_kvm_support, .disabled_by_bios = vmx_disabled_by_bios, @@ -2791,6 +3130,7 @@ static struct kvm_x86_ops vmx_x86_ops = { .inject_pending_vectors = do_interrupt_requests, .set_tss_addr = vmx_set_tss_addr, + .get_tdp_level = get_ept_level, }; static int __init vmx_init(void) @@ -2843,9 +3183,14 @@ static int __init vmx_init(void) vmx_disable_intercept_for_msr(vmx_msr_bitmap, MSR_IA32_SYSENTER_ESP); vmx_disable_intercept_for_msr(vmx_msr_bitmap, MSR_IA32_SYSENTER_EIP); + if (cpu_has_vmx_ept()) + bypass_guest_pf = 0; + if (bypass_guest_pf) kvm_mmu_set_nonpresent_ptes(~0xffeull, 0ull); + ept_sync_global(); + return 0; out2: diff --git a/arch/x86/kvm/vmx.h b/arch/x86/kvm/vmx.h index 5dff4606b98..79d94c610df 100644 --- a/arch/x86/kvm/vmx.h +++ b/arch/x86/kvm/vmx.h @@ -35,6 +35,8 @@ #define CPU_BASED_MWAIT_EXITING 0x00000400 #define CPU_BASED_RDPMC_EXITING 0x00000800 #define CPU_BASED_RDTSC_EXITING 0x00001000 +#define CPU_BASED_CR3_LOAD_EXITING 0x00008000 +#define CPU_BASED_CR3_STORE_EXITING 0x00010000 #define CPU_BASED_CR8_LOAD_EXITING 0x00080000 #define CPU_BASED_CR8_STORE_EXITING 0x00100000 #define CPU_BASED_TPR_SHADOW 0x00200000 @@ -49,6 +51,7 @@ * Definitions of Secondary Processor-Based VM-Execution Controls. */ #define SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES 0x00000001 +#define SECONDARY_EXEC_ENABLE_EPT 0x00000002 #define SECONDARY_EXEC_ENABLE_VPID 0x00000020 #define SECONDARY_EXEC_WBINVD_EXITING 0x00000040 @@ -100,10 +103,22 @@ enum vmcs_field { VIRTUAL_APIC_PAGE_ADDR_HIGH = 0x00002013, APIC_ACCESS_ADDR = 0x00002014, APIC_ACCESS_ADDR_HIGH = 0x00002015, + EPT_POINTER = 0x0000201a, + EPT_POINTER_HIGH = 0x0000201b, + GUEST_PHYSICAL_ADDRESS = 0x00002400, + GUEST_PHYSICAL_ADDRESS_HIGH = 0x00002401, VMCS_LINK_POINTER = 0x00002800, VMCS_LINK_POINTER_HIGH = 0x00002801, GUEST_IA32_DEBUGCTL = 0x00002802, GUEST_IA32_DEBUGCTL_HIGH = 0x00002803, + GUEST_PDPTR0 = 0x0000280a, + GUEST_PDPTR0_HIGH = 0x0000280b, + GUEST_PDPTR1 = 0x0000280c, + GUEST_PDPTR1_HIGH = 0x0000280d, + GUEST_PDPTR2 = 0x0000280e, + GUEST_PDPTR2_HIGH = 0x0000280f, + GUEST_PDPTR3 = 0x00002810, + GUEST_PDPTR3_HIGH = 0x00002811, PIN_BASED_VM_EXEC_CONTROL = 0x00004000, CPU_BASED_VM_EXEC_CONTROL = 0x00004002, EXCEPTION_BITMAP = 0x00004004, @@ -226,6 +241,8 @@ enum vmcs_field { #define EXIT_REASON_MWAIT_INSTRUCTION 36 #define EXIT_REASON_TPR_BELOW_THRESHOLD 43 #define EXIT_REASON_APIC_ACCESS 44 +#define EXIT_REASON_EPT_VIOLATION 48 +#define EXIT_REASON_EPT_MISCONFIG 49 #define EXIT_REASON_WBINVD 54 /* @@ -316,15 +333,36 @@ enum vmcs_field { #define MSR_IA32_VMX_CR4_FIXED1 0x489 #define MSR_IA32_VMX_VMCS_ENUM 0x48a #define MSR_IA32_VMX_PROCBASED_CTLS2 0x48b +#define MSR_IA32_VMX_EPT_VPID_CAP 0x48c #define MSR_IA32_FEATURE_CONTROL 0x3a #define MSR_IA32_FEATURE_CONTROL_LOCKED 0x1 #define MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED 0x4 #define APIC_ACCESS_PAGE_PRIVATE_MEMSLOT 9 +#define IDENTITY_PAGETABLE_PRIVATE_MEMSLOT 10 #define VMX_NR_VPIDS (1 << 16) #define VMX_VPID_EXTENT_SINGLE_CONTEXT 1 #define VMX_VPID_EXTENT_ALL_CONTEXT 2 +#define VMX_EPT_EXTENT_INDIVIDUAL_ADDR 0 +#define VMX_EPT_EXTENT_CONTEXT 1 +#define VMX_EPT_EXTENT_GLOBAL 2 +#define VMX_EPT_EXTENT_INDIVIDUAL_BIT (1ull << 24) +#define VMX_EPT_EXTENT_CONTEXT_BIT (1ull << 25) +#define VMX_EPT_EXTENT_GLOBAL_BIT (1ull << 26) +#define VMX_EPT_DEFAULT_GAW 3 +#define VMX_EPT_MAX_GAW 0x4 +#define VMX_EPT_MT_EPTE_SHIFT 3 +#define VMX_EPT_GAW_EPTP_SHIFT 3 +#define VMX_EPT_DEFAULT_MT 0x6ull +#define VMX_EPT_READABLE_MASK 0x1ull +#define VMX_EPT_WRITABLE_MASK 0x2ull +#define VMX_EPT_EXECUTABLE_MASK 0x4ull +#define VMX_EPT_FAKE_ACCESSED_MASK (1ull << 62) +#define VMX_EPT_FAKE_DIRTY_MASK (1ull << 63) + +#define VMX_EPT_IDENTITY_PAGETABLE_ADDR 0xfffbc000ul + #endif diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 0ce556372a4..21338bdb28f 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2417,6 +2417,9 @@ int kvm_arch_init(void *opaque) kvm_x86_ops = ops; kvm_mmu_set_nonpresent_ptes(0ull, 0ull); + kvm_mmu_set_base_ptes(PT_PRESENT_MASK); + kvm_mmu_set_mask_ptes(PT_USER_MASK, PT_ACCESSED_MASK, + PT_DIRTY_MASK, PT64_NX_MASK, 0); return 0; out: @@ -3019,6 +3022,8 @@ int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) kvm_x86_ops->decache_regs(vcpu); + vcpu->arch.exception.pending = false; + vcpu_put(vcpu); return 0; @@ -3481,7 +3486,7 @@ int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason) } if (reason == TASK_SWITCH_IRET || reason == TASK_SWITCH_JMP) { - cseg_desc.type &= ~(1 << 8); //clear the B flag + cseg_desc.type &= ~(1 << 1); //clear the B flag save_guest_segment_descriptor(vcpu, tr_seg.selector, &cseg_desc); } @@ -3507,7 +3512,7 @@ int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason) } if (reason != TASK_SWITCH_IRET) { - nseg_desc.type |= (1 << 8); + nseg_desc.type |= (1 << 1); save_guest_segment_descriptor(vcpu, tss_selector, &nseg_desc); } @@ -3698,10 +3703,19 @@ void fx_init(struct kvm_vcpu *vcpu) { unsigned after_mxcsr_mask; + /* + * Touch the fpu the first time in non atomic context as if + * this is the first fpu instruction the exception handler + * will fire before the instruction returns and it'll have to + * allocate ram with GFP_KERNEL. + */ + if (!used_math()) + fx_save(&vcpu->arch.host_fx_image); + /* Initialize guest FPU by resetting ours and saving into guest's */ preempt_disable(); fx_save(&vcpu->arch.host_fx_image); - fpu_init(); + fx_finit(); fx_save(&vcpu->arch.guest_fx_image); fx_restore(&vcpu->arch.host_fx_image); preempt_enable(); @@ -3906,6 +3920,8 @@ void kvm_arch_destroy_vm(struct kvm *kvm) kvm_free_physmem(kvm); if (kvm->arch.apic_access_page) put_page(kvm->arch.apic_access_page); + if (kvm->arch.ept_identity_pagetable) + put_page(kvm->arch.ept_identity_pagetable); kfree(kvm); } diff --git a/arch/x86/kvm/x86_emulate.c b/arch/x86/kvm/x86_emulate.c index 2ca08386f99..f2a696d6a24 100644 --- a/arch/x86/kvm/x86_emulate.c +++ b/arch/x86/kvm/x86_emulate.c @@ -1761,6 +1761,7 @@ twobyte_insn: case 6: /* lmsw */ realmode_lmsw(ctxt->vcpu, (u16)c->src.val, &ctxt->eflags); + c->dst.type = OP_NONE; break; case 7: /* invlpg*/ emulate_invlpg(ctxt->vcpu, memop); diff --git a/arch/x86/mm/discontig_32.c b/arch/x86/mm/discontig_32.c index 18378850e25..914ccf98368 100644 --- a/arch/x86/mm/discontig_32.c +++ b/arch/x86/mm/discontig_32.c @@ -476,29 +476,3 @@ int memory_add_physaddr_to_nid(u64 addr) EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid); #endif - -#ifndef CONFIG_HAVE_ARCH_PARSE_SRAT -/* - * XXX FIXME: Make SLIT table parsing available to 32-bit NUMA - * - * These stub functions are needed to compile 32-bit NUMA when SRAT is - * not set. There are functions in srat_64.c for parsing this table - * and it may be possible to make them common functions. - */ -void acpi_numa_slit_init (struct acpi_table_slit *slit) -{ - printk(KERN_INFO "ACPI: No support for parsing SLIT table\n"); -} - -void acpi_numa_processor_affinity_init (struct acpi_srat_cpu_affinity *pa) -{ -} - -void acpi_numa_memory_affinity_init (struct acpi_srat_mem_affinity *ma) -{ -} - -void acpi_numa_arch_fixup(void) -{ -} -#endif /* CONFIG_HAVE_ARCH_PARSE_SRAT */ diff --git a/arch/x86/mm/pgtable_32.c b/arch/x86/mm/pgtable_32.c index 9ee007be914..369cf065b6a 100644 --- a/arch/x86/mm/pgtable_32.c +++ b/arch/x86/mm/pgtable_32.c @@ -172,10 +172,3 @@ void reserve_top_address(unsigned long reserve) __FIXADDR_TOP = -reserve - PAGE_SIZE; __VMALLOC_RESERVE += reserve; } - -int pmd_bad(pmd_t pmd) -{ - WARN_ON_ONCE(pmd_bad_v1(pmd) != pmd_bad_v2(pmd)); - - return pmd_bad_v1(pmd); -} diff --git a/arch/x86/pci/Makefile_32 b/arch/x86/pci/Makefile_32 index 7fa519868d7..89ec35d00ef 100644 --- a/arch/x86/pci/Makefile_32 +++ b/arch/x86/pci/Makefile_32 @@ -6,11 +6,19 @@ obj-$(CONFIG_PCI_DIRECT) += direct.o obj-$(CONFIG_PCI_OLPC) += olpc.o pci-y := fixup.o + +# Do not change the ordering here. There is a nasty init function +# ordering dependency which breaks when you move acpi.o below +# legacy/irq.o pci-$(CONFIG_ACPI) += acpi.o pci-y += legacy.o irq.o -pci-$(CONFIG_X86_VISWS) += visws.o fixup.o -pci-$(CONFIG_X86_NUMAQ) += numa.o irq.o +# Careful: VISWS and NUMAQ overrule the pci-y above. The colons are +# therefor correct. This needs a proper fix by distangling the code. +pci-$(CONFIG_X86_VISWS) := visws.o fixup.o +pci-$(CONFIG_X86_NUMAQ) := numa.o irq.o + +# Necessary for NUMAQ as well pci-$(CONFIG_NUMA) += mp_bus_to_node.o obj-y += $(pci-y) common.o early.o diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c index 1a9c0c6a1a1..d95de2f199c 100644 --- a/arch/x86/pci/acpi.c +++ b/arch/x86/pci/acpi.c @@ -6,45 +6,6 @@ #include <asm/numa.h> #include "pci.h" -static int __devinit can_skip_ioresource_align(const struct dmi_system_id *d) -{ - pci_probe |= PCI_CAN_SKIP_ISA_ALIGN; - printk(KERN_INFO "PCI: %s detected, can skip ISA alignment\n", d->ident); - return 0; -} - -static struct dmi_system_id acpi_pciprobe_dmi_table[] __devinitdata = { -/* - * Systems where PCI IO resource ISA alignment can be skipped - * when the ISA enable bit in the bridge control is not set - */ - { - .callback = can_skip_ioresource_align, - .ident = "IBM System x3800", - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "IBM"), - DMI_MATCH(DMI_PRODUCT_NAME, "x3800"), - }, - }, - { - .callback = can_skip_ioresource_align, - .ident = "IBM System x3850", - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "IBM"), - DMI_MATCH(DMI_PRODUCT_NAME, "x3850"), - }, - }, - { - .callback = can_skip_ioresource_align, - .ident = "IBM System x3950", - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "IBM"), - DMI_MATCH(DMI_PRODUCT_NAME, "x3950"), - }, - }, - {} -}; - struct pci_root_info { char *name; unsigned int res_num; @@ -196,8 +157,6 @@ struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_device *device, int do int pxm; #endif - dmi_check_system(acpi_pciprobe_dmi_table); - if (domain && !pci_domains_supported) { printk(KERN_WARNING "PCI: Multiple domains not supported " "(dom %d, bus %d)\n", domain, busnum); diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c index 2a4d751818b..8545c8a9d10 100644 --- a/arch/x86/pci/common.c +++ b/arch/x86/pci/common.c @@ -77,17 +77,48 @@ int pcibios_scanned; */ DEFINE_SPINLOCK(pci_config_lock); -static void __devinit pcibios_fixup_device_resources(struct pci_dev *dev) +static int __devinit can_skip_ioresource_align(const struct dmi_system_id *d) { - struct resource *rom_r = &dev->resource[PCI_ROM_RESOURCE]; - - if (rom_r->parent) - return; - if (rom_r->start) - /* we deal with BIOS assigned ROM later */ - return; - if (!(pci_probe & PCI_ASSIGN_ROMS)) - rom_r->start = rom_r->end = rom_r->flags = 0; + pci_probe |= PCI_CAN_SKIP_ISA_ALIGN; + printk(KERN_INFO "PCI: %s detected, can skip ISA alignment\n", d->ident); + return 0; +} + +static struct dmi_system_id can_skip_pciprobe_dmi_table[] __devinitdata = { +/* + * Systems where PCI IO resource ISA alignment can be skipped + * when the ISA enable bit in the bridge control is not set + */ + { + .callback = can_skip_ioresource_align, + .ident = "IBM System x3800", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "IBM"), + DMI_MATCH(DMI_PRODUCT_NAME, "x3800"), + }, + }, + { + .callback = can_skip_ioresource_align, + .ident = "IBM System x3850", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "IBM"), + DMI_MATCH(DMI_PRODUCT_NAME, "x3850"), + }, + }, + { + .callback = can_skip_ioresource_align, + .ident = "IBM System x3950", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "IBM"), + DMI_MATCH(DMI_PRODUCT_NAME, "x3950"), + }, + }, + {} +}; + +void __init dmi_check_skip_isa_align(void) +{ + dmi_check_system(can_skip_pciprobe_dmi_table); } /* @@ -97,11 +128,7 @@ static void __devinit pcibios_fixup_device_resources(struct pci_dev *dev) void __devinit pcibios_fixup_bus(struct pci_bus *b) { - struct pci_dev *dev; - pci_read_bridge_bases(b); - list_for_each_entry(dev, &b->devices, bus_list) - pcibios_fixup_device_resources(dev); } /* @@ -318,13 +345,16 @@ static struct dmi_system_id __devinitdata pciprobe_dmi_table[] = { {} }; +void __init dmi_check_pciprobe(void) +{ + dmi_check_system(pciprobe_dmi_table); +} + struct pci_bus * __devinit pcibios_scan_root(int busnum) { struct pci_bus *bus = NULL; struct pci_sysdata *sd; - dmi_check_system(pciprobe_dmi_table); - while ((bus = pci_find_next_bus(bus)) != NULL) { if (bus->number == busnum) { /* Already scanned */ @@ -462,6 +492,9 @@ char * __devinit pcibios_setup(char *str) } else if (!strcmp(str, "routeirq")) { pci_routeirq = 1; return NULL; + } else if (!strcmp(str, "skip_isa_align")) { + pci_probe |= PCI_CAN_SKIP_ISA_ALIGN; + return NULL; } return str; } @@ -489,7 +522,7 @@ void pcibios_disable_device (struct pci_dev *dev) pcibios_disable_irq(dev); } -struct pci_bus *pci_scan_bus_on_node(int busno, struct pci_ops *ops, int node) +struct pci_bus * __devinit pci_scan_bus_on_node(int busno, struct pci_ops *ops, int node) { struct pci_bus *bus = NULL; struct pci_sysdata *sd; @@ -512,7 +545,7 @@ struct pci_bus *pci_scan_bus_on_node(int busno, struct pci_ops *ops, int node) return bus; } -struct pci_bus *pci_scan_bus_with_sysdata(int busno) +struct pci_bus * __devinit pci_scan_bus_with_sysdata(int busno) { return pci_scan_bus_on_node(busno, &pci_root_ops, -1); } diff --git a/arch/x86/pci/fixup.c b/arch/x86/pci/fixup.c index b60b2abd480..ff3a6a33634 100644 --- a/arch/x86/pci/fixup.c +++ b/arch/x86/pci/fixup.c @@ -502,7 +502,7 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_SIEMENS, 0x0015, */ static void fam10h_pci_cfg_space_size(struct pci_dev *dev) { - dev->cfg_size = pci_cfg_space_size_ext(dev, 0); + dev->cfg_size = pci_cfg_space_size_ext(dev); } DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1200, fam10h_pci_cfg_space_size); diff --git a/arch/x86/pci/init.c b/arch/x86/pci/init.c index dd30c6076b5..e70b9c57b88 100644 --- a/arch/x86/pci/init.c +++ b/arch/x86/pci/init.c @@ -33,6 +33,10 @@ static __init int pci_access_init(void) printk(KERN_ERR "PCI: Fatal: No config space access function found\n"); + dmi_check_pciprobe(); + + dmi_check_skip_isa_align(); + return 0; } arch_initcall(pci_access_init); diff --git a/arch/x86/pci/pci.h b/arch/x86/pci/pci.h index c58805a92db..f3972b12c60 100644 --- a/arch/x86/pci/pci.h +++ b/arch/x86/pci/pci.h @@ -38,6 +38,9 @@ enum pci_bf_sort_state { pci_dmi_bf, }; +extern void __init dmi_check_pciprobe(void); +extern void __init dmi_check_skip_isa_align(void); + /* pci-i386.c */ extern unsigned int pcibios_max_latency; diff --git a/arch/x86/vdso/vdso32-setup.c b/arch/x86/vdso/vdso32-setup.c index 4dceeb1fc5e..cf058fecfce 100644 --- a/arch/x86/vdso/vdso32-setup.c +++ b/arch/x86/vdso/vdso32-setup.c @@ -162,7 +162,7 @@ static __init void relocate_vdso(Elf32_Ehdr *ehdr) Elf32_Shdr *shdr; int i; - BUG_ON(memcmp(ehdr->e_ident, ELFMAG, 4) != 0 || + BUG_ON(memcmp(ehdr->e_ident, ELFMAG, SELFMAG) != 0 || !elf_check_arch_ia32(ehdr) || ehdr->e_type != ET_DYN); diff --git a/arch/x86/video/fbdev.c b/arch/x86/video/fbdev.c index 4db42bff8c6..69527688f79 100644 --- a/arch/x86/video/fbdev.c +++ b/arch/x86/video/fbdev.c @@ -1,5 +1,4 @@ /* - * * Copyright (C) 2007 Antonino Daplas <adaplas@gmail.com> * * This file is subject to the terms and conditions of the GNU General Public @@ -29,3 +28,4 @@ int fb_is_primary_device(struct fb_info *info) return retval; } EXPORT_SYMBOL(fb_is_primary_device); +MODULE_LICENSE("GPL"); diff --git a/block/blk-core.c b/block/blk-core.c index b754a4a2f9b..2987fe47b5e 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -54,15 +54,16 @@ static DEFINE_PER_CPU(struct list_head, blk_cpu_done); static void drive_stat_acct(struct request *rq, int new_io) { + struct hd_struct *part; int rw = rq_data_dir(rq); if (!blk_fs_request(rq) || !rq->rq_disk) return; - if (!new_io) { - __all_stat_inc(rq->rq_disk, merges[rw], rq->sector); - } else { - struct hd_struct *part = get_part(rq->rq_disk, rq->sector); + part = get_part(rq->rq_disk, rq->sector); + if (!new_io) + __all_stat_inc(rq->rq_disk, part, merges[rw], rq->sector); + else { disk_round_stats(rq->rq_disk); rq->rq_disk->in_flight++; if (part) { @@ -253,9 +254,11 @@ EXPORT_SYMBOL(__generic_unplug_device); **/ void generic_unplug_device(struct request_queue *q) { - spin_lock_irq(q->queue_lock); - __generic_unplug_device(q); - spin_unlock_irq(q->queue_lock); + if (blk_queue_plugged(q)) { + spin_lock_irq(q->queue_lock); + __generic_unplug_device(q); + spin_unlock_irq(q->queue_lock); + } } EXPORT_SYMBOL(generic_unplug_device); @@ -1536,10 +1539,11 @@ static int __end_that_request_first(struct request *req, int error, } if (blk_fs_request(req) && req->rq_disk) { + struct hd_struct *part = get_part(req->rq_disk, req->sector); const int rw = rq_data_dir(req); - all_stat_add(req->rq_disk, sectors[rw], - nr_bytes >> 9, req->sector); + all_stat_add(req->rq_disk, part, sectors[rw], + nr_bytes >> 9, req->sector); } total_bytes = bio_nbytes = 0; @@ -1725,8 +1729,8 @@ static void end_that_request_last(struct request *req, int error) const int rw = rq_data_dir(req); struct hd_struct *part = get_part(disk, req->sector); - __all_stat_inc(disk, ios[rw], req->sector); - __all_stat_add(disk, ticks[rw], duration, req->sector); + __all_stat_inc(disk, part, ios[rw], req->sector); + __all_stat_add(disk, part, ticks[rw], duration, req->sector); disk_round_stats(disk); disk->in_flight--; if (part) { diff --git a/block/blk-ioc.c b/block/blk-ioc.c index e34df7c9fc3..012f065ac8e 100644 --- a/block/blk-ioc.c +++ b/block/blk-ioc.c @@ -41,8 +41,8 @@ int put_io_context(struct io_context *ioc) rcu_read_lock(); if (ioc->aic && ioc->aic->dtor) ioc->aic->dtor(ioc->aic); - rcu_read_unlock(); cfq_dtor(ioc); + rcu_read_unlock(); kmem_cache_free(iocontext_cachep, ioc); return 1; diff --git a/block/blk-merge.c b/block/blk-merge.c index 73b23562af2..651136aae76 100644 --- a/block/blk-merge.c +++ b/block/blk-merge.c @@ -149,9 +149,9 @@ static int blk_phys_contig_segment(struct request_queue *q, struct bio *bio, static int blk_hw_contig_segment(struct request_queue *q, struct bio *bio, struct bio *nxt) { - if (unlikely(!bio_flagged(bio, BIO_SEG_VALID))) + if (!bio_flagged(bio, BIO_SEG_VALID)) blk_recount_segments(q, bio); - if (unlikely(!bio_flagged(nxt, BIO_SEG_VALID))) + if (!bio_flagged(nxt, BIO_SEG_VALID)) blk_recount_segments(q, nxt); if (!BIOVEC_VIRT_MERGEABLE(__BVEC_END(bio), __BVEC_START(nxt)) || BIOVEC_VIRT_OVERSIZE(bio->bi_hw_back_size + nxt->bi_hw_front_size)) @@ -312,9 +312,9 @@ int ll_back_merge_fn(struct request_queue *q, struct request *req, q->last_merge = NULL; return 0; } - if (unlikely(!bio_flagged(req->biotail, BIO_SEG_VALID))) + if (!bio_flagged(req->biotail, BIO_SEG_VALID)) blk_recount_segments(q, req->biotail); - if (unlikely(!bio_flagged(bio, BIO_SEG_VALID))) + if (!bio_flagged(bio, BIO_SEG_VALID)) blk_recount_segments(q, bio); len = req->biotail->bi_hw_back_size + bio->bi_hw_front_size; if (BIOVEC_VIRT_MERGEABLE(__BVEC_END(req->biotail), __BVEC_START(bio)) @@ -352,9 +352,9 @@ int ll_front_merge_fn(struct request_queue *q, struct request *req, return 0; } len = bio->bi_hw_back_size + req->bio->bi_hw_front_size; - if (unlikely(!bio_flagged(bio, BIO_SEG_VALID))) + if (!bio_flagged(bio, BIO_SEG_VALID)) blk_recount_segments(q, bio); - if (unlikely(!bio_flagged(req->bio, BIO_SEG_VALID))) + if (!bio_flagged(req->bio, BIO_SEG_VALID)) blk_recount_segments(q, req->bio); if (BIOVEC_VIRT_MERGEABLE(__BVEC_END(bio), __BVEC_START(req->bio)) && !BIOVEC_VIRT_OVERSIZE(len)) { diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index e85c4013e8a..304ec73ab82 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -146,11 +146,13 @@ static ssize_t queue_nomerges_store(struct request_queue *q, const char *page, unsigned long nm; ssize_t ret = queue_var_store(&nm, page, count); + spin_lock_irq(q->queue_lock); if (nm) - set_bit(QUEUE_FLAG_NOMERGES, &q->queue_flags); + queue_flag_set(QUEUE_FLAG_NOMERGES, q); else - clear_bit(QUEUE_FLAG_NOMERGES, &q->queue_flags); + queue_flag_clear(QUEUE_FLAG_NOMERGES, q); + spin_unlock_irq(q->queue_lock); return ret; } diff --git a/block/blk-tag.c b/block/blk-tag.c index de64e042997..32667beb03e 100644 --- a/block/blk-tag.c +++ b/block/blk-tag.c @@ -70,7 +70,7 @@ void __blk_queue_free_tags(struct request_queue *q) __blk_free_tags(bqt); q->queue_tags = NULL; - queue_flag_clear(QUEUE_FLAG_QUEUED, q); + queue_flag_clear_unlocked(QUEUE_FLAG_QUEUED, q); } /** @@ -98,7 +98,7 @@ EXPORT_SYMBOL(blk_free_tags); **/ void blk_queue_free_tags(struct request_queue *q) { - queue_flag_clear(QUEUE_FLAG_QUEUED, q); + queue_flag_clear_unlocked(QUEUE_FLAG_QUEUED, q); } EXPORT_SYMBOL(blk_queue_free_tags); @@ -171,6 +171,9 @@ EXPORT_SYMBOL(blk_init_tags); * @q: the request queue for the device * @depth: the maximum queue depth supported * @tags: the tag to use + * + * Queue lock must be held here if the function is called to resize an + * existing map. **/ int blk_queue_init_tags(struct request_queue *q, int depth, struct blk_queue_tag *tags) @@ -197,7 +200,7 @@ int blk_queue_init_tags(struct request_queue *q, int depth, * assign it, all done */ q->queue_tags = tags; - queue_flag_set(QUEUE_FLAG_QUEUED, q); + queue_flag_set_unlocked(QUEUE_FLAG_QUEUED, q); INIT_LIST_HEAD(&q->tag_busy_list); return 0; fail: diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index f4e1006c253..b399c62936e 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -1142,6 +1142,17 @@ static void cfq_put_queue(struct cfq_queue *cfqq) kmem_cache_free(cfq_pool, cfqq); } +static void +__call_for_each_cic(struct io_context *ioc, + void (*func)(struct io_context *, struct cfq_io_context *)) +{ + struct cfq_io_context *cic; + struct hlist_node *n; + + hlist_for_each_entry_rcu(cic, n, &ioc->cic_list, cic_list) + func(ioc, cic); +} + /* * Call func for each cic attached to this ioc. */ @@ -1149,12 +1160,8 @@ static void call_for_each_cic(struct io_context *ioc, void (*func)(struct io_context *, struct cfq_io_context *)) { - struct cfq_io_context *cic; - struct hlist_node *n; - rcu_read_lock(); - hlist_for_each_entry_rcu(cic, n, &ioc->cic_list, cic_list) - func(ioc, cic); + __call_for_each_cic(ioc, func); rcu_read_unlock(); } @@ -1198,7 +1205,7 @@ static void cfq_free_io_context(struct io_context *ioc) * should be ok to iterate over the known list, we will see all cic's * since no new ones are added. */ - call_for_each_cic(ioc, cic_free_func); + __call_for_each_cic(ioc, cic_free_func); } static void cfq_exit_cfqq(struct cfq_data *cfqd, struct cfq_queue *cfqq) @@ -1296,10 +1303,10 @@ static void cfq_init_prio_data(struct cfq_queue *cfqq, struct io_context *ioc) printk(KERN_ERR "cfq: bad prio %x\n", ioprio_class); case IOPRIO_CLASS_NONE: /* - * no prio set, place us in the middle of the BE classes + * no prio set, inherit CPU scheduling settings */ cfqq->ioprio = task_nice_ioprio(tsk); - cfqq->ioprio_class = IOPRIO_CLASS_BE; + cfqq->ioprio_class = task_nice_ioclass(tsk); break; case IOPRIO_CLASS_RT: cfqq->ioprio = task_ioprio(ioc); diff --git a/drivers/accessibility/Kconfig b/drivers/accessibility/Kconfig index 1264c4b9809..ef3b65bfdd0 100644 --- a/drivers/accessibility/Kconfig +++ b/drivers/accessibility/Kconfig @@ -1,7 +1,17 @@ menuconfig ACCESSIBILITY bool "Accessibility support" ---help--- - Enable a submenu where accessibility items may be enabled. + Accessibility handles all special kinds of hardware devices or + software adapters which help people with disabilities (e.g. + blindness) to use computers. + + That includes braille devices, speech synthesis, keyboard + remapping, etc. + + Say Y here to get to see options for accessibility. + This option alone does not add any kernel code. + + If you say N, all options in this submenu will be skipped and disabled. If unsure, say N. diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig index 1c11df9a5f3..9bf2986a278 100644 --- a/drivers/ata/Kconfig +++ b/drivers/ata/Kconfig @@ -205,8 +205,8 @@ config SATA_VITESSE If unsure, say N. config SATA_INIC162X - tristate "Initio 162x SATA support (HIGHLY EXPERIMENTAL)" - depends on PCI && EXPERIMENTAL + tristate "Initio 162x SATA support" + depends on PCI help This option enables support for Initio 162x Serial ATA. @@ -697,6 +697,15 @@ config PATA_SCC If unsure, say N. +config PATA_SCH + tristate "Intel SCH PATA support" + depends on PCI + help + This option enables support for Intel SCH PATA on the Intel + SCH (US15W, US15L, UL11L) series host controllers. + + If unsure, say N. + config PATA_BF54X tristate "Blackfin 54x ATAPI support" depends on BF542 || BF548 || BF549 diff --git a/drivers/ata/Makefile b/drivers/ata/Makefile index b693d829383..674965fa326 100644 --- a/drivers/ata/Makefile +++ b/drivers/ata/Makefile @@ -67,6 +67,7 @@ obj-$(CONFIG_PATA_SIS) += pata_sis.o obj-$(CONFIG_PATA_TRIFLEX) += pata_triflex.o obj-$(CONFIG_PATA_IXP4XX_CF) += pata_ixp4xx_cf.o obj-$(CONFIG_PATA_SCC) += pata_scc.o +obj-$(CONFIG_PATA_SCH) += pata_sch.o obj-$(CONFIG_PATA_BF54X) += pata_bf54x.o obj-$(CONFIG_PATA_PLATFORM) += pata_platform.o obj-$(CONFIG_PATA_OF_PLATFORM) += pata_of_platform.o diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index 8cace9aa9c0..97f83fb2ee2 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -1267,9 +1267,7 @@ static int ahci_check_ready(struct ata_link *link) void __iomem *port_mmio = ahci_port_base(link->ap); u8 status = readl(port_mmio + PORT_TFDATA) & 0xFF; - if (!(status & ATA_BUSY)) - return 1; - return 0; + return ata_check_ready(status); } static int ahci_softreset(struct ata_link *link, unsigned int *class, diff --git a/drivers/ata/ata_generic.c b/drivers/ata/ata_generic.c index 47aeccd52fa..75a406f5e69 100644 --- a/drivers/ata/ata_generic.c +++ b/drivers/ata/ata_generic.c @@ -152,6 +152,12 @@ static int ata_generic_init_one(struct pci_dev *dev, const struct pci_device_id if (dev->vendor == PCI_VENDOR_ID_AL) ata_pci_bmdma_clear_simplex(dev); + if (dev->vendor == PCI_VENDOR_ID_ATI) { + int rc = pcim_enable_device(dev); + if (rc < 0) + return rc; + pcim_pin_device(dev); + } return ata_pci_sff_init_one(dev, ppi, &generic_sht, NULL); } diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c index ea2c7649d39..a9027b8fbdd 100644 --- a/drivers/ata/ata_piix.c +++ b/drivers/ata/ata_piix.c @@ -1348,6 +1348,8 @@ static void __devinit piix_init_sidpr(struct ata_host *host) { struct pci_dev *pdev = to_pci_dev(host->dev); struct piix_host_priv *hpriv = host->private_data; + struct ata_device *dev0 = &host->ports[0]->link.device[0]; + u32 scontrol; int i; /* check for availability */ @@ -1366,6 +1368,29 @@ static void __devinit piix_init_sidpr(struct ata_host *host) return; hpriv->sidpr = pcim_iomap_table(pdev)[PIIX_SIDPR_BAR]; + + /* SCR access via SIDPR doesn't work on some configurations. + * Give it a test drive by inhibiting power save modes which + * we'll do anyway. + */ + scontrol = piix_sidpr_read(dev0, SCR_CONTROL); + + /* if IPM is already 3, SCR access is probably working. Don't + * un-inhibit power save modes as BIOS might have inhibited + * them for a reason. + */ + if ((scontrol & 0xf00) != 0x300) { + scontrol |= 0x300; + piix_sidpr_write(dev0, SCR_CONTROL, scontrol); + scontrol = piix_sidpr_read(dev0, SCR_CONTROL); + + if ((scontrol & 0xf00) != 0x300) { + dev_printk(KERN_INFO, host->dev, "SCR access via " + "SIDPR is available but doesn't work\n"); + return; + } + } + host->ports[0]->ops = &piix_sidpr_sata_ops; host->ports[1]->ops = &piix_sidpr_sata_ops; } diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 3bc48853820..927b692d723 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -6292,6 +6292,7 @@ EXPORT_SYMBOL_GPL(ata_eh_freeze_port); EXPORT_SYMBOL_GPL(ata_eh_thaw_port); EXPORT_SYMBOL_GPL(ata_eh_qc_complete); EXPORT_SYMBOL_GPL(ata_eh_qc_retry); +EXPORT_SYMBOL_GPL(ata_eh_analyze_ncq_error); EXPORT_SYMBOL_GPL(ata_do_eh); EXPORT_SYMBOL_GPL(ata_std_error_handler); diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index 61dcd0026c6..62e033146be 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -1357,7 +1357,7 @@ static void ata_eh_analyze_serror(struct ata_link *link) * LOCKING: * Kernel thread context (may sleep). */ -static void ata_eh_analyze_ncq_error(struct ata_link *link) +void ata_eh_analyze_ncq_error(struct ata_link *link) { struct ata_port *ap = link->ap; struct ata_eh_context *ehc = &link->eh_context; diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c index 2ec65a8fda7..3c2d2289f85 100644 --- a/drivers/ata/libata-sff.c +++ b/drivers/ata/libata-sff.c @@ -314,11 +314,7 @@ static int ata_sff_check_ready(struct ata_link *link) { u8 status = link->ap->ops->sff_check_status(link->ap); - if (!(status & ATA_BUSY)) - return 1; - if (status == 0xff) - return -ENODEV; - return 0; + return ata_check_ready(status); } /** diff --git a/drivers/ata/pata_acpi.c b/drivers/ata/pata_acpi.c index c5f91e62994..fbe60571155 100644 --- a/drivers/ata/pata_acpi.c +++ b/drivers/ata/pata_acpi.c @@ -259,6 +259,12 @@ static int pacpi_init_one (struct pci_dev *pdev, const struct pci_device_id *id) .port_ops = &pacpi_ops, }; const struct ata_port_info *ppi[] = { &info, NULL }; + if (pdev->vendor == PCI_VENDOR_ID_ATI) { + int rc = pcim_enable_device(pdev); + if (rc < 0) + return rc; + pcim_pin_device(pdev); + } return ata_pci_sff_init_one(pdev, ppi, &pacpi_sht, NULL); } diff --git a/drivers/ata/pata_sch.c b/drivers/ata/pata_sch.c new file mode 100644 index 00000000000..c8cc027789f --- /dev/null +++ b/drivers/ata/pata_sch.c @@ -0,0 +1,206 @@ +/* + * pata_sch.c - Intel SCH PATA controllers + * + * Copyright (c) 2008 Alek Du <alek.du@intel.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License 2 as published + * by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; see the file COPYING. If not, write to + * the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. + * + */ + +/* + * Supports: + * Intel SCH (AF82US15W, AF82US15L, AF82UL11L) chipsets -- see spec at: + * http://download.intel.com/design/chipsets/embedded/datashts/319537.pdf + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/pci.h> +#include <linux/init.h> +#include <linux/blkdev.h> +#include <linux/delay.h> +#include <linux/device.h> +#include <scsi/scsi_host.h> +#include <linux/libata.h> +#include <linux/dmi.h> + +#define DRV_NAME "pata_sch" +#define DRV_VERSION "0.2" + +/* see SCH datasheet page 351 */ +enum { + D0TIM = 0x80, /* Device 0 Timing Register */ + D1TIM = 0x84, /* Device 1 Timing Register */ + PM = 0x07, /* PIO Mode Bit Mask */ + MDM = (0x03 << 8), /* Multi-word DMA Mode Bit Mask */ + UDM = (0x07 << 16), /* Ultra DMA Mode Bit Mask */ + PPE = (1 << 30), /* Prefetch/Post Enable */ + USD = (1 << 31), /* Use Synchronous DMA */ +}; + +static int sch_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent); +static void sch_set_piomode(struct ata_port *ap, struct ata_device *adev); +static void sch_set_dmamode(struct ata_port *ap, struct ata_device *adev); + +static const struct pci_device_id sch_pci_tbl[] = { + /* Intel SCH PATA Controller */ + { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_SCH_IDE), 0 }, + { } /* terminate list */ +}; + +static struct pci_driver sch_pci_driver = { + .name = DRV_NAME, + .id_table = sch_pci_tbl, + .probe = sch_init_one, + .remove = ata_pci_remove_one, +#ifdef CONFIG_PM + .suspend = ata_pci_device_suspend, + .resume = ata_pci_device_resume, +#endif +}; + +static struct scsi_host_template sch_sht = { + ATA_BMDMA_SHT(DRV_NAME), +}; + +static struct ata_port_operations sch_pata_ops = { + .inherits = &ata_bmdma_port_ops, + .cable_detect = ata_cable_unknown, + .set_piomode = sch_set_piomode, + .set_dmamode = sch_set_dmamode, +}; + +static struct ata_port_info sch_port_info = { + .flags = 0, + .pio_mask = ATA_PIO4, /* pio0-4 */ + .mwdma_mask = ATA_MWDMA2, /* mwdma0-2 */ + .udma_mask = ATA_UDMA5, /* udma0-5 */ + .port_ops = &sch_pata_ops, +}; + +MODULE_AUTHOR("Alek Du <alek.du@intel.com>"); +MODULE_DESCRIPTION("SCSI low-level driver for Intel SCH PATA controllers"); +MODULE_LICENSE("GPL"); +MODULE_DEVICE_TABLE(pci, sch_pci_tbl); +MODULE_VERSION(DRV_VERSION); + +/** + * sch_set_piomode - Initialize host controller PATA PIO timings + * @ap: Port whose timings we are configuring + * @adev: ATA device + * + * Set PIO mode for device, in host controller PCI config space. + * + * LOCKING: + * None (inherited from caller). + */ + +static void sch_set_piomode(struct ata_port *ap, struct ata_device *adev) +{ + unsigned int pio = adev->pio_mode - XFER_PIO_0; + struct pci_dev *dev = to_pci_dev(ap->host->dev); + unsigned int port = adev->devno ? D1TIM : D0TIM; + unsigned int data; + + pci_read_config_dword(dev, port, &data); + /* see SCH datasheet page 351 */ + /* set PIO mode */ + data &= ~(PM | PPE); + data |= pio; + /* enable PPE for block device */ + if (adev->class == ATA_DEV_ATA) + data |= PPE; + pci_write_config_dword(dev, port, data); +} + +/** + * sch_set_dmamode - Initialize host controller PATA DMA timings + * @ap: Port whose timings we are configuring + * @adev: ATA device + * + * Set MW/UDMA mode for device, in host controller PCI config space. + * + * LOCKING: + * None (inherited from caller). + */ + +static void sch_set_dmamode(struct ata_port *ap, struct ata_device *adev) +{ + unsigned int dma_mode = adev->dma_mode; + struct pci_dev *dev = to_pci_dev(ap->host->dev); + unsigned int port = adev->devno ? D1TIM : D0TIM; + unsigned int data; + + pci_read_config_dword(dev, port, &data); + /* see SCH datasheet page 351 */ + if (dma_mode >= XFER_UDMA_0) { + /* enable Synchronous DMA mode */ + data |= USD; + data &= ~UDM; + data |= (dma_mode - XFER_UDMA_0) << 16; + } else { /* must be MWDMA mode, since we masked SWDMA already */ + data &= ~(USD | MDM); + data |= (dma_mode - XFER_MW_DMA_0) << 8; + } + pci_write_config_dword(dev, port, data); +} + +/** + * sch_init_one - Register SCH ATA PCI device with kernel services + * @pdev: PCI device to register + * @ent: Entry in sch_pci_tbl matching with @pdev + * + * LOCKING: + * Inherited from PCI layer (may sleep). + * + * RETURNS: + * Zero on success, or -ERRNO value. + */ + +static int __devinit sch_init_one(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + static int printed_version; + const struct ata_port_info *ppi[] = { &sch_port_info, NULL }; + struct ata_host *host; + int rc; + + if (!printed_version++) + dev_printk(KERN_DEBUG, &pdev->dev, + "version " DRV_VERSION "\n"); + + /* enable device and prepare host */ + rc = pcim_enable_device(pdev); + if (rc) + return rc; + rc = ata_pci_sff_prepare_host(pdev, ppi, &host); + if (rc) + return rc; + pci_set_master(pdev); + return ata_pci_sff_activate_host(host, ata_sff_interrupt, &sch_sht); +} + +static int __init sch_init(void) +{ + return pci_register_driver(&sch_pci_driver); +} + +static void __exit sch_exit(void) +{ + pci_unregister_driver(&sch_pci_driver); +} + +module_init(sch_init); +module_exit(sch_exit); diff --git a/drivers/ata/sata_inic162x.c b/drivers/ata/sata_inic162x.c index d27bb9a2568..3ead02fe379 100644 --- a/drivers/ata/sata_inic162x.c +++ b/drivers/ata/sata_inic162x.c @@ -10,13 +10,33 @@ * right. Documentation is available at initio's website but it only * documents registers (not programming model). * - * - ATA disks work. - * - Hotplug works. - * - ATAPI read works but burning doesn't. This thing is really - * peculiar about ATAPI and I couldn't figure out how ATAPI PIO and - * ATAPI DMA WRITE should be programmed. If you've got a clue, be - * my guest. - * - Both STR and STD work. + * This driver has interesting history. The first version was written + * from the documentation and a 2.4 IDE driver posted on a Taiwan + * company, which didn't use any IDMA features and couldn't handle + * LBA48. The resulting driver couldn't handle LBA48 devices either + * making it pretty useless. + * + * After a while, initio picked the driver up, renamed it to + * sata_initio162x, updated it to use IDMA for ATA DMA commands and + * posted it on their website. It only used ATA_PROT_DMA for IDMA and + * attaching both devices and issuing IDMA and !IDMA commands + * simultaneously broke it due to PIRQ masking interaction but it did + * show how to use the IDMA (ADMA + some initio specific twists) + * engine. + * + * Then, I picked up their changes again and here's the usable driver + * which uses IDMA for everything. Everything works now including + * LBA48, CD/DVD burning, suspend/resume and hotplug. There are some + * issues tho. Result Tf is not resported properly, NCQ isn't + * supported yet and CD/DVD writing works with DMA assisted PIO + * protocol (which, for native SATA devices, shouldn't cause any + * noticeable difference). + * + * Anyways, so, here's finally a working driver for inic162x. Enjoy! + * + * initio: If you guys wanna improve the driver regarding result TF + * access and other stuff, please feel free to contact me. I'll be + * happy to assist. */ #include <linux/kernel.h> @@ -28,13 +48,19 @@ #include <scsi/scsi_device.h> #define DRV_NAME "sata_inic162x" -#define DRV_VERSION "0.3" +#define DRV_VERSION "0.4" enum { - MMIO_BAR = 5, + MMIO_BAR_PCI = 5, + MMIO_BAR_CARDBUS = 1, NR_PORTS = 2, + IDMA_CPB_TBL_SIZE = 4 * 32, + + INIC_DMA_BOUNDARY = 0xffffff, + + HOST_ACTRL = 0x08, HOST_CTL = 0x7c, HOST_STAT = 0x7e, HOST_IRQ_STAT = 0xbc, @@ -43,22 +69,37 @@ enum { PORT_SIZE = 0x40, /* registers for ATA TF operation */ - PORT_TF = 0x00, - PORT_ALT_STAT = 0x08, + PORT_TF_DATA = 0x00, + PORT_TF_FEATURE = 0x01, + PORT_TF_NSECT = 0x02, + PORT_TF_LBAL = 0x03, + PORT_TF_LBAM = 0x04, + PORT_TF_LBAH = 0x05, + PORT_TF_DEVICE = 0x06, + PORT_TF_COMMAND = 0x07, + PORT_TF_ALT_STAT = 0x08, PORT_IRQ_STAT = 0x09, PORT_IRQ_MASK = 0x0a, PORT_PRD_CTL = 0x0b, PORT_PRD_ADDR = 0x0c, PORT_PRD_XFERLEN = 0x10, + PORT_CPB_CPBLAR = 0x18, + PORT_CPB_PTQFIFO = 0x1c, /* IDMA register */ PORT_IDMA_CTL = 0x14, + PORT_IDMA_STAT = 0x16, + + PORT_RPQ_FIFO = 0x1e, + PORT_RPQ_CNT = 0x1f, PORT_SCR = 0x20, /* HOST_CTL bits */ HCTL_IRQOFF = (1 << 8), /* global IRQ off */ - HCTL_PWRDWN = (1 << 13), /* power down PHYs */ + HCTL_FTHD0 = (1 << 10), /* fifo threshold 0 */ + HCTL_FTHD1 = (1 << 11), /* fifo threshold 1*/ + HCTL_PWRDWN = (1 << 12), /* power down PHYs */ HCTL_SOFTRST = (1 << 13), /* global reset (no phy reset) */ HCTL_RPGSEL = (1 << 15), /* register page select */ @@ -81,9 +122,7 @@ enum { PIRQ_PENDING = (1 << 7), /* port IRQ pending (STAT only) */ PIRQ_ERR = PIRQ_OFFLINE | PIRQ_ONLINE | PIRQ_FATAL, - - PIRQ_MASK_DMA_READ = PIRQ_REPLY | PIRQ_ATA, - PIRQ_MASK_OTHER = PIRQ_REPLY | PIRQ_COMPLETE, + PIRQ_MASK_DEFAULT = PIRQ_REPLY | PIRQ_ATA, PIRQ_MASK_FREEZE = 0xff, /* PORT_PRD_CTL bits */ @@ -96,20 +135,104 @@ enum { IDMA_CTL_RST_IDMA = (1 << 5), /* reset IDMA machinary */ IDMA_CTL_GO = (1 << 7), /* IDMA mode go */ IDMA_CTL_ATA_NIEN = (1 << 8), /* ATA IRQ disable */ + + /* PORT_IDMA_STAT bits */ + IDMA_STAT_PERR = (1 << 0), /* PCI ERROR MODE */ + IDMA_STAT_CPBERR = (1 << 1), /* ADMA CPB error */ + IDMA_STAT_LGCY = (1 << 3), /* ADMA legacy */ + IDMA_STAT_UIRQ = (1 << 4), /* ADMA unsolicited irq */ + IDMA_STAT_STPD = (1 << 5), /* ADMA stopped */ + IDMA_STAT_PSD = (1 << 6), /* ADMA pause */ + IDMA_STAT_DONE = (1 << 7), /* ADMA done */ + + IDMA_STAT_ERR = IDMA_STAT_PERR | IDMA_STAT_CPBERR, + + /* CPB Control Flags*/ + CPB_CTL_VALID = (1 << 0), /* CPB valid */ + CPB_CTL_QUEUED = (1 << 1), /* queued command */ + CPB_CTL_DATA = (1 << 2), /* data, rsvd in datasheet */ + CPB_CTL_IEN = (1 << 3), /* PCI interrupt enable */ + CPB_CTL_DEVDIR = (1 << 4), /* device direction control */ + + /* CPB Response Flags */ + CPB_RESP_DONE = (1 << 0), /* ATA command complete */ + CPB_RESP_REL = (1 << 1), /* ATA release */ + CPB_RESP_IGNORED = (1 << 2), /* CPB ignored */ + CPB_RESP_ATA_ERR = (1 << 3), /* ATA command error */ + CPB_RESP_SPURIOUS = (1 << 4), /* ATA spurious interrupt error */ + CPB_RESP_UNDERFLOW = (1 << 5), /* APRD deficiency length error */ + CPB_RESP_OVERFLOW = (1 << 6), /* APRD exccess length error */ + CPB_RESP_CPB_ERR = (1 << 7), /* CPB error flag */ + + /* PRD Control Flags */ + PRD_DRAIN = (1 << 1), /* ignore data excess */ + PRD_CDB = (1 << 2), /* atapi packet command pointer */ + PRD_DIRECT_INTR = (1 << 3), /* direct interrupt */ + PRD_DMA = (1 << 4), /* data transfer method */ + PRD_WRITE = (1 << 5), /* data dir, rsvd in datasheet */ + PRD_IOM = (1 << 6), /* io/memory transfer */ + PRD_END = (1 << 7), /* APRD chain end */ }; +/* Comman Parameter Block */ +struct inic_cpb { + u8 resp_flags; /* Response Flags */ + u8 error; /* ATA Error */ + u8 status; /* ATA Status */ + u8 ctl_flags; /* Control Flags */ + __le32 len; /* Total Transfer Length */ + __le32 prd; /* First PRD pointer */ + u8 rsvd[4]; + /* 16 bytes */ + u8 feature; /* ATA Feature */ + u8 hob_feature; /* ATA Ex. Feature */ + u8 device; /* ATA Device/Head */ + u8 mirctl; /* Mirror Control */ + u8 nsect; /* ATA Sector Count */ + u8 hob_nsect; /* ATA Ex. Sector Count */ + u8 lbal; /* ATA Sector Number */ + u8 hob_lbal; /* ATA Ex. Sector Number */ + u8 lbam; /* ATA Cylinder Low */ + u8 hob_lbam; /* ATA Ex. Cylinder Low */ + u8 lbah; /* ATA Cylinder High */ + u8 hob_lbah; /* ATA Ex. Cylinder High */ + u8 command; /* ATA Command */ + u8 ctl; /* ATA Control */ + u8 slave_error; /* Slave ATA Error */ + u8 slave_status; /* Slave ATA Status */ + /* 32 bytes */ +} __packed; + +/* Physical Region Descriptor */ +struct inic_prd { + __le32 mad; /* Physical Memory Address */ + __le16 len; /* Transfer Length */ + u8 rsvd; + u8 flags; /* Control Flags */ +} __packed; + +struct inic_pkt { + struct inic_cpb cpb; + struct inic_prd prd[LIBATA_MAX_PRD + 1]; /* + 1 for cdb */ + u8 cdb[ATAPI_CDB_LEN]; +} __packed; + struct inic_host_priv { - u16 cached_hctl; + void __iomem *mmio_base; + u16 cached_hctl; }; struct inic_port_priv { - u8 dfl_prdctl; - u8 cached_prdctl; - u8 cached_pirq_mask; + struct inic_pkt *pkt; + dma_addr_t pkt_dma; + u32 *cpb_tbl; + dma_addr_t cpb_tbl_dma; }; static struct scsi_host_template inic_sht = { - ATA_BMDMA_SHT(DRV_NAME), + ATA_BASE_SHT(DRV_NAME), + .sg_tablesize = LIBATA_MAX_PRD, /* maybe it can be larger? */ + .dma_boundary = INIC_DMA_BOUNDARY, }; static const int scr_map[] = { @@ -120,54 +243,34 @@ static const int scr_map[] = { static void __iomem *inic_port_base(struct ata_port *ap) { - return ap->host->iomap[MMIO_BAR] + ap->port_no * PORT_SIZE; -} - -static void __inic_set_pirq_mask(struct ata_port *ap, u8 mask) -{ - void __iomem *port_base = inic_port_base(ap); - struct inic_port_priv *pp = ap->private_data; + struct inic_host_priv *hpriv = ap->host->private_data; - writeb(mask, port_base + PORT_IRQ_MASK); - pp->cached_pirq_mask = mask; -} - -static void inic_set_pirq_mask(struct ata_port *ap, u8 mask) -{ - struct inic_port_priv *pp = ap->private_data; - - if (pp->cached_pirq_mask != mask) - __inic_set_pirq_mask(ap, mask); + return hpriv->mmio_base + ap->port_no * PORT_SIZE; } static void inic_reset_port(void __iomem *port_base) { void __iomem *idma_ctl = port_base + PORT_IDMA_CTL; - u16 ctl; - ctl = readw(idma_ctl); - ctl &= ~(IDMA_CTL_RST_IDMA | IDMA_CTL_ATA_NIEN | IDMA_CTL_GO); + /* stop IDMA engine */ + readw(idma_ctl); /* flush */ + msleep(1); /* mask IRQ and assert reset */ - writew(ctl | IDMA_CTL_RST_IDMA | IDMA_CTL_ATA_NIEN, idma_ctl); + writew(IDMA_CTL_RST_IDMA, idma_ctl); readw(idma_ctl); /* flush */ - - /* give it some time */ msleep(1); /* release reset */ - writew(ctl | IDMA_CTL_ATA_NIEN, idma_ctl); + writew(0, idma_ctl); /* clear irq */ writeb(0xff, port_base + PORT_IRQ_STAT); - - /* reenable ATA IRQ, turn off IDMA mode */ - writew(ctl, idma_ctl); } static int inic_scr_read(struct ata_port *ap, unsigned sc_reg, u32 *val) { - void __iomem *scr_addr = ap->ioaddr.scr_addr; + void __iomem *scr_addr = inic_port_base(ap) + PORT_SCR; void __iomem *addr; if (unlikely(sc_reg >= ARRAY_SIZE(scr_map))) @@ -184,120 +287,126 @@ static int inic_scr_read(struct ata_port *ap, unsigned sc_reg, u32 *val) static int inic_scr_write(struct ata_port *ap, unsigned sc_reg, u32 val) { - void __iomem *scr_addr = ap->ioaddr.scr_addr; - void __iomem *addr; + void __iomem *scr_addr = inic_port_base(ap) + PORT_SCR; if (unlikely(sc_reg >= ARRAY_SIZE(scr_map))) return -EINVAL; - addr = scr_addr + scr_map[sc_reg] * 4; writel(val, scr_addr + scr_map[sc_reg] * 4); return 0; } -/* - * In TF mode, inic162x is very similar to SFF device. TF registers - * function the same. DMA engine behaves similary using the same PRD - * format as BMDMA but different command register, interrupt and event - * notification methods are used. The following inic_bmdma_*() - * functions do the impedance matching. - */ -static void inic_bmdma_setup(struct ata_queued_cmd *qc) +static void inic_stop_idma(struct ata_port *ap) { - struct ata_port *ap = qc->ap; - struct inic_port_priv *pp = ap->private_data; void __iomem *port_base = inic_port_base(ap); - int rw = qc->tf.flags & ATA_TFLAG_WRITE; - - /* make sure device sees PRD table writes */ - wmb(); - - /* load transfer length */ - writel(qc->nbytes, port_base + PORT_PRD_XFERLEN); - - /* turn on DMA and specify data direction */ - pp->cached_prdctl = pp->dfl_prdctl | PRD_CTL_DMAEN; - if (!rw) - pp->cached_prdctl |= PRD_CTL_WR; - writeb(pp->cached_prdctl, port_base + PORT_PRD_CTL); - /* issue r/w command */ - ap->ops->sff_exec_command(ap, &qc->tf); + readb(port_base + PORT_RPQ_FIFO); + readb(port_base + PORT_RPQ_CNT); + writew(0, port_base + PORT_IDMA_CTL); } -static void inic_bmdma_start(struct ata_queued_cmd *qc) +static void inic_host_err_intr(struct ata_port *ap, u8 irq_stat, u16 idma_stat) { - struct ata_port *ap = qc->ap; + struct ata_eh_info *ehi = &ap->link.eh_info; struct inic_port_priv *pp = ap->private_data; - void __iomem *port_base = inic_port_base(ap); + struct inic_cpb *cpb = &pp->pkt->cpb; + bool freeze = false; - /* start host DMA transaction */ - pp->cached_prdctl |= PRD_CTL_START; - writeb(pp->cached_prdctl, port_base + PORT_PRD_CTL); -} + ata_ehi_clear_desc(ehi); + ata_ehi_push_desc(ehi, "irq_stat=0x%x idma_stat=0x%x", + irq_stat, idma_stat); -static void inic_bmdma_stop(struct ata_queued_cmd *qc) -{ - struct ata_port *ap = qc->ap; - struct inic_port_priv *pp = ap->private_data; - void __iomem *port_base = inic_port_base(ap); + inic_stop_idma(ap); - /* stop DMA engine */ - writeb(pp->dfl_prdctl, port_base + PORT_PRD_CTL); -} + if (irq_stat & (PIRQ_OFFLINE | PIRQ_ONLINE)) { + ata_ehi_push_desc(ehi, "hotplug"); + ata_ehi_hotplugged(ehi); + freeze = true; + } -static u8 inic_bmdma_status(struct ata_port *ap) -{ - /* event is already verified by the interrupt handler */ - return ATA_DMA_INTR; + if (idma_stat & IDMA_STAT_PERR) { + ata_ehi_push_desc(ehi, "PCI error"); + freeze = true; + } + + if (idma_stat & IDMA_STAT_CPBERR) { + ata_ehi_push_desc(ehi, "CPB error"); + + if (cpb->resp_flags & CPB_RESP_IGNORED) { + __ata_ehi_push_desc(ehi, " ignored"); + ehi->err_mask |= AC_ERR_INVALID; + freeze = true; + } + + if (cpb->resp_flags & CPB_RESP_ATA_ERR) + ehi->err_mask |= AC_ERR_DEV; + + if (cpb->resp_flags & CPB_RESP_SPURIOUS) { + __ata_ehi_push_desc(ehi, " spurious-intr"); + ehi->err_mask |= AC_ERR_HSM; + freeze = true; + } + + if (cpb->resp_flags & + (CPB_RESP_UNDERFLOW | CPB_RESP_OVERFLOW)) { + __ata_ehi_push_desc(ehi, " data-over/underflow"); + ehi->err_mask |= AC_ERR_HSM; + freeze = true; + } + } + + if (freeze) + ata_port_freeze(ap); + else + ata_port_abort(ap); } static void inic_host_intr(struct ata_port *ap) { void __iomem *port_base = inic_port_base(ap); - struct ata_eh_info *ehi = &ap->link.eh_info; + struct ata_queued_cmd *qc = ata_qc_from_tag(ap, ap->link.active_tag); u8 irq_stat; + u16 idma_stat; - /* fetch and clear irq */ + /* read and clear IRQ status */ irq_stat = readb(port_base + PORT_IRQ_STAT); writeb(irq_stat, port_base + PORT_IRQ_STAT); + idma_stat = readw(port_base + PORT_IDMA_STAT); - if (likely(!(irq_stat & PIRQ_ERR))) { - struct ata_queued_cmd *qc = - ata_qc_from_tag(ap, ap->link.active_tag); + if (unlikely((irq_stat & PIRQ_ERR) || (idma_stat & IDMA_STAT_ERR))) + inic_host_err_intr(ap, irq_stat, idma_stat); - if (unlikely(!qc || (qc->tf.flags & ATA_TFLAG_POLLING))) { - ap->ops->sff_check_status(ap); /* clear ATA interrupt */ - return; - } + if (unlikely(!qc)) + goto spurious; - if (likely(ata_sff_host_intr(ap, qc))) - return; + if (likely(idma_stat & IDMA_STAT_DONE)) { + inic_stop_idma(ap); - ap->ops->sff_check_status(ap); /* clear ATA interrupt */ - ata_port_printk(ap, KERN_WARNING, "unhandled " - "interrupt, irq_stat=%x\n", irq_stat); + /* Depending on circumstances, device error + * isn't reported by IDMA, check it explicitly. + */ + if (unlikely(readb(port_base + PORT_TF_COMMAND) & + (ATA_DF | ATA_ERR))) + qc->err_mask |= AC_ERR_DEV; + + ata_qc_complete(qc); return; } - /* error */ - ata_ehi_push_desc(ehi, "irq_stat=0x%x", irq_stat); - - if (irq_stat & (PIRQ_OFFLINE | PIRQ_ONLINE)) { - ata_ehi_hotplugged(ehi); - ata_port_freeze(ap); - } else - ata_port_abort(ap); + spurious: + ata_port_printk(ap, KERN_WARNING, "unhandled interrupt: " + "cmd=0x%x irq_stat=0x%x idma_stat=0x%x\n", + qc ? qc->tf.command : 0xff, irq_stat, idma_stat); } static irqreturn_t inic_interrupt(int irq, void *dev_instance) { struct ata_host *host = dev_instance; - void __iomem *mmio_base = host->iomap[MMIO_BAR]; + struct inic_host_priv *hpriv = host->private_data; u16 host_irq_stat; int i, handled = 0;; - host_irq_stat = readw(mmio_base + HOST_IRQ_STAT); + host_irq_stat = readw(hpriv->mmio_base + HOST_IRQ_STAT); if (unlikely(!(host_irq_stat & HIRQ_GLOBAL))) goto out; @@ -327,60 +436,173 @@ static irqreturn_t inic_interrupt(int irq, void *dev_instance) return IRQ_RETVAL(handled); } +static int inic_check_atapi_dma(struct ata_queued_cmd *qc) +{ + /* For some reason ATAPI_PROT_DMA doesn't work for some + * commands including writes and other misc ops. Use PIO + * protocol instead, which BTW is driven by the DMA engine + * anyway, so it shouldn't make much difference for native + * SATA devices. + */ + if (atapi_cmd_type(qc->cdb[0]) == READ) + return 0; + return 1; +} + +static void inic_fill_sg(struct inic_prd *prd, struct ata_queued_cmd *qc) +{ + struct scatterlist *sg; + unsigned int si; + u8 flags = 0; + + if (qc->tf.flags & ATA_TFLAG_WRITE) + flags |= PRD_WRITE; + + if (ata_is_dma(qc->tf.protocol)) + flags |= PRD_DMA; + + for_each_sg(qc->sg, sg, qc->n_elem, si) { + prd->mad = cpu_to_le32(sg_dma_address(sg)); + prd->len = cpu_to_le16(sg_dma_len(sg)); + prd->flags = flags; + prd++; + } + + WARN_ON(!si); + prd[-1].flags |= PRD_END; +} + +static void inic_qc_prep(struct ata_queued_cmd *qc) +{ + struct inic_port_priv *pp = qc->ap->private_data; + struct inic_pkt *pkt = pp->pkt; + struct inic_cpb *cpb = &pkt->cpb; + struct inic_prd *prd = pkt->prd; + bool is_atapi = ata_is_atapi(qc->tf.protocol); + bool is_data = ata_is_data(qc->tf.protocol); + unsigned int cdb_len = 0; + + VPRINTK("ENTER\n"); + + if (is_atapi) + cdb_len = qc->dev->cdb_len; + + /* prepare packet, based on initio driver */ + memset(pkt, 0, sizeof(struct inic_pkt)); + + cpb->ctl_flags = CPB_CTL_VALID | CPB_CTL_IEN; + if (is_atapi || is_data) + cpb->ctl_flags |= CPB_CTL_DATA; + + cpb->len = cpu_to_le32(qc->nbytes + cdb_len); + cpb->prd = cpu_to_le32(pp->pkt_dma + offsetof(struct inic_pkt, prd)); + + cpb->device = qc->tf.device; + cpb->feature = qc->tf.feature; + cpb->nsect = qc->tf.nsect; + cpb->lbal = qc->tf.lbal; + cpb->lbam = qc->tf.lbam; + cpb->lbah = qc->tf.lbah; + + if (qc->tf.flags & ATA_TFLAG_LBA48) { + cpb->hob_feature = qc->tf.hob_feature; + cpb->hob_nsect = qc->tf.hob_nsect; + cpb->hob_lbal = qc->tf.hob_lbal; + cpb->hob_lbam = qc->tf.hob_lbam; + cpb->hob_lbah = qc->tf.hob_lbah; + } + + cpb->command = qc->tf.command; + /* don't load ctl - dunno why. it's like that in the initio driver */ + + /* setup PRD for CDB */ + if (is_atapi) { + memcpy(pkt->cdb, qc->cdb, ATAPI_CDB_LEN); + prd->mad = cpu_to_le32(pp->pkt_dma + + offsetof(struct inic_pkt, cdb)); + prd->len = cpu_to_le16(cdb_len); + prd->flags = PRD_CDB | PRD_WRITE; + if (!is_data) + prd->flags |= PRD_END; + prd++; + } + + /* setup sg table */ + if (is_data) + inic_fill_sg(prd, qc); + + pp->cpb_tbl[0] = pp->pkt_dma; +} + static unsigned int inic_qc_issue(struct ata_queued_cmd *qc) { struct ata_port *ap = qc->ap; + void __iomem *port_base = inic_port_base(ap); - /* ATA IRQ doesn't wait for DMA transfer completion and vice - * versa. Mask IRQ selectively to detect command completion. - * Without it, ATA DMA read command can cause data corruption. - * - * Something similar might be needed for ATAPI writes. I - * tried a lot of combinations but couldn't find the solution. - */ - if (qc->tf.protocol == ATA_PROT_DMA && - !(qc->tf.flags & ATA_TFLAG_WRITE)) - inic_set_pirq_mask(ap, PIRQ_MASK_DMA_READ); - else - inic_set_pirq_mask(ap, PIRQ_MASK_OTHER); + /* fire up the ADMA engine */ + writew(HCTL_FTHD0, port_base + HOST_CTL); + writew(IDMA_CTL_GO, port_base + PORT_IDMA_CTL); + writeb(0, port_base + PORT_CPB_PTQFIFO); + + return 0; +} + +static void inic_tf_read(struct ata_port *ap, struct ata_taskfile *tf) +{ + void __iomem *port_base = inic_port_base(ap); + + tf->feature = readb(port_base + PORT_TF_FEATURE); + tf->nsect = readb(port_base + PORT_TF_NSECT); + tf->lbal = readb(port_base + PORT_TF_LBAL); + tf->lbam = readb(port_base + PORT_TF_LBAM); + tf->lbah = readb(port_base + PORT_TF_LBAH); + tf->device = readb(port_base + PORT_TF_DEVICE); + tf->command = readb(port_base + PORT_TF_COMMAND); +} - /* Issuing a command to yet uninitialized port locks up the - * controller. Most of the time, this happens for the first - * command after reset which are ATA and ATAPI IDENTIFYs. - * Fast fail if stat is 0x7f or 0xff for those commands. +static bool inic_qc_fill_rtf(struct ata_queued_cmd *qc) +{ + struct ata_taskfile *rtf = &qc->result_tf; + struct ata_taskfile tf; + + /* FIXME: Except for status and error, result TF access + * doesn't work. I tried reading from BAR0/2, CPB and BAR5. + * None works regardless of which command interface is used. + * For now return true iff status indicates device error. + * This means that we're reporting bogus sector for RW + * failures. Eeekk.... */ - if (unlikely(qc->tf.command == ATA_CMD_ID_ATA || - qc->tf.command == ATA_CMD_ID_ATAPI)) { - u8 stat = ap->ops->sff_check_status(ap); - if (stat == 0x7f || stat == 0xff) - return AC_ERR_HSM; - } + inic_tf_read(qc->ap, &tf); - return ata_sff_qc_issue(qc); + if (!(tf.command & ATA_ERR)) + return false; + + rtf->command = tf.command; + rtf->feature = tf.feature; + return true; } static void inic_freeze(struct ata_port *ap) { void __iomem *port_base = inic_port_base(ap); - __inic_set_pirq_mask(ap, PIRQ_MASK_FREEZE); - - ap->ops->sff_check_status(ap); + writeb(PIRQ_MASK_FREEZE, port_base + PORT_IRQ_MASK); writeb(0xff, port_base + PORT_IRQ_STAT); - - readb(port_base + PORT_IRQ_STAT); /* flush */ } static void inic_thaw(struct ata_port *ap) { void __iomem *port_base = inic_port_base(ap); - ap->ops->sff_check_status(ap); writeb(0xff, port_base + PORT_IRQ_STAT); + writeb(PIRQ_MASK_DEFAULT, port_base + PORT_IRQ_MASK); +} - __inic_set_pirq_mask(ap, PIRQ_MASK_OTHER); +static int inic_check_ready(struct ata_link *link) +{ + void __iomem *port_base = inic_port_base(link->ap); - readb(port_base + PORT_IRQ_STAT); /* flush */ + return ata_check_ready(readb(port_base + PORT_TF_COMMAND)); } /* @@ -394,17 +616,15 @@ static int inic_hardreset(struct ata_link *link, unsigned int *class, void __iomem *port_base = inic_port_base(ap); void __iomem *idma_ctl = port_base + PORT_IDMA_CTL; const unsigned long *timing = sata_ehc_deb_timing(&link->eh_context); - u16 val; int rc; /* hammer it into sane state */ inic_reset_port(port_base); - val = readw(idma_ctl); - writew(val | IDMA_CTL_RST_ATA, idma_ctl); + writew(IDMA_CTL_RST_ATA, idma_ctl); readw(idma_ctl); /* flush */ msleep(1); - writew(val & ~IDMA_CTL_RST_ATA, idma_ctl); + writew(0, idma_ctl); rc = sata_link_resume(link, timing, deadline); if (rc) { @@ -418,7 +638,7 @@ static int inic_hardreset(struct ata_link *link, unsigned int *class, struct ata_taskfile tf; /* wait for link to become ready */ - rc = ata_sff_wait_after_reset(link, 1, deadline); + rc = ata_wait_after_reset(link, deadline, inic_check_ready); /* link occupied, -ENODEV too is an error */ if (rc) { ata_link_printk(link, KERN_WARNING, "device not ready " @@ -426,7 +646,7 @@ static int inic_hardreset(struct ata_link *link, unsigned int *class, return rc; } - ata_sff_tf_read(ap, &tf); + inic_tf_read(ap, &tf); *class = ata_dev_classify(&tf); } @@ -436,18 +656,8 @@ static int inic_hardreset(struct ata_link *link, unsigned int *class, static void inic_error_handler(struct ata_port *ap) { void __iomem *port_base = inic_port_base(ap); - struct inic_port_priv *pp = ap->private_data; - unsigned long flags; - /* reset PIO HSM and stop DMA engine */ inic_reset_port(port_base); - - spin_lock_irqsave(ap->lock, flags); - ap->hsm_task_state = HSM_ST_IDLE; - writeb(pp->dfl_prdctl, port_base + PORT_PRD_CTL); - spin_unlock_irqrestore(ap->lock, flags); - - /* PIO and DMA engines have been stopped, perform recovery */ ata_std_error_handler(ap); } @@ -458,26 +668,18 @@ static void inic_post_internal_cmd(struct ata_queued_cmd *qc) inic_reset_port(inic_port_base(qc->ap)); } -static void inic_dev_config(struct ata_device *dev) -{ - /* inic can only handle upto LBA28 max sectors */ - if (dev->max_sectors > ATA_MAX_SECTORS) - dev->max_sectors = ATA_MAX_SECTORS; - - if (dev->n_sectors >= 1 << 28) { - ata_dev_printk(dev, KERN_ERR, - "ERROR: This driver doesn't support LBA48 yet and may cause\n" - " data corruption on such devices. Disabling.\n"); - ata_dev_disable(dev); - } -} - static void init_port(struct ata_port *ap) { void __iomem *port_base = inic_port_base(ap); + struct inic_port_priv *pp = ap->private_data; - /* Setup PRD address */ + /* clear packet and CPB table */ + memset(pp->pkt, 0, sizeof(struct inic_pkt)); + memset(pp->cpb_tbl, 0, IDMA_CPB_TBL_SIZE); + + /* setup PRD and CPB lookup table addresses */ writel(ap->prd_dma, port_base + PORT_PRD_ADDR); + writel(pp->cpb_tbl_dma, port_base + PORT_CPB_CPBLAR); } static int inic_port_resume(struct ata_port *ap) @@ -488,28 +690,30 @@ static int inic_port_resume(struct ata_port *ap) static int inic_port_start(struct ata_port *ap) { - void __iomem *port_base = inic_port_base(ap); + struct device *dev = ap->host->dev; struct inic_port_priv *pp; - u8 tmp; int rc; /* alloc and initialize private data */ - pp = devm_kzalloc(ap->host->dev, sizeof(*pp), GFP_KERNEL); + pp = devm_kzalloc(dev, sizeof(*pp), GFP_KERNEL); if (!pp) return -ENOMEM; ap->private_data = pp; - /* default PRD_CTL value, DMAEN, WR and START off */ - tmp = readb(port_base + PORT_PRD_CTL); - tmp &= ~(PRD_CTL_DMAEN | PRD_CTL_WR | PRD_CTL_START); - pp->dfl_prdctl = tmp; - /* Alloc resources */ rc = ata_port_start(ap); - if (rc) { - kfree(pp); + if (rc) return rc; - } + + pp->pkt = dmam_alloc_coherent(dev, sizeof(struct inic_pkt), + &pp->pkt_dma, GFP_KERNEL); + if (!pp->pkt) + return -ENOMEM; + + pp->cpb_tbl = dmam_alloc_coherent(dev, IDMA_CPB_TBL_SIZE, + &pp->cpb_tbl_dma, GFP_KERNEL); + if (!pp->cpb_tbl) + return -ENOMEM; init_port(ap); @@ -517,21 +721,18 @@ static int inic_port_start(struct ata_port *ap) } static struct ata_port_operations inic_port_ops = { - .inherits = &ata_sff_port_ops, + .inherits = &sata_port_ops, - .bmdma_setup = inic_bmdma_setup, - .bmdma_start = inic_bmdma_start, - .bmdma_stop = inic_bmdma_stop, - .bmdma_status = inic_bmdma_status, + .check_atapi_dma = inic_check_atapi_dma, + .qc_prep = inic_qc_prep, .qc_issue = inic_qc_issue, + .qc_fill_rtf = inic_qc_fill_rtf, .freeze = inic_freeze, .thaw = inic_thaw, - .softreset = ATA_OP_NULL, /* softreset is broken */ .hardreset = inic_hardreset, .error_handler = inic_error_handler, .post_internal_cmd = inic_post_internal_cmd, - .dev_config = inic_dev_config, .scr_read = inic_scr_read, .scr_write = inic_scr_write, @@ -541,12 +742,6 @@ static struct ata_port_operations inic_port_ops = { }; static struct ata_port_info inic_port_info = { - /* For some reason, ATAPI_PROT_PIO is broken on this - * controller, and no, PIO_POLLING does't fix it. It somehow - * manages to report the wrong ireason and ignoring ireason - * results in machine lock up. Tell libata to always prefer - * DMA. - */ .flags = ATA_FLAG_SATA | ATA_FLAG_PIO_DMA, .pio_mask = 0x1f, /* pio0-4 */ .mwdma_mask = 0x07, /* mwdma0-2 */ @@ -599,7 +794,6 @@ static int inic_pci_device_resume(struct pci_dev *pdev) { struct ata_host *host = dev_get_drvdata(&pdev->dev); struct inic_host_priv *hpriv = host->private_data; - void __iomem *mmio_base = host->iomap[MMIO_BAR]; int rc; rc = ata_pci_device_do_resume(pdev); @@ -607,7 +801,7 @@ static int inic_pci_device_resume(struct pci_dev *pdev) return rc; if (pdev->dev.power.power_state.event == PM_EVENT_SUSPEND) { - rc = init_controller(mmio_base, hpriv->cached_hctl); + rc = init_controller(hpriv->mmio_base, hpriv->cached_hctl); if (rc) return rc; } @@ -625,6 +819,7 @@ static int inic_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) struct ata_host *host; struct inic_host_priv *hpriv; void __iomem * const *iomap; + int mmio_bar; int i, rc; if (!printed_version++) @@ -638,38 +833,31 @@ static int inic_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) host->private_data = hpriv; - /* acquire resources and fill host */ + /* Acquire resources and fill host. Note that PCI and cardbus + * use different BARs. + */ rc = pcim_enable_device(pdev); if (rc) return rc; - rc = pcim_iomap_regions(pdev, 0x3f, DRV_NAME); + if (pci_resource_flags(pdev, MMIO_BAR_PCI) & IORESOURCE_MEM) + mmio_bar = MMIO_BAR_PCI; + else + mmio_bar = MMIO_BAR_CARDBUS; + + rc = pcim_iomap_regions(pdev, 1 << mmio_bar, DRV_NAME); if (rc) return rc; host->iomap = iomap = pcim_iomap_table(pdev); + hpriv->mmio_base = iomap[mmio_bar]; + hpriv->cached_hctl = readw(hpriv->mmio_base + HOST_CTL); for (i = 0; i < NR_PORTS; i++) { struct ata_port *ap = host->ports[i]; - struct ata_ioports *port = &ap->ioaddr; - unsigned int offset = i * PORT_SIZE; - - port->cmd_addr = iomap[2 * i]; - port->altstatus_addr = - port->ctl_addr = (void __iomem *) - ((unsigned long)iomap[2 * i + 1] | ATA_PCI_CTL_OFS); - port->scr_addr = iomap[MMIO_BAR] + offset + PORT_SCR; - - ata_sff_std_ports(port); - - ata_port_pbar_desc(ap, MMIO_BAR, -1, "mmio"); - ata_port_pbar_desc(ap, MMIO_BAR, offset, "port"); - ata_port_desc(ap, "cmd 0x%llx ctl 0x%llx", - (unsigned long long)pci_resource_start(pdev, 2 * i), - (unsigned long long)pci_resource_start(pdev, (2 * i + 1)) | - ATA_PCI_CTL_OFS); - } - hpriv->cached_hctl = readw(iomap[MMIO_BAR] + HOST_CTL); + ata_port_pbar_desc(ap, mmio_bar, -1, "mmio"); + ata_port_pbar_desc(ap, mmio_bar, i * PORT_SIZE, "port"); + } /* Set dma_mask. This devices doesn't support 64bit addressing. */ rc = pci_set_dma_mask(pdev, DMA_32BIT_MASK); @@ -698,7 +886,7 @@ static int inic_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) return rc; } - rc = init_controller(iomap[MMIO_BAR], hpriv->cached_hctl); + rc = init_controller(hpriv->mmio_base, hpriv->cached_hctl); if (rc) { dev_printk(KERN_ERR, &pdev->dev, "failed to initialize controller\n"); diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c index 842b1a15b78..bb73b222262 100644 --- a/drivers/ata/sata_mv.c +++ b/drivers/ata/sata_mv.c @@ -65,6 +65,7 @@ #include <linux/platform_device.h> #include <linux/ata_platform.h> #include <linux/mbus.h> +#include <linux/bitops.h> #include <scsi/scsi_host.h> #include <scsi/scsi_cmnd.h> #include <scsi/scsi_device.h> @@ -91,9 +92,9 @@ enum { MV_IRQ_COAL_TIME_THRESHOLD = (MV_IRQ_COAL_REG_BASE + 0xd0), MV_SATAHC0_REG_BASE = 0x20000, - MV_FLASH_CTL = 0x1046c, - MV_GPIO_PORT_CTL = 0x104f0, - MV_RESET_CFG = 0x180d8, + MV_FLASH_CTL_OFS = 0x1046c, + MV_GPIO_PORT_CTL_OFS = 0x104f0, + MV_RESET_CFG_OFS = 0x180d8, MV_PCI_REG_SZ = MV_MAJOR_REG_AREA_SZ, MV_SATAHC_REG_SZ = MV_MAJOR_REG_AREA_SZ, @@ -147,18 +148,21 @@ enum { /* PCI interface registers */ PCI_COMMAND_OFS = 0xc00, + PCI_COMMAND_MRDTRIG = (1 << 7), /* PCI Master Read Trigger */ PCI_MAIN_CMD_STS_OFS = 0xd30, STOP_PCI_MASTER = (1 << 2), PCI_MASTER_EMPTY = (1 << 3), GLOB_SFT_RST = (1 << 4), - MV_PCI_MODE = 0xd00, + MV_PCI_MODE_OFS = 0xd00, + MV_PCI_MODE_MASK = 0x30, + MV_PCI_EXP_ROM_BAR_CTL = 0xd2c, MV_PCI_DISC_TIMER = 0xd04, MV_PCI_MSI_TRIGGER = 0xc38, MV_PCI_SERR_MASK = 0xc28, - MV_PCI_XBAR_TMOUT = 0x1d04, + MV_PCI_XBAR_TMOUT_OFS = 0x1d04, MV_PCI_ERR_LOW_ADDRESS = 0x1d40, MV_PCI_ERR_HIGH_ADDRESS = 0x1d44, MV_PCI_ERR_ATTRIBUTE = 0x1d48, @@ -225,16 +229,18 @@ enum { PHY_MODE4 = 0x314, PHY_MODE2 = 0x330, SATA_IFCTL_OFS = 0x344, + SATA_TESTCTL_OFS = 0x348, SATA_IFSTAT_OFS = 0x34c, VENDOR_UNIQUE_FIS_OFS = 0x35c, - FIS_CFG_OFS = 0x360, - FIS_CFG_SINGLE_SYNC = (1 << 16), /* SYNC on DMA activation */ + FISCFG_OFS = 0x360, + FISCFG_WAIT_DEV_ERR = (1 << 8), /* wait for host on DevErr */ + FISCFG_SINGLE_SYNC = (1 << 16), /* SYNC on DMA activation */ MV5_PHY_MODE = 0x74, - MV5_LT_MODE = 0x30, - MV5_PHY_CTL = 0x0C, - SATA_INTERFACE_CFG = 0x050, + MV5_LTMODE_OFS = 0x30, + MV5_PHY_CTL_OFS = 0x0C, + SATA_INTERFACE_CFG_OFS = 0x050, MV_M2_PREAMP_MASK = 0x7e0, @@ -332,10 +338,16 @@ enum { EDMA_CMD_OFS = 0x28, /* EDMA command register */ EDMA_EN = (1 << 0), /* enable EDMA */ EDMA_DS = (1 << 1), /* disable EDMA; self-negated */ - ATA_RST = (1 << 2), /* reset trans/link/phy */ + EDMA_RESET = (1 << 2), /* reset eng/trans/link/phy */ + + EDMA_STATUS_OFS = 0x30, /* EDMA engine status */ + EDMA_STATUS_CACHE_EMPTY = (1 << 6), /* GenIIe command cache empty */ + EDMA_STATUS_IDLE = (1 << 7), /* GenIIe EDMA enabled/idle */ - EDMA_IORDY_TMOUT = 0x34, - EDMA_ARB_CFG = 0x38, + EDMA_IORDY_TMOUT_OFS = 0x34, + EDMA_ARB_CFG_OFS = 0x38, + + EDMA_HALTCOND_OFS = 0x60, /* GenIIe halt conditions */ GEN_II_NCQ_MAX_SECTORS = 256, /* max sects/io on Gen2 w/NCQ */ @@ -350,15 +362,19 @@ enum { MV_HP_GEN_II = (1 << 7), /* Generation II: 60xx */ MV_HP_GEN_IIE = (1 << 8), /* Generation IIE: 6042/7042 */ MV_HP_PCIE = (1 << 9), /* PCIe bus/regs: 7042 */ + MV_HP_CUT_THROUGH = (1 << 10), /* can use EDMA cut-through */ /* Port private flags (pp_flags) */ MV_PP_FLAG_EDMA_EN = (1 << 0), /* is EDMA engine enabled? */ MV_PP_FLAG_NCQ_EN = (1 << 1), /* is EDMA set up for NCQ? */ + MV_PP_FLAG_FBS_EN = (1 << 2), /* is EDMA set up for FBS? */ + MV_PP_FLAG_DELAYED_EH = (1 << 3), /* delayed dev err handling */ }; #define IS_GEN_I(hpriv) ((hpriv)->hp_flags & MV_HP_GEN_I) #define IS_GEN_II(hpriv) ((hpriv)->hp_flags & MV_HP_GEN_II) #define IS_GEN_IIE(hpriv) ((hpriv)->hp_flags & MV_HP_GEN_IIE) +#define IS_PCIE(hpriv) ((hpriv)->hp_flags & MV_HP_PCIE) #define HAS_PCI(host) (!((host)->ports[0]->flags & MV_FLAG_SOC)) #define WINDOW_CTRL(i) (0x20030 + ((i) << 4)) @@ -433,6 +449,7 @@ struct mv_port_priv { unsigned int resp_idx; u32 pp_flags; + unsigned int delayed_eh_pmp_map; }; struct mv_port_signal { @@ -479,6 +496,7 @@ static int mv5_scr_read(struct ata_port *ap, unsigned int sc_reg_in, u32 *val); static int mv5_scr_write(struct ata_port *ap, unsigned int sc_reg_in, u32 val); static int mv_port_start(struct ata_port *ap); static void mv_port_stop(struct ata_port *ap); +static int mv_qc_defer(struct ata_queued_cmd *qc); static void mv_qc_prep(struct ata_queued_cmd *qc); static void mv_qc_prep_iie(struct ata_queued_cmd *qc); static unsigned int mv_qc_issue(struct ata_queued_cmd *qc); @@ -527,6 +545,9 @@ static int mv_pmp_hardreset(struct ata_link *link, unsigned int *class, unsigned long deadline); static int mv_softreset(struct ata_link *link, unsigned int *class, unsigned long deadline); +static void mv_pmp_error_handler(struct ata_port *ap); +static void mv_process_crpb_entries(struct ata_port *ap, + struct mv_port_priv *pp); /* .sg_tablesize is (MV_MAX_SG_CT / 2) in the structures below * because we have to allow room for worst case splitting of @@ -548,6 +569,7 @@ static struct scsi_host_template mv6_sht = { static struct ata_port_operations mv5_ops = { .inherits = &ata_sff_port_ops, + .qc_defer = mv_qc_defer, .qc_prep = mv_qc_prep, .qc_issue = mv_qc_issue, @@ -566,7 +588,6 @@ static struct ata_port_operations mv5_ops = { static struct ata_port_operations mv6_ops = { .inherits = &mv5_ops, - .qc_defer = sata_pmp_qc_defer_cmd_switch, .dev_config = mv6_dev_config, .scr_read = mv_scr_read, .scr_write = mv_scr_write, @@ -574,12 +595,11 @@ static struct ata_port_operations mv6_ops = { .pmp_hardreset = mv_pmp_hardreset, .pmp_softreset = mv_softreset, .softreset = mv_softreset, - .error_handler = sata_pmp_error_handler, + .error_handler = mv_pmp_error_handler, }; static struct ata_port_operations mv_iie_ops = { .inherits = &mv6_ops, - .qc_defer = ata_std_qc_defer, /* FIS-based switching */ .dev_config = ATA_OP_NULL, .qc_prep = mv_qc_prep_iie, }; @@ -875,6 +895,29 @@ static void mv_start_dma(struct ata_port *ap, void __iomem *port_mmio, } } +static void mv_wait_for_edma_empty_idle(struct ata_port *ap) +{ + void __iomem *port_mmio = mv_ap_base(ap); + const u32 empty_idle = (EDMA_STATUS_CACHE_EMPTY | EDMA_STATUS_IDLE); + const int per_loop = 5, timeout = (15 * 1000 / per_loop); + int i; + + /* + * Wait for the EDMA engine to finish transactions in progress. + * No idea what a good "timeout" value might be, but measurements + * indicate that it often requires hundreds of microseconds + * with two drives in-use. So we use the 15msec value above + * as a rough guess at what even more drives might require. + */ + for (i = 0; i < timeout; ++i) { + u32 edma_stat = readl(port_mmio + EDMA_STATUS_OFS); + if ((edma_stat & empty_idle) == empty_idle) + break; + udelay(per_loop); + } + /* ata_port_printk(ap, KERN_INFO, "%s: %u+ usecs\n", __func__, i); */ +} + /** * mv_stop_edma_engine - Disable eDMA engine * @port_mmio: io base address @@ -907,6 +950,7 @@ static int mv_stop_edma(struct ata_port *ap) if (!(pp->pp_flags & MV_PP_FLAG_EDMA_EN)) return 0; pp->pp_flags &= ~MV_PP_FLAG_EDMA_EN; + mv_wait_for_edma_empty_idle(ap); if (mv_stop_edma_engine(port_mmio)) { ata_port_printk(ap, KERN_ERR, "Unable to stop eDMA\n"); return -EIO; @@ -1057,26 +1101,95 @@ static void mv6_dev_config(struct ata_device *adev) } } -static void mv_config_fbs(void __iomem *port_mmio, int enable_fbs) +static int mv_qc_defer(struct ata_queued_cmd *qc) { - u32 old_fcfg, new_fcfg, old_ltmode, new_ltmode; + struct ata_link *link = qc->dev->link; + struct ata_port *ap = link->ap; + struct mv_port_priv *pp = ap->private_data; + + /* + * Don't allow new commands if we're in a delayed EH state + * for NCQ and/or FIS-based switching. + */ + if (pp->pp_flags & MV_PP_FLAG_DELAYED_EH) + return ATA_DEFER_PORT; /* - * Various bit settings required for operation - * in FIS-based switching (fbs) mode on GenIIe: + * If the port is completely idle, then allow the new qc. */ - old_fcfg = readl(port_mmio + FIS_CFG_OFS); - old_ltmode = readl(port_mmio + LTMODE_OFS); - if (enable_fbs) { - new_fcfg = old_fcfg | FIS_CFG_SINGLE_SYNC; - new_ltmode = old_ltmode | LTMODE_BIT8; - } else { /* disable fbs */ - new_fcfg = old_fcfg & ~FIS_CFG_SINGLE_SYNC; - new_ltmode = old_ltmode & ~LTMODE_BIT8; - } - if (new_fcfg != old_fcfg) - writelfl(new_fcfg, port_mmio + FIS_CFG_OFS); + if (ap->nr_active_links == 0) + return 0; + + if (pp->pp_flags & MV_PP_FLAG_EDMA_EN) { + /* + * The port is operating in host queuing mode (EDMA). + * It can accomodate a new qc if the qc protocol + * is compatible with the current host queue mode. + */ + if (pp->pp_flags & MV_PP_FLAG_NCQ_EN) { + /* + * The host queue (EDMA) is in NCQ mode. + * If the new qc is also an NCQ command, + * then allow the new qc. + */ + if (qc->tf.protocol == ATA_PROT_NCQ) + return 0; + } else { + /* + * The host queue (EDMA) is in non-NCQ, DMA mode. + * If the new qc is also a non-NCQ, DMA command, + * then allow the new qc. + */ + if (qc->tf.protocol == ATA_PROT_DMA) + return 0; + } + } + return ATA_DEFER_PORT; +} + +static void mv_config_fbs(void __iomem *port_mmio, int want_ncq, int want_fbs) +{ + u32 new_fiscfg, old_fiscfg; + u32 new_ltmode, old_ltmode; + u32 new_haltcond, old_haltcond; + + old_fiscfg = readl(port_mmio + FISCFG_OFS); + old_ltmode = readl(port_mmio + LTMODE_OFS); + old_haltcond = readl(port_mmio + EDMA_HALTCOND_OFS); + + new_fiscfg = old_fiscfg & ~(FISCFG_SINGLE_SYNC | FISCFG_WAIT_DEV_ERR); + new_ltmode = old_ltmode & ~LTMODE_BIT8; + new_haltcond = old_haltcond | EDMA_ERR_DEV; + + if (want_fbs) { + new_fiscfg = old_fiscfg | FISCFG_SINGLE_SYNC; + new_ltmode = old_ltmode | LTMODE_BIT8; + if (want_ncq) + new_haltcond &= ~EDMA_ERR_DEV; + else + new_fiscfg |= FISCFG_WAIT_DEV_ERR; + } + + if (new_fiscfg != old_fiscfg) + writelfl(new_fiscfg, port_mmio + FISCFG_OFS); if (new_ltmode != old_ltmode) writelfl(new_ltmode, port_mmio + LTMODE_OFS); + if (new_haltcond != old_haltcond) + writelfl(new_haltcond, port_mmio + EDMA_HALTCOND_OFS); +} + +static void mv_60x1_errata_sata25(struct ata_port *ap, int want_ncq) +{ + struct mv_host_priv *hpriv = ap->host->private_data; + u32 old, new; + + /* workaround for 88SX60x1 FEr SATA#25 (part 1) */ + old = readl(hpriv->base + MV_GPIO_PORT_CTL_OFS); + if (want_ncq) + new = old | (1 << 22); + else + new = old & ~(1 << 22); + if (new != old) + writel(new, hpriv->base + MV_GPIO_PORT_CTL_OFS); } static void mv_edma_cfg(struct ata_port *ap, int want_ncq) @@ -1088,25 +1201,40 @@ static void mv_edma_cfg(struct ata_port *ap, int want_ncq) /* set up non-NCQ EDMA configuration */ cfg = EDMA_CFG_Q_DEPTH; /* always 0x1f for *all* chips */ + pp->pp_flags &= ~MV_PP_FLAG_FBS_EN; if (IS_GEN_I(hpriv)) cfg |= (1 << 8); /* enab config burst size mask */ - else if (IS_GEN_II(hpriv)) + else if (IS_GEN_II(hpriv)) { cfg |= EDMA_CFG_RD_BRST_EXT | EDMA_CFG_WR_BUFF_LEN; + mv_60x1_errata_sata25(ap, want_ncq); - else if (IS_GEN_IIE(hpriv)) { - cfg |= (1 << 23); /* do not mask PM field in rx'd FIS */ - cfg |= (1 << 22); /* enab 4-entry host queue cache */ - cfg |= (1 << 18); /* enab early completion */ - cfg |= (1 << 17); /* enab cut-through (dis stor&forwrd) */ + } else if (IS_GEN_IIE(hpriv)) { + int want_fbs = sata_pmp_attached(ap); + /* + * Possible future enhancement: + * + * The chip can use FBS with non-NCQ, if we allow it, + * But first we need to have the error handling in place + * for this mode (datasheet section 7.3.15.4.2.3). + * So disallow non-NCQ FBS for now. + */ + want_fbs &= want_ncq; + + mv_config_fbs(port_mmio, want_ncq, want_fbs); - if (want_ncq && sata_pmp_attached(ap)) { + if (want_fbs) { + pp->pp_flags |= MV_PP_FLAG_FBS_EN; cfg |= EDMA_CFG_EDMA_FBS; /* FIS-based switching */ - mv_config_fbs(port_mmio, 1); - } else { - mv_config_fbs(port_mmio, 0); } + + cfg |= (1 << 23); /* do not mask PM field in rx'd FIS */ + cfg |= (1 << 22); /* enab 4-entry host queue cache */ + if (HAS_PCI(ap->host)) + cfg |= (1 << 18); /* enab early completion */ + if (hpriv->hp_flags & MV_HP_CUT_THROUGH) + cfg |= (1 << 17); /* enab cut-thru (dis stor&forwrd) */ } if (want_ncq) { @@ -1483,25 +1611,186 @@ static struct ata_queued_cmd *mv_get_active_qc(struct ata_port *ap) return qc; } -static void mv_unexpected_intr(struct ata_port *ap) +static void mv_pmp_error_handler(struct ata_port *ap) { + unsigned int pmp, pmp_map; struct mv_port_priv *pp = ap->private_data; - struct ata_eh_info *ehi = &ap->link.eh_info; - char *when = ""; + if (pp->pp_flags & MV_PP_FLAG_DELAYED_EH) { + /* + * Perform NCQ error analysis on failed PMPs + * before we freeze the port entirely. + * + * The failed PMPs are marked earlier by mv_pmp_eh_prep(). + */ + pmp_map = pp->delayed_eh_pmp_map; + pp->pp_flags &= ~MV_PP_FLAG_DELAYED_EH; + for (pmp = 0; pmp_map != 0; pmp++) { + unsigned int this_pmp = (1 << pmp); + if (pmp_map & this_pmp) { + struct ata_link *link = &ap->pmp_link[pmp]; + pmp_map &= ~this_pmp; + ata_eh_analyze_ncq_error(link); + } + } + ata_port_freeze(ap); + } + sata_pmp_error_handler(ap); +} + +static unsigned int mv_get_err_pmp_map(struct ata_port *ap) +{ + void __iomem *port_mmio = mv_ap_base(ap); + + return readl(port_mmio + SATA_TESTCTL_OFS) >> 16; +} + +static void mv_pmp_eh_prep(struct ata_port *ap, unsigned int pmp_map) +{ + struct ata_eh_info *ehi; + unsigned int pmp; + + /* + * Initialize EH info for PMPs which saw device errors + */ + ehi = &ap->link.eh_info; + for (pmp = 0; pmp_map != 0; pmp++) { + unsigned int this_pmp = (1 << pmp); + if (pmp_map & this_pmp) { + struct ata_link *link = &ap->pmp_link[pmp]; + + pmp_map &= ~this_pmp; + ehi = &link->eh_info; + ata_ehi_clear_desc(ehi); + ata_ehi_push_desc(ehi, "dev err"); + ehi->err_mask |= AC_ERR_DEV; + ehi->action |= ATA_EH_RESET; + ata_link_abort(link); + } + } +} + +static int mv_handle_fbs_ncq_dev_err(struct ata_port *ap) +{ + struct mv_port_priv *pp = ap->private_data; + int failed_links; + unsigned int old_map, new_map; + + /* + * Device error during FBS+NCQ operation: + * + * Set a port flag to prevent further I/O being enqueued. + * Leave the EDMA running to drain outstanding commands from this port. + * Perform the post-mortem/EH only when all responses are complete. + * Follow recovery sequence from 6042/7042 datasheet (7.3.15.4.2.2). + */ + if (!(pp->pp_flags & MV_PP_FLAG_DELAYED_EH)) { + pp->pp_flags |= MV_PP_FLAG_DELAYED_EH; + pp->delayed_eh_pmp_map = 0; + } + old_map = pp->delayed_eh_pmp_map; + new_map = old_map | mv_get_err_pmp_map(ap); + + if (old_map != new_map) { + pp->delayed_eh_pmp_map = new_map; + mv_pmp_eh_prep(ap, new_map & ~old_map); + } + failed_links = hweight16(new_map); + + ata_port_printk(ap, KERN_INFO, "%s: pmp_map=%04x qc_map=%04x " + "failed_links=%d nr_active_links=%d\n", + __func__, pp->delayed_eh_pmp_map, + ap->qc_active, failed_links, + ap->nr_active_links); + + if (ap->nr_active_links <= failed_links) { + mv_process_crpb_entries(ap, pp); + mv_stop_edma(ap); + mv_eh_freeze(ap); + ata_port_printk(ap, KERN_INFO, "%s: done\n", __func__); + return 1; /* handled */ + } + ata_port_printk(ap, KERN_INFO, "%s: waiting\n", __func__); + return 1; /* handled */ +} + +static int mv_handle_fbs_non_ncq_dev_err(struct ata_port *ap) +{ /* - * We got a device interrupt from something that - * was supposed to be using EDMA or polling. + * Possible future enhancement: + * + * FBS+non-NCQ operation is not yet implemented. + * See related notes in mv_edma_cfg(). + * + * Device error during FBS+non-NCQ operation: + * + * We need to snapshot the shadow registers for each failed command. + * Follow recovery sequence from 6042/7042 datasheet (7.3.15.4.2.3). */ + return 0; /* not handled */ +} + +static int mv_handle_dev_err(struct ata_port *ap, u32 edma_err_cause) +{ + struct mv_port_priv *pp = ap->private_data; + + if (!(pp->pp_flags & MV_PP_FLAG_EDMA_EN)) + return 0; /* EDMA was not active: not handled */ + if (!(pp->pp_flags & MV_PP_FLAG_FBS_EN)) + return 0; /* FBS was not active: not handled */ + + if (!(edma_err_cause & EDMA_ERR_DEV)) + return 0; /* non DEV error: not handled */ + edma_err_cause &= ~EDMA_ERR_IRQ_TRANSIENT; + if (edma_err_cause & ~(EDMA_ERR_DEV | EDMA_ERR_SELF_DIS)) + return 0; /* other problems: not handled */ + + if (pp->pp_flags & MV_PP_FLAG_NCQ_EN) { + /* + * EDMA should NOT have self-disabled for this case. + * If it did, then something is wrong elsewhere, + * and we cannot handle it here. + */ + if (edma_err_cause & EDMA_ERR_SELF_DIS) { + ata_port_printk(ap, KERN_WARNING, + "%s: err_cause=0x%x pp_flags=0x%x\n", + __func__, edma_err_cause, pp->pp_flags); + return 0; /* not handled */ + } + return mv_handle_fbs_ncq_dev_err(ap); + } else { + /* + * EDMA should have self-disabled for this case. + * If it did not, then something is wrong elsewhere, + * and we cannot handle it here. + */ + if (!(edma_err_cause & EDMA_ERR_SELF_DIS)) { + ata_port_printk(ap, KERN_WARNING, + "%s: err_cause=0x%x pp_flags=0x%x\n", + __func__, edma_err_cause, pp->pp_flags); + return 0; /* not handled */ + } + return mv_handle_fbs_non_ncq_dev_err(ap); + } + return 0; /* not handled */ +} + +static void mv_unexpected_intr(struct ata_port *ap, int edma_was_enabled) +{ + struct ata_eh_info *ehi = &ap->link.eh_info; + char *when = "idle"; + ata_ehi_clear_desc(ehi); - if (pp->pp_flags & MV_PP_FLAG_EDMA_EN) { - when = " while EDMA enabled"; + if (!ap || (ap->flags & ATA_FLAG_DISABLED)) { + when = "disabled"; + } else if (edma_was_enabled) { + when = "EDMA enabled"; } else { struct ata_queued_cmd *qc = ata_qc_from_tag(ap, ap->link.active_tag); if (qc && (qc->tf.flags & ATA_TFLAG_POLLING)) - when = " while polling"; + when = "polling"; } - ata_ehi_push_desc(ehi, "unexpected device interrupt%s", when); + ata_ehi_push_desc(ehi, "unexpected device interrupt while %s", when); ehi->err_mask |= AC_ERR_OTHER; ehi->action |= ATA_EH_RESET; ata_port_freeze(ap); @@ -1519,7 +1808,7 @@ static void mv_unexpected_intr(struct ata_port *ap) * LOCKING: * Inherited from caller. */ -static void mv_err_intr(struct ata_port *ap, struct ata_queued_cmd *qc) +static void mv_err_intr(struct ata_port *ap) { void __iomem *port_mmio = mv_ap_base(ap); u32 edma_err_cause, eh_freeze_mask, serr = 0; @@ -1527,24 +1816,42 @@ static void mv_err_intr(struct ata_port *ap, struct ata_queued_cmd *qc) struct mv_host_priv *hpriv = ap->host->private_data; unsigned int action = 0, err_mask = 0; struct ata_eh_info *ehi = &ap->link.eh_info; - - ata_ehi_clear_desc(ehi); + struct ata_queued_cmd *qc; + int abort = 0; /* - * Read and clear the err_cause bits. This won't actually - * clear for some errors (eg. SError), but we will be doing - * a hard reset in those cases regardless, which *will* clear it. + * Read and clear the SError and err_cause bits. */ + sata_scr_read(&ap->link, SCR_ERROR, &serr); + sata_scr_write_flush(&ap->link, SCR_ERROR, serr); + edma_err_cause = readl(port_mmio + EDMA_ERR_IRQ_CAUSE_OFS); writelfl(~edma_err_cause, port_mmio + EDMA_ERR_IRQ_CAUSE_OFS); - ata_ehi_push_desc(ehi, "edma_err_cause=%08x", edma_err_cause); + ata_port_printk(ap, KERN_INFO, "%s: err_cause=%08x pp_flags=0x%x\n", + __func__, edma_err_cause, pp->pp_flags); + + if (edma_err_cause & EDMA_ERR_DEV) { + /* + * Device errors during FIS-based switching operation + * require special handling. + */ + if (mv_handle_dev_err(ap, edma_err_cause)) + return; + } + qc = mv_get_active_qc(ap); + ata_ehi_clear_desc(ehi); + ata_ehi_push_desc(ehi, "edma_err_cause=%08x pp_flags=%08x", + edma_err_cause, pp->pp_flags); /* * All generations share these EDMA error cause bits: */ - if (edma_err_cause & EDMA_ERR_DEV) + if (edma_err_cause & EDMA_ERR_DEV) { err_mask |= AC_ERR_DEV; + action |= ATA_EH_RESET; + ata_ehi_push_desc(ehi, "dev error"); + } if (edma_err_cause & (EDMA_ERR_D_PAR | EDMA_ERR_PRD_PAR | EDMA_ERR_CRQB_PAR | EDMA_ERR_CRPB_PAR | EDMA_ERR_INTRL_PAR)) { @@ -1576,13 +1883,6 @@ static void mv_err_intr(struct ata_port *ap, struct ata_queued_cmd *qc) ata_ehi_push_desc(ehi, "EDMA self-disable"); } if (edma_err_cause & EDMA_ERR_SERR) { - /* - * Ensure that we read our own SCR, not a pmp link SCR: - */ - ap->ops->scr_read(ap, SCR_ERROR, &serr); - /* - * Don't clear SError here; leave it for libata-eh: - */ ata_ehi_push_desc(ehi, "SError=%08x", serr); err_mask |= AC_ERR_ATA_BUS; action |= ATA_EH_RESET; @@ -1602,10 +1902,29 @@ static void mv_err_intr(struct ata_port *ap, struct ata_queued_cmd *qc) else ehi->err_mask |= err_mask; - if (edma_err_cause & eh_freeze_mask) + if (err_mask == AC_ERR_DEV) { + /* + * Cannot do ata_port_freeze() here, + * because it would kill PIO access, + * which is needed for further diagnosis. + */ + mv_eh_freeze(ap); + abort = 1; + } else if (edma_err_cause & eh_freeze_mask) { + /* + * Note to self: ata_port_freeze() calls ata_port_abort() + */ ata_port_freeze(ap); - else - ata_port_abort(ap); + } else { + abort = 1; + } + + if (abort) { + if (qc) + ata_link_abort(qc->dev->link); + else + ata_port_abort(ap); + } } static void mv_process_crpb_response(struct ata_port *ap, @@ -1632,8 +1951,9 @@ static void mv_process_crpb_response(struct ata_port *ap, } } ata_status = edma_status >> CRPB_FLAG_STATUS_SHIFT; - qc->err_mask |= ac_err_mask(ata_status); - ata_qc_complete(qc); + if (!ac_err_mask(ata_status)) + ata_qc_complete(qc); + /* else: leave it for mv_err_intr() */ } else { ata_port_printk(ap, KERN_ERR, "%s: no qc for tag=%d\n", __func__, tag); @@ -1677,6 +1997,44 @@ static void mv_process_crpb_entries(struct ata_port *ap, struct mv_port_priv *pp port_mmio + EDMA_RSP_Q_OUT_PTR_OFS); } +static void mv_port_intr(struct ata_port *ap, u32 port_cause) +{ + struct mv_port_priv *pp; + int edma_was_enabled; + + if (!ap || (ap->flags & ATA_FLAG_DISABLED)) { + mv_unexpected_intr(ap, 0); + return; + } + /* + * Grab a snapshot of the EDMA_EN flag setting, + * so that we have a consistent view for this port, + * even if something we call of our routines changes it. + */ + pp = ap->private_data; + edma_was_enabled = (pp->pp_flags & MV_PP_FLAG_EDMA_EN); + /* + * Process completed CRPB response(s) before other events. + */ + if (edma_was_enabled && (port_cause & DONE_IRQ)) { + mv_process_crpb_entries(ap, pp); + if (pp->pp_flags & MV_PP_FLAG_DELAYED_EH) + mv_handle_fbs_ncq_dev_err(ap); + } + /* + * Handle chip-reported errors, or continue on to handle PIO. + */ + if (unlikely(port_cause & ERR_IRQ)) { + mv_err_intr(ap); + } else if (!edma_was_enabled) { + struct ata_queued_cmd *qc = mv_get_active_qc(ap); + if (qc) + ata_sff_host_intr(ap, qc); + else + mv_unexpected_intr(ap, edma_was_enabled); + } +} + /** * mv_host_intr - Handle all interrupts on the given host controller * @host: host specific structure @@ -1688,66 +2046,58 @@ static void mv_process_crpb_entries(struct ata_port *ap, struct mv_port_priv *pp static int mv_host_intr(struct ata_host *host, u32 main_irq_cause) { struct mv_host_priv *hpriv = host->private_data; - void __iomem *mmio = hpriv->base, *hc_mmio = NULL; - u32 hc_irq_cause = 0; + void __iomem *mmio = hpriv->base, *hc_mmio; unsigned int handled = 0, port; for (port = 0; port < hpriv->n_ports; port++) { struct ata_port *ap = host->ports[port]; - struct mv_port_priv *pp; - unsigned int shift, hardport, port_cause; - /* - * When we move to the second hc, flag our cached - * copies of hc_mmio (and hc_irq_cause) as invalid again. - */ - if (port == MV_PORTS_PER_HC) - hc_mmio = NULL; - /* - * Do nothing if port is not interrupting or is disabled: - */ + unsigned int p, shift, hardport, port_cause; + MV_PORT_TO_SHIFT_AND_HARDPORT(port, shift, hardport); - port_cause = (main_irq_cause >> shift) & (DONE_IRQ | ERR_IRQ); - if (!port_cause || !ap || (ap->flags & ATA_FLAG_DISABLED)) - continue; /* - * Each hc within the host has its own hc_irq_cause register. - * We defer reading it until we know we need it, right now: - * - * FIXME later: we don't really need to read this register - * (some logic changes required below if we go that way), - * because it doesn't tell us anything new. But we do need - * to write to it, outside the top of this loop, - * to reset the interrupt triggers for next time. + * Each hc within the host has its own hc_irq_cause register, + * where the interrupting ports bits get ack'd. */ - if (!hc_mmio) { + if (hardport == 0) { /* first port on this hc ? */ + u32 hc_cause = (main_irq_cause >> shift) & HC0_IRQ_PEND; + u32 port_mask, ack_irqs; + /* + * Skip this entire hc if nothing pending for any ports + */ + if (!hc_cause) { + port += MV_PORTS_PER_HC - 1; + continue; + } + /* + * We don't need/want to read the hc_irq_cause register, + * because doing so hurts performance, and + * main_irq_cause already gives us everything we need. + * + * But we do have to *write* to the hc_irq_cause to ack + * the ports that we are handling this time through. + * + * This requires that we create a bitmap for those + * ports which interrupted us, and use that bitmap + * to ack (only) those ports via hc_irq_cause. + */ + ack_irqs = 0; + for (p = 0; p < MV_PORTS_PER_HC; ++p) { + if ((port + p) >= hpriv->n_ports) + break; + port_mask = (DONE_IRQ | ERR_IRQ) << (p * 2); + if (hc_cause & port_mask) + ack_irqs |= (DMA_IRQ | DEV_IRQ) << p; + } hc_mmio = mv_hc_base_from_port(mmio, port); - hc_irq_cause = readl(hc_mmio + HC_IRQ_CAUSE_OFS); - writelfl(~hc_irq_cause, hc_mmio + HC_IRQ_CAUSE_OFS); + writelfl(~ack_irqs, hc_mmio + HC_IRQ_CAUSE_OFS); handled = 1; } /* - * Process completed CRPB response(s) before other events. - */ - pp = ap->private_data; - if (hc_irq_cause & (DMA_IRQ << hardport)) { - if (pp->pp_flags & MV_PP_FLAG_EDMA_EN) - mv_process_crpb_entries(ap, pp); - } - /* - * Handle chip-reported errors, or continue on to handle PIO. + * Handle interrupts signalled for this port: */ - if (unlikely(port_cause & ERR_IRQ)) { - mv_err_intr(ap, mv_get_active_qc(ap)); - } else if (hc_irq_cause & (DEV_IRQ << hardport)) { - if (!(pp->pp_flags & MV_PP_FLAG_EDMA_EN)) { - struct ata_queued_cmd *qc = mv_get_active_qc(ap); - if (qc) { - ata_sff_host_intr(ap, qc); - continue; - } - } - mv_unexpected_intr(ap); - } + port_cause = (main_irq_cause >> shift) & (DONE_IRQ | ERR_IRQ); + if (port_cause) + mv_port_intr(ap, port_cause); } return handled; } @@ -1894,7 +2244,7 @@ static void mv5_reset_bus(struct ata_host *host, void __iomem *mmio) static void mv5_reset_flash(struct mv_host_priv *hpriv, void __iomem *mmio) { - writel(0x0fcfffff, mmio + MV_FLASH_CTL); + writel(0x0fcfffff, mmio + MV_FLASH_CTL_OFS); } static void mv5_read_preamp(struct mv_host_priv *hpriv, int idx, @@ -1913,7 +2263,7 @@ static void mv5_enable_leds(struct mv_host_priv *hpriv, void __iomem *mmio) { u32 tmp; - writel(0, mmio + MV_GPIO_PORT_CTL); + writel(0, mmio + MV_GPIO_PORT_CTL_OFS); /* FIXME: handle MV_HP_ERRATA_50XXB2 errata */ @@ -1931,14 +2281,14 @@ static void mv5_phy_errata(struct mv_host_priv *hpriv, void __iomem *mmio, int fix_apm_sq = (hpriv->hp_flags & MV_HP_ERRATA_50XXB0); if (fix_apm_sq) { - tmp = readl(phy_mmio + MV5_LT_MODE); + tmp = readl(phy_mmio + MV5_LTMODE_OFS); tmp |= (1 << 19); - writel(tmp, phy_mmio + MV5_LT_MODE); + writel(tmp, phy_mmio + MV5_LTMODE_OFS); - tmp = readl(phy_mmio + MV5_PHY_CTL); + tmp = readl(phy_mmio + MV5_PHY_CTL_OFS); tmp &= ~0x3; tmp |= 0x1; - writel(tmp, phy_mmio + MV5_PHY_CTL); + writel(tmp, phy_mmio + MV5_PHY_CTL_OFS); } tmp = readl(phy_mmio + MV5_PHY_MODE); @@ -1956,11 +2306,6 @@ static void mv5_reset_hc_port(struct mv_host_priv *hpriv, void __iomem *mmio, { void __iomem *port_mmio = mv_port_base(mmio, port); - /* - * The datasheet warns against setting ATA_RST when EDMA is active - * (but doesn't say what the problem might be). So we first try - * to disable the EDMA engine before doing the ATA_RST operation. - */ mv_reset_channel(hpriv, mmio, port); ZERO(0x028); /* command */ @@ -1975,7 +2320,7 @@ static void mv5_reset_hc_port(struct mv_host_priv *hpriv, void __iomem *mmio, ZERO(0x024); /* respq outp */ ZERO(0x020); /* respq inp */ ZERO(0x02c); /* test control */ - writel(0xbc, port_mmio + EDMA_IORDY_TMOUT); + writel(0xbc, port_mmio + EDMA_IORDY_TMOUT_OFS); } #undef ZERO @@ -2021,13 +2366,13 @@ static void mv_reset_pci_bus(struct ata_host *host, void __iomem *mmio) struct mv_host_priv *hpriv = host->private_data; u32 tmp; - tmp = readl(mmio + MV_PCI_MODE); + tmp = readl(mmio + MV_PCI_MODE_OFS); tmp &= 0xff00ffff; - writel(tmp, mmio + MV_PCI_MODE); + writel(tmp, mmio + MV_PCI_MODE_OFS); ZERO(MV_PCI_DISC_TIMER); ZERO(MV_PCI_MSI_TRIGGER); - writel(0x000100ff, mmio + MV_PCI_XBAR_TMOUT); + writel(0x000100ff, mmio + MV_PCI_XBAR_TMOUT_OFS); ZERO(PCI_HC_MAIN_IRQ_MASK_OFS); ZERO(MV_PCI_SERR_MASK); ZERO(hpriv->irq_cause_ofs); @@ -2045,10 +2390,10 @@ static void mv6_reset_flash(struct mv_host_priv *hpriv, void __iomem *mmio) mv5_reset_flash(hpriv, mmio); - tmp = readl(mmio + MV_GPIO_PORT_CTL); + tmp = readl(mmio + MV_GPIO_PORT_CTL_OFS); tmp &= 0x3; tmp |= (1 << 5) | (1 << 6); - writel(tmp, mmio + MV_GPIO_PORT_CTL); + writel(tmp, mmio + MV_GPIO_PORT_CTL_OFS); } /** @@ -2121,7 +2466,7 @@ static void mv6_read_preamp(struct mv_host_priv *hpriv, int idx, void __iomem *port_mmio; u32 tmp; - tmp = readl(mmio + MV_RESET_CFG); + tmp = readl(mmio + MV_RESET_CFG_OFS); if ((tmp & (1 << 0)) == 0) { hpriv->signal[idx].amps = 0x7 << 8; hpriv->signal[idx].pre = 0x1 << 5; @@ -2137,7 +2482,7 @@ static void mv6_read_preamp(struct mv_host_priv *hpriv, int idx, static void mv6_enable_leds(struct mv_host_priv *hpriv, void __iomem *mmio) { - writel(0x00000060, mmio + MV_GPIO_PORT_CTL); + writel(0x00000060, mmio + MV_GPIO_PORT_CTL_OFS); } static void mv6_phy_errata(struct mv_host_priv *hpriv, void __iomem *mmio, @@ -2235,11 +2580,6 @@ static void mv_soc_reset_hc_port(struct mv_host_priv *hpriv, { void __iomem *port_mmio = mv_port_base(mmio, port); - /* - * The datasheet warns against setting ATA_RST when EDMA is active - * (but doesn't say what the problem might be). So we first try - * to disable the EDMA engine before doing the ATA_RST operation. - */ mv_reset_channel(hpriv, mmio, port); ZERO(0x028); /* command */ @@ -2254,7 +2594,7 @@ static void mv_soc_reset_hc_port(struct mv_host_priv *hpriv, ZERO(0x024); /* respq outp */ ZERO(0x020); /* respq inp */ ZERO(0x02c); /* test control */ - writel(0xbc, port_mmio + EDMA_IORDY_TMOUT); + writel(0xbc, port_mmio + EDMA_IORDY_TMOUT_OFS); } #undef ZERO @@ -2297,38 +2637,39 @@ static void mv_soc_reset_bus(struct ata_host *host, void __iomem *mmio) return; } -static void mv_setup_ifctl(void __iomem *port_mmio, int want_gen2i) +static void mv_setup_ifcfg(void __iomem *port_mmio, int want_gen2i) { - u32 ifctl = readl(port_mmio + SATA_INTERFACE_CFG); + u32 ifcfg = readl(port_mmio + SATA_INTERFACE_CFG_OFS); - ifctl = (ifctl & 0xf7f) | 0x9b1000; /* from chip spec */ + ifcfg = (ifcfg & 0xf7f) | 0x9b1000; /* from chip spec */ if (want_gen2i) - ifctl |= (1 << 7); /* enable gen2i speed */ - writelfl(ifctl, port_mmio + SATA_INTERFACE_CFG); + ifcfg |= (1 << 7); /* enable gen2i speed */ + writelfl(ifcfg, port_mmio + SATA_INTERFACE_CFG_OFS); } -/* - * Caller must ensure that EDMA is not active, - * by first doing mv_stop_edma() where needed. - */ static void mv_reset_channel(struct mv_host_priv *hpriv, void __iomem *mmio, unsigned int port_no) { void __iomem *port_mmio = mv_port_base(mmio, port_no); + /* + * The datasheet warns against setting EDMA_RESET when EDMA is active + * (but doesn't say what the problem might be). So we first try + * to disable the EDMA engine before doing the EDMA_RESET operation. + */ mv_stop_edma_engine(port_mmio); - writelfl(ATA_RST, port_mmio + EDMA_CMD_OFS); + writelfl(EDMA_RESET, port_mmio + EDMA_CMD_OFS); if (!IS_GEN_I(hpriv)) { - /* Enable 3.0gb/s link speed */ - mv_setup_ifctl(port_mmio, 1); + /* Enable 3.0gb/s link speed: this survives EDMA_RESET */ + mv_setup_ifcfg(port_mmio, 1); } /* - * Strobing ATA_RST here causes a hard reset of the SATA transport, + * Strobing EDMA_RESET here causes a hard reset of the SATA transport, * link, and physical layers. It resets all SATA interface registers * (except for SATA_INTERFACE_CFG), and issues a COMRESET to the dev. */ - writelfl(ATA_RST, port_mmio + EDMA_CMD_OFS); + writelfl(EDMA_RESET, port_mmio + EDMA_CMD_OFS); udelay(25); /* allow reset propagation */ writelfl(0, port_mmio + EDMA_CMD_OFS); @@ -2392,7 +2733,7 @@ static int mv_hardreset(struct ata_link *link, unsigned int *class, sata_scr_read(link, SCR_STATUS, &sstatus); if (!IS_GEN_I(hpriv) && ++attempts >= 5 && sstatus == 0x121) { /* Force 1.5gb/s link speed and try again */ - mv_setup_ifctl(mv_ap_base(ap), 0); + mv_setup_ifcfg(mv_ap_base(ap), 0); if (time_after(jiffies + HZ, deadline)) extra = HZ; /* only extend it once, max */ } @@ -2493,6 +2834,34 @@ static void mv_port_init(struct ata_ioports *port, void __iomem *port_mmio) readl(port_mmio + EDMA_ERR_IRQ_MASK_OFS)); } +static unsigned int mv_in_pcix_mode(struct ata_host *host) +{ + struct mv_host_priv *hpriv = host->private_data; + void __iomem *mmio = hpriv->base; + u32 reg; + + if (!HAS_PCI(host) || !IS_PCIE(hpriv)) + return 0; /* not PCI-X capable */ + reg = readl(mmio + MV_PCI_MODE_OFS); + if ((reg & MV_PCI_MODE_MASK) == 0) + return 0; /* conventional PCI mode */ + return 1; /* chip is in PCI-X mode */ +} + +static int mv_pci_cut_through_okay(struct ata_host *host) +{ + struct mv_host_priv *hpriv = host->private_data; + void __iomem *mmio = hpriv->base; + u32 reg; + + if (!mv_in_pcix_mode(host)) { + reg = readl(mmio + PCI_COMMAND_OFS); + if (reg & PCI_COMMAND_MRDTRIG) + return 0; /* not okay */ + } + return 1; /* okay */ +} + static int mv_chip_id(struct ata_host *host, unsigned int board_idx) { struct pci_dev *pdev = to_pci_dev(host->dev); @@ -2560,7 +2929,7 @@ static int mv_chip_id(struct ata_host *host, unsigned int board_idx) break; case chip_7042: - hp_flags |= MV_HP_PCIE; + hp_flags |= MV_HP_PCIE | MV_HP_CUT_THROUGH; if (pdev->vendor == PCI_VENDOR_ID_TTI && (pdev->device == 0x2300 || pdev->device == 0x2310)) { @@ -2590,9 +2959,12 @@ static int mv_chip_id(struct ata_host *host, unsigned int board_idx) " and avoid the final two gigabytes on" " all RocketRAID BIOS initialized drives.\n"); } + /* drop through */ case chip_6042: hpriv->ops = &mv6xxx_ops; hp_flags |= MV_HP_GEN_IIE; + if (board_idx == chip_6042 && mv_pci_cut_through_okay(host)) + hp_flags |= MV_HP_CUT_THROUGH; switch (pdev->revision) { case 0x0: diff --git a/drivers/base/sys.c b/drivers/base/sys.c index 4fbb56bcb1e..358bb0be3c0 100644 --- a/drivers/base/sys.c +++ b/drivers/base/sys.c @@ -175,8 +175,7 @@ int sysdev_driver_register(struct sysdev_class *cls, struct sysdev_driver *drv) } /* Check whether this driver has already been added to a class. */ - if ((drv->entry.next != drv->entry.prev) || - (drv->entry.next != NULL)) { + if (drv->entry.next && !list_empty(&drv->entry)) { printk(KERN_WARNING "sysdev: class %s: driver (%p) has already" " been registered to a class, something is wrong, but " "will forge on!\n", cls->name, drv); diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index 8fc429cf82b..41f818be2f7 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -755,11 +755,13 @@ diskstats(struct gendisk *disk, struct bio *bio, ulong duration, sector_t sector { unsigned long n_sect = bio->bi_size >> 9; const int rw = bio_data_dir(bio); + struct hd_struct *part; - all_stat_inc(disk, ios[rw], sector); - all_stat_add(disk, ticks[rw], duration, sector); - all_stat_add(disk, sectors[rw], n_sect, sector); - all_stat_add(disk, io_ticks, duration, sector); + part = get_part(disk, sector); + all_stat_inc(disk, part, ios[rw], sector); + all_stat_add(disk, part, ticks[rw], duration, sector); + all_stat_add(disk, part, sectors[rw], n_sect, sector); + all_stat_add(disk, part, io_ticks, duration, sector); } void diff --git a/drivers/char/serial167.c b/drivers/char/serial167.c index fd2db07a50f..3b23270eaa6 100644 --- a/drivers/char/serial167.c +++ b/drivers/char/serial167.c @@ -1073,7 +1073,7 @@ static int cy_put_char(struct tty_struct *tty, unsigned char ch) return 0; if (!info->xmit_buf) - return; + return 0; local_irq_save(flags); if (info->xmit_cnt >= PAGE_SIZE - 1) { diff --git a/drivers/char/sx.c b/drivers/char/sx.c index f39f6fd8935..b1a7a8cb65e 100644 --- a/drivers/char/sx.c +++ b/drivers/char/sx.c @@ -970,7 +970,8 @@ static int sx_set_real_termios(void *ptr) sx_write_channel_byte(port, hi_mask, 0x1f); break; default: - printk(KERN_INFO "sx: Invalid wordsize: %u\n", CFLAG & CSIZE); + printk(KERN_INFO "sx: Invalid wordsize: %u\n", + (unsigned int)CFLAG & CSIZE); break; } @@ -997,7 +998,8 @@ static int sx_set_real_termios(void *ptr) set_bit(TTY_HW_COOK_IN, &port->gs.tty->flags); } sx_dprintk(SX_DEBUG_TERMIOS, "iflags: %x(%d) ", - port->gs.tty->termios->c_iflag, I_OTHER(port->gs.tty)); + (unsigned int)port->gs.tty->termios->c_iflag, + I_OTHER(port->gs.tty)); /* Tell line discipline whether we will do output cooking. * If OPOST is set and no other output flags are set then we can do output @@ -1010,7 +1012,8 @@ static int sx_set_real_termios(void *ptr) clear_bit(TTY_HW_COOK_OUT, &port->gs.tty->flags); } sx_dprintk(SX_DEBUG_TERMIOS, "oflags: %x(%d)\n", - port->gs.tty->termios->c_oflag, O_OTHER(port->gs.tty)); + (unsigned int)port->gs.tty->termios->c_oflag, + O_OTHER(port->gs.tty)); /* port->c_dcd = sx_get_CD (port); */ func_exit(); return 0; diff --git a/drivers/char/vt.c b/drivers/char/vt.c index e458b08139a..fa1ffbf2c62 100644 --- a/drivers/char/vt.c +++ b/drivers/char/vt.c @@ -2742,6 +2742,10 @@ static int con_open(struct tty_struct *tty, struct file *filp) tty->winsize.ws_row = vc_cons[currcons].d->vc_rows; tty->winsize.ws_col = vc_cons[currcons].d->vc_cols; } + if (vc->vc_utf) + tty->termios->c_iflag |= IUTF8; + else + tty->termios->c_iflag &= ~IUTF8; release_console_sem(); vcs_make_sysfs(tty); return ret; @@ -2918,6 +2922,8 @@ int __init vty_init(void) console_driver->minor_start = 1; console_driver->type = TTY_DRIVER_TYPE_CONSOLE; console_driver->init_termios = tty_std_termios; + if (default_utf8) + console_driver->init_termios.c_iflag |= IUTF8; console_driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_RESET_TERMIOS; tty_set_operations(console_driver, &con_ops); if (tty_register_driver(console_driver)) diff --git a/drivers/edac/edac_core.h b/drivers/edac/edac_core.h index a9aa845dbe7..b27b13c5eb5 100644 --- a/drivers/edac/edac_core.h +++ b/drivers/edac/edac_core.h @@ -97,7 +97,7 @@ extern int edac_debug_level; #define PCI_VEND_DEV(vend, dev) PCI_VENDOR_ID_ ## vend, \ PCI_DEVICE_ID_ ## vend ## _ ## dev -#define dev_name(dev) (dev)->dev_name +#define edac_dev_name(dev) (dev)->dev_name /* memory devices */ enum dev_type { diff --git a/drivers/edac/edac_device.c b/drivers/edac/edac_device.c index 63372fa7ecf..5fcd3d89c75 100644 --- a/drivers/edac/edac_device.c +++ b/drivers/edac/edac_device.c @@ -333,7 +333,7 @@ static int add_edac_dev_to_global_list(struct edac_device_ctl_info *edac_dev) fail0: edac_printk(KERN_WARNING, EDAC_MC, "%s (%s) %s %s already assigned %d\n", - rover->dev->bus_id, dev_name(rover), + rover->dev->bus_id, edac_dev_name(rover), rover->mod_name, rover->ctl_name, rover->dev_idx); return 1; @@ -538,7 +538,7 @@ int edac_device_add_device(struct edac_device_ctl_info *edac_dev) "'%s': DEV '%s' (%s)\n", edac_dev->mod_name, edac_dev->ctl_name, - dev_name(edac_dev), + edac_dev_name(edac_dev), edac_op_state_to_string(edac_dev->op_state)); mutex_unlock(&device_ctls_mutex); @@ -599,7 +599,7 @@ struct edac_device_ctl_info *edac_device_del_device(struct device *dev) edac_printk(KERN_INFO, EDAC_MC, "Removed device %d for %s %s: DEV %s\n", edac_dev->dev_idx, - edac_dev->mod_name, edac_dev->ctl_name, dev_name(edac_dev)); + edac_dev->mod_name, edac_dev->ctl_name, edac_dev_name(edac_dev)); return edac_dev; } diff --git a/drivers/edac/edac_mc.c b/drivers/edac/edac_mc.c index a4cf1645f58..d110392d48f 100644 --- a/drivers/edac/edac_mc.c +++ b/drivers/edac/edac_mc.c @@ -402,7 +402,7 @@ static int add_mc_to_global_list(struct mem_ctl_info *mci) fail0: edac_printk(KERN_WARNING, EDAC_MC, "%s (%s) %s %s already assigned %d\n", p->dev->bus_id, - dev_name(mci), p->mod_name, p->ctl_name, p->mc_idx); + edac_dev_name(mci), p->mod_name, p->ctl_name, p->mc_idx); return 1; fail1: @@ -517,7 +517,7 @@ int edac_mc_add_mc(struct mem_ctl_info *mci) /* Report action taken */ edac_mc_printk(mci, KERN_INFO, "Giving out device to '%s' '%s':" - " DEV %s\n", mci->mod_name, mci->ctl_name, dev_name(mci)); + " DEV %s\n", mci->mod_name, mci->ctl_name, edac_dev_name(mci)); mutex_unlock(&mem_ctls_mutex); return 0; @@ -565,7 +565,7 @@ struct mem_ctl_info *edac_mc_del_mc(struct device *dev) edac_printk(KERN_INFO, EDAC_MC, "Removed device %d for %s %s: DEV %s\n", mci->mc_idx, - mci->mod_name, mci->ctl_name, dev_name(mci)); + mci->mod_name, mci->ctl_name, edac_dev_name(mci)); return mci; } diff --git a/drivers/edac/edac_pci.c b/drivers/edac/edac_pci.c index 9b24340b52e..22ec9d5d431 100644 --- a/drivers/edac/edac_pci.c +++ b/drivers/edac/edac_pci.c @@ -150,7 +150,7 @@ static int add_edac_pci_to_global_list(struct edac_pci_ctl_info *pci) fail0: edac_printk(KERN_WARNING, EDAC_PCI, "%s (%s) %s %s already assigned %d\n", - rover->dev->bus_id, dev_name(rover), + rover->dev->bus_id, edac_dev_name(rover), rover->mod_name, rover->ctl_name, rover->pci_idx); return 1; @@ -360,7 +360,7 @@ int edac_pci_add_device(struct edac_pci_ctl_info *pci, int edac_idx) " DEV '%s' (%s)\n", pci->mod_name, pci->ctl_name, - dev_name(pci), edac_op_state_to_string(pci->op_state)); + edac_dev_name(pci), edac_op_state_to_string(pci->op_state)); mutex_unlock(&edac_pci_ctls_mutex); return 0; @@ -415,7 +415,7 @@ struct edac_pci_ctl_info *edac_pci_del_device(struct device *dev) edac_printk(KERN_INFO, EDAC_PCI, "Removed device %d for %s %s: DEV %s\n", - pci->pci_idx, pci->mod_name, pci->ctl_name, dev_name(pci)); + pci->pci_idx, pci->mod_name, pci->ctl_name, edac_dev_name(pci)); return pci; } diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c index 591deda3f86..34b0d4f26b5 100644 --- a/drivers/ide/ide-probe.c +++ b/drivers/ide/ide-probe.c @@ -1355,12 +1355,6 @@ static void ide_init_port(ide_hwif_t *hwif, unsigned int port, if (hwif->chipset != ide_dtc2278 || hwif->channel == 0) hwif->port_ops = d->port_ops; - if ((d->host_flags & IDE_HFLAG_SERIALIZE) || - ((d->host_flags & IDE_HFLAG_SERIALIZE_DMA) && hwif->dma_base)) { - if (hwif->mate) - hwif->mate->serialized = hwif->serialized = 1; - } - hwif->swdma_mask = d->swdma_mask; hwif->mwdma_mask = d->mwdma_mask; hwif->ultra_mask = d->udma_mask; @@ -1382,6 +1376,12 @@ static void ide_init_port(ide_hwif_t *hwif, unsigned int port, hwif->dma_ops = d->dma_ops; } + if ((d->host_flags & IDE_HFLAG_SERIALIZE) || + ((d->host_flags & IDE_HFLAG_SERIALIZE_DMA) && hwif->dma_base)) { + if (hwif->mate) + hwif->mate->serialized = hwif->serialized = 1; + } + if (d->host_flags & IDE_HFLAG_RQSIZE_256) hwif->rqsize = 256; diff --git a/drivers/ide/legacy/falconide.c b/drivers/ide/legacy/falconide.c index 83555ca513b..9e449a0c623 100644 --- a/drivers/ide/legacy/falconide.c +++ b/drivers/ide/legacy/falconide.c @@ -61,7 +61,7 @@ static void falconide_output_data(ide_drive_t *drive, struct request *rq, unsigned long data_addr = drive->hwif->io_ports.data_addr; if (drive->media == ide_disk && rq && rq->cmd_type == REQ_TYPE_FS) - return outsw(data_adr, buf, (len + 1) / 2); + return outsw(data_addr, buf, (len + 1) / 2); outsw_swapw(data_addr, buf, (len + 1) / 2); } diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.c b/drivers/infiniband/hw/cxgb3/cxio_hal.c index ed2ee4ba4b7..ebf9d3043f8 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_hal.c +++ b/drivers/infiniband/hw/cxgb3/cxio_hal.c @@ -359,9 +359,10 @@ static void insert_recv_cqe(struct t3_wq *wq, struct t3_cq *cq) cq->sw_wptr++; } -void cxio_flush_rq(struct t3_wq *wq, struct t3_cq *cq, int count) +int cxio_flush_rq(struct t3_wq *wq, struct t3_cq *cq, int count) { u32 ptr; + int flushed = 0; PDBG("%s wq %p cq %p\n", __func__, wq, cq); @@ -369,8 +370,11 @@ void cxio_flush_rq(struct t3_wq *wq, struct t3_cq *cq, int count) PDBG("%s rq_rptr %u rq_wptr %u skip count %u\n", __func__, wq->rq_rptr, wq->rq_wptr, count); ptr = wq->rq_rptr + count; - while (ptr++ != wq->rq_wptr) + while (ptr++ != wq->rq_wptr) { insert_recv_cqe(wq, cq); + flushed++; + } + return flushed; } static void insert_sq_cqe(struct t3_wq *wq, struct t3_cq *cq, @@ -394,9 +398,10 @@ static void insert_sq_cqe(struct t3_wq *wq, struct t3_cq *cq, cq->sw_wptr++; } -void cxio_flush_sq(struct t3_wq *wq, struct t3_cq *cq, int count) +int cxio_flush_sq(struct t3_wq *wq, struct t3_cq *cq, int count) { __u32 ptr; + int flushed = 0; struct t3_swsq *sqp = wq->sq + Q_PTR2IDX(wq->sq_rptr, wq->sq_size_log2); ptr = wq->sq_rptr + count; @@ -405,7 +410,9 @@ void cxio_flush_sq(struct t3_wq *wq, struct t3_cq *cq, int count) insert_sq_cqe(wq, cq, sqp); sqp++; ptr++; + flushed++; } + return flushed; } /* @@ -581,7 +588,7 @@ static int cxio_hal_destroy_ctrl_qp(struct cxio_rdev *rdev_p) * caller aquires the ctrl_qp lock before the call */ static int cxio_hal_ctrl_qp_write_mem(struct cxio_rdev *rdev_p, u32 addr, - u32 len, void *data, int completion) + u32 len, void *data) { u32 i, nr_wqe, copy_len; u8 *copy_data; @@ -617,7 +624,7 @@ static int cxio_hal_ctrl_qp_write_mem(struct cxio_rdev *rdev_p, u32 addr, flag = 0; if (i == (nr_wqe - 1)) { /* last WQE */ - flag = completion ? T3_COMPLETION_FLAG : 0; + flag = T3_COMPLETION_FLAG; if (len % 32) utx_len = len / 32 + 1; else @@ -676,21 +683,20 @@ static int cxio_hal_ctrl_qp_write_mem(struct cxio_rdev *rdev_p, u32 addr, return 0; } -/* IN: stag key, pdid, perm, zbva, to, len, page_size, pbl, and pbl_size - * OUT: stag index, actual pbl_size, pbl_addr allocated. +/* IN: stag key, pdid, perm, zbva, to, len, page_size, pbl_size and pbl_addr + * OUT: stag index * TBD: shared memory region support */ static int __cxio_tpt_op(struct cxio_rdev *rdev_p, u32 reset_tpt_entry, u32 *stag, u8 stag_state, u32 pdid, enum tpt_mem_type type, enum tpt_mem_perm perm, - u32 zbva, u64 to, u32 len, u8 page_size, __be64 *pbl, - u32 *pbl_size, u32 *pbl_addr) + u32 zbva, u64 to, u32 len, u8 page_size, + u32 pbl_size, u32 pbl_addr) { int err; struct tpt_entry tpt; u32 stag_idx; u32 wptr; - int rereg = (*stag != T3_STAG_UNSET); stag_state = stag_state > 0; stag_idx = (*stag) >> 8; @@ -704,30 +710,8 @@ static int __cxio_tpt_op(struct cxio_rdev *rdev_p, u32 reset_tpt_entry, PDBG("%s stag_state 0x%0x type 0x%0x pdid 0x%0x, stag_idx 0x%x\n", __func__, stag_state, type, pdid, stag_idx); - if (reset_tpt_entry) - cxio_hal_pblpool_free(rdev_p, *pbl_addr, *pbl_size << 3); - else if (!rereg) { - *pbl_addr = cxio_hal_pblpool_alloc(rdev_p, *pbl_size << 3); - if (!*pbl_addr) { - return -ENOMEM; - } - } - mutex_lock(&rdev_p->ctrl_qp.lock); - /* write PBL first if any - update pbl only if pbl list exist */ - if (pbl) { - - PDBG("%s *pdb_addr 0x%x, pbl_base 0x%x, pbl_size %d\n", - __func__, *pbl_addr, rdev_p->rnic_info.pbl_base, - *pbl_size); - err = cxio_hal_ctrl_qp_write_mem(rdev_p, - (*pbl_addr >> 5), - (*pbl_size << 3), pbl, 0); - if (err) - goto ret; - } - /* write TPT entry */ if (reset_tpt_entry) memset(&tpt, 0, sizeof(tpt)); @@ -742,23 +726,23 @@ static int __cxio_tpt_op(struct cxio_rdev *rdev_p, u32 reset_tpt_entry, V_TPT_ADDR_TYPE((zbva ? TPT_ZBTO : TPT_VATO)) | V_TPT_PAGE_SIZE(page_size)); tpt.rsvd_pbl_addr = reset_tpt_entry ? 0 : - cpu_to_be32(V_TPT_PBL_ADDR(PBL_OFF(rdev_p, *pbl_addr)>>3)); + cpu_to_be32(V_TPT_PBL_ADDR(PBL_OFF(rdev_p, pbl_addr)>>3)); tpt.len = cpu_to_be32(len); tpt.va_hi = cpu_to_be32((u32) (to >> 32)); tpt.va_low_or_fbo = cpu_to_be32((u32) (to & 0xFFFFFFFFULL)); tpt.rsvd_bind_cnt_or_pstag = 0; tpt.rsvd_pbl_size = reset_tpt_entry ? 0 : - cpu_to_be32(V_TPT_PBL_SIZE((*pbl_size) >> 2)); + cpu_to_be32(V_TPT_PBL_SIZE(pbl_size >> 2)); } err = cxio_hal_ctrl_qp_write_mem(rdev_p, stag_idx + (rdev_p->rnic_info.tpt_base >> 5), - sizeof(tpt), &tpt, 1); + sizeof(tpt), &tpt); /* release the stag index to free pool */ if (reset_tpt_entry) cxio_hal_put_stag(rdev_p->rscp, stag_idx); -ret: + wptr = rdev_p->ctrl_qp.wptr; mutex_unlock(&rdev_p->ctrl_qp.lock); if (!err) @@ -769,44 +753,67 @@ ret: return err; } +int cxio_write_pbl(struct cxio_rdev *rdev_p, __be64 *pbl, + u32 pbl_addr, u32 pbl_size) +{ + u32 wptr; + int err; + + PDBG("%s *pdb_addr 0x%x, pbl_base 0x%x, pbl_size %d\n", + __func__, pbl_addr, rdev_p->rnic_info.pbl_base, + pbl_size); + + mutex_lock(&rdev_p->ctrl_qp.lock); + err = cxio_hal_ctrl_qp_write_mem(rdev_p, pbl_addr >> 5, pbl_size << 3, + pbl); + wptr = rdev_p->ctrl_qp.wptr; + mutex_unlock(&rdev_p->ctrl_qp.lock); + if (err) + return err; + + if (wait_event_interruptible(rdev_p->ctrl_qp.waitq, + SEQ32_GE(rdev_p->ctrl_qp.rptr, + wptr))) + return -ERESTARTSYS; + + return 0; +} + int cxio_register_phys_mem(struct cxio_rdev *rdev_p, u32 *stag, u32 pdid, enum tpt_mem_perm perm, u32 zbva, u64 to, u32 len, - u8 page_size, __be64 *pbl, u32 *pbl_size, - u32 *pbl_addr) + u8 page_size, u32 pbl_size, u32 pbl_addr) { *stag = T3_STAG_UNSET; return __cxio_tpt_op(rdev_p, 0, stag, 1, pdid, TPT_NON_SHARED_MR, perm, - zbva, to, len, page_size, pbl, pbl_size, pbl_addr); + zbva, to, len, page_size, pbl_size, pbl_addr); } int cxio_reregister_phys_mem(struct cxio_rdev *rdev_p, u32 *stag, u32 pdid, enum tpt_mem_perm perm, u32 zbva, u64 to, u32 len, - u8 page_size, __be64 *pbl, u32 *pbl_size, - u32 *pbl_addr) + u8 page_size, u32 pbl_size, u32 pbl_addr) { return __cxio_tpt_op(rdev_p, 0, stag, 1, pdid, TPT_NON_SHARED_MR, perm, - zbva, to, len, page_size, pbl, pbl_size, pbl_addr); + zbva, to, len, page_size, pbl_size, pbl_addr); } int cxio_dereg_mem(struct cxio_rdev *rdev_p, u32 stag, u32 pbl_size, u32 pbl_addr) { - return __cxio_tpt_op(rdev_p, 1, &stag, 0, 0, 0, 0, 0, 0ULL, 0, 0, NULL, - &pbl_size, &pbl_addr); + return __cxio_tpt_op(rdev_p, 1, &stag, 0, 0, 0, 0, 0, 0ULL, 0, 0, + pbl_size, pbl_addr); } int cxio_allocate_window(struct cxio_rdev *rdev_p, u32 * stag, u32 pdid) { - u32 pbl_size = 0; *stag = T3_STAG_UNSET; return __cxio_tpt_op(rdev_p, 0, stag, 0, pdid, TPT_MW, 0, 0, 0ULL, 0, 0, - NULL, &pbl_size, NULL); + 0, 0); } int cxio_deallocate_window(struct cxio_rdev *rdev_p, u32 stag) { - return __cxio_tpt_op(rdev_p, 1, &stag, 0, 0, 0, 0, 0, 0ULL, 0, 0, NULL, - NULL, NULL); + return __cxio_tpt_op(rdev_p, 1, &stag, 0, 0, 0, 0, 0, 0ULL, 0, 0, + 0, 0); } int cxio_rdma_init(struct cxio_rdev *rdev_p, struct t3_rdma_init_attr *attr) diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.h b/drivers/infiniband/hw/cxgb3/cxio_hal.h index 2bcff7f5046..6e128f6bab0 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_hal.h +++ b/drivers/infiniband/hw/cxgb3/cxio_hal.h @@ -154,14 +154,14 @@ int cxio_create_qp(struct cxio_rdev *rdev, u32 kernel_domain, struct t3_wq *wq, int cxio_destroy_qp(struct cxio_rdev *rdev, struct t3_wq *wq, struct cxio_ucontext *uctx); int cxio_peek_cq(struct t3_wq *wr, struct t3_cq *cq, int opcode); +int cxio_write_pbl(struct cxio_rdev *rdev_p, __be64 *pbl, + u32 pbl_addr, u32 pbl_size); int cxio_register_phys_mem(struct cxio_rdev *rdev, u32 * stag, u32 pdid, enum tpt_mem_perm perm, u32 zbva, u64 to, u32 len, - u8 page_size, __be64 *pbl, u32 *pbl_size, - u32 *pbl_addr); + u8 page_size, u32 pbl_size, u32 pbl_addr); int cxio_reregister_phys_mem(struct cxio_rdev *rdev, u32 * stag, u32 pdid, enum tpt_mem_perm perm, u32 zbva, u64 to, u32 len, - u8 page_size, __be64 *pbl, u32 *pbl_size, - u32 *pbl_addr); + u8 page_size, u32 pbl_size, u32 pbl_addr); int cxio_dereg_mem(struct cxio_rdev *rdev, u32 stag, u32 pbl_size, u32 pbl_addr); int cxio_allocate_window(struct cxio_rdev *rdev, u32 * stag, u32 pdid); @@ -173,8 +173,8 @@ u32 cxio_hal_get_pdid(struct cxio_hal_resource *rscp); void cxio_hal_put_pdid(struct cxio_hal_resource *rscp, u32 pdid); int __init cxio_hal_init(void); void __exit cxio_hal_exit(void); -void cxio_flush_rq(struct t3_wq *wq, struct t3_cq *cq, int count); -void cxio_flush_sq(struct t3_wq *wq, struct t3_cq *cq, int count); +int cxio_flush_rq(struct t3_wq *wq, struct t3_cq *cq, int count); +int cxio_flush_sq(struct t3_wq *wq, struct t3_cq *cq, int count); void cxio_count_rcqes(struct t3_cq *cq, struct t3_wq *wq, int *count); void cxio_count_scqes(struct t3_cq *cq, struct t3_wq *wq, int *count); void cxio_flush_hw_cq(struct t3_cq *cq); diff --git a/drivers/infiniband/hw/cxgb3/cxio_resource.c b/drivers/infiniband/hw/cxgb3/cxio_resource.c index 45ed4f25ef7..bd233c08765 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_resource.c +++ b/drivers/infiniband/hw/cxgb3/cxio_resource.c @@ -250,7 +250,6 @@ void cxio_hal_destroy_resource(struct cxio_hal_resource *rscp) */ #define MIN_PBL_SHIFT 8 /* 256B == min PBL size (32 entries) */ -#define PBL_CHUNK 2*1024*1024 u32 cxio_hal_pblpool_alloc(struct cxio_rdev *rdev_p, int size) { @@ -267,14 +266,35 @@ void cxio_hal_pblpool_free(struct cxio_rdev *rdev_p, u32 addr, int size) int cxio_hal_pblpool_create(struct cxio_rdev *rdev_p) { - unsigned long i; + unsigned pbl_start, pbl_chunk; + rdev_p->pbl_pool = gen_pool_create(MIN_PBL_SHIFT, -1); - if (rdev_p->pbl_pool) - for (i = rdev_p->rnic_info.pbl_base; - i <= rdev_p->rnic_info.pbl_top - PBL_CHUNK + 1; - i += PBL_CHUNK) - gen_pool_add(rdev_p->pbl_pool, i, PBL_CHUNK, -1); - return rdev_p->pbl_pool ? 0 : -ENOMEM; + if (!rdev_p->pbl_pool) + return -ENOMEM; + + pbl_start = rdev_p->rnic_info.pbl_base; + pbl_chunk = rdev_p->rnic_info.pbl_top - pbl_start + 1; + + while (pbl_start < rdev_p->rnic_info.pbl_top) { + pbl_chunk = min(rdev_p->rnic_info.pbl_top - pbl_start + 1, + pbl_chunk); + if (gen_pool_add(rdev_p->pbl_pool, pbl_start, pbl_chunk, -1)) { + PDBG("%s failed to add PBL chunk (%x/%x)\n", + __func__, pbl_start, pbl_chunk); + if (pbl_chunk <= 1024 << MIN_PBL_SHIFT) { + printk(KERN_WARNING MOD "%s: Failed to add all PBL chunks (%x/%x)\n", + __func__, pbl_start, rdev_p->rnic_info.pbl_top - pbl_start); + return 0; + } + pbl_chunk >>= 1; + } else { + PDBG("%s added PBL chunk (%x/%x)\n", + __func__, pbl_start, pbl_chunk); + pbl_start += pbl_chunk; + } + } + + return 0; } void cxio_hal_pblpool_destroy(struct cxio_rdev *rdev_p) diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.c b/drivers/infiniband/hw/cxgb3/iwch_cm.c index d44a6df9ad8..c325c44807e 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_cm.c +++ b/drivers/infiniband/hw/cxgb3/iwch_cm.c @@ -67,10 +67,10 @@ int peer2peer = 0; module_param(peer2peer, int, 0644); MODULE_PARM_DESC(peer2peer, "Support peer2peer ULPs (default=0)"); -static int ep_timeout_secs = 10; +static int ep_timeout_secs = 60; module_param(ep_timeout_secs, int, 0644); MODULE_PARM_DESC(ep_timeout_secs, "CM Endpoint operation timeout " - "in seconds (default=10)"); + "in seconds (default=60)"); static int mpa_rev = 1; module_param(mpa_rev, int, 0644); @@ -1650,8 +1650,8 @@ static int close_con_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) release = 1; break; case ABORTING: - break; case DEAD: + break; default: BUG_ON(1); break; diff --git a/drivers/infiniband/hw/cxgb3/iwch_mem.c b/drivers/infiniband/hw/cxgb3/iwch_mem.c index 58c3d61bcd1..ec49a5cbdeb 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_mem.c +++ b/drivers/infiniband/hw/cxgb3/iwch_mem.c @@ -35,17 +35,26 @@ #include <rdma/ib_verbs.h> #include "cxio_hal.h" +#include "cxio_resource.h" #include "iwch.h" #include "iwch_provider.h" -int iwch_register_mem(struct iwch_dev *rhp, struct iwch_pd *php, - struct iwch_mr *mhp, - int shift, - __be64 *page_list) +static void iwch_finish_mem_reg(struct iwch_mr *mhp, u32 stag) { - u32 stag; u32 mmid; + mhp->attr.state = 1; + mhp->attr.stag = stag; + mmid = stag >> 8; + mhp->ibmr.rkey = mhp->ibmr.lkey = stag; + insert_handle(mhp->rhp, &mhp->rhp->mmidr, mhp, mmid); + PDBG("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp); +} + +int iwch_register_mem(struct iwch_dev *rhp, struct iwch_pd *php, + struct iwch_mr *mhp, int shift) +{ + u32 stag; if (cxio_register_phys_mem(&rhp->rdev, &stag, mhp->attr.pdid, @@ -53,28 +62,21 @@ int iwch_register_mem(struct iwch_dev *rhp, struct iwch_pd *php, mhp->attr.zbva, mhp->attr.va_fbo, mhp->attr.len, - shift-12, - page_list, - &mhp->attr.pbl_size, &mhp->attr.pbl_addr)) + shift - 12, + mhp->attr.pbl_size, mhp->attr.pbl_addr)) return -ENOMEM; - mhp->attr.state = 1; - mhp->attr.stag = stag; - mmid = stag >> 8; - mhp->ibmr.rkey = mhp->ibmr.lkey = stag; - insert_handle(rhp, &rhp->mmidr, mhp, mmid); - PDBG("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp); + + iwch_finish_mem_reg(mhp, stag); + return 0; } int iwch_reregister_mem(struct iwch_dev *rhp, struct iwch_pd *php, struct iwch_mr *mhp, int shift, - __be64 *page_list, int npages) { u32 stag; - u32 mmid; - /* We could support this... */ if (npages > mhp->attr.pbl_size) @@ -87,19 +89,40 @@ int iwch_reregister_mem(struct iwch_dev *rhp, struct iwch_pd *php, mhp->attr.zbva, mhp->attr.va_fbo, mhp->attr.len, - shift-12, - page_list, - &mhp->attr.pbl_size, &mhp->attr.pbl_addr)) + shift - 12, + mhp->attr.pbl_size, mhp->attr.pbl_addr)) return -ENOMEM; - mhp->attr.state = 1; - mhp->attr.stag = stag; - mmid = stag >> 8; - mhp->ibmr.rkey = mhp->ibmr.lkey = stag; - insert_handle(rhp, &rhp->mmidr, mhp, mmid); - PDBG("%s mmid 0x%x mhp %p\n", __func__, mmid, mhp); + + iwch_finish_mem_reg(mhp, stag); + + return 0; +} + +int iwch_alloc_pbl(struct iwch_mr *mhp, int npages) +{ + mhp->attr.pbl_addr = cxio_hal_pblpool_alloc(&mhp->rhp->rdev, + npages << 3); + + if (!mhp->attr.pbl_addr) + return -ENOMEM; + + mhp->attr.pbl_size = npages; + return 0; } +void iwch_free_pbl(struct iwch_mr *mhp) +{ + cxio_hal_pblpool_free(&mhp->rhp->rdev, mhp->attr.pbl_addr, + mhp->attr.pbl_size << 3); +} + +int iwch_write_pbl(struct iwch_mr *mhp, __be64 *pages, int npages, int offset) +{ + return cxio_write_pbl(&mhp->rhp->rdev, pages, + mhp->attr.pbl_addr + (offset << 3), npages); +} + int build_phys_page_list(struct ib_phys_buf *buffer_list, int num_phys_buf, u64 *iova_start, diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.c b/drivers/infiniband/hw/cxgb3/iwch_provider.c index d07d3a377b5..8934178a23e 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_provider.c +++ b/drivers/infiniband/hw/cxgb3/iwch_provider.c @@ -442,6 +442,7 @@ static int iwch_dereg_mr(struct ib_mr *ib_mr) mmid = mhp->attr.stag >> 8; cxio_dereg_mem(&rhp->rdev, mhp->attr.stag, mhp->attr.pbl_size, mhp->attr.pbl_addr); + iwch_free_pbl(mhp); remove_handle(rhp, &rhp->mmidr, mmid); if (mhp->kva) kfree((void *) (unsigned long) mhp->kva); @@ -475,6 +476,8 @@ static struct ib_mr *iwch_register_phys_mem(struct ib_pd *pd, if (!mhp) return ERR_PTR(-ENOMEM); + mhp->rhp = rhp; + /* First check that we have enough alignment */ if ((*iova_start & ~PAGE_MASK) != (buffer_list[0].addr & ~PAGE_MASK)) { ret = -EINVAL; @@ -492,7 +495,17 @@ static struct ib_mr *iwch_register_phys_mem(struct ib_pd *pd, if (ret) goto err; - mhp->rhp = rhp; + ret = iwch_alloc_pbl(mhp, npages); + if (ret) { + kfree(page_list); + goto err_pbl; + } + + ret = iwch_write_pbl(mhp, page_list, npages, 0); + kfree(page_list); + if (ret) + goto err_pbl; + mhp->attr.pdid = php->pdid; mhp->attr.zbva = 0; @@ -502,12 +515,15 @@ static struct ib_mr *iwch_register_phys_mem(struct ib_pd *pd, mhp->attr.len = (u32) total_size; mhp->attr.pbl_size = npages; - ret = iwch_register_mem(rhp, php, mhp, shift, page_list); - kfree(page_list); - if (ret) { - goto err; - } + ret = iwch_register_mem(rhp, php, mhp, shift); + if (ret) + goto err_pbl; + return &mhp->ibmr; + +err_pbl: + iwch_free_pbl(mhp); + err: kfree(mhp); return ERR_PTR(ret); @@ -560,7 +576,7 @@ static int iwch_reregister_phys_mem(struct ib_mr *mr, return ret; } - ret = iwch_reregister_mem(rhp, php, &mh, shift, page_list, npages); + ret = iwch_reregister_mem(rhp, php, &mh, shift, npages); kfree(page_list); if (ret) { return ret; @@ -602,6 +618,8 @@ static struct ib_mr *iwch_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, if (!mhp) return ERR_PTR(-ENOMEM); + mhp->rhp = rhp; + mhp->umem = ib_umem_get(pd->uobject->context, start, length, acc, 0); if (IS_ERR(mhp->umem)) { err = PTR_ERR(mhp->umem); @@ -615,10 +633,14 @@ static struct ib_mr *iwch_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, list_for_each_entry(chunk, &mhp->umem->chunk_list, list) n += chunk->nents; - pages = kmalloc(n * sizeof(u64), GFP_KERNEL); + err = iwch_alloc_pbl(mhp, n); + if (err) + goto err; + + pages = (__be64 *) __get_free_page(GFP_KERNEL); if (!pages) { err = -ENOMEM; - goto err; + goto err_pbl; } i = n = 0; @@ -630,25 +652,38 @@ static struct ib_mr *iwch_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, pages[i++] = cpu_to_be64(sg_dma_address( &chunk->page_list[j]) + mhp->umem->page_size * k); + if (i == PAGE_SIZE / sizeof *pages) { + err = iwch_write_pbl(mhp, pages, i, n); + if (err) + goto pbl_done; + n += i; + i = 0; + } } } - mhp->rhp = rhp; + if (i) + err = iwch_write_pbl(mhp, pages, i, n); + +pbl_done: + free_page((unsigned long) pages); + if (err) + goto err_pbl; + mhp->attr.pdid = php->pdid; mhp->attr.zbva = 0; mhp->attr.perms = iwch_ib_to_tpt_access(acc); mhp->attr.va_fbo = virt; mhp->attr.page_size = shift - 12; mhp->attr.len = (u32) length; - mhp->attr.pbl_size = i; - err = iwch_register_mem(rhp, php, mhp, shift, pages); - kfree(pages); + + err = iwch_register_mem(rhp, php, mhp, shift); if (err) - goto err; + goto err_pbl; if (udata && !t3a_device(rhp)) { uresp.pbl_addr = (mhp->attr.pbl_addr - - rhp->rdev.rnic_info.pbl_base) >> 3; + rhp->rdev.rnic_info.pbl_base) >> 3; PDBG("%s user resp pbl_addr 0x%x\n", __func__, uresp.pbl_addr); @@ -661,6 +696,9 @@ static struct ib_mr *iwch_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, return &mhp->ibmr; +err_pbl: + iwch_free_pbl(mhp); + err: ib_umem_release(mhp->umem); kfree(mhp); diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.h b/drivers/infiniband/hw/cxgb3/iwch_provider.h index db5100d27ca..836163fc542 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_provider.h +++ b/drivers/infiniband/hw/cxgb3/iwch_provider.h @@ -340,14 +340,14 @@ int iwch_quiesce_qps(struct iwch_cq *chp); int iwch_resume_qps(struct iwch_cq *chp); void stop_read_rep_timer(struct iwch_qp *qhp); int iwch_register_mem(struct iwch_dev *rhp, struct iwch_pd *php, - struct iwch_mr *mhp, - int shift, - __be64 *page_list); + struct iwch_mr *mhp, int shift); int iwch_reregister_mem(struct iwch_dev *rhp, struct iwch_pd *php, struct iwch_mr *mhp, int shift, - __be64 *page_list, int npages); +int iwch_alloc_pbl(struct iwch_mr *mhp, int npages); +void iwch_free_pbl(struct iwch_mr *mhp); +int iwch_write_pbl(struct iwch_mr *mhp, __be64 *pages, int npages, int offset); int build_phys_page_list(struct ib_phys_buf *buffer_list, int num_phys_buf, u64 *iova_start, diff --git a/drivers/infiniband/hw/cxgb3/iwch_qp.c b/drivers/infiniband/hw/cxgb3/iwch_qp.c index 9b4be889c58..79dbe5beae5 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_qp.c +++ b/drivers/infiniband/hw/cxgb3/iwch_qp.c @@ -655,6 +655,7 @@ static void __flush_qp(struct iwch_qp *qhp, unsigned long *flag) { struct iwch_cq *rchp, *schp; int count; + int flushed; rchp = get_chp(qhp->rhp, qhp->attr.rcq); schp = get_chp(qhp->rhp, qhp->attr.scq); @@ -669,20 +670,22 @@ static void __flush_qp(struct iwch_qp *qhp, unsigned long *flag) spin_lock(&qhp->lock); cxio_flush_hw_cq(&rchp->cq); cxio_count_rcqes(&rchp->cq, &qhp->wq, &count); - cxio_flush_rq(&qhp->wq, &rchp->cq, count); + flushed = cxio_flush_rq(&qhp->wq, &rchp->cq, count); spin_unlock(&qhp->lock); spin_unlock_irqrestore(&rchp->lock, *flag); - (*rchp->ibcq.comp_handler)(&rchp->ibcq, rchp->ibcq.cq_context); + if (flushed) + (*rchp->ibcq.comp_handler)(&rchp->ibcq, rchp->ibcq.cq_context); /* locking heirarchy: cq lock first, then qp lock. */ spin_lock_irqsave(&schp->lock, *flag); spin_lock(&qhp->lock); cxio_flush_hw_cq(&schp->cq); cxio_count_scqes(&schp->cq, &qhp->wq, &count); - cxio_flush_sq(&qhp->wq, &schp->cq, count); + flushed = cxio_flush_sq(&qhp->wq, &schp->cq, count); spin_unlock(&qhp->lock); spin_unlock_irqrestore(&schp->lock, *flag); - (*schp->ibcq.comp_handler)(&schp->ibcq, schp->ibcq.cq_context); + if (flushed) + (*schp->ibcq.comp_handler)(&schp->ibcq, schp->ibcq.cq_context); /* deref */ if (atomic_dec_and_test(&qhp->refcnt)) @@ -880,7 +883,6 @@ int iwch_modify_qp(struct iwch_dev *rhp, struct iwch_qp *qhp, ep = qhp->ep; get_ep(&ep->com); } - flush_qp(qhp, &flag); break; case IWCH_QP_STATE_TERMINATE: qhp->attr.state = IWCH_QP_STATE_TERMINATE; @@ -911,6 +913,7 @@ int iwch_modify_qp(struct iwch_dev *rhp, struct iwch_qp *qhp, } switch (attrs->next_state) { case IWCH_QP_STATE_IDLE: + flush_qp(qhp, &flag); qhp->attr.state = IWCH_QP_STATE_IDLE; qhp->attr.llp_stream_handle = NULL; put_ep(&qhp->ep->com); diff --git a/drivers/infiniband/hw/ehca/ehca_classes.h b/drivers/infiniband/hw/ehca/ehca_classes.h index 00bab60f6de..1e9e99a1393 100644 --- a/drivers/infiniband/hw/ehca/ehca_classes.h +++ b/drivers/infiniband/hw/ehca/ehca_classes.h @@ -192,6 +192,8 @@ struct ehca_qp { int mtu_shift; u32 message_count; u32 packet_count; + atomic_t nr_events; /* events seen */ + wait_queue_head_t wait_completion; }; #define IS_SRQ(qp) (qp->ext_type == EQPT_SRQ) diff --git a/drivers/infiniband/hw/ehca/ehca_hca.c b/drivers/infiniband/hw/ehca/ehca_hca.c index 2515cbde7e6..bc3b37d2070 100644 --- a/drivers/infiniband/hw/ehca/ehca_hca.c +++ b/drivers/infiniband/hw/ehca/ehca_hca.c @@ -101,7 +101,6 @@ int ehca_query_device(struct ib_device *ibdev, struct ib_device_attr *props) props->max_ee = limit_uint(rblock->max_rd_ee_context); props->max_rdd = limit_uint(rblock->max_rd_domain); props->max_fmr = limit_uint(rblock->max_mr); - props->local_ca_ack_delay = limit_uint(rblock->local_ca_ack_delay); props->max_qp_rd_atom = limit_uint(rblock->max_rr_qp); props->max_ee_rd_atom = limit_uint(rblock->max_rr_ee_context); props->max_res_rd_atom = limit_uint(rblock->max_rr_hca); @@ -115,7 +114,7 @@ int ehca_query_device(struct ib_device *ibdev, struct ib_device_attr *props) } props->max_pkeys = 16; - props->local_ca_ack_delay = limit_uint(rblock->local_ca_ack_delay); + props->local_ca_ack_delay = min_t(u8, rblock->local_ca_ack_delay, 255); props->max_raw_ipv6_qp = limit_uint(rblock->max_raw_ipv6_qp); props->max_raw_ethy_qp = limit_uint(rblock->max_raw_ethy_qp); props->max_mcast_grp = limit_uint(rblock->max_mcast_grp); @@ -136,7 +135,7 @@ query_device1: return ret; } -static int map_mtu(struct ehca_shca *shca, u32 fw_mtu) +static enum ib_mtu map_mtu(struct ehca_shca *shca, u32 fw_mtu) { switch (fw_mtu) { case 0x1: @@ -156,7 +155,7 @@ static int map_mtu(struct ehca_shca *shca, u32 fw_mtu) } } -static int map_number_of_vls(struct ehca_shca *shca, u32 vl_cap) +static u8 map_number_of_vls(struct ehca_shca *shca, u32 vl_cap) { switch (vl_cap) { case 0x1: diff --git a/drivers/infiniband/hw/ehca/ehca_irq.c b/drivers/infiniband/hw/ehca/ehca_irq.c index ca5eb0cb628..ce1ab0571be 100644 --- a/drivers/infiniband/hw/ehca/ehca_irq.c +++ b/drivers/infiniband/hw/ehca/ehca_irq.c @@ -204,6 +204,8 @@ static void qp_event_callback(struct ehca_shca *shca, u64 eqe, read_lock(&ehca_qp_idr_lock); qp = idr_find(&ehca_qp_idr, token); + if (qp) + atomic_inc(&qp->nr_events); read_unlock(&ehca_qp_idr_lock); if (!qp) @@ -223,6 +225,8 @@ static void qp_event_callback(struct ehca_shca *shca, u64 eqe, if (fatal && qp->ext_type == EQPT_SRQBASE) dispatch_qp_event(shca, qp, IB_EVENT_QP_LAST_WQE_REACHED); + if (atomic_dec_and_test(&qp->nr_events)) + wake_up(&qp->wait_completion); return; } diff --git a/drivers/infiniband/hw/ehca/ehca_qp.c b/drivers/infiniband/hw/ehca/ehca_qp.c index 18fba92fa7a..3f59587338e 100644 --- a/drivers/infiniband/hw/ehca/ehca_qp.c +++ b/drivers/infiniband/hw/ehca/ehca_qp.c @@ -566,6 +566,8 @@ static struct ehca_qp *internal_create_qp( return ERR_PTR(-ENOMEM); } + atomic_set(&my_qp->nr_events, 0); + init_waitqueue_head(&my_qp->wait_completion); spin_lock_init(&my_qp->spinlock_s); spin_lock_init(&my_qp->spinlock_r); my_qp->qp_type = qp_type; @@ -1934,6 +1936,9 @@ static int internal_destroy_qp(struct ib_device *dev, struct ehca_qp *my_qp, idr_remove(&ehca_qp_idr, my_qp->token); write_unlock_irqrestore(&ehca_qp_idr_lock, flags); + /* now wait until all pending events have completed */ + wait_event(my_qp->wait_completion, !atomic_read(&my_qp->nr_events)); + h_ret = hipz_h_destroy_qp(shca->ipz_hca_handle, my_qp); if (h_ret != H_SUCCESS) { ehca_err(dev, "hipz_h_destroy_qp() failed h_ret=%li " diff --git a/drivers/infiniband/hw/ipath/ipath_driver.c b/drivers/infiniband/hw/ipath/ipath_driver.c index acf30c06a0c..ce7b7c34360 100644 --- a/drivers/infiniband/hw/ipath/ipath_driver.c +++ b/drivers/infiniband/hw/ipath/ipath_driver.c @@ -1197,7 +1197,7 @@ void ipath_kreceive(struct ipath_portdata *pd) } reloop: - for (last = 0, i = 1; !last; i++) { + for (last = 0, i = 1; !last; i += !last) { hdr = dd->ipath_f_get_msgheader(dd, rhf_addr); eflags = ipath_hdrget_err_flags(rhf_addr); etype = ipath_hdrget_rcv_type(rhf_addr); @@ -1428,6 +1428,40 @@ static void ipath_update_pio_bufs(struct ipath_devdata *dd) spin_unlock_irqrestore(&ipath_pioavail_lock, flags); } +/* + * used to force update of pioavailshadow if we can't get a pio buffer. + * Needed primarily due to exitting freeze mode after recovering + * from errors. Done lazily, because it's safer (known to not + * be writing pio buffers). + */ +static void ipath_reset_availshadow(struct ipath_devdata *dd) +{ + int i, im; + unsigned long flags; + + spin_lock_irqsave(&ipath_pioavail_lock, flags); + for (i = 0; i < dd->ipath_pioavregs; i++) { + u64 val, oldval; + /* deal with 6110 chip bug on high register #s */ + im = (i > 3 && (dd->ipath_flags & IPATH_SWAP_PIOBUFS)) ? + i ^ 1 : i; + val = le64_to_cpu(dd->ipath_pioavailregs_dma[im]); + /* + * busy out the buffers not in the kernel avail list, + * without changing the generation bits. + */ + oldval = dd->ipath_pioavailshadow[i]; + dd->ipath_pioavailshadow[i] = val | + ((~dd->ipath_pioavailkernel[i] << + INFINIPATH_SENDPIOAVAIL_BUSY_SHIFT) & + 0xaaaaaaaaaaaaaaaaULL); /* All BUSY bits in qword */ + if (oldval != dd->ipath_pioavailshadow[i]) + ipath_dbg("shadow[%d] was %Lx, now %lx\n", + i, oldval, dd->ipath_pioavailshadow[i]); + } + spin_unlock_irqrestore(&ipath_pioavail_lock, flags); +} + /** * ipath_setrcvhdrsize - set the receive header size * @dd: the infinipath device @@ -1482,9 +1516,12 @@ static noinline void no_pio_bufs(struct ipath_devdata *dd) */ ipath_stats.sps_nopiobufs++; if (!(++dd->ipath_consec_nopiobuf % 100000)) { - ipath_dbg("%u pio sends with no bufavail; dmacopy: " - "%llx %llx %llx %llx; shadow: %lx %lx %lx %lx\n", + ipath_force_pio_avail_update(dd); /* at start */ + ipath_dbg("%u tries no piobufavail ts%lx; dmacopy: " + "%llx %llx %llx %llx\n" + "ipath shadow: %lx %lx %lx %lx\n", dd->ipath_consec_nopiobuf, + (unsigned long)get_cycles(), (unsigned long long) le64_to_cpu(dma[0]), (unsigned long long) le64_to_cpu(dma[1]), (unsigned long long) le64_to_cpu(dma[2]), @@ -1496,14 +1533,17 @@ static noinline void no_pio_bufs(struct ipath_devdata *dd) */ if ((dd->ipath_piobcnt2k + dd->ipath_piobcnt4k) > (sizeof(shadow[0]) * 4 * 4)) - ipath_dbg("2nd group: dmacopy: %llx %llx " - "%llx %llx; shadow: %lx %lx %lx %lx\n", + ipath_dbg("2nd group: dmacopy: " + "%llx %llx %llx %llx\n" + "ipath shadow: %lx %lx %lx %lx\n", (unsigned long long)le64_to_cpu(dma[4]), (unsigned long long)le64_to_cpu(dma[5]), (unsigned long long)le64_to_cpu(dma[6]), (unsigned long long)le64_to_cpu(dma[7]), - shadow[4], shadow[5], shadow[6], - shadow[7]); + shadow[4], shadow[5], shadow[6], shadow[7]); + + /* at end, so update likely happened */ + ipath_reset_availshadow(dd); } } @@ -1652,19 +1692,46 @@ void ipath_chg_pioavailkernel(struct ipath_devdata *dd, unsigned start, unsigned len, int avail) { unsigned long flags; - unsigned end; + unsigned end, cnt = 0, next; /* There are two bits per send buffer (busy and generation) */ start *= 2; - len *= 2; - end = start + len; + end = start + len * 2; - /* Set or clear the generation bits. */ spin_lock_irqsave(&ipath_pioavail_lock, flags); + /* Set or clear the busy bit in the shadow. */ while (start < end) { if (avail) { - __clear_bit(start + INFINIPATH_SENDPIOAVAIL_BUSY_SHIFT, - dd->ipath_pioavailshadow); + unsigned long dma; + int i, im; + /* + * the BUSY bit will never be set, because we disarm + * the user buffers before we hand them back to the + * kernel. We do have to make sure the generation + * bit is set correctly in shadow, since it could + * have changed many times while allocated to user. + * We can't use the bitmap functions on the full + * dma array because it is always little-endian, so + * we have to flip to host-order first. + * BITS_PER_LONG is slightly wrong, since it's + * always 64 bits per register in chip... + * We only work on 64 bit kernels, so that's OK. + */ + /* deal with 6110 chip bug on high register #s */ + i = start / BITS_PER_LONG; + im = (i > 3 && (dd->ipath_flags & IPATH_SWAP_PIOBUFS)) ? + i ^ 1 : i; + __clear_bit(INFINIPATH_SENDPIOAVAIL_BUSY_SHIFT + + start, dd->ipath_pioavailshadow); + dma = (unsigned long) le64_to_cpu( + dd->ipath_pioavailregs_dma[im]); + if (test_bit((INFINIPATH_SENDPIOAVAIL_CHECK_SHIFT + + start) % BITS_PER_LONG, &dma)) + __set_bit(INFINIPATH_SENDPIOAVAIL_CHECK_SHIFT + + start, dd->ipath_pioavailshadow); + else + __clear_bit(INFINIPATH_SENDPIOAVAIL_CHECK_SHIFT + + start, dd->ipath_pioavailshadow); __set_bit(start, dd->ipath_pioavailkernel); } else { __set_bit(start + INFINIPATH_SENDPIOAVAIL_BUSY_SHIFT, @@ -1673,7 +1740,44 @@ void ipath_chg_pioavailkernel(struct ipath_devdata *dd, unsigned start, } start += 2; } + + if (dd->ipath_pioupd_thresh) { + end = 2 * (dd->ipath_piobcnt2k + dd->ipath_piobcnt4k); + next = find_first_bit(dd->ipath_pioavailkernel, end); + while (next < end) { + cnt++; + next = find_next_bit(dd->ipath_pioavailkernel, end, + next + 1); + } + } spin_unlock_irqrestore(&ipath_pioavail_lock, flags); + + /* + * When moving buffers from kernel to user, if number assigned to + * the user is less than the pio update threshold, and threshold + * is supported (cnt was computed > 0), drop the update threshold + * so we update at least once per allocated number of buffers. + * In any case, if the kernel buffers are less than the threshold, + * drop the threshold. We don't bother increasing it, having once + * decreased it, since it would typically just cycle back and forth. + * If we don't decrease below buffers in use, we can wait a long + * time for an update, until some other context uses PIO buffers. + */ + if (!avail && len < cnt) + cnt = len; + if (cnt < dd->ipath_pioupd_thresh) { + dd->ipath_pioupd_thresh = cnt; + ipath_dbg("Decreased pio update threshold to %u\n", + dd->ipath_pioupd_thresh); + spin_lock_irqsave(&dd->ipath_sendctrl_lock, flags); + dd->ipath_sendctrl &= ~(INFINIPATH_S_UPDTHRESH_MASK + << INFINIPATH_S_UPDTHRESH_SHIFT); + dd->ipath_sendctrl |= dd->ipath_pioupd_thresh + << INFINIPATH_S_UPDTHRESH_SHIFT; + ipath_write_kreg(dd, dd->ipath_kregs->kr_sendctrl, + dd->ipath_sendctrl); + spin_unlock_irqrestore(&dd->ipath_sendctrl_lock, flags); + } } /** @@ -1794,8 +1898,8 @@ void ipath_cancel_sends(struct ipath_devdata *dd, int restore_sendctrl) spin_lock_irqsave(&dd->ipath_sdma_lock, flags); skip_cancel = - !test_bit(IPATH_SDMA_DISABLED, statp) && - test_and_set_bit(IPATH_SDMA_ABORTING, statp); + test_and_set_bit(IPATH_SDMA_ABORTING, statp) + && !test_bit(IPATH_SDMA_DISABLED, statp); spin_unlock_irqrestore(&dd->ipath_sdma_lock, flags); if (skip_cancel) goto bail; @@ -1826,6 +1930,9 @@ void ipath_cancel_sends(struct ipath_devdata *dd, int restore_sendctrl) ipath_disarm_piobufs(dd, 0, dd->ipath_piobcnt2k + dd->ipath_piobcnt4k); + if (dd->ipath_flags & IPATH_HAS_SEND_DMA) + set_bit(IPATH_SDMA_DISARMED, &dd->ipath_sdma_status); + if (restore_sendctrl) { /* else done by caller later if needed */ spin_lock_irqsave(&dd->ipath_sendctrl_lock, flags); @@ -1845,7 +1952,6 @@ void ipath_cancel_sends(struct ipath_devdata *dd, int restore_sendctrl) /* only wait so long for intr */ dd->ipath_sdma_abort_intr_timeout = jiffies + HZ; dd->ipath_sdma_reset_wait = 200; - __set_bit(IPATH_SDMA_DISARMED, &dd->ipath_sdma_status); if (!test_bit(IPATH_SDMA_SHUTDOWN, &dd->ipath_sdma_status)) tasklet_hi_schedule(&dd->ipath_sdma_abort_task); spin_unlock_irqrestore(&dd->ipath_sdma_lock, flags); diff --git a/drivers/infiniband/hw/ipath/ipath_file_ops.c b/drivers/infiniband/hw/ipath/ipath_file_ops.c index 8b1752202e7..3295177c937 100644 --- a/drivers/infiniband/hw/ipath/ipath_file_ops.c +++ b/drivers/infiniband/hw/ipath/ipath_file_ops.c @@ -173,47 +173,25 @@ static int ipath_get_base_info(struct file *fp, (void *) dd->ipath_statusp - (void *) dd->ipath_pioavailregs_dma; if (!shared) { - kinfo->spi_piocnt = dd->ipath_pbufsport; + kinfo->spi_piocnt = pd->port_piocnt; kinfo->spi_piobufbase = (u64) pd->port_piobufs; kinfo->__spi_uregbase = (u64) dd->ipath_uregbase + dd->ipath_ureg_align * pd->port_port; } else if (master) { - kinfo->spi_piocnt = (dd->ipath_pbufsport / subport_cnt) + - (dd->ipath_pbufsport % subport_cnt); + kinfo->spi_piocnt = (pd->port_piocnt / subport_cnt) + + (pd->port_piocnt % subport_cnt); /* Master's PIO buffers are after all the slave's */ kinfo->spi_piobufbase = (u64) pd->port_piobufs + dd->ipath_palign * - (dd->ipath_pbufsport - kinfo->spi_piocnt); + (pd->port_piocnt - kinfo->spi_piocnt); } else { unsigned slave = subport_fp(fp) - 1; - kinfo->spi_piocnt = dd->ipath_pbufsport / subport_cnt; + kinfo->spi_piocnt = pd->port_piocnt / subport_cnt; kinfo->spi_piobufbase = (u64) pd->port_piobufs + dd->ipath_palign * kinfo->spi_piocnt * slave; } - /* - * Set the PIO avail update threshold to no larger - * than the number of buffers per process. Note that - * we decrease it here, but won't ever increase it. - */ - if (dd->ipath_pioupd_thresh && - kinfo->spi_piocnt < dd->ipath_pioupd_thresh) { - unsigned long flags; - - dd->ipath_pioupd_thresh = kinfo->spi_piocnt; - ipath_dbg("Decreased pio update threshold to %u\n", - dd->ipath_pioupd_thresh); - spin_lock_irqsave(&dd->ipath_sendctrl_lock, flags); - dd->ipath_sendctrl &= ~(INFINIPATH_S_UPDTHRESH_MASK - << INFINIPATH_S_UPDTHRESH_SHIFT); - dd->ipath_sendctrl |= dd->ipath_pioupd_thresh - << INFINIPATH_S_UPDTHRESH_SHIFT; - ipath_write_kreg(dd, dd->ipath_kregs->kr_sendctrl, - dd->ipath_sendctrl); - spin_unlock_irqrestore(&dd->ipath_sendctrl_lock, flags); - } - if (shared) { kinfo->spi_port_uregbase = (u64) dd->ipath_uregbase + dd->ipath_ureg_align * pd->port_port; @@ -1309,19 +1287,19 @@ static int ipath_mmap(struct file *fp, struct vm_area_struct *vma) ureg = dd->ipath_uregbase + dd->ipath_ureg_align * pd->port_port; if (!pd->port_subport_cnt) { /* port is not shared */ - piocnt = dd->ipath_pbufsport; + piocnt = pd->port_piocnt; piobufs = pd->port_piobufs; } else if (!subport_fp(fp)) { /* caller is the master */ - piocnt = (dd->ipath_pbufsport / pd->port_subport_cnt) + - (dd->ipath_pbufsport % pd->port_subport_cnt); + piocnt = (pd->port_piocnt / pd->port_subport_cnt) + + (pd->port_piocnt % pd->port_subport_cnt); piobufs = pd->port_piobufs + - dd->ipath_palign * (dd->ipath_pbufsport - piocnt); + dd->ipath_palign * (pd->port_piocnt - piocnt); } else { unsigned slave = subport_fp(fp) - 1; /* caller is a slave */ - piocnt = dd->ipath_pbufsport / pd->port_subport_cnt; + piocnt = pd->port_piocnt / pd->port_subport_cnt; piobufs = pd->port_piobufs + dd->ipath_palign * piocnt * slave; } @@ -1633,9 +1611,6 @@ static int try_alloc_port(struct ipath_devdata *dd, int port, port_fp(fp) = pd; pd->port_pid = current->pid; strncpy(pd->port_comm, current->comm, sizeof(pd->port_comm)); - ipath_chg_pioavailkernel(dd, - dd->ipath_pbufsport * (pd->port_port - 1), - dd->ipath_pbufsport, 0); ipath_stats.sps_ports++; ret = 0; } else @@ -1938,11 +1913,25 @@ static int ipath_do_user_init(struct file *fp, /* for now we do nothing with rcvhdrcnt: uinfo->spu_rcvhdrcnt */ + /* some ports may get extra buffers, calculate that here */ + if (pd->port_port <= dd->ipath_ports_extrabuf) + pd->port_piocnt = dd->ipath_pbufsport + 1; + else + pd->port_piocnt = dd->ipath_pbufsport; + /* for right now, kernel piobufs are at end, so port 1 is at 0 */ + if (pd->port_port <= dd->ipath_ports_extrabuf) + pd->port_pio_base = (dd->ipath_pbufsport + 1) + * (pd->port_port - 1); + else + pd->port_pio_base = dd->ipath_ports_extrabuf + + dd->ipath_pbufsport * (pd->port_port - 1); pd->port_piobufs = dd->ipath_piobufbase + - dd->ipath_pbufsport * (pd->port_port - 1) * dd->ipath_palign; - ipath_cdbg(VERBOSE, "Set base of piobufs for port %u to 0x%x\n", - pd->port_port, pd->port_piobufs); + pd->port_pio_base * dd->ipath_palign; + ipath_cdbg(VERBOSE, "piobuf base for port %u is 0x%x, piocnt %u," + " first pio %u\n", pd->port_port, pd->port_piobufs, + pd->port_piocnt, pd->port_pio_base); + ipath_chg_pioavailkernel(dd, pd->port_pio_base, pd->port_piocnt, 0); /* * Now allocate the rcvhdr Q and eager TIDs; skip the TID @@ -2107,7 +2096,6 @@ static int ipath_close(struct inode *in, struct file *fp) } if (dd->ipath_kregbase) { - int i; /* atomically clear receive enable port and intr avail. */ clear_bit(dd->ipath_r_portenable_shift + port, &dd->ipath_rcvctrl); @@ -2136,9 +2124,9 @@ static int ipath_close(struct inode *in, struct file *fp) ipath_write_kreg_port(dd, dd->ipath_kregs->kr_rcvhdraddr, pd->port_port, dd->ipath_dummy_hdrq_phys); - i = dd->ipath_pbufsport * (port - 1); - ipath_disarm_piobufs(dd, i, dd->ipath_pbufsport); - ipath_chg_pioavailkernel(dd, i, dd->ipath_pbufsport, 1); + ipath_disarm_piobufs(dd, pd->port_pio_base, pd->port_piocnt); + ipath_chg_pioavailkernel(dd, pd->port_pio_base, + pd->port_piocnt, 1); dd->ipath_f_clear_tids(dd, pd->port_port); diff --git a/drivers/infiniband/hw/ipath/ipath_iba7220.c b/drivers/infiniband/hw/ipath/ipath_iba7220.c index e3ec0d1bdf5..8eee7830f04 100644 --- a/drivers/infiniband/hw/ipath/ipath_iba7220.c +++ b/drivers/infiniband/hw/ipath/ipath_iba7220.c @@ -595,7 +595,7 @@ static void ipath_7220_txe_recover(struct ipath_devdata *dd) dev_info(&dd->pcidev->dev, "Recovering from TXE PIO parity error\n"); - ipath_disarm_senderrbufs(dd, 1); + ipath_disarm_senderrbufs(dd); } @@ -675,10 +675,8 @@ static void ipath_7220_handle_hwerrors(struct ipath_devdata *dd, char *msg, ctrl = ipath_read_kreg32(dd, dd->ipath_kregs->kr_control); if ((ctrl & INFINIPATH_C_FREEZEMODE) && !ipath_diag_inuse) { /* - * Parity errors in send memory are recoverable, - * just cancel the send (if indicated in * sendbuffererror), - * count the occurrence, unfreeze (if no other handled - * hardware error bits are set), and continue. + * Parity errors in send memory are recoverable by h/w + * just do housekeeping, exit freeze mode and continue. */ if (hwerrs & ((INFINIPATH_HWE_TXEMEMPARITYERR_PIOBUF | INFINIPATH_HWE_TXEMEMPARITYERR_PIOPBC) @@ -687,13 +685,6 @@ static void ipath_7220_handle_hwerrors(struct ipath_devdata *dd, char *msg, hwerrs &= ~((INFINIPATH_HWE_TXEMEMPARITYERR_PIOBUF | INFINIPATH_HWE_TXEMEMPARITYERR_PIOPBC) << INFINIPATH_HWE_TXEMEMPARITYERR_SHIFT); - if (!hwerrs) { - /* else leave in freeze mode */ - ipath_write_kreg(dd, - dd->ipath_kregs->kr_control, - dd->ipath_control); - goto bail; - } } if (hwerrs) { /* @@ -723,8 +714,8 @@ static void ipath_7220_handle_hwerrors(struct ipath_devdata *dd, char *msg, *dd->ipath_statusp |= IPATH_STATUS_HWERROR; dd->ipath_flags &= ~IPATH_INITTED; } else { - ipath_dbg("Clearing freezemode on ignored hardware " - "error\n"); + ipath_dbg("Clearing freezemode on ignored or " + "recovered hardware error\n"); ipath_clear_freeze(dd); } } @@ -870,8 +861,9 @@ static int ipath_7220_boardname(struct ipath_devdata *dd, char *name, "revision %u.%u!\n", dd->ipath_majrev, dd->ipath_minrev); ret = 1; - } else if (dd->ipath_minrev == 1) { - /* Rev1 chips are prototype. Complain, but allow use */ + } else if (dd->ipath_minrev == 1 && + !(dd->ipath_flags & IPATH_INITTED)) { + /* Rev1 chips are prototype. Complain at init, but allow use */ ipath_dev_err(dd, "Unsupported hardware " "revision %u.%u, Contact support@qlogic.com\n", dd->ipath_majrev, dd->ipath_minrev); @@ -1966,7 +1958,7 @@ static void ipath_7220_config_ports(struct ipath_devdata *dd, ushort cfgports) dd->ipath_rcvctrl); dd->ipath_p0_rcvegrcnt = 2048; /* always */ if (dd->ipath_flags & IPATH_HAS_SEND_DMA) - dd->ipath_pioreserved = 1; /* reserve a buffer */ + dd->ipath_pioreserved = 3; /* kpiobufs used for PIO */ } diff --git a/drivers/infiniband/hw/ipath/ipath_init_chip.c b/drivers/infiniband/hw/ipath/ipath_init_chip.c index 27dd8947666..3e5baa43fc8 100644 --- a/drivers/infiniband/hw/ipath/ipath_init_chip.c +++ b/drivers/infiniband/hw/ipath/ipath_init_chip.c @@ -41,7 +41,7 @@ /* * min buffers we want to have per port, after driver */ -#define IPATH_MIN_USER_PORT_BUFCNT 8 +#define IPATH_MIN_USER_PORT_BUFCNT 7 /* * Number of ports we are configured to use (to allow for more pio @@ -54,13 +54,9 @@ MODULE_PARM_DESC(cfgports, "Set max number of ports to use"); /* * Number of buffers reserved for driver (verbs and layered drivers.) - * Reserved at end of buffer list. Initialized based on - * number of PIO buffers if not set via module interface. + * Initialized based on number of PIO buffers if not set via module interface. * The problem with this is that it's global, but we'll use different - * numbers for different chip types. So the default value is not - * very useful. I've redefined it for the 1.3 release so that it's - * zero unless set by the user to something else, in which case we - * try to respect it. + * numbers for different chip types. */ static ushort ipath_kpiobufs; @@ -546,9 +542,12 @@ static void enable_chip(struct ipath_devdata *dd, int reinit) pioavail = dd->ipath_pioavailregs_dma[i ^ 1]; else pioavail = dd->ipath_pioavailregs_dma[i]; - dd->ipath_pioavailshadow[i] = le64_to_cpu(pioavail) | - (~dd->ipath_pioavailkernel[i] << - INFINIPATH_SENDPIOAVAIL_BUSY_SHIFT); + /* + * don't need to worry about ipath_pioavailkernel here + * because we will call ipath_chg_pioavailkernel() later + * in initialization, to busy out buffers as needed + */ + dd->ipath_pioavailshadow[i] = le64_to_cpu(pioavail); } /* can get counters, stats, etc. */ dd->ipath_flags |= IPATH_PRESENT; @@ -708,12 +707,11 @@ static void verify_interrupt(unsigned long opaque) int ipath_init_chip(struct ipath_devdata *dd, int reinit) { int ret = 0; - u32 val32, kpiobufs; + u32 kpiobufs, defkbufs; u32 piobufs, uports; u64 val; struct ipath_portdata *pd; gfp_t gfp_flags = GFP_USER | __GFP_COMP; - unsigned long flags; ret = init_housekeeping(dd, reinit); if (ret) @@ -753,56 +751,46 @@ int ipath_init_chip(struct ipath_devdata *dd, int reinit) dd->ipath_pioavregs = ALIGN(piobufs, sizeof(u64) * BITS_PER_BYTE / 2) / (sizeof(u64) * BITS_PER_BYTE / 2); uports = dd->ipath_cfgports ? dd->ipath_cfgports - 1 : 0; - if (ipath_kpiobufs == 0) { - /* not set by user (this is default) */ - if (piobufs > 144) - kpiobufs = 32; - else - kpiobufs = 16; - } + if (piobufs > 144) + defkbufs = 32 + dd->ipath_pioreserved; else - kpiobufs = ipath_kpiobufs; + defkbufs = 16 + dd->ipath_pioreserved; - if (kpiobufs + (uports * IPATH_MIN_USER_PORT_BUFCNT) > piobufs) { + if (ipath_kpiobufs && (ipath_kpiobufs + + (uports * IPATH_MIN_USER_PORT_BUFCNT)) > piobufs) { int i = (int) piobufs - (int) (uports * IPATH_MIN_USER_PORT_BUFCNT); if (i < 1) i = 1; dev_info(&dd->pcidev->dev, "Allocating %d PIO bufs of " "%d for kernel leaves too few for %d user ports " - "(%d each); using %u\n", kpiobufs, + "(%d each); using %u\n", ipath_kpiobufs, piobufs, uports, IPATH_MIN_USER_PORT_BUFCNT, i); /* * shouldn't change ipath_kpiobufs, because could be * different for different devices... */ kpiobufs = i; - } + } else if (ipath_kpiobufs) + kpiobufs = ipath_kpiobufs; + else + kpiobufs = defkbufs; dd->ipath_lastport_piobuf = piobufs - kpiobufs; dd->ipath_pbufsport = uports ? dd->ipath_lastport_piobuf / uports : 0; - val32 = dd->ipath_lastport_piobuf - (dd->ipath_pbufsport * uports); - if (val32 > 0) { - ipath_dbg("allocating %u pbufs/port leaves %u unused, " - "add to kernel\n", dd->ipath_pbufsport, val32); - dd->ipath_lastport_piobuf -= val32; - kpiobufs += val32; - ipath_dbg("%u pbufs/port leaves %u unused, add to kernel\n", - dd->ipath_pbufsport, val32); - } + /* if not an even divisor, some user ports get extra buffers */ + dd->ipath_ports_extrabuf = dd->ipath_lastport_piobuf - + (dd->ipath_pbufsport * uports); + if (dd->ipath_ports_extrabuf) + ipath_dbg("%u pbufs/port leaves some unused, add 1 buffer to " + "ports <= %u\n", dd->ipath_pbufsport, + dd->ipath_ports_extrabuf); dd->ipath_lastpioindex = 0; dd->ipath_lastpioindexl = dd->ipath_piobcnt2k; - ipath_chg_pioavailkernel(dd, 0, piobufs, 1); + /* ipath_pioavailshadow initialized earlier */ ipath_cdbg(VERBOSE, "%d PIO bufs for kernel out of %d total %u " "each for %u user ports\n", kpiobufs, piobufs, dd->ipath_pbufsport, uports); - if (dd->ipath_pioupd_thresh) { - if (dd->ipath_pbufsport < dd->ipath_pioupd_thresh) - dd->ipath_pioupd_thresh = dd->ipath_pbufsport; - if (kpiobufs < dd->ipath_pioupd_thresh) - dd->ipath_pioupd_thresh = kpiobufs; - } - ret = dd->ipath_f_early_init(dd); if (ret) { ipath_dev_err(dd, "Early initialization failure\n"); @@ -810,13 +798,6 @@ int ipath_init_chip(struct ipath_devdata *dd, int reinit) } /* - * Cancel any possible active sends from early driver load. - * Follows early_init because some chips have to initialize - * PIO buffers in early_init to avoid false parity errors. - */ - ipath_cancel_sends(dd, 0); - - /* * Early_init sets rcvhdrentsize and rcvhdrsize, so this must be * done after early_init. */ @@ -836,6 +817,7 @@ int ipath_init_chip(struct ipath_devdata *dd, int reinit) ipath_write_kreg(dd, dd->ipath_kregs->kr_sendpioavailaddr, dd->ipath_pioavailregs_phys); + /* * this is to detect s/w errors, which the h/w works around by * ignoring the low 6 bits of address, if it wasn't aligned. @@ -862,12 +844,6 @@ int ipath_init_chip(struct ipath_devdata *dd, int reinit) ~0ULL&~INFINIPATH_HWE_MEMBISTFAILED); ipath_write_kreg(dd, dd->ipath_kregs->kr_control, 0ULL); - spin_lock_irqsave(&dd->ipath_sendctrl_lock, flags); - dd->ipath_sendctrl = INFINIPATH_S_PIOENABLE; - ipath_write_kreg(dd, dd->ipath_kregs->kr_sendctrl, dd->ipath_sendctrl); - ipath_read_kreg64(dd, dd->ipath_kregs->kr_scratch); - spin_unlock_irqrestore(&dd->ipath_sendctrl_lock, flags); - /* * before error clears, since we expect serdes pll errors during * this, the first time after reset @@ -940,6 +916,19 @@ int ipath_init_chip(struct ipath_devdata *dd, int reinit) else enable_chip(dd, reinit); + /* after enable_chip, so pioavailshadow setup */ + ipath_chg_pioavailkernel(dd, 0, piobufs, 1); + + /* + * Cancel any possible active sends from early driver load. + * Follows early_init because some chips have to initialize + * PIO buffers in early_init to avoid false parity errors. + * After enable and ipath_chg_pioavailkernel so we can safely + * enable pioavail updates and PIOENABLE; packets are now + * ready to go out. + */ + ipath_cancel_sends(dd, 1); + if (!reinit) { /* * Used when we close a port, for DMA already in flight diff --git a/drivers/infiniband/hw/ipath/ipath_intr.c b/drivers/infiniband/hw/ipath/ipath_intr.c index 1b58f4737c7..26900b3b7a4 100644 --- a/drivers/infiniband/hw/ipath/ipath_intr.c +++ b/drivers/infiniband/hw/ipath/ipath_intr.c @@ -38,42 +38,12 @@ #include "ipath_verbs.h" #include "ipath_common.h" -/* - * clear (write) a pio buffer, to clear a parity error. This routine - * should only be called when in freeze mode, and the buffer should be - * canceled afterwards. - */ -static void ipath_clrpiobuf(struct ipath_devdata *dd, u32 pnum) -{ - u32 __iomem *pbuf; - u32 dwcnt; /* dword count to write */ - if (pnum < dd->ipath_piobcnt2k) { - pbuf = (u32 __iomem *) (dd->ipath_pio2kbase + pnum * - dd->ipath_palign); - dwcnt = dd->ipath_piosize2k >> 2; - } - else { - pbuf = (u32 __iomem *) (dd->ipath_pio4kbase + - (pnum - dd->ipath_piobcnt2k) * dd->ipath_4kalign); - dwcnt = dd->ipath_piosize4k >> 2; - } - dev_info(&dd->pcidev->dev, - "Rewrite PIO buffer %u, to recover from parity error\n", - pnum); - - /* no flush required, since already in freeze */ - writel(dwcnt + 1, pbuf); - while (--dwcnt) - writel(0, pbuf++); -} /* * Called when we might have an error that is specific to a particular * PIO buffer, and may need to cancel that buffer, so it can be re-used. - * If rewrite is true, and bits are set in the sendbufferror registers, - * we'll write to the buffer, for error recovery on parity errors. */ -void ipath_disarm_senderrbufs(struct ipath_devdata *dd, int rewrite) +void ipath_disarm_senderrbufs(struct ipath_devdata *dd) { u32 piobcnt; unsigned long sbuf[4]; @@ -109,11 +79,8 @@ void ipath_disarm_senderrbufs(struct ipath_devdata *dd, int rewrite) } for (i = 0; i < piobcnt; i++) - if (test_bit(i, sbuf)) { - if (rewrite) - ipath_clrpiobuf(dd, i); + if (test_bit(i, sbuf)) ipath_disarm_piobufs(dd, i, 1); - } /* ignore armlaunch errs for a bit */ dd->ipath_lastcancel = jiffies+3; } @@ -164,7 +131,7 @@ static u64 handle_e_sum_errs(struct ipath_devdata *dd, ipath_err_t errs) { u64 ignore_this_time = 0; - ipath_disarm_senderrbufs(dd, 0); + ipath_disarm_senderrbufs(dd); if ((errs & E_SUM_LINK_PKTERRS) && !(dd->ipath_flags & IPATH_LINKACTIVE)) { /* @@ -909,8 +876,8 @@ static int handle_errors(struct ipath_devdata *dd, ipath_err_t errs) * processes (causing armlaunch), send errors due to going into freeze mode, * etc., and try to avoid causing extra interrupts while doing so. * Forcibly update the in-memory pioavail register copies after cleanup - * because the chip won't do it for anything changing while in freeze mode - * (we don't want to wait for the next pio buffer state change). + * because the chip won't do it while in freeze mode (the register values + * themselves are kept correct). * Make sure that we don't lose any important interrupts by using the chip * feature that says that writing 0 to a bit in *clear that is set in * *status will cause an interrupt to be generated again (if allowed by @@ -918,44 +885,23 @@ static int handle_errors(struct ipath_devdata *dd, ipath_err_t errs) */ void ipath_clear_freeze(struct ipath_devdata *dd) { - int i, im; - u64 val; - /* disable error interrupts, to avoid confusion */ ipath_write_kreg(dd, dd->ipath_kregs->kr_errormask, 0ULL); /* also disable interrupts; errormask is sometimes overwriten */ ipath_write_kreg(dd, dd->ipath_kregs->kr_intmask, 0ULL); - /* - * clear all sends, because they have may been - * completed by usercode while in freeze mode, and - * therefore would not be sent, and eventually - * might cause the process to run out of bufs - */ - ipath_cancel_sends(dd, 0); + ipath_cancel_sends(dd, 1); + + /* clear the freeze, and be sure chip saw it */ ipath_write_kreg(dd, dd->ipath_kregs->kr_control, dd->ipath_control); + ipath_read_kreg64(dd, dd->ipath_kregs->kr_scratch); - /* ensure pio avail updates continue */ + /* force in-memory update now we are out of freeze */ ipath_force_pio_avail_update(dd); /* - * We just enabled pioavailupdate, so dma copy is almost certainly - * not yet right, so read the registers directly. Similar to init - */ - for (i = 0; i < dd->ipath_pioavregs; i++) { - /* deal with 6110 chip bug */ - im = (i > 3 && (dd->ipath_flags & IPATH_SWAP_PIOBUFS)) ? - i ^ 1 : i; - val = ipath_read_kreg64(dd, (0x1000 / sizeof(u64)) + im); - dd->ipath_pioavailregs_dma[i] = cpu_to_le64(val); - dd->ipath_pioavailshadow[i] = val | - (~dd->ipath_pioavailkernel[i] << - INFINIPATH_SENDPIOAVAIL_BUSY_SHIFT); - } - - /* * force new interrupt if any hwerr, error or interrupt bits are * still set, and clear "safe" send packet errors related to freeze * and cancelling sends. Re-enable error interrupts before possible @@ -1312,10 +1258,8 @@ irqreturn_t ipath_intr(int irq, void *data) ipath_read_kreg64(dd, dd->ipath_kregs->kr_scratch); spin_unlock_irqrestore(&dd->ipath_sendctrl_lock, flags); - if (!(dd->ipath_flags & IPATH_HAS_SEND_DMA)) - handle_layer_pioavail(dd); - else - ipath_dbg("unexpected BUFAVAIL intr\n"); + /* always process; sdma verbs uses PIO for acks and VL15 */ + handle_layer_pioavail(dd); } ret = IRQ_HANDLED; diff --git a/drivers/infiniband/hw/ipath/ipath_kernel.h b/drivers/infiniband/hw/ipath/ipath_kernel.h index 202337ae90d..02b24a34059 100644 --- a/drivers/infiniband/hw/ipath/ipath_kernel.h +++ b/drivers/infiniband/hw/ipath/ipath_kernel.h @@ -117,6 +117,10 @@ struct ipath_portdata { u16 port_subport_cnt; /* non-zero if port is being shared. */ u16 port_subport_id; + /* number of pio bufs for this port (all procs, if shared) */ + u32 port_piocnt; + /* first pio buffer for this port */ + u32 port_pio_base; /* chip offset of PIO buffers for this port */ u32 port_piobufs; /* how many alloc_pages() chunks in port_rcvegrbuf_pages */ @@ -384,6 +388,8 @@ struct ipath_devdata { u32 ipath_lastrpkts; /* pio bufs allocated per port */ u32 ipath_pbufsport; + /* if remainder on bufs/port, ports < extrabuf get 1 extra */ + u32 ipath_ports_extrabuf; u32 ipath_pioupd_thresh; /* update threshold, some chips */ /* * number of ports configured as max; zero is set to number chip @@ -1011,7 +1017,7 @@ void ipath_get_eeprom_info(struct ipath_devdata *); int ipath_update_eeprom_log(struct ipath_devdata *dd); void ipath_inc_eeprom_err(struct ipath_devdata *dd, u32 eidx, u32 incr); u64 ipath_snap_cntr(struct ipath_devdata *, ipath_creg); -void ipath_disarm_senderrbufs(struct ipath_devdata *, int); +void ipath_disarm_senderrbufs(struct ipath_devdata *); void ipath_force_pio_avail_update(struct ipath_devdata *); void signal_ib_event(struct ipath_devdata *dd, enum ib_event_type ev); diff --git a/drivers/infiniband/hw/ipath/ipath_rc.c b/drivers/infiniband/hw/ipath/ipath_rc.c index c405dfba553..08b11b56761 100644 --- a/drivers/infiniband/hw/ipath/ipath_rc.c +++ b/drivers/infiniband/hw/ipath/ipath_rc.c @@ -1746,7 +1746,11 @@ void ipath_rc_rcv(struct ipath_ibdev *dev, struct ipath_ib_header *hdr, qp->r_wrid_valid = 0; wc.wr_id = qp->r_wr_id; wc.status = IB_WC_SUCCESS; - wc.opcode = IB_WC_RECV; + if (opcode == OP(RDMA_WRITE_LAST_WITH_IMMEDIATE) || + opcode == OP(RDMA_WRITE_ONLY_WITH_IMMEDIATE)) + wc.opcode = IB_WC_RECV_RDMA_WITH_IMM; + else + wc.opcode = IB_WC_RECV; wc.vendor_err = 0; wc.qp = &qp->ibqp; wc.src_qp = qp->remote_qpn; diff --git a/drivers/infiniband/hw/ipath/ipath_ruc.c b/drivers/infiniband/hw/ipath/ipath_ruc.c index 8ac5c1d82cc..9e3fe61cbd0 100644 --- a/drivers/infiniband/hw/ipath/ipath_ruc.c +++ b/drivers/infiniband/hw/ipath/ipath_ruc.c @@ -481,9 +481,10 @@ done: wake_up(&qp->wait); } -static void want_buffer(struct ipath_devdata *dd) +static void want_buffer(struct ipath_devdata *dd, struct ipath_qp *qp) { - if (!(dd->ipath_flags & IPATH_HAS_SEND_DMA)) { + if (!(dd->ipath_flags & IPATH_HAS_SEND_DMA) || + qp->ibqp.qp_type == IB_QPT_SMI) { unsigned long flags; spin_lock_irqsave(&dd->ipath_sendctrl_lock, flags); @@ -519,7 +520,7 @@ static void ipath_no_bufs_available(struct ipath_qp *qp, spin_lock_irqsave(&dev->pending_lock, flags); list_add_tail(&qp->piowait, &dev->piowait); spin_unlock_irqrestore(&dev->pending_lock, flags); - want_buffer(dev->dd); + want_buffer(dev->dd, qp); dev->n_piowait++; } diff --git a/drivers/infiniband/hw/ipath/ipath_sdma.c b/drivers/infiniband/hw/ipath/ipath_sdma.c index 1974df7a9f7..3697449c1ba 100644 --- a/drivers/infiniband/hw/ipath/ipath_sdma.c +++ b/drivers/infiniband/hw/ipath/ipath_sdma.c @@ -308,13 +308,15 @@ static void sdma_abort_task(unsigned long opaque) spin_unlock_irqrestore(&dd->ipath_sdma_lock, flags); /* - * Don't restart sdma here. Wait until link is up to ACTIVE. - * VL15 MADs used to bring the link up use PIO, and multiple - * link transitions otherwise cause the sdma engine to be + * Don't restart sdma here (with the exception + * below). Wait until link is up to ACTIVE. VL15 MADs + * used to bring the link up use PIO, and multiple link + * transitions otherwise cause the sdma engine to be * stopped and started multiple times. - * The disable is done here, including the shadow, so the - * state is kept consistent. - * See ipath_restart_sdma() for the actual starting of sdma. + * The disable is done here, including the shadow, + * so the state is kept consistent. + * See ipath_restart_sdma() for the actual starting + * of sdma. */ spin_lock_irqsave(&dd->ipath_sendctrl_lock, flags); dd->ipath_sendctrl &= ~INFINIPATH_S_SDMAENABLE; @@ -326,6 +328,13 @@ static void sdma_abort_task(unsigned long opaque) /* make sure I see next message */ dd->ipath_sdma_abort_jiffies = 0; + /* + * Not everything that takes SDMA offline is a link + * status change. If the link was up, restart SDMA. + */ + if (dd->ipath_flags & IPATH_LINKACTIVE) + ipath_restart_sdma(dd); + goto done; } @@ -427,7 +436,12 @@ int setup_sdma(struct ipath_devdata *dd) goto done; } - dd->ipath_sdma_status = 0; + /* + * Set initial status as if we had been up, then gone down. + * This lets initial start on transition to ACTIVE be the + * same as restart after link flap. + */ + dd->ipath_sdma_status = IPATH_SDMA_ABORT_ABORTED; dd->ipath_sdma_abort_jiffies = 0; dd->ipath_sdma_generation = 0; dd->ipath_sdma_descq_tail = 0; @@ -449,16 +463,19 @@ int setup_sdma(struct ipath_devdata *dd) ipath_write_kreg(dd, dd->ipath_kregs->kr_senddmaheadaddr, dd->ipath_sdma_head_phys); - /* Reserve all the former "kernel" piobufs */ - n = dd->ipath_piobcnt2k + dd->ipath_piobcnt4k - dd->ipath_pioreserved; - for (i = dd->ipath_lastport_piobuf; i < n; ++i) { + /* + * Reserve all the former "kernel" piobufs, using high number range + * so we get as many 4K buffers as possible + */ + n = dd->ipath_piobcnt2k + dd->ipath_piobcnt4k; + i = dd->ipath_lastport_piobuf + dd->ipath_pioreserved; + ipath_chg_pioavailkernel(dd, i, n - i , 0); + for (; i < n; ++i) { unsigned word = i / 64; unsigned bit = i & 63; BUG_ON(word >= 3); senddmabufmask[word] |= 1ULL << bit; } - ipath_chg_pioavailkernel(dd, dd->ipath_lastport_piobuf, - n - dd->ipath_lastport_piobuf, 0); ipath_write_kreg(dd, dd->ipath_kregs->kr_senddmabufmask0, senddmabufmask[0]); ipath_write_kreg(dd, dd->ipath_kregs->kr_senddmabufmask1, @@ -615,6 +632,9 @@ void ipath_restart_sdma(struct ipath_devdata *dd) ipath_read_kreg64(dd, dd->ipath_kregs->kr_scratch); spin_unlock_irqrestore(&dd->ipath_sendctrl_lock, flags); + /* notify upper layers */ + ipath_ib_piobufavail(dd->verbs_dev); + bail: return; } diff --git a/drivers/infiniband/hw/ipath/ipath_verbs.c b/drivers/infiniband/hw/ipath/ipath_verbs.c index e63927cce5b..5015cd2e57b 100644 --- a/drivers/infiniband/hw/ipath/ipath_verbs.c +++ b/drivers/infiniband/hw/ipath/ipath_verbs.c @@ -396,7 +396,6 @@ static int ipath_post_one_send(struct ipath_qp *qp, struct ib_send_wr *wr) wqe = get_swqe_ptr(qp, qp->s_head); wqe->wr = *wr; - wqe->ssn = qp->s_ssn++; wqe->length = 0; if (wr->num_sge) { acc = wr->opcode >= IB_WR_RDMA_READ ? @@ -422,6 +421,7 @@ static int ipath_post_one_send(struct ipath_qp *qp, struct ib_send_wr *wr) goto bail_inval; } else if (wqe->length > to_idev(qp->ibqp.device)->dd->ipath_ibmtu) goto bail_inval; + wqe->ssn = qp->s_ssn++; qp->s_head = next; ret = 0; diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c index 2f199c5c4a7..4521319b140 100644 --- a/drivers/infiniband/hw/mlx4/cq.c +++ b/drivers/infiniband/hw/mlx4/cq.c @@ -246,7 +246,7 @@ err_mtt: if (context) ib_umem_release(cq->umem); else - mlx4_ib_free_cq_buf(dev, &cq->buf, entries); + mlx4_ib_free_cq_buf(dev, &cq->buf, cq->ibcq.cqe); err_db: if (!context) @@ -434,7 +434,7 @@ int mlx4_ib_destroy_cq(struct ib_cq *cq) mlx4_ib_db_unmap_user(to_mucontext(cq->uobject->context), &mcq->db); ib_umem_release(mcq->umem); } else { - mlx4_ib_free_cq_buf(dev, &mcq->buf, cq->cqe + 1); + mlx4_ib_free_cq_buf(dev, &mcq->buf, cq->cqe); mlx4_db_free(dev->dev, &mcq->db); } diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index 9044f880353..ca126fc2b85 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -334,6 +334,7 @@ struct ipoib_dev_priv { #endif int hca_caps; struct ipoib_ethtool_st ethtool; + struct timer_list poll_timer; }; struct ipoib_ah { @@ -404,6 +405,7 @@ extern struct workqueue_struct *ipoib_workqueue; int ipoib_poll(struct napi_struct *napi, int budget); void ipoib_ib_completion(struct ib_cq *cq, void *dev_ptr); +void ipoib_send_comp_handler(struct ib_cq *cq, void *dev_ptr); struct ipoib_ah *ipoib_create_ah(struct net_device *dev, struct ib_pd *pd, struct ib_ah_attr *attr); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index 97b815c1a3f..f429bce24c2 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -461,6 +461,26 @@ void ipoib_ib_completion(struct ib_cq *cq, void *dev_ptr) netif_rx_schedule(dev, &priv->napi); } +static void drain_tx_cq(struct net_device *dev) +{ + struct ipoib_dev_priv *priv = netdev_priv(dev); + unsigned long flags; + + spin_lock_irqsave(&priv->tx_lock, flags); + while (poll_tx(priv)) + ; /* nothing */ + + if (netif_queue_stopped(dev)) + mod_timer(&priv->poll_timer, jiffies + 1); + + spin_unlock_irqrestore(&priv->tx_lock, flags); +} + +void ipoib_send_comp_handler(struct ib_cq *cq, void *dev_ptr) +{ + drain_tx_cq((struct net_device *)dev_ptr); +} + static inline int post_send(struct ipoib_dev_priv *priv, unsigned int wr_id, struct ib_ah *address, u32 qpn, @@ -555,12 +575,22 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb, else priv->tx_wr.send_flags &= ~IB_SEND_IP_CSUM; + if (++priv->tx_outstanding == ipoib_sendq_size) { + ipoib_dbg(priv, "TX ring full, stopping kernel net queue\n"); + if (ib_req_notify_cq(priv->send_cq, IB_CQ_NEXT_COMP)) + ipoib_warn(priv, "request notify on send CQ failed\n"); + netif_stop_queue(dev); + } + if (unlikely(post_send(priv, priv->tx_head & (ipoib_sendq_size - 1), address->ah, qpn, tx_req, phead, hlen))) { ipoib_warn(priv, "post_send failed\n"); ++dev->stats.tx_errors; + --priv->tx_outstanding; ipoib_dma_unmap_tx(priv->ca, tx_req); dev_kfree_skb_any(skb); + if (netif_queue_stopped(dev)) + netif_wake_queue(dev); } else { dev->trans_start = jiffies; @@ -568,14 +598,11 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb, ++priv->tx_head; skb_orphan(skb); - if (++priv->tx_outstanding == ipoib_sendq_size) { - ipoib_dbg(priv, "TX ring full, stopping kernel net queue\n"); - netif_stop_queue(dev); - } } if (unlikely(priv->tx_outstanding > MAX_SEND_CQE)) - poll_tx(priv); + while (poll_tx(priv)) + ; /* nothing */ } static void __ipoib_reap_ah(struct net_device *dev) @@ -609,6 +636,11 @@ void ipoib_reap_ah(struct work_struct *work) round_jiffies_relative(HZ)); } +static void ipoib_ib_tx_timer_func(unsigned long ctx) +{ + drain_tx_cq((struct net_device *)ctx); +} + int ipoib_ib_dev_open(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); @@ -645,6 +677,10 @@ int ipoib_ib_dev_open(struct net_device *dev) queue_delayed_work(ipoib_workqueue, &priv->ah_reap_task, round_jiffies_relative(HZ)); + init_timer(&priv->poll_timer); + priv->poll_timer.function = ipoib_ib_tx_timer_func; + priv->poll_timer.data = (unsigned long)dev; + set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags); return 0; @@ -810,6 +846,7 @@ int ipoib_ib_dev_stop(struct net_device *dev, int flush) ipoib_dbg(priv, "All sends and receives done.\n"); timeout: + del_timer_sync(&priv->poll_timer); qp_attr.qp_state = IB_QPS_RESET; if (ib_modify_qp(priv->qp, &qp_attr, IB_QP_STATE)) ipoib_warn(priv, "Failed to modify QP to RESET state\n"); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c index c1e7ece1fd4..8766d29ce3b 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c @@ -187,7 +187,8 @@ int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca) goto out_free_mr; } - priv->send_cq = ib_create_cq(priv->ca, NULL, NULL, dev, ipoib_sendq_size, 0); + priv->send_cq = ib_create_cq(priv->ca, ipoib_send_comp_handler, NULL, + dev, ipoib_sendq_size, 0); if (IS_ERR(priv->send_cq)) { printk(KERN_WARNING "%s: failed to create send CQ\n", ca->name); goto out_free_recv_cq; diff --git a/drivers/input/misc/Kconfig b/drivers/input/misc/Kconfig index 92b683411d5..3ad8bd9f754 100644 --- a/drivers/input/misc/Kconfig +++ b/drivers/input/misc/Kconfig @@ -14,7 +14,7 @@ if INPUT_MISC config INPUT_PCSPKR tristate "PC Speaker support" - depends on ALPHA || X86 || MIPS || PPC_PREP || PPC_CHRP || PPC_PSERIES + depends on PCSPKR_PLATFORM depends on SND_PCSP=n help Say Y here if you want the standard PC Speaker to be used for diff --git a/drivers/input/serio/hp_sdc.c b/drivers/input/serio/hp_sdc.c index 02b3ad8c082..edfedd9a166 100644 --- a/drivers/input/serio/hp_sdc.c +++ b/drivers/input/serio/hp_sdc.c @@ -69,6 +69,7 @@ #include <linux/time.h> #include <linux/slab.h> #include <linux/hil.h> +#include <linux/semaphore.h> #include <asm/io.h> #include <asm/system.h> diff --git a/drivers/macintosh/adb.c b/drivers/macintosh/adb.c index 20978205cd0..b8b9e44f7f4 100644 --- a/drivers/macintosh/adb.c +++ b/drivers/macintosh/adb.c @@ -37,7 +37,7 @@ #include <linux/device.h> #include <linux/kthread.h> #include <linux/platform_device.h> -#include <linux/semaphore.h> +#include <linux/mutex.h> #include <asm/uaccess.h> #ifdef CONFIG_PPC @@ -102,7 +102,7 @@ static struct adb_handler { } adb_handler[16]; /* - * The adb_handler_sem mutex protects all accesses to the original_address + * The adb_handler_mutex mutex protects all accesses to the original_address * and handler_id fields of adb_handler[i] for all i, and changes to the * handler field. * Accesses to the handler field are protected by the adb_handler_lock @@ -110,7 +110,7 @@ static struct adb_handler { * time adb_unregister returns, we know that the old handler isn't being * called. */ -static DECLARE_MUTEX(adb_handler_sem); +static DEFINE_MUTEX(adb_handler_mutex); static DEFINE_RWLOCK(adb_handler_lock); #if 0 @@ -355,7 +355,7 @@ do_adb_reset_bus(void) msleep(500); } - down(&adb_handler_sem); + mutex_lock(&adb_handler_mutex); write_lock_irq(&adb_handler_lock); memset(adb_handler, 0, sizeof(adb_handler)); write_unlock_irq(&adb_handler_lock); @@ -376,7 +376,7 @@ do_adb_reset_bus(void) if (adb_controller->autopoll) adb_controller->autopoll(autopoll_devs); } - up(&adb_handler_sem); + mutex_unlock(&adb_handler_mutex); blocking_notifier_call_chain(&adb_client_list, ADB_MSG_POST_RESET, NULL); @@ -454,7 +454,7 @@ adb_register(int default_id, int handler_id, struct adb_ids *ids, { int i; - down(&adb_handler_sem); + mutex_lock(&adb_handler_mutex); ids->nids = 0; for (i = 1; i < 16; i++) { if ((adb_handler[i].original_address == default_id) && @@ -472,7 +472,7 @@ adb_register(int default_id, int handler_id, struct adb_ids *ids, ids->id[ids->nids++] = i; } } - up(&adb_handler_sem); + mutex_unlock(&adb_handler_mutex); return ids->nids; } @@ -481,7 +481,7 @@ adb_unregister(int index) { int ret = -ENODEV; - down(&adb_handler_sem); + mutex_lock(&adb_handler_mutex); write_lock_irq(&adb_handler_lock); if (adb_handler[index].handler) { while(adb_handler[index].busy) { @@ -493,7 +493,7 @@ adb_unregister(int index) adb_handler[index].handler = NULL; } write_unlock_irq(&adb_handler_lock); - up(&adb_handler_sem); + mutex_unlock(&adb_handler_mutex); return ret; } @@ -557,19 +557,19 @@ adb_try_handler_change(int address, int new_id) { int ret; - down(&adb_handler_sem); + mutex_lock(&adb_handler_mutex); ret = try_handler_change(address, new_id); - up(&adb_handler_sem); + mutex_unlock(&adb_handler_mutex); return ret; } int adb_get_infos(int address, int *original_address, int *handler_id) { - down(&adb_handler_sem); + mutex_lock(&adb_handler_mutex); *original_address = adb_handler[address].original_address; *handler_id = adb_handler[address].handler_id; - up(&adb_handler_sem); + mutex_unlock(&adb_handler_mutex); return (*original_address != 0); } @@ -628,10 +628,10 @@ do_adb_query(struct adb_request *req) case ADB_QUERY_GETDEVINFO: if (req->nbytes < 3) break; - down(&adb_handler_sem); + mutex_lock(&adb_handler_mutex); req->reply[0] = adb_handler[req->data[2]].original_address; req->reply[1] = adb_handler[req->data[2]].handler_id; - up(&adb_handler_sem); + mutex_unlock(&adb_handler_mutex); req->complete = 1; req->reply_len = 2; adb_write_done(req); diff --git a/drivers/macintosh/therm_pm72.c b/drivers/macintosh/therm_pm72.c index 1e0a69a5e81..ddfb426a9ab 100644 --- a/drivers/macintosh/therm_pm72.c +++ b/drivers/macintosh/therm_pm72.c @@ -122,6 +122,7 @@ #include <linux/kmod.h> #include <linux/i2c.h> #include <linux/kthread.h> +#include <linux/mutex.h> #include <asm/prom.h> #include <asm/machdep.h> #include <asm/io.h> @@ -169,7 +170,7 @@ static int rackmac; static s32 dimm_output_clamp; static int fcu_rpm_shift; static int fcu_tickle_ticks; -static DECLARE_MUTEX(driver_lock); +static DEFINE_MUTEX(driver_lock); /* * We have 3 types of CPU PID control. One is "split" old style control @@ -729,9 +730,9 @@ static void fetch_cpu_pumps_minmax(void) static ssize_t show_##name(struct device *dev, struct device_attribute *attr, char *buf) \ { \ ssize_t r; \ - down(&driver_lock); \ + mutex_lock(&driver_lock); \ r = sprintf(buf, "%d.%03d", FIX32TOPRINT(data)); \ - up(&driver_lock); \ + mutex_unlock(&driver_lock); \ return r; \ } #define BUILD_SHOW_FUNC_INT(name, data) \ @@ -1803,11 +1804,11 @@ static int main_control_loop(void *x) { DBG("main_control_loop started\n"); - down(&driver_lock); + mutex_lock(&driver_lock); if (start_fcu() < 0) { printk(KERN_ERR "kfand: failed to start FCU\n"); - up(&driver_lock); + mutex_unlock(&driver_lock); goto out; } @@ -1822,14 +1823,14 @@ static int main_control_loop(void *x) fcu_tickle_ticks = FCU_TICKLE_TICKS; - up(&driver_lock); + mutex_unlock(&driver_lock); while (state == state_attached) { unsigned long elapsed, start; start = jiffies; - down(&driver_lock); + mutex_lock(&driver_lock); /* Tickle the FCU just in case */ if (--fcu_tickle_ticks < 0) { @@ -1861,7 +1862,7 @@ static int main_control_loop(void *x) do_monitor_slots(&slots_state); else do_monitor_drives(&drives_state); - up(&driver_lock); + mutex_unlock(&driver_lock); if (critical_state == 1) { printk(KERN_WARNING "Temperature control detected a critical condition\n"); @@ -2019,13 +2020,13 @@ static void detach_fcu(void) */ static int therm_pm72_attach(struct i2c_adapter *adapter) { - down(&driver_lock); + mutex_lock(&driver_lock); /* Check state */ if (state == state_detached) state = state_attaching; if (state != state_attaching) { - up(&driver_lock); + mutex_unlock(&driver_lock); return 0; } @@ -2054,7 +2055,7 @@ static int therm_pm72_attach(struct i2c_adapter *adapter) state = state_attached; start_control_loops(); } - up(&driver_lock); + mutex_unlock(&driver_lock); return 0; } @@ -2065,16 +2066,16 @@ static int therm_pm72_attach(struct i2c_adapter *adapter) */ static int therm_pm72_detach(struct i2c_adapter *adapter) { - down(&driver_lock); + mutex_lock(&driver_lock); if (state != state_detached) state = state_detaching; /* Stop control loops if any */ DBG("stopping control loops\n"); - up(&driver_lock); + mutex_unlock(&driver_lock); stop_control_loops(); - down(&driver_lock); + mutex_lock(&driver_lock); if (u3_0 != NULL && !strcmp(adapter->name, "u3 0")) { DBG("lost U3-0, disposing control loops\n"); @@ -2090,7 +2091,7 @@ static int therm_pm72_detach(struct i2c_adapter *adapter) if (u3_0 == NULL && u3_1 == NULL) state = state_detached; - up(&driver_lock); + mutex_unlock(&driver_lock); return 0; } diff --git a/drivers/macintosh/windfarm_smu_sat.c b/drivers/macintosh/windfarm_smu_sat.c index 797918d0e59..7f2be4baaed 100644 --- a/drivers/macintosh/windfarm_smu_sat.c +++ b/drivers/macintosh/windfarm_smu_sat.c @@ -13,7 +13,7 @@ #include <linux/init.h> #include <linux/wait.h> #include <linux/i2c.h> -#include <linux/semaphore.h> +#include <linux/mutex.h> #include <asm/prom.h> #include <asm/smu.h> #include <asm/pmac_low_i2c.h> @@ -36,7 +36,7 @@ struct wf_sat { int nr; atomic_t refcnt; - struct semaphore mutex; + struct mutex mutex; unsigned long last_read; /* jiffies when cache last updated */ u8 cache[16]; struct i2c_client i2c; @@ -163,7 +163,7 @@ static int wf_sat_get(struct wf_sensor *sr, s32 *value) if (sat->i2c.adapter == NULL) return -ENODEV; - down(&sat->mutex); + mutex_lock(&sat->mutex); if (time_after(jiffies, (sat->last_read + MAX_AGE))) { err = wf_sat_read_cache(sat); if (err) @@ -182,7 +182,7 @@ static int wf_sat_get(struct wf_sensor *sr, s32 *value) err = 0; fail: - up(&sat->mutex); + mutex_unlock(&sat->mutex); return err; } @@ -233,7 +233,7 @@ static void wf_sat_create(struct i2c_adapter *adapter, struct device_node *dev) sat->nr = -1; sat->node = of_node_get(dev); atomic_set(&sat->refcnt, 0); - init_MUTEX(&sat->mutex); + mutex_init(&sat->mutex); sat->i2c.addr = (addr >> 1) & 0x7f; sat->i2c.adapter = adapter; sat->i2c.driver = &wf_sat_driver; diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 5938fa96292..faf3d891297 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -886,7 +886,7 @@ static int make_request(struct request_queue *q, struct bio * bio) */ raid10_find_phys(conf, r10_bio); retry_write: - blocked_rdev = 0; + blocked_rdev = NULL; rcu_read_lock(); for (i = 0; i < conf->copies; i++) { int d = r10_bio->devs[i].devnum; diff --git a/drivers/media/Makefile b/drivers/media/Makefile index 73f742c7e81..cc11c4c0e7e 100644 --- a/drivers/media/Makefile +++ b/drivers/media/Makefile @@ -2,6 +2,8 @@ # Makefile for the kernel multimedia device drivers. # +obj-y := common/ + obj-$(CONFIG_VIDEO_MEDIA) += common/ # Since hybrid devices are here, should be compiled if DVB and/or V4L diff --git a/drivers/media/video/cx18/cx18-driver.c b/drivers/media/video/cx18/cx18-driver.c index 8f5ed9b4bf8..3f55d47bc4b 100644 --- a/drivers/media/video/cx18/cx18-driver.c +++ b/drivers/media/video/cx18/cx18-driver.c @@ -613,7 +613,7 @@ static int __devinit cx18_probe(struct pci_dev *dev, } cx = kzalloc(sizeof(struct cx18), GFP_ATOMIC); - if (cx == 0) { + if (!cx) { spin_unlock(&cx18_cards_lock); return -ENOMEM; } diff --git a/drivers/media/video/saa7134/saa7134-video.c b/drivers/media/video/saa7134/saa7134-video.c index a0baf2d0ba7..48e1a01718e 100644 --- a/drivers/media/video/saa7134/saa7134-video.c +++ b/drivers/media/video/saa7134/saa7134-video.c @@ -1634,7 +1634,7 @@ static int saa7134_s_fmt_overlay(struct file *file, void *priv, struct saa7134_fh *fh = priv; struct saa7134_dev *dev = fh->dev; int err; - unsigned int flags; + unsigned long flags; if (saa7134_no_overlay > 0) { printk(KERN_ERR "V4L2_BUF_TYPE_VIDEO_OVERLAY: no_overlay\n"); diff --git a/drivers/misc/kgdbts.c b/drivers/misc/kgdbts.c index 30a1af857c7..fa394104339 100644 --- a/drivers/misc/kgdbts.c +++ b/drivers/misc/kgdbts.c @@ -47,6 +47,7 @@ * to test the HW NMI watchdog * F## = Break at do_fork for ## iterations * S## = Break at sys_open for ## iterations + * I## = Run the single step test ## iterations * * NOTE: that the do_fork and sys_open tests are mutually exclusive. * @@ -375,7 +376,7 @@ static void emul_sstep_get(char *arg) break; case 1: /* set breakpoint */ - break_helper("Z0", 0, sstep_addr); + break_helper("Z0", NULL, sstep_addr); break; case 2: /* Continue */ @@ -383,7 +384,7 @@ static void emul_sstep_get(char *arg) break; case 3: /* Clear breakpoint */ - break_helper("z0", 0, sstep_addr); + break_helper("z0", NULL, sstep_addr); break; default: eprintk("kgdbts: ERROR failed sstep get emulation\n"); @@ -465,11 +466,11 @@ static struct test_struct sw_breakpoint_test[] = { { "?", "S0*" }, /* Clear break points */ { "kgdbts_break_test", "OK", sw_break, }, /* set sw breakpoint */ { "c", "T0*", }, /* Continue */ - { "g", "kgdbts_break_test", 0, check_and_rewind_pc }, + { "g", "kgdbts_break_test", NULL, check_and_rewind_pc }, { "write", "OK", write_regs }, { "kgdbts_break_test", "OK", sw_rem_break }, /*remove breakpoint */ { "D", "OK" }, /* Detach */ - { "D", "OK", 0, got_break }, /* If the test worked we made it here */ + { "D", "OK", NULL, got_break }, /* On success we made it here */ { "", "" }, }; @@ -499,14 +500,14 @@ static struct test_struct singlestep_break_test[] = { { "?", "S0*" }, /* Clear break points */ { "kgdbts_break_test", "OK", sw_break, }, /* set sw breakpoint */ { "c", "T0*", }, /* Continue */ - { "g", "kgdbts_break_test", 0, check_and_rewind_pc }, + { "g", "kgdbts_break_test", NULL, check_and_rewind_pc }, { "write", "OK", write_regs }, /* Write registers */ { "kgdbts_break_test", "OK", sw_rem_break }, /*remove breakpoint */ { "s", "T0*", emul_sstep_get, emul_sstep_put }, /* Single step */ - { "g", "kgdbts_break_test", 0, check_single_step }, + { "g", "kgdbts_break_test", NULL, check_single_step }, { "kgdbts_break_test", "OK", sw_break, }, /* set sw breakpoint */ { "c", "T0*", }, /* Continue */ - { "g", "kgdbts_break_test", 0, check_and_rewind_pc }, + { "g", "kgdbts_break_test", NULL, check_and_rewind_pc }, { "write", "OK", write_regs }, /* Write registers */ { "D", "OK" }, /* Remove all breakpoints and continues */ { "", "" }, @@ -520,14 +521,14 @@ static struct test_struct do_fork_test[] = { { "?", "S0*" }, /* Clear break points */ { "do_fork", "OK", sw_break, }, /* set sw breakpoint */ { "c", "T0*", }, /* Continue */ - { "g", "do_fork", 0, check_and_rewind_pc }, /* check location */ + { "g", "do_fork", NULL, check_and_rewind_pc }, /* check location */ { "write", "OK", write_regs }, /* Write registers */ { "do_fork", "OK", sw_rem_break }, /*remove breakpoint */ { "s", "T0*", emul_sstep_get, emul_sstep_put }, /* Single step */ - { "g", "do_fork", 0, check_single_step }, + { "g", "do_fork", NULL, check_single_step }, { "do_fork", "OK", sw_break, }, /* set sw breakpoint */ { "7", "T0*", skip_back_repeat_test }, /* Loop based on repeat_test */ - { "D", "OK", 0, final_ack_set }, /* detach and unregister I/O */ + { "D", "OK", NULL, final_ack_set }, /* detach and unregister I/O */ { "", "" }, }; @@ -538,14 +539,14 @@ static struct test_struct sys_open_test[] = { { "?", "S0*" }, /* Clear break points */ { "sys_open", "OK", sw_break, }, /* set sw breakpoint */ { "c", "T0*", }, /* Continue */ - { "g", "sys_open", 0, check_and_rewind_pc }, /* check location */ + { "g", "sys_open", NULL, check_and_rewind_pc }, /* check location */ { "write", "OK", write_regs }, /* Write registers */ { "sys_open", "OK", sw_rem_break }, /*remove breakpoint */ { "s", "T0*", emul_sstep_get, emul_sstep_put }, /* Single step */ - { "g", "sys_open", 0, check_single_step }, + { "g", "sys_open", NULL, check_single_step }, { "sys_open", "OK", sw_break, }, /* set sw breakpoint */ { "7", "T0*", skip_back_repeat_test }, /* Loop based on repeat_test */ - { "D", "OK", 0, final_ack_set }, /* detach and unregister I/O */ + { "D", "OK", NULL, final_ack_set }, /* detach and unregister I/O */ { "", "" }, }; @@ -556,11 +557,11 @@ static struct test_struct hw_breakpoint_test[] = { { "?", "S0*" }, /* Clear break points */ { "kgdbts_break_test", "OK", hw_break, }, /* set hw breakpoint */ { "c", "T0*", }, /* Continue */ - { "g", "kgdbts_break_test", 0, check_and_rewind_pc }, + { "g", "kgdbts_break_test", NULL, check_and_rewind_pc }, { "write", "OK", write_regs }, { "kgdbts_break_test", "OK", hw_rem_break }, /*remove breakpoint */ { "D", "OK" }, /* Detach */ - { "D", "OK", 0, got_break }, /* If the test worked we made it here */ + { "D", "OK", NULL, got_break }, /* On success we made it here */ { "", "" }, }; @@ -570,12 +571,12 @@ static struct test_struct hw_breakpoint_test[] = { static struct test_struct hw_write_break_test[] = { { "?", "S0*" }, /* Clear break points */ { "hw_break_val", "OK", hw_write_break, }, /* set hw breakpoint */ - { "c", "T0*", 0, got_break }, /* Continue */ - { "g", "silent", 0, check_and_rewind_pc }, + { "c", "T0*", NULL, got_break }, /* Continue */ + { "g", "silent", NULL, check_and_rewind_pc }, { "write", "OK", write_regs }, { "hw_break_val", "OK", hw_rem_write_break }, /*remove breakpoint */ { "D", "OK" }, /* Detach */ - { "D", "OK", 0, got_break }, /* If the test worked we made it here */ + { "D", "OK", NULL, got_break }, /* On success we made it here */ { "", "" }, }; @@ -585,12 +586,12 @@ static struct test_struct hw_write_break_test[] = { static struct test_struct hw_access_break_test[] = { { "?", "S0*" }, /* Clear break points */ { "hw_break_val", "OK", hw_access_break, }, /* set hw breakpoint */ - { "c", "T0*", 0, got_break }, /* Continue */ - { "g", "silent", 0, check_and_rewind_pc }, + { "c", "T0*", NULL, got_break }, /* Continue */ + { "g", "silent", NULL, check_and_rewind_pc }, { "write", "OK", write_regs }, { "hw_break_val", "OK", hw_rem_access_break }, /*remove breakpoint */ { "D", "OK" }, /* Detach */ - { "D", "OK", 0, got_break }, /* If the test worked we made it here */ + { "D", "OK", NULL, got_break }, /* On success we made it here */ { "", "" }, }; @@ -599,9 +600,9 @@ static struct test_struct hw_access_break_test[] = { */ static struct test_struct nmi_sleep_test[] = { { "?", "S0*" }, /* Clear break points */ - { "c", "T0*", 0, got_break }, /* Continue */ + { "c", "T0*", NULL, got_break }, /* Continue */ { "D", "OK" }, /* Detach */ - { "D", "OK", 0, got_break }, /* If the test worked we made it here */ + { "D", "OK", NULL, got_break }, /* On success we made it here */ { "", "" }, }; @@ -874,18 +875,23 @@ static void kgdbts_run_tests(void) { char *ptr; int fork_test = 0; - int sys_open_test = 0; + int do_sys_open_test = 0; + int sstep_test = 1000; int nmi_sleep = 0; + int i; ptr = strstr(config, "F"); if (ptr) - fork_test = simple_strtol(ptr+1, NULL, 10); + fork_test = simple_strtol(ptr + 1, NULL, 10); ptr = strstr(config, "S"); if (ptr) - sys_open_test = simple_strtol(ptr+1, NULL, 10); + do_sys_open_test = simple_strtol(ptr + 1, NULL, 10); ptr = strstr(config, "N"); if (ptr) nmi_sleep = simple_strtol(ptr+1, NULL, 10); + ptr = strstr(config, "I"); + if (ptr) + sstep_test = simple_strtol(ptr+1, NULL, 10); /* required internal KGDB tests */ v1printk("kgdbts:RUN plant and detach test\n"); @@ -894,8 +900,13 @@ static void kgdbts_run_tests(void) run_breakpoint_test(0); v1printk("kgdbts:RUN bad memory access test\n"); run_bad_read_test(); - v1printk("kgdbts:RUN singlestep breakpoint test\n"); - run_singlestep_break_test(); + v1printk("kgdbts:RUN singlestep test %i iterations\n", sstep_test); + for (i = 0; i < sstep_test; i++) { + run_singlestep_break_test(); + if (i % 100 == 0) + v1printk("kgdbts:RUN singlestep [%i/%i]\n", + i, sstep_test); + } /* ===Optional tests=== */ @@ -922,7 +933,7 @@ static void kgdbts_run_tests(void) repeat_test = fork_test; printk(KERN_INFO "kgdbts:RUN do_fork for %i breakpoints\n", repeat_test); - kthread_run(kgdbts_unreg_thread, 0, "kgdbts_unreg"); + kthread_run(kgdbts_unreg_thread, NULL, "kgdbts_unreg"); run_do_fork_test(); return; } @@ -931,11 +942,11 @@ static void kgdbts_run_tests(void) * executed because a kernel thread will be spawned at the very * end to unregister the debug hooks. */ - if (sys_open_test) { - repeat_test = sys_open_test; + if (do_sys_open_test) { + repeat_test = do_sys_open_test; printk(KERN_INFO "kgdbts:RUN sys_open for %i breakpoints\n", repeat_test); - kthread_run(kgdbts_unreg_thread, 0, "kgdbts_unreg"); + kthread_run(kgdbts_unreg_thread, NULL, "kgdbts_unreg"); run_sys_open_test(); return; } diff --git a/drivers/net/3c59x.c b/drivers/net/3c59x.c index 6f8e7d4cf74..2edda8cc7f9 100644 --- a/drivers/net/3c59x.c +++ b/drivers/net/3c59x.c @@ -319,7 +319,7 @@ static struct vortex_chip_info { {"3c920B-EMB-WNM (ATI Radeon 9100 IGP)", PCI_USES_MASTER, IS_TORNADO|HAS_MII|HAS_HWCKSM, 128, }, {"3c980 Cyclone", - PCI_USES_MASTER, IS_CYCLONE|HAS_HWCKSM, 128, }, + PCI_USES_MASTER, IS_CYCLONE|HAS_HWCKSM|EXTRA_PREAMBLE, 128, }, {"3c980C Python-T", PCI_USES_MASTER, IS_CYCLONE|HAS_NWAY|HAS_HWCKSM, 128, }, @@ -600,7 +600,6 @@ struct vortex_private { struct sk_buff* tx_skbuff[TX_RING_SIZE]; unsigned int cur_rx, cur_tx; /* The next free ring entry */ unsigned int dirty_rx, dirty_tx; /* The ring entries to be free()ed. */ - struct net_device_stats stats; /* Generic stats */ struct vortex_extra_stats xstats; /* NIC-specific extra stats */ struct sk_buff *tx_skb; /* Packet being eaten by bus master ctrl. */ dma_addr_t tx_skb_dma; /* Allocated DMA address for bus master ctrl DMA. */ @@ -1875,7 +1874,7 @@ static void vortex_tx_timeout(struct net_device *dev) issue_and_wait(dev, TxReset); - vp->stats.tx_errors++; + dev->stats.tx_errors++; if (vp->full_bus_master_tx) { printk(KERN_DEBUG "%s: Resetting the Tx ring pointer.\n", dev->name); if (vp->cur_tx - vp->dirty_tx > 0 && ioread32(ioaddr + DownListPtr) == 0) @@ -1887,7 +1886,7 @@ static void vortex_tx_timeout(struct net_device *dev) iowrite8(PKT_BUF_SZ>>8, ioaddr + TxFreeThreshold); iowrite16(DownUnstall, ioaddr + EL3_CMD); } else { - vp->stats.tx_dropped++; + dev->stats.tx_dropped++; netif_wake_queue(dev); } @@ -1928,8 +1927,8 @@ vortex_error(struct net_device *dev, int status) } dump_tx_ring(dev); } - if (tx_status & 0x14) vp->stats.tx_fifo_errors++; - if (tx_status & 0x38) vp->stats.tx_aborted_errors++; + if (tx_status & 0x14) dev->stats.tx_fifo_errors++; + if (tx_status & 0x38) dev->stats.tx_aborted_errors++; if (tx_status & 0x08) vp->xstats.tx_max_collisions++; iowrite8(0, ioaddr + TxStatus); if (tx_status & 0x30) { /* txJabber or txUnderrun */ @@ -2051,8 +2050,8 @@ vortex_start_xmit(struct sk_buff *skb, struct net_device *dev) if (vortex_debug > 2) printk(KERN_DEBUG "%s: Tx error, status %2.2x.\n", dev->name, tx_status); - if (tx_status & 0x04) vp->stats.tx_fifo_errors++; - if (tx_status & 0x38) vp->stats.tx_aborted_errors++; + if (tx_status & 0x04) dev->stats.tx_fifo_errors++; + if (tx_status & 0x38) dev->stats.tx_aborted_errors++; if (tx_status & 0x30) { issue_and_wait(dev, TxReset); } @@ -2350,7 +2349,7 @@ boomerang_interrupt(int irq, void *dev_id) } else { printk(KERN_DEBUG "boomerang_interrupt: no skb!\n"); } - /* vp->stats.tx_packets++; Counted below. */ + /* dev->stats.tx_packets++; Counted below. */ dirty_tx++; } vp->dirty_tx = dirty_tx; @@ -2409,12 +2408,12 @@ static int vortex_rx(struct net_device *dev) unsigned char rx_error = ioread8(ioaddr + RxErrors); if (vortex_debug > 2) printk(KERN_DEBUG " Rx error: status %2.2x.\n", rx_error); - vp->stats.rx_errors++; - if (rx_error & 0x01) vp->stats.rx_over_errors++; - if (rx_error & 0x02) vp->stats.rx_length_errors++; - if (rx_error & 0x04) vp->stats.rx_frame_errors++; - if (rx_error & 0x08) vp->stats.rx_crc_errors++; - if (rx_error & 0x10) vp->stats.rx_length_errors++; + dev->stats.rx_errors++; + if (rx_error & 0x01) dev->stats.rx_over_errors++; + if (rx_error & 0x02) dev->stats.rx_length_errors++; + if (rx_error & 0x04) dev->stats.rx_frame_errors++; + if (rx_error & 0x08) dev->stats.rx_crc_errors++; + if (rx_error & 0x10) dev->stats.rx_length_errors++; } else { /* The packet length: up to 4.5K!. */ int pkt_len = rx_status & 0x1fff; @@ -2446,7 +2445,7 @@ static int vortex_rx(struct net_device *dev) skb->protocol = eth_type_trans(skb, dev); netif_rx(skb); dev->last_rx = jiffies; - vp->stats.rx_packets++; + dev->stats.rx_packets++; /* Wait a limited time to go to next packet. */ for (i = 200; i >= 0; i--) if ( ! (ioread16(ioaddr + EL3_STATUS) & CmdInProgress)) @@ -2455,7 +2454,7 @@ static int vortex_rx(struct net_device *dev) } else if (vortex_debug > 0) printk(KERN_NOTICE "%s: No memory to allocate a sk_buff of " "size %d.\n", dev->name, pkt_len); - vp->stats.rx_dropped++; + dev->stats.rx_dropped++; } issue_and_wait(dev, RxDiscard); } @@ -2482,12 +2481,12 @@ boomerang_rx(struct net_device *dev) unsigned char rx_error = rx_status >> 16; if (vortex_debug > 2) printk(KERN_DEBUG " Rx error: status %2.2x.\n", rx_error); - vp->stats.rx_errors++; - if (rx_error & 0x01) vp->stats.rx_over_errors++; - if (rx_error & 0x02) vp->stats.rx_length_errors++; - if (rx_error & 0x04) vp->stats.rx_frame_errors++; - if (rx_error & 0x08) vp->stats.rx_crc_errors++; - if (rx_error & 0x10) vp->stats.rx_length_errors++; + dev->stats.rx_errors++; + if (rx_error & 0x01) dev->stats.rx_over_errors++; + if (rx_error & 0x02) dev->stats.rx_length_errors++; + if (rx_error & 0x04) dev->stats.rx_frame_errors++; + if (rx_error & 0x08) dev->stats.rx_crc_errors++; + if (rx_error & 0x10) dev->stats.rx_length_errors++; } else { /* The packet length: up to 4.5K!. */ int pkt_len = rx_status & 0x1fff; @@ -2529,7 +2528,7 @@ boomerang_rx(struct net_device *dev) } netif_rx(skb); dev->last_rx = jiffies; - vp->stats.rx_packets++; + dev->stats.rx_packets++; } entry = (++vp->cur_rx) % RX_RING_SIZE; } @@ -2591,7 +2590,7 @@ vortex_down(struct net_device *dev, int final_down) del_timer_sync(&vp->rx_oom_timer); del_timer_sync(&vp->timer); - /* Turn off statistics ASAP. We update vp->stats below. */ + /* Turn off statistics ASAP. We update dev->stats below. */ iowrite16(StatsDisable, ioaddr + EL3_CMD); /* Disable the receiver and transmitter. */ @@ -2728,7 +2727,7 @@ static struct net_device_stats *vortex_get_stats(struct net_device *dev) update_stats(ioaddr, dev); spin_unlock_irqrestore (&vp->lock, flags); } - return &vp->stats; + return &dev->stats; } /* Update statistics. @@ -2748,18 +2747,18 @@ static void update_stats(void __iomem *ioaddr, struct net_device *dev) /* Unlike the 3c5x9 we need not turn off stats updates while reading. */ /* Switch to the stats window, and read everything. */ EL3WINDOW(6); - vp->stats.tx_carrier_errors += ioread8(ioaddr + 0); - vp->stats.tx_heartbeat_errors += ioread8(ioaddr + 1); - vp->stats.tx_window_errors += ioread8(ioaddr + 4); - vp->stats.rx_fifo_errors += ioread8(ioaddr + 5); - vp->stats.tx_packets += ioread8(ioaddr + 6); - vp->stats.tx_packets += (ioread8(ioaddr + 9)&0x30) << 4; + dev->stats.tx_carrier_errors += ioread8(ioaddr + 0); + dev->stats.tx_heartbeat_errors += ioread8(ioaddr + 1); + dev->stats.tx_window_errors += ioread8(ioaddr + 4); + dev->stats.rx_fifo_errors += ioread8(ioaddr + 5); + dev->stats.tx_packets += ioread8(ioaddr + 6); + dev->stats.tx_packets += (ioread8(ioaddr + 9)&0x30) << 4; /* Rx packets */ ioread8(ioaddr + 7); /* Must read to clear */ /* Don't bother with register 9, an extension of registers 6&7. If we do use the 6&7 values the atomic update assumption above is invalid. */ - vp->stats.rx_bytes += ioread16(ioaddr + 10); - vp->stats.tx_bytes += ioread16(ioaddr + 12); + dev->stats.rx_bytes += ioread16(ioaddr + 10); + dev->stats.tx_bytes += ioread16(ioaddr + 12); /* Extra stats for get_ethtool_stats() */ vp->xstats.tx_multiple_collisions += ioread8(ioaddr + 2); vp->xstats.tx_single_collisions += ioread8(ioaddr + 3); @@ -2767,14 +2766,14 @@ static void update_stats(void __iomem *ioaddr, struct net_device *dev) EL3WINDOW(4); vp->xstats.rx_bad_ssd += ioread8(ioaddr + 12); - vp->stats.collisions = vp->xstats.tx_multiple_collisions + dev->stats.collisions = vp->xstats.tx_multiple_collisions + vp->xstats.tx_single_collisions + vp->xstats.tx_max_collisions; { u8 up = ioread8(ioaddr + 13); - vp->stats.rx_bytes += (up & 0x0f) << 16; - vp->stats.tx_bytes += (up & 0xf0) << 12; + dev->stats.rx_bytes += (up & 0x0f) << 16; + dev->stats.tx_bytes += (up & 0xf0) << 12; } EL3WINDOW(old_window >> 13); diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index af46341827f..d27f54a2df7 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -1273,20 +1273,6 @@ config PCNET32 To compile this driver as a module, choose M here. The module will be called pcnet32. -config PCNET32_NAPI - bool "Use RX polling (NAPI)" - depends on PCNET32 - help - NAPI is a new driver API designed to reduce CPU and interrupt load - when the driver is receiving lots of packets from the card. It is - still somewhat experimental and thus not yet enabled by default. - - If your estimated Rx load is 10kpps or more, or if the card will be - deployed on potentially unfriendly networks (e.g. in a firewall), - then say Y here. - - If in doubt, say N. - config AMD8111_ETH tristate "AMD 8111 (new PCI lance) support" depends on NET_PCI && PCI diff --git a/drivers/net/appletalk/cops.c b/drivers/net/appletalk/cops.c index 82e9a5bd0dd..a0b4c851607 100644 --- a/drivers/net/appletalk/cops.c +++ b/drivers/net/appletalk/cops.c @@ -499,19 +499,13 @@ static void cops_reset(struct net_device *dev, int sleep) { outb(0, ioaddr+DAYNA_RESET); /* Assert the reset port */ inb(ioaddr+DAYNA_RESET); /* Clear the reset */ - if(sleep) - { - long snap=jiffies; - - /* Let card finish initializing, about 1/3 second */ - while (time_before(jiffies, snap + HZ/3)) - schedule(); - } - else - mdelay(333); + if (sleep) + msleep(333); + else + mdelay(333); } + netif_wake_queue(dev); - return; } static void cops_load (struct net_device *dev) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 6425603bc37..50a40e43315 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1425,13 +1425,13 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) res = netdev_set_master(slave_dev, bond_dev); if (res) { dprintk("Error %d calling netdev_set_master\n", res); - goto err_close; + goto err_restore_mac; } /* open the slave since the application closed it */ res = dev_open(slave_dev); if (res) { dprintk("Openning slave %s failed\n", slave_dev->name); - goto err_restore_mac; + goto err_unset_master; } new_slave->dev = slave_dev; @@ -1444,7 +1444,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) */ res = bond_alb_init_slave(bond, new_slave); if (res) { - goto err_unset_master; + goto err_close; } } @@ -1619,7 +1619,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) res = bond_create_slave_symlinks(bond_dev, slave_dev); if (res) - goto err_unset_master; + goto err_close; printk(KERN_INFO DRV_NAME ": %s: enslaving %s as a%s interface with a%s link.\n", @@ -1631,12 +1631,12 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) return 0; /* Undo stages on error */ -err_unset_master: - netdev_set_master(slave_dev, NULL); - err_close: dev_close(slave_dev); +err_unset_master: + netdev_set_master(slave_dev, NULL); + err_restore_mac: if (!bond->params.fail_over_mac) { memcpy(addr.sa_data, new_slave->perm_hwaddr, ETH_ALEN); @@ -4936,7 +4936,9 @@ int bond_create(char *name, struct bond_params *params, struct bonding **newbond if (res < 0) { rtnl_lock(); down_write(&bonding_rwsem); - goto out_bond; + bond_deinit(bond_dev); + unregister_netdevice(bond_dev); + goto out_rtnl; } return 0; @@ -4990,9 +4992,10 @@ err: destroy_workqueue(bond->wq); } + bond_destroy_sysfs(); + rtnl_lock(); bond_free_all(); - bond_destroy_sysfs(); rtnl_unlock(); out: return res; @@ -5004,9 +5007,10 @@ static void __exit bonding_exit(void) unregister_netdevice_notifier(&bond_netdev_notifier); unregister_inetaddr_notifier(&bond_inetaddr_notifier); + bond_destroy_sysfs(); + rtnl_lock(); bond_free_all(); - bond_destroy_sysfs(); rtnl_unlock(); } diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c index 979c2d05ff9..68c41a00d93 100644 --- a/drivers/net/bonding/bond_sysfs.c +++ b/drivers/net/bonding/bond_sysfs.c @@ -146,29 +146,29 @@ static ssize_t bonding_store_bonds(struct class *cls, const char *buffer, size_t ": Unable remove bond %s due to open references.\n", ifname); res = -EPERM; - goto out; + goto out_unlock; } printk(KERN_INFO DRV_NAME ": %s is being deleted...\n", bond->dev->name); bond_destroy(bond); - up_write(&bonding_rwsem); - rtnl_unlock(); - goto out; + goto out_unlock; } printk(KERN_ERR DRV_NAME ": unable to delete non-existent bond %s\n", ifname); res = -ENODEV; - up_write(&bonding_rwsem); - rtnl_unlock(); - goto out; + goto out_unlock; } err_no_cmd: printk(KERN_ERR DRV_NAME ": no command found in bonding_masters. Use +ifname or -ifname.\n"); - res = -EPERM; + return -EPERM; + +out_unlock: + up_write(&bonding_rwsem); + rtnl_unlock(); /* Always return either count or an error. If you return 0, you'll * get called forever, which is bad. diff --git a/drivers/net/cxgb3/cxgb3_main.c b/drivers/net/cxgb3/cxgb3_main.c index 05e5f59e87f..ce949d5fae3 100644 --- a/drivers/net/cxgb3/cxgb3_main.c +++ b/drivers/net/cxgb3/cxgb3_main.c @@ -1894,11 +1894,11 @@ static int cxgb_extension_ioctl(struct net_device *dev, void __user *useraddr) u8 *fw_data; struct ch_mem_range t; - if (!capable(CAP_NET_ADMIN)) + if (!capable(CAP_SYS_RAWIO)) return -EPERM; if (copy_from_user(&t, useraddr, sizeof(t))) return -EFAULT; - + /* Check t.len sanity ? */ fw_data = kmalloc(t.len, GFP_KERNEL); if (!fw_data) return -ENOMEM; diff --git a/drivers/net/e1000e/defines.h b/drivers/net/e1000e/defines.h index 2a53875cddb..f823b8ba578 100644 --- a/drivers/net/e1000e/defines.h +++ b/drivers/net/e1000e/defines.h @@ -648,6 +648,8 @@ #define IFE_E_PHY_ID 0x02A80330 #define IFE_PLUS_E_PHY_ID 0x02A80320 #define IFE_C_E_PHY_ID 0x02A80310 +#define BME1000_E_PHY_ID 0x01410CB0 +#define BME1000_E_PHY_ID_R2 0x01410CB1 /* M88E1000 Specific Registers */ #define M88E1000_PHY_SPEC_CTRL 0x10 /* PHY Specific Control Register */ @@ -701,6 +703,14 @@ #define M88EC018_EPSCR_DOWNSHIFT_COUNTER_MASK 0x0E00 #define M88EC018_EPSCR_DOWNSHIFT_COUNTER_5X 0x0800 +/* BME1000 PHY Specific Control Register */ +#define BME1000_PSCR_ENABLE_DOWNSHIFT 0x0800 /* 1 = enable downshift */ + + +#define PHY_PAGE_SHIFT 5 +#define PHY_REG(page, reg) (((page) << PHY_PAGE_SHIFT) | \ + ((reg) & MAX_PHY_REG_ADDRESS)) + /* * Bits... * 15-5: page diff --git a/drivers/net/e1000e/e1000.h b/drivers/net/e1000e/e1000.h index 38bfd0d261f..d3bc6f8101f 100644 --- a/drivers/net/e1000e/e1000.h +++ b/drivers/net/e1000e/e1000.h @@ -127,7 +127,7 @@ struct e1000_buffer { /* arrays of page information for packet split */ struct e1000_ps_page *ps_pages; }; - + struct page *page; }; struct e1000_ring { @@ -304,6 +304,7 @@ struct e1000_info { #define FLAG_HAS_CTRLEXT_ON_LOAD (1 << 5) #define FLAG_HAS_SWSM_ON_LOAD (1 << 6) #define FLAG_HAS_JUMBO_FRAMES (1 << 7) +#define FLAG_IS_ICH (1 << 9) #define FLAG_HAS_SMART_POWER_DOWN (1 << 11) #define FLAG_IS_QUAD_PORT_A (1 << 12) #define FLAG_IS_QUAD_PORT (1 << 13) @@ -386,6 +387,7 @@ extern void e1000e_set_kmrn_lock_loss_workaround_ich8lan(struct e1000_hw *hw, bool state); extern void e1000e_igp3_phy_powerdown_workaround_ich8lan(struct e1000_hw *hw); extern void e1000e_gig_downshift_workaround_ich8lan(struct e1000_hw *hw); +extern void e1000e_disable_gig_wol_ich8lan(struct e1000_hw *hw); extern s32 e1000e_check_for_copper_link(struct e1000_hw *hw); extern s32 e1000e_check_for_fiber_link(struct e1000_hw *hw); @@ -443,6 +445,9 @@ extern s32 e1000e_get_phy_info_m88(struct e1000_hw *hw); extern s32 e1000e_read_phy_reg_m88(struct e1000_hw *hw, u32 offset, u16 *data); extern s32 e1000e_write_phy_reg_m88(struct e1000_hw *hw, u32 offset, u16 data); extern enum e1000_phy_type e1000e_get_phy_type_from_id(u32 phy_id); +extern s32 e1000e_determine_phy_address(struct e1000_hw *hw); +extern s32 e1000e_write_phy_reg_bm(struct e1000_hw *hw, u32 offset, u16 data); +extern s32 e1000e_read_phy_reg_bm(struct e1000_hw *hw, u32 offset, u16 *data); extern void e1000e_phy_force_speed_duplex_setup(struct e1000_hw *hw, u16 *phy_ctrl); extern s32 e1000e_write_kmrn_reg(struct e1000_hw *hw, u32 offset, u16 data); extern s32 e1000e_read_kmrn_reg(struct e1000_hw *hw, u32 offset, u16 *data); diff --git a/drivers/net/e1000e/ethtool.c b/drivers/net/e1000e/ethtool.c index ce045acce63..a14561f40db 100644 --- a/drivers/net/e1000e/ethtool.c +++ b/drivers/net/e1000e/ethtool.c @@ -494,8 +494,12 @@ static int e1000_get_eeprom(struct net_device *netdev, for (i = 0; i < last_word - first_word + 1; i++) { ret_val = e1000_read_nvm(hw, first_word + i, 1, &eeprom_buff[i]); - if (ret_val) + if (ret_val) { + /* a read error occurred, throw away the + * result */ + memset(eeprom_buff, 0xff, sizeof(eeprom_buff)); break; + } } } @@ -803,8 +807,7 @@ static int e1000_reg_test(struct e1000_adapter *adapter, u64 *data) /* restore previous status */ ew32(STATUS, before); - if ((mac->type != e1000_ich8lan) && - (mac->type != e1000_ich9lan)) { + if (!(adapter->flags & FLAG_IS_ICH)) { REG_PATTERN_TEST(E1000_FCAL, 0xFFFFFFFF, 0xFFFFFFFF); REG_PATTERN_TEST(E1000_FCAH, 0x0000FFFF, 0xFFFFFFFF); REG_PATTERN_TEST(E1000_FCT, 0x0000FFFF, 0xFFFFFFFF); @@ -824,15 +827,13 @@ static int e1000_reg_test(struct e1000_adapter *adapter, u64 *data) REG_SET_AND_CHECK(E1000_RCTL, 0xFFFFFFFF, 0x00000000); - before = (((mac->type == e1000_ich8lan) || - (mac->type == e1000_ich9lan)) ? 0x06C3B33E : 0x06DFB3FE); + before = ((adapter->flags & FLAG_IS_ICH) ? 0x06C3B33E : 0x06DFB3FE); REG_SET_AND_CHECK(E1000_RCTL, before, 0x003FFFFB); REG_SET_AND_CHECK(E1000_TCTL, 0xFFFFFFFF, 0x00000000); REG_SET_AND_CHECK(E1000_RCTL, before, 0xFFFFFFFF); REG_PATTERN_TEST(E1000_RDBAL, 0xFFFFFFF0, 0xFFFFFFFF); - if ((mac->type != e1000_ich8lan) && - (mac->type != e1000_ich9lan)) + if (!(adapter->flags & FLAG_IS_ICH)) REG_PATTERN_TEST(E1000_TXCW, 0xC000FFFF, 0x0000FFFF); REG_PATTERN_TEST(E1000_TDBAL, 0xFFFFFFF0, 0xFFFFFFFF); REG_PATTERN_TEST(E1000_TIDV, 0x0000FFFF, 0x0000FFFF); @@ -911,9 +912,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data) /* Test each interrupt */ for (i = 0; i < 10; i++) { - - if (((adapter->hw.mac.type == e1000_ich8lan) || - (adapter->hw.mac.type == e1000_ich9lan)) && i == 8) + if ((adapter->flags & FLAG_IS_ICH) && (i == 8)) continue; /* Interrupt to test */ @@ -1184,6 +1183,7 @@ static int e1000_integrated_phy_loopback(struct e1000_adapter *adapter) struct e1000_hw *hw = &adapter->hw; u32 ctrl_reg = 0; u32 stat_reg = 0; + u16 phy_reg = 0; hw->mac.autoneg = 0; @@ -1211,6 +1211,28 @@ static int e1000_integrated_phy_loopback(struct e1000_adapter *adapter) E1000_CTRL_SPD_100 |/* Force Speed to 100 */ E1000_CTRL_FD); /* Force Duplex to FULL */ break; + case e1000_phy_bm: + /* Set Default MAC Interface speed to 1GB */ + e1e_rphy(hw, PHY_REG(2, 21), &phy_reg); + phy_reg &= ~0x0007; + phy_reg |= 0x006; + e1e_wphy(hw, PHY_REG(2, 21), phy_reg); + /* Assert SW reset for above settings to take effect */ + e1000e_commit_phy(hw); + mdelay(1); + /* Force Full Duplex */ + e1e_rphy(hw, PHY_REG(769, 16), &phy_reg); + e1e_wphy(hw, PHY_REG(769, 16), phy_reg | 0x000C); + /* Set Link Up (in force link) */ + e1e_rphy(hw, PHY_REG(776, 16), &phy_reg); + e1e_wphy(hw, PHY_REG(776, 16), phy_reg | 0x0040); + /* Force Link */ + e1e_rphy(hw, PHY_REG(769, 16), &phy_reg); + e1e_wphy(hw, PHY_REG(769, 16), phy_reg | 0x0040); + /* Set Early Link Enable */ + e1e_rphy(hw, PHY_REG(769, 20), &phy_reg); + e1e_wphy(hw, PHY_REG(769, 20), phy_reg | 0x0400); + /* fall through */ default: /* force 1000, set loopback */ e1e_wphy(hw, PHY_CONTROL, 0x4140); @@ -1224,8 +1246,7 @@ static int e1000_integrated_phy_loopback(struct e1000_adapter *adapter) E1000_CTRL_SPD_1000 |/* Force Speed to 1000 */ E1000_CTRL_FD); /* Force Duplex to FULL */ - if ((adapter->hw.mac.type == e1000_ich8lan) || - (adapter->hw.mac.type == e1000_ich9lan)) + if (adapter->flags & FLAG_IS_ICH) ctrl_reg |= E1000_CTRL_SLU; /* Set Link Up */ } diff --git a/drivers/net/e1000e/hw.h b/drivers/net/e1000e/hw.h index a930e6d9cf0..74f263acb17 100644 --- a/drivers/net/e1000e/hw.h +++ b/drivers/net/e1000e/hw.h @@ -216,6 +216,21 @@ enum e1e_registers { #define IGP01E1000_PHY_LINK_HEALTH 0x13 /* PHY Link Health */ #define IGP02E1000_PHY_POWER_MGMT 0x19 /* Power Management */ #define IGP01E1000_PHY_PAGE_SELECT 0x1F /* Page Select */ +#define BM_PHY_PAGE_SELECT 22 /* Page Select for BM */ +#define IGP_PAGE_SHIFT 5 +#define PHY_REG_MASK 0x1F + +#define BM_WUC_PAGE 800 +#define BM_WUC_ADDRESS_OPCODE 0x11 +#define BM_WUC_DATA_OPCODE 0x12 +#define BM_WUC_ENABLE_PAGE 769 +#define BM_WUC_ENABLE_REG 17 +#define BM_WUC_ENABLE_BIT (1 << 2) +#define BM_WUC_HOST_WU_BIT (1 << 4) + +#define BM_WUC PHY_REG(BM_WUC_PAGE, 1) +#define BM_WUFC PHY_REG(BM_WUC_PAGE, 2) +#define BM_WUS PHY_REG(BM_WUC_PAGE, 3) #define IGP01E1000_PHY_PCS_INIT_REG 0x00B4 #define IGP01E1000_PHY_POLARITY_MASK 0x0078 @@ -331,10 +346,16 @@ enum e1e_registers { #define E1000_DEV_ID_ICH8_IFE_G 0x10C5 #define E1000_DEV_ID_ICH8_IGP_M 0x104D #define E1000_DEV_ID_ICH9_IGP_AMT 0x10BD +#define E1000_DEV_ID_ICH9_IGP_M_AMT 0x10F5 +#define E1000_DEV_ID_ICH9_IGP_M 0x10BF +#define E1000_DEV_ID_ICH9_IGP_M_V 0x10CB #define E1000_DEV_ID_ICH9_IGP_C 0x294C #define E1000_DEV_ID_ICH9_IFE 0x10C0 #define E1000_DEV_ID_ICH9_IFE_GT 0x10C3 #define E1000_DEV_ID_ICH9_IFE_G 0x10C2 +#define E1000_DEV_ID_ICH10_R_BM_LM 0x10CC +#define E1000_DEV_ID_ICH10_R_BM_LF 0x10CD +#define E1000_DEV_ID_ICH10_R_BM_V 0x10CE #define E1000_FUNC_1 1 @@ -378,6 +399,7 @@ enum e1000_phy_type { e1000_phy_gg82563, e1000_phy_igp_3, e1000_phy_ife, + e1000_phy_bm, }; enum e1000_bus_width { diff --git a/drivers/net/e1000e/ich8lan.c b/drivers/net/e1000e/ich8lan.c index 768485dbb2c..9e38452a738 100644 --- a/drivers/net/e1000e/ich8lan.c +++ b/drivers/net/e1000e/ich8lan.c @@ -38,6 +38,12 @@ * 82566DM Gigabit Network Connection * 82566MC Gigabit Network Connection * 82566MM Gigabit Network Connection + * 82567LM Gigabit Network Connection + * 82567LF Gigabit Network Connection + * 82567LM-2 Gigabit Network Connection + * 82567LF-2 Gigabit Network Connection + * 82567V-2 Gigabit Network Connection + * 82562GT-3 10/100 Network Connection */ #include <linux/netdevice.h> @@ -198,6 +204,19 @@ static s32 e1000_init_phy_params_ich8lan(struct e1000_hw *hw) phy->addr = 1; phy->reset_delay_us = 100; + /* + * We may need to do this twice - once for IGP and if that fails, + * we'll set BM func pointers and try again + */ + ret_val = e1000e_determine_phy_address(hw); + if (ret_val) { + hw->phy.ops.write_phy_reg = e1000e_write_phy_reg_bm; + hw->phy.ops.read_phy_reg = e1000e_read_phy_reg_bm; + ret_val = e1000e_determine_phy_address(hw); + if (ret_val) + return ret_val; + } + phy->id = 0; while ((e1000_phy_unknown == e1000e_get_phy_type_from_id(phy->id)) && (i++ < 100)) { @@ -219,6 +238,13 @@ static s32 e1000_init_phy_params_ich8lan(struct e1000_hw *hw) phy->type = e1000_phy_ife; phy->autoneg_mask = E1000_ALL_NOT_GIG; break; + case BME1000_E_PHY_ID: + phy->type = e1000_phy_bm; + phy->autoneg_mask = AUTONEG_ADVERTISE_SPEED_DEFAULT; + hw->phy.ops.read_phy_reg = e1000e_read_phy_reg_bm; + hw->phy.ops.write_phy_reg = e1000e_write_phy_reg_bm; + hw->phy.ops.commit_phy = e1000e_phy_sw_reset; + break; default: return -E1000_ERR_PHY; break; @@ -664,6 +690,7 @@ static s32 e1000_get_phy_info_ich8lan(struct e1000_hw *hw) return e1000_get_phy_info_ife_ich8lan(hw); break; case e1000_phy_igp_3: + case e1000_phy_bm: return e1000e_get_phy_info_igp(hw); break; default: @@ -728,7 +755,7 @@ static s32 e1000_set_d0_lplu_state_ich8lan(struct e1000_hw *hw, bool active) s32 ret_val = 0; u16 data; - if (phy->type != e1000_phy_igp_3) + if (phy->type == e1000_phy_ife) return ret_val; phy_ctrl = er32(PHY_CTRL); @@ -1918,8 +1945,35 @@ static s32 e1000_setup_copper_link_ich8lan(struct e1000_hw *hw) ret_val = e1000e_copper_link_setup_igp(hw); if (ret_val) return ret_val; + } else if (hw->phy.type == e1000_phy_bm) { + ret_val = e1000e_copper_link_setup_m88(hw); + if (ret_val) + return ret_val; } + if (hw->phy.type == e1000_phy_ife) { + ret_val = e1e_rphy(hw, IFE_PHY_MDIX_CONTROL, ®_data); + if (ret_val) + return ret_val; + + reg_data &= ~IFE_PMC_AUTO_MDIX; + + switch (hw->phy.mdix) { + case 1: + reg_data &= ~IFE_PMC_FORCE_MDIX; + break; + case 2: + reg_data |= IFE_PMC_FORCE_MDIX; + break; + case 0: + default: + reg_data |= IFE_PMC_AUTO_MDIX; + break; + } + ret_val = e1e_wphy(hw, IFE_PHY_MDIX_CONTROL, reg_data); + if (ret_val) + return ret_val; + } return e1000e_setup_copper_link(hw); } @@ -2127,6 +2181,31 @@ void e1000e_gig_downshift_workaround_ich8lan(struct e1000_hw *hw) } /** + * e1000e_disable_gig_wol_ich8lan - disable gig during WoL + * @hw: pointer to the HW structure + * + * During S0 to Sx transition, it is possible the link remains at gig + * instead of negotiating to a lower speed. Before going to Sx, set + * 'LPLU Enabled' and 'Gig Disable' to force link speed negotiation + * to a lower speed. + * + * Should only be called for ICH9 devices. + **/ +void e1000e_disable_gig_wol_ich8lan(struct e1000_hw *hw) +{ + u32 phy_ctrl; + + if (hw->mac.type == e1000_ich9lan) { + phy_ctrl = er32(PHY_CTRL); + phy_ctrl |= E1000_PHY_CTRL_D0A_LPLU | + E1000_PHY_CTRL_GBE_DISABLE; + ew32(PHY_CTRL, phy_ctrl); + } + + return; +} + +/** * e1000_cleanup_led_ich8lan - Restore the default LED operation * @hw: pointer to the HW structure * @@ -2247,6 +2326,7 @@ static struct e1000_nvm_operations ich8_nvm_ops = { struct e1000_info e1000_ich8_info = { .mac = e1000_ich8lan, .flags = FLAG_HAS_WOL + | FLAG_IS_ICH | FLAG_RX_CSUM_ENABLED | FLAG_HAS_CTRLEXT_ON_LOAD | FLAG_HAS_AMT @@ -2262,6 +2342,7 @@ struct e1000_info e1000_ich8_info = { struct e1000_info e1000_ich9_info = { .mac = e1000_ich9lan, .flags = FLAG_HAS_JUMBO_FRAMES + | FLAG_IS_ICH | FLAG_HAS_WOL | FLAG_RX_CSUM_ENABLED | FLAG_HAS_CTRLEXT_ON_LOAD diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c index 8991ab8911e..8cbb40f3a50 100644 --- a/drivers/net/e1000e/netdev.c +++ b/drivers/net/e1000e/netdev.c @@ -43,10 +43,11 @@ #include <linux/if_vlan.h> #include <linux/cpu.h> #include <linux/smp.h> +#include <linux/pm_qos_params.h> #include "e1000.h" -#define DRV_VERSION "0.2.1" +#define DRV_VERSION "0.3.3.3-k2" char e1000e_driver_name[] = "e1000e"; const char e1000e_driver_version[] = DRV_VERSION; @@ -341,6 +342,89 @@ no_buffers: } /** + * e1000_alloc_jumbo_rx_buffers - Replace used jumbo receive buffers + * @adapter: address of board private structure + * @rx_ring: pointer to receive ring structure + * @cleaned_count: number of buffers to allocate this pass + **/ + +static void e1000_alloc_jumbo_rx_buffers(struct e1000_adapter *adapter, + int cleaned_count) +{ + struct net_device *netdev = adapter->netdev; + struct pci_dev *pdev = adapter->pdev; + struct e1000_rx_desc *rx_desc; + struct e1000_ring *rx_ring = adapter->rx_ring; + struct e1000_buffer *buffer_info; + struct sk_buff *skb; + unsigned int i; + unsigned int bufsz = 256 - + 16 /* for skb_reserve */ - + NET_IP_ALIGN; + + i = rx_ring->next_to_use; + buffer_info = &rx_ring->buffer_info[i]; + + while (cleaned_count--) { + skb = buffer_info->skb; + if (skb) { + skb_trim(skb, 0); + goto check_page; + } + + skb = netdev_alloc_skb(netdev, bufsz); + if (unlikely(!skb)) { + /* Better luck next round */ + adapter->alloc_rx_buff_failed++; + break; + } + + /* Make buffer alignment 2 beyond a 16 byte boundary + * this will result in a 16 byte aligned IP header after + * the 14 byte MAC header is removed + */ + skb_reserve(skb, NET_IP_ALIGN); + + buffer_info->skb = skb; +check_page: + /* allocate a new page if necessary */ + if (!buffer_info->page) { + buffer_info->page = alloc_page(GFP_ATOMIC); + if (unlikely(!buffer_info->page)) { + adapter->alloc_rx_buff_failed++; + break; + } + } + + if (!buffer_info->dma) + buffer_info->dma = pci_map_page(pdev, + buffer_info->page, 0, + PAGE_SIZE, + PCI_DMA_FROMDEVICE); + + rx_desc = E1000_RX_DESC(*rx_ring, i); + rx_desc->buffer_addr = cpu_to_le64(buffer_info->dma); + + if (unlikely(++i == rx_ring->count)) + i = 0; + buffer_info = &rx_ring->buffer_info[i]; + } + + if (likely(rx_ring->next_to_use != i)) { + rx_ring->next_to_use = i; + if (unlikely(i-- == 0)) + i = (rx_ring->count - 1); + + /* Force memory writes to complete before letting h/w + * know there are new descriptors to fetch. (Only + * applicable for weak-ordered memory model archs, + * such as IA-64). */ + wmb(); + writel(i, adapter->hw.hw_addr + rx_ring->tail); + } +} + +/** * e1000_clean_rx_irq - Send received data up the network stack; legacy * @adapter: board private structure * @@ -783,6 +867,186 @@ next_desc: } /** + * e1000_consume_page - helper function + **/ +static void e1000_consume_page(struct e1000_buffer *bi, struct sk_buff *skb, + u16 length) +{ + bi->page = NULL; + skb->len += length; + skb->data_len += length; + skb->truesize += length; +} + +/** + * e1000_clean_jumbo_rx_irq - Send received data up the network stack; legacy + * @adapter: board private structure + * + * the return value indicates whether actual cleaning was done, there + * is no guarantee that everything was cleaned + **/ + +static bool e1000_clean_jumbo_rx_irq(struct e1000_adapter *adapter, + int *work_done, int work_to_do) +{ + struct net_device *netdev = adapter->netdev; + struct pci_dev *pdev = adapter->pdev; + struct e1000_ring *rx_ring = adapter->rx_ring; + struct e1000_rx_desc *rx_desc, *next_rxd; + struct e1000_buffer *buffer_info, *next_buffer; + u32 length; + unsigned int i; + int cleaned_count = 0; + bool cleaned = false; + unsigned int total_rx_bytes=0, total_rx_packets=0; + + i = rx_ring->next_to_clean; + rx_desc = E1000_RX_DESC(*rx_ring, i); + buffer_info = &rx_ring->buffer_info[i]; + + while (rx_desc->status & E1000_RXD_STAT_DD) { + struct sk_buff *skb; + u8 status; + + if (*work_done >= work_to_do) + break; + (*work_done)++; + + status = rx_desc->status; + skb = buffer_info->skb; + buffer_info->skb = NULL; + + ++i; + if (i == rx_ring->count) + i = 0; + next_rxd = E1000_RX_DESC(*rx_ring, i); + prefetch(next_rxd); + + next_buffer = &rx_ring->buffer_info[i]; + + cleaned = true; + cleaned_count++; + pci_unmap_page(pdev, buffer_info->dma, PAGE_SIZE, + PCI_DMA_FROMDEVICE); + buffer_info->dma = 0; + + length = le16_to_cpu(rx_desc->length); + + /* errors is only valid for DD + EOP descriptors */ + if (unlikely((status & E1000_RXD_STAT_EOP) && + (rx_desc->errors & E1000_RXD_ERR_FRAME_ERR_MASK))) { + /* recycle both page and skb */ + buffer_info->skb = skb; + /* an error means any chain goes out the window + * too */ + if (rx_ring->rx_skb_top) + dev_kfree_skb(rx_ring->rx_skb_top); + rx_ring->rx_skb_top = NULL; + goto next_desc; + } + +#define rxtop rx_ring->rx_skb_top + if (!(status & E1000_RXD_STAT_EOP)) { + /* this descriptor is only the beginning (or middle) */ + if (!rxtop) { + /* this is the beginning of a chain */ + rxtop = skb; + skb_fill_page_desc(rxtop, 0, buffer_info->page, + 0, length); + } else { + /* this is the middle of a chain */ + skb_fill_page_desc(rxtop, + skb_shinfo(rxtop)->nr_frags, + buffer_info->page, 0, length); + /* re-use the skb, only consumed the page */ + buffer_info->skb = skb; + } + e1000_consume_page(buffer_info, rxtop, length); + goto next_desc; + } else { + if (rxtop) { + /* end of the chain */ + skb_fill_page_desc(rxtop, + skb_shinfo(rxtop)->nr_frags, + buffer_info->page, 0, length); + /* re-use the current skb, we only consumed the + * page */ + buffer_info->skb = skb; + skb = rxtop; + rxtop = NULL; + e1000_consume_page(buffer_info, skb, length); + } else { + /* no chain, got EOP, this buf is the packet + * copybreak to save the put_page/alloc_page */ + if (length <= copybreak && + skb_tailroom(skb) >= length) { + u8 *vaddr; + vaddr = kmap_atomic(buffer_info->page, + KM_SKB_DATA_SOFTIRQ); + memcpy(skb_tail_pointer(skb), vaddr, + length); + kunmap_atomic(vaddr, + KM_SKB_DATA_SOFTIRQ); + /* re-use the page, so don't erase + * buffer_info->page */ + skb_put(skb, length); + } else { + skb_fill_page_desc(skb, 0, + buffer_info->page, 0, + length); + e1000_consume_page(buffer_info, skb, + length); + } + } + } + + /* Receive Checksum Offload XXX recompute due to CRC strip? */ + e1000_rx_checksum(adapter, + (u32)(status) | + ((u32)(rx_desc->errors) << 24), + le16_to_cpu(rx_desc->csum), skb); + + /* probably a little skewed due to removing CRC */ + total_rx_bytes += skb->len; + total_rx_packets++; + + /* eth type trans needs skb->data to point to something */ + if (!pskb_may_pull(skb, ETH_HLEN)) { + ndev_err(netdev, "pskb_may_pull failed.\n"); + dev_kfree_skb(skb); + goto next_desc; + } + + e1000_receive_skb(adapter, netdev, skb, status, + rx_desc->special); + +next_desc: + rx_desc->status = 0; + + /* return some buffers to hardware, one at a time is too slow */ + if (unlikely(cleaned_count >= E1000_RX_BUFFER_WRITE)) { + adapter->alloc_rx_buf(adapter, cleaned_count); + cleaned_count = 0; + } + + /* use prefetched values */ + rx_desc = next_rxd; + buffer_info = next_buffer; + } + rx_ring->next_to_clean = i; + + cleaned_count = e1000_desc_unused(rx_ring); + if (cleaned_count) + adapter->alloc_rx_buf(adapter, cleaned_count); + + adapter->total_rx_bytes += total_rx_bytes; + adapter->total_rx_packets += total_rx_packets; + adapter->net_stats.rx_bytes += total_rx_bytes; + adapter->net_stats.rx_packets += total_rx_packets; + return cleaned; +} + +/** * e1000_clean_rx_ring - Free Rx Buffers per Queue * @adapter: board private structure **/ @@ -802,6 +1066,10 @@ static void e1000_clean_rx_ring(struct e1000_adapter *adapter) pci_unmap_single(pdev, buffer_info->dma, adapter->rx_buffer_len, PCI_DMA_FROMDEVICE); + else if (adapter->clean_rx == e1000_clean_jumbo_rx_irq) + pci_unmap_page(pdev, buffer_info->dma, + PAGE_SIZE, + PCI_DMA_FROMDEVICE); else if (adapter->clean_rx == e1000_clean_rx_irq_ps) pci_unmap_single(pdev, buffer_info->dma, adapter->rx_ps_bsize0, @@ -809,6 +1077,11 @@ static void e1000_clean_rx_ring(struct e1000_adapter *adapter) buffer_info->dma = 0; } + if (buffer_info->page) { + put_page(buffer_info->page); + buffer_info->page = NULL; + } + if (buffer_info->skb) { dev_kfree_skb(buffer_info->skb); buffer_info->skb = NULL; @@ -1755,10 +2028,12 @@ static void e1000_setup_rctl(struct e1000_adapter *adapter) * a lot of memory, since we allocate 3 pages at all times * per packet. */ - adapter->rx_ps_pages = 0; pages = PAGE_USE_COUNT(adapter->netdev->mtu); - if ((pages <= 3) && (PAGE_SIZE <= 16384) && (rctl & E1000_RCTL_LPE)) + if (!(adapter->flags & FLAG_IS_ICH) && (pages <= 3) && + (PAGE_SIZE <= 16384) && (rctl & E1000_RCTL_LPE)) adapter->rx_ps_pages = pages; + else + adapter->rx_ps_pages = 0; if (adapter->rx_ps_pages) { /* Configure extra packet-split registers */ @@ -1819,9 +2094,12 @@ static void e1000_configure_rx(struct e1000_adapter *adapter) sizeof(union e1000_rx_desc_packet_split); adapter->clean_rx = e1000_clean_rx_irq_ps; adapter->alloc_rx_buf = e1000_alloc_rx_buffers_ps; + } else if (adapter->netdev->mtu > ETH_FRAME_LEN + ETH_FCS_LEN) { + rdlen = rx_ring->count * sizeof(struct e1000_rx_desc); + adapter->clean_rx = e1000_clean_jumbo_rx_irq; + adapter->alloc_rx_buf = e1000_alloc_jumbo_rx_buffers; } else { - rdlen = rx_ring->count * - sizeof(struct e1000_rx_desc); + rdlen = rx_ring->count * sizeof(struct e1000_rx_desc); adapter->clean_rx = e1000_clean_rx_irq; adapter->alloc_rx_buf = e1000_alloc_rx_buffers; } @@ -1885,8 +2163,21 @@ static void e1000_configure_rx(struct e1000_adapter *adapter) * units), e.g. using jumbo frames when setting to E1000_ERT_2048 */ if ((adapter->flags & FLAG_HAS_ERT) && - (adapter->netdev->mtu > ETH_DATA_LEN)) - ew32(ERT, E1000_ERT_2048); + (adapter->netdev->mtu > ETH_DATA_LEN)) { + u32 rxdctl = er32(RXDCTL(0)); + ew32(RXDCTL(0), rxdctl | 0x3); + ew32(ERT, E1000_ERT_2048 | (1 << 13)); + /* + * With jumbo frames and early-receive enabled, excessive + * C4->C2 latencies result in dropped transactions. + */ + pm_qos_update_requirement(PM_QOS_CPU_DMA_LATENCY, + e1000e_driver_name, 55); + } else { + pm_qos_update_requirement(PM_QOS_CPU_DMA_LATENCY, + e1000e_driver_name, + PM_QOS_DEFAULT_VALUE); + } /* Enable Receives */ ew32(RCTL, rctl); @@ -2155,6 +2446,14 @@ void e1000e_reset(struct e1000_adapter *adapter) /* Allow time for pending master requests to run */ mac->ops.reset_hw(hw); + + /* + * For parts with AMT enabled, let the firmware know + * that the network interface is in control + */ + if ((adapter->flags & FLAG_HAS_AMT) && e1000e_check_mng_mode(hw)) + e1000_get_hw_control(adapter); + ew32(WUC, 0); if (mac->ops.init_hw(hw)) @@ -3469,6 +3768,8 @@ static int e1000_change_mtu(struct net_device *netdev, int new_mtu) * means we reserve 2 more, this pushes us to allocate from the next * larger slab size. * i.e. RXBUFFER_2048 --> size-4096 slab + * However with the new *_jumbo_rx* routines, jumbo receives will use + * fragmented skbs */ if (max_frame <= 256) @@ -3626,6 +3927,9 @@ static int e1000_suspend(struct pci_dev *pdev, pm_message_t state) ew32(CTRL_EXT, ctrl_ext); } + if (adapter->flags & FLAG_IS_ICH) + e1000e_disable_gig_wol_ich8lan(&adapter->hw); + /* Allow time for pending master requests to run */ e1000e_disable_pcie_master(&adapter->hw); @@ -4292,6 +4596,13 @@ static struct pci_device_id e1000_pci_tbl[] = { { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IFE_GT), board_ich9lan }, { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_AMT), board_ich9lan }, { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_C), board_ich9lan }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M), board_ich9lan }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M_AMT), board_ich9lan }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M_V), board_ich9lan }, + + { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_R_BM_LM), board_ich9lan }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_R_BM_LF), board_ich9lan }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_R_BM_V), board_ich9lan }, { } /* terminate list */ }; @@ -4326,7 +4637,9 @@ static int __init e1000_init_module(void) printk(KERN_INFO "%s: Copyright (c) 1999-2008 Intel Corporation.\n", e1000e_driver_name); ret = pci_register_driver(&e1000_driver); - + pm_qos_add_requirement(PM_QOS_CPU_DMA_LATENCY, e1000e_driver_name, + PM_QOS_DEFAULT_VALUE); + return ret; } module_init(e1000_init_module); @@ -4340,6 +4653,7 @@ module_init(e1000_init_module); static void __exit e1000_exit_module(void) { pci_unregister_driver(&e1000_driver); + pm_qos_remove_requirement(PM_QOS_CPU_DMA_LATENCY, e1000e_driver_name); } module_exit(e1000_exit_module); diff --git a/drivers/net/e1000e/phy.c b/drivers/net/e1000e/phy.c index e102332a6be..b133dcf0e95 100644 --- a/drivers/net/e1000e/phy.c +++ b/drivers/net/e1000e/phy.c @@ -34,6 +34,9 @@ static s32 e1000_get_phy_cfg_done(struct e1000_hw *hw); static s32 e1000_phy_force_speed_duplex(struct e1000_hw *hw); static s32 e1000_set_d0_lplu_state(struct e1000_hw *hw, bool active); static s32 e1000_wait_autoneg(struct e1000_hw *hw); +static u32 e1000_get_phy_addr_for_bm_page(u32 page, u32 reg); +static s32 e1000_access_phy_wakeup_reg_bm(struct e1000_hw *hw, u32 offset, + u16 *data, bool read); /* Cable length tables */ static const u16 e1000_m88_cable_length_table[] = @@ -465,6 +468,10 @@ s32 e1000e_copper_link_setup_m88(struct e1000_hw *hw) if (phy->disable_polarity_correction == 1) phy_data |= M88E1000_PSCR_POLARITY_REVERSAL; + /* Enable downshift on BM (disabled by default) */ + if (phy->type == e1000_phy_bm) + phy_data |= BME1000_PSCR_ENABLE_DOWNSHIFT; + ret_val = e1e_wphy(hw, M88E1000_PHY_SPEC_CTRL, phy_data); if (ret_val) return ret_val; @@ -1776,6 +1783,10 @@ enum e1000_phy_type e1000e_get_phy_type_from_id(u32 phy_id) case IFE_C_E_PHY_ID: phy_type = e1000_phy_ife; break; + case BME1000_E_PHY_ID: + case BME1000_E_PHY_ID_R2: + phy_type = e1000_phy_bm; + break; default: phy_type = e1000_phy_unknown; break; @@ -1784,6 +1795,273 @@ enum e1000_phy_type e1000e_get_phy_type_from_id(u32 phy_id) } /** + * e1000e_determine_phy_address - Determines PHY address. + * @hw: pointer to the HW structure + * + * This uses a trial and error method to loop through possible PHY + * addresses. It tests each by reading the PHY ID registers and + * checking for a match. + **/ +s32 e1000e_determine_phy_address(struct e1000_hw *hw) +{ + s32 ret_val = -E1000_ERR_PHY_TYPE; + u32 phy_addr= 0; + u32 i = 0; + enum e1000_phy_type phy_type = e1000_phy_unknown; + + do { + for (phy_addr = 0; phy_addr < 4; phy_addr++) { + hw->phy.addr = phy_addr; + e1000e_get_phy_id(hw); + phy_type = e1000e_get_phy_type_from_id(hw->phy.id); + + /* + * If phy_type is valid, break - we found our + * PHY address + */ + if (phy_type != e1000_phy_unknown) { + ret_val = 0; + break; + } + } + i++; + } while ((ret_val != 0) && (i < 100)); + + return ret_val; +} + +/** + * e1000_get_phy_addr_for_bm_page - Retrieve PHY page address + * @page: page to access + * + * Returns the phy address for the page requested. + **/ +static u32 e1000_get_phy_addr_for_bm_page(u32 page, u32 reg) +{ + u32 phy_addr = 2; + + if ((page >= 768) || (page == 0 && reg == 25) || (reg == 31)) + phy_addr = 1; + + return phy_addr; +} + +/** + * e1000e_write_phy_reg_bm - Write BM PHY register + * @hw: pointer to the HW structure + * @offset: register offset to write to + * @data: data to write at register offset + * + * Acquires semaphore, if necessary, then writes the data to PHY register + * at the offset. Release any acquired semaphores before exiting. + **/ +s32 e1000e_write_phy_reg_bm(struct e1000_hw *hw, u32 offset, u16 data) +{ + s32 ret_val; + u32 page_select = 0; + u32 page = offset >> IGP_PAGE_SHIFT; + u32 page_shift = 0; + + /* Page 800 works differently than the rest so it has its own func */ + if (page == BM_WUC_PAGE) { + ret_val = e1000_access_phy_wakeup_reg_bm(hw, offset, &data, + false); + goto out; + } + + ret_val = hw->phy.ops.acquire_phy(hw); + if (ret_val) + goto out; + + hw->phy.addr = e1000_get_phy_addr_for_bm_page(page, offset); + + if (offset > MAX_PHY_MULTI_PAGE_REG) { + /* + * Page select is register 31 for phy address 1 and 22 for + * phy address 2 and 3. Page select is shifted only for + * phy address 1. + */ + if (hw->phy.addr == 1) { + page_shift = IGP_PAGE_SHIFT; + page_select = IGP01E1000_PHY_PAGE_SELECT; + } else { + page_shift = 0; + page_select = BM_PHY_PAGE_SELECT; + } + + /* Page is shifted left, PHY expects (page x 32) */ + ret_val = e1000e_write_phy_reg_mdic(hw, page_select, + (page << page_shift)); + if (ret_val) { + hw->phy.ops.release_phy(hw); + goto out; + } + } + + ret_val = e1000e_write_phy_reg_mdic(hw, MAX_PHY_REG_ADDRESS & offset, + data); + + hw->phy.ops.release_phy(hw); + +out: + return ret_val; +} + +/** + * e1000e_read_phy_reg_bm - Read BM PHY register + * @hw: pointer to the HW structure + * @offset: register offset to be read + * @data: pointer to the read data + * + * Acquires semaphore, if necessary, then reads the PHY register at offset + * and storing the retrieved information in data. Release any acquired + * semaphores before exiting. + **/ +s32 e1000e_read_phy_reg_bm(struct e1000_hw *hw, u32 offset, u16 *data) +{ + s32 ret_val; + u32 page_select = 0; + u32 page = offset >> IGP_PAGE_SHIFT; + u32 page_shift = 0; + + /* Page 800 works differently than the rest so it has its own func */ + if (page == BM_WUC_PAGE) { + ret_val = e1000_access_phy_wakeup_reg_bm(hw, offset, data, + true); + goto out; + } + + ret_val = hw->phy.ops.acquire_phy(hw); + if (ret_val) + goto out; + + hw->phy.addr = e1000_get_phy_addr_for_bm_page(page, offset); + + if (offset > MAX_PHY_MULTI_PAGE_REG) { + /* + * Page select is register 31 for phy address 1 and 22 for + * phy address 2 and 3. Page select is shifted only for + * phy address 1. + */ + if (hw->phy.addr == 1) { + page_shift = IGP_PAGE_SHIFT; + page_select = IGP01E1000_PHY_PAGE_SELECT; + } else { + page_shift = 0; + page_select = BM_PHY_PAGE_SELECT; + } + + /* Page is shifted left, PHY expects (page x 32) */ + ret_val = e1000e_write_phy_reg_mdic(hw, page_select, + (page << page_shift)); + if (ret_val) { + hw->phy.ops.release_phy(hw); + goto out; + } + } + + ret_val = e1000e_read_phy_reg_mdic(hw, MAX_PHY_REG_ADDRESS & offset, + data); + hw->phy.ops.release_phy(hw); + +out: + return ret_val; +} + +/** + * e1000_access_phy_wakeup_reg_bm - Read BM PHY wakeup register + * @hw: pointer to the HW structure + * @offset: register offset to be read or written + * @data: pointer to the data to read or write + * @read: determines if operation is read or write + * + * Acquires semaphore, if necessary, then reads the PHY register at offset + * and storing the retrieved information in data. Release any acquired + * semaphores before exiting. Note that procedure to read the wakeup + * registers are different. It works as such: + * 1) Set page 769, register 17, bit 2 = 1 + * 2) Set page to 800 for host (801 if we were manageability) + * 3) Write the address using the address opcode (0x11) + * 4) Read or write the data using the data opcode (0x12) + * 5) Restore 769_17.2 to its original value + **/ +static s32 e1000_access_phy_wakeup_reg_bm(struct e1000_hw *hw, u32 offset, + u16 *data, bool read) +{ + s32 ret_val; + u16 reg = ((u16)offset) & PHY_REG_MASK; + u16 phy_reg = 0; + u8 phy_acquired = 1; + + + ret_val = hw->phy.ops.acquire_phy(hw); + if (ret_val) { + phy_acquired = 0; + goto out; + } + + /* All operations in this function are phy address 1 */ + hw->phy.addr = 1; + + /* Set page 769 */ + e1000e_write_phy_reg_mdic(hw, IGP01E1000_PHY_PAGE_SELECT, + (BM_WUC_ENABLE_PAGE << IGP_PAGE_SHIFT)); + + ret_val = e1000e_read_phy_reg_mdic(hw, BM_WUC_ENABLE_REG, &phy_reg); + if (ret_val) + goto out; + + /* First clear bit 4 to avoid a power state change */ + phy_reg &= ~(BM_WUC_HOST_WU_BIT); + ret_val = e1000e_write_phy_reg_mdic(hw, BM_WUC_ENABLE_REG, phy_reg); + if (ret_val) + goto out; + + /* Write bit 2 = 1, and clear bit 4 to 769_17 */ + ret_val = e1000e_write_phy_reg_mdic(hw, BM_WUC_ENABLE_REG, + phy_reg | BM_WUC_ENABLE_BIT); + if (ret_val) + goto out; + + /* Select page 800 */ + ret_val = e1000e_write_phy_reg_mdic(hw, IGP01E1000_PHY_PAGE_SELECT, + (BM_WUC_PAGE << IGP_PAGE_SHIFT)); + + /* Write the page 800 offset value using opcode 0x11 */ + ret_val = e1000e_write_phy_reg_mdic(hw, BM_WUC_ADDRESS_OPCODE, reg); + if (ret_val) + goto out; + + if (read) { + /* Read the page 800 value using opcode 0x12 */ + ret_val = e1000e_read_phy_reg_mdic(hw, BM_WUC_DATA_OPCODE, + data); + } else { + /* Read the page 800 value using opcode 0x12 */ + ret_val = e1000e_write_phy_reg_mdic(hw, BM_WUC_DATA_OPCODE, + *data); + } + + if (ret_val) + goto out; + + /* + * Restore 769_17.2 to its original value + * Set page 769 + */ + e1000e_write_phy_reg_mdic(hw, IGP01E1000_PHY_PAGE_SELECT, + (BM_WUC_ENABLE_PAGE << IGP_PAGE_SHIFT)); + + /* Clear 769_17.2 */ + ret_val = e1000e_write_phy_reg_mdic(hw, BM_WUC_ENABLE_REG, phy_reg); + +out: + if (phy_acquired == 1) + hw->phy.ops.release_phy(hw); + return ret_val; +} + +/** * e1000e_commit_phy - Soft PHY reset * @hw: pointer to the HW structure * diff --git a/drivers/net/eexpress.c b/drivers/net/eexpress.c index 2eb82aba4a8..795c594a4b7 100644 --- a/drivers/net/eexpress.c +++ b/drivers/net/eexpress.c @@ -202,7 +202,7 @@ static unsigned short start_code[] = { 0x0000,Cmd_MCast, 0x0076, /* link to next command */ #define CONF_NR_MULTICAST 0x44 - 0x0000, /* number of multicast addresses */ + 0x0000, /* number of bytes in multicast address(es) */ #define CONF_MULTICAST 0x46 0x0000, 0x0000, 0x0000, /* some addresses */ 0x0000, 0x0000, 0x0000, @@ -1569,7 +1569,7 @@ static void eexp_hw_init586(struct net_device *dev) static void eexp_setup_filter(struct net_device *dev) { - struct dev_mc_list *dmi = dev->mc_list; + struct dev_mc_list *dmi; unsigned short ioaddr = dev->base_addr; int count = dev->mc_count; int i; @@ -1580,9 +1580,9 @@ static void eexp_setup_filter(struct net_device *dev) } outw(CONF_NR_MULTICAST & ~31, ioaddr+SM_PTR); - outw(count, ioaddr+SHADOW(CONF_NR_MULTICAST)); - for (i = 0; i < count; i++) { - unsigned short *data = (unsigned short *)dmi->dmi_addr; + outw(6*count, ioaddr+SHADOW(CONF_NR_MULTICAST)); + for (i = 0, dmi = dev->mc_list; i < count; i++, dmi = dmi->next) { + unsigned short *data; if (!dmi) { printk(KERN_INFO "%s: too few multicast addresses\n", dev->name); break; @@ -1591,6 +1591,7 @@ static void eexp_setup_filter(struct net_device *dev) printk(KERN_INFO "%s: invalid multicast address length given.\n", dev->name); continue; } + data = (unsigned short *)dmi->dmi_addr; outw((CONF_MULTICAST+(6*i)) & ~31, ioaddr+SM_PTR); outw(data[0], ioaddr+SHADOW(CONF_MULTICAST+(6*i))); outw((CONF_MULTICAST+(6*i)+2) & ~31, ioaddr+SM_PTR); diff --git a/drivers/net/fs_enet/mii-fec.c b/drivers/net/fs_enet/mii-fec.c index ba75efc9f5b..f0014cfbb27 100644 --- a/drivers/net/fs_enet/mii-fec.c +++ b/drivers/net/fs_enet/mii-fec.c @@ -194,7 +194,7 @@ static int __devinit fs_enet_mdio_probe(struct of_device *ofdev, ret = of_address_to_resource(ofdev->node, 0, &res); if (ret) - return ret; + goto out_res; snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", res.start); @@ -236,6 +236,7 @@ out_free_irqs: kfree(new_bus->irq); out_unmap_regs: iounmap(fec->fecp); +out_res: out_fec: kfree(fec); out_mii: diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c index 587afe7be68..6f22f068d6e 100644 --- a/drivers/net/gianfar.c +++ b/drivers/net/gianfar.c @@ -138,6 +138,7 @@ static int gfar_poll(struct napi_struct *napi, int budget); static void gfar_netpoll(struct net_device *dev); #endif int gfar_clean_rx_ring(struct net_device *dev, int rx_work_limit); +static int gfar_clean_tx_ring(struct net_device *dev); static int gfar_process_frame(struct net_device *dev, struct sk_buff *skb, int length); static void gfar_vlan_rx_register(struct net_device *netdev, struct vlan_group *grp); @@ -1141,7 +1142,7 @@ static int gfar_close(struct net_device *dev) } /* Changes the mac address if the controller is not running. */ -int gfar_set_mac_address(struct net_device *dev) +static int gfar_set_mac_address(struct net_device *dev) { gfar_set_mac_for_addr(dev, 0, dev->dev_addr); @@ -1260,7 +1261,7 @@ static void gfar_timeout(struct net_device *dev) } /* Interrupt Handler for Transmit complete */ -int gfar_clean_tx_ring(struct net_device *dev) +static int gfar_clean_tx_ring(struct net_device *dev) { struct txbd8 *bdp; struct gfar_private *priv = netdev_priv(dev); diff --git a/drivers/net/gianfar.h b/drivers/net/gianfar.h index fd487be3993..27f37c81e52 100644 --- a/drivers/net/gianfar.h +++ b/drivers/net/gianfar.h @@ -782,5 +782,8 @@ extern void gfar_halt(struct net_device *dev); extern void gfar_phy_test(struct mii_bus *bus, struct phy_device *phydev, int enable, u32 regnum, u32 read); void gfar_init_sysfs(struct net_device *dev); +int gfar_local_mdio_write(struct gfar_mii __iomem *regs, int mii_id, + int regnum, u16 value); +int gfar_local_mdio_read(struct gfar_mii __iomem *regs, int mii_id, int regnum); #endif /* __GIANFAR_H */ diff --git a/drivers/net/gianfar_sysfs.c b/drivers/net/gianfar_sysfs.c index 230878b9419..5116f68e01b 100644 --- a/drivers/net/gianfar_sysfs.c +++ b/drivers/net/gianfar_sysfs.c @@ -103,10 +103,10 @@ static ssize_t gfar_set_rx_stash_size(struct device *dev, spin_lock_irqsave(&priv->rxlock, flags); if (length > priv->rx_buffer_size) - return count; + goto out; if (length == priv->rx_stash_size) - return count; + goto out; priv->rx_stash_size = length; @@ -125,6 +125,7 @@ static ssize_t gfar_set_rx_stash_size(struct device *dev, gfar_write(&priv->regs->attr, temp); +out: spin_unlock_irqrestore(&priv->rxlock, flags); return count; @@ -154,10 +155,10 @@ static ssize_t gfar_set_rx_stash_index(struct device *dev, spin_lock_irqsave(&priv->rxlock, flags); if (index > priv->rx_stash_size) - return count; + goto out; if (index == priv->rx_stash_index) - return count; + goto out; priv->rx_stash_index = index; @@ -166,6 +167,7 @@ static ssize_t gfar_set_rx_stash_index(struct device *dev, temp |= ATTRELI_EI(index); gfar_write(&priv->regs->attreli, flags); +out: spin_unlock_irqrestore(&priv->rxlock, flags); return count; diff --git a/drivers/net/irda/nsc-ircc.c b/drivers/net/irda/nsc-ircc.c index a873d2b315c..a7714da7c28 100644 --- a/drivers/net/irda/nsc-ircc.c +++ b/drivers/net/irda/nsc-ircc.c @@ -100,7 +100,9 @@ static int nsc_ircc_probe_39x(nsc_chip_t *chip, chipio_t *info); static int nsc_ircc_init_108(nsc_chip_t *chip, chipio_t *info); static int nsc_ircc_init_338(nsc_chip_t *chip, chipio_t *info); static int nsc_ircc_init_39x(nsc_chip_t *chip, chipio_t *info); +#ifdef CONFIG_PNP static int nsc_ircc_pnp_probe(struct pnp_dev *dev, const struct pnp_device_id *id); +#endif /* These are the known NSC chips */ static nsc_chip_t chips[] = { @@ -156,9 +158,11 @@ static const struct pnp_device_id nsc_ircc_pnp_table[] = { MODULE_DEVICE_TABLE(pnp, nsc_ircc_pnp_table); static struct pnp_driver nsc_ircc_pnp_driver = { +#ifdef CONFIG_PNP .name = "nsc-ircc", .id_table = nsc_ircc_pnp_table, .probe = nsc_ircc_pnp_probe, +#endif }; /* Some prototypes */ @@ -916,6 +920,7 @@ static int nsc_ircc_probe_39x(nsc_chip_t *chip, chipio_t *info) return 0; } +#ifdef CONFIG_PNP /* PNP probing */ static int nsc_ircc_pnp_probe(struct pnp_dev *dev, const struct pnp_device_id *id) { @@ -952,6 +957,7 @@ static int nsc_ircc_pnp_probe(struct pnp_dev *dev, const struct pnp_device_id *i return 0; } +#endif /* * Function nsc_ircc_setup (info) diff --git a/drivers/net/irda/smsc-ircc2.c b/drivers/net/irda/smsc-ircc2.c index 1f26da761e9..cfe0194fef7 100644 --- a/drivers/net/irda/smsc-ircc2.c +++ b/drivers/net/irda/smsc-ircc2.c @@ -376,6 +376,7 @@ MODULE_DEVICE_TABLE(pnp, smsc_ircc_pnp_table); static int pnp_driver_registered; +#ifdef CONFIG_PNP static int __init smsc_ircc_pnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id) { @@ -402,7 +403,9 @@ static struct pnp_driver smsc_ircc_pnp_driver = { .id_table = smsc_ircc_pnp_table, .probe = smsc_ircc_pnp_probe, }; - +#else /* CONFIG_PNP */ +static struct pnp_driver smsc_ircc_pnp_driver; +#endif /******************************************************************************* * diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index 2056cfc624d..c36a03ae9bf 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -450,7 +450,7 @@ static void macvlan_dellink(struct net_device *dev) unregister_netdevice(dev); if (list_empty(&port->vlans)) - macvlan_port_destroy(dev); + macvlan_port_destroy(port->dev); } static struct rtnl_link_ops macvlan_link_ops __read_mostly = { diff --git a/drivers/net/mlx4/mr.c b/drivers/net/mlx4/mr.c index cb46446b269..03a9abcce52 100644 --- a/drivers/net/mlx4/mr.c +++ b/drivers/net/mlx4/mr.c @@ -551,7 +551,7 @@ int mlx4_fmr_alloc(struct mlx4_dev *dev, u32 pd, u32 access, int max_pages, u64 mtt_seg; int err = -ENOMEM; - if (page_shift < 12 || page_shift >= 32) + if (page_shift < (ffs(dev->caps.page_size_cap) - 1) || page_shift >= 32) return -EINVAL; /* All MTTs must fit in the same page */ diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index 381b36e5f64..b7915cdcc6a 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -91,6 +91,11 @@ */ #define PHY_ADDR_REG 0x0000 #define SMI_REG 0x0004 +#define WINDOW_BASE(i) (0x0200 + ((i) << 3)) +#define WINDOW_SIZE(i) (0x0204 + ((i) << 3)) +#define WINDOW_REMAP_HIGH(i) (0x0280 + ((i) << 2)) +#define WINDOW_BAR_ENABLE 0x0290 +#define WINDOW_PROTECT(i) (0x0294 + ((i) << 4)) /* * Per-port registers. @@ -507,9 +512,23 @@ struct mv643xx_mib_counters { u32 late_collision; }; +struct mv643xx_shared_private { + void __iomem *eth_base; + + /* used to protect SMI_REG, which is shared across ports */ + spinlock_t phy_lock; + + u32 win_protect; + + unsigned int t_clk; +}; + struct mv643xx_private { + struct mv643xx_shared_private *shared; int port_num; /* User Ethernet port number */ + struct mv643xx_shared_private *shared_smi; + u32 rx_sram_addr; /* Base address of rx sram area */ u32 rx_sram_size; /* Size of rx sram area */ u32 tx_sram_addr; /* Base address of tx sram area */ @@ -614,19 +633,14 @@ static const struct ethtool_ops mv643xx_ethtool_ops; static char mv643xx_driver_name[] = "mv643xx_eth"; static char mv643xx_driver_version[] = "1.0"; -static void __iomem *mv643xx_eth_base; - -/* used to protect SMI_REG, which is shared across ports */ -static DEFINE_SPINLOCK(mv643xx_eth_phy_lock); - static inline u32 rdl(struct mv643xx_private *mp, int offset) { - return readl(mv643xx_eth_base + offset); + return readl(mp->shared->eth_base + offset); } static inline void wrl(struct mv643xx_private *mp, int offset, u32 data) { - writel(data, mv643xx_eth_base + offset); + writel(data, mp->shared->eth_base + offset); } /* @@ -1119,7 +1133,6 @@ static irqreturn_t mv643xx_eth_int_handler(int irq, void *dev_id) * * INPUT: * struct mv643xx_private *mp Ethernet port - * unsigned int t_clk t_clk of the MV-643xx chip in HZ units * unsigned int delay Delay in usec * * OUTPUT: @@ -1130,10 +1143,10 @@ static irqreturn_t mv643xx_eth_int_handler(int irq, void *dev_id) * */ static unsigned int eth_port_set_rx_coal(struct mv643xx_private *mp, - unsigned int t_clk, unsigned int delay) + unsigned int delay) { unsigned int port_num = mp->port_num; - unsigned int coal = ((t_clk / 1000000) * delay) / 64; + unsigned int coal = ((mp->shared->t_clk / 1000000) * delay) / 64; /* Set RX Coalescing mechanism */ wrl(mp, SDMA_CONFIG_REG(port_num), @@ -1158,7 +1171,6 @@ static unsigned int eth_port_set_rx_coal(struct mv643xx_private *mp, * * INPUT: * struct mv643xx_private *mp Ethernet port - * unsigned int t_clk t_clk of the MV-643xx chip in HZ units * unsigned int delay Delay in uSeconds * * OUTPUT: @@ -1169,9 +1181,9 @@ static unsigned int eth_port_set_rx_coal(struct mv643xx_private *mp, * */ static unsigned int eth_port_set_tx_coal(struct mv643xx_private *mp, - unsigned int t_clk, unsigned int delay) + unsigned int delay) { - unsigned int coal = ((t_clk / 1000000) * delay) / 64; + unsigned int coal = ((mp->shared->t_clk / 1000000) * delay) / 64; /* Set TX Coalescing mechanism */ wrl(mp, TX_FIFO_URGENT_THRESHOLD_REG(mp->port_num), coal << 4); @@ -1413,11 +1425,11 @@ static int mv643xx_eth_open(struct net_device *dev) #ifdef MV643XX_COAL mp->rx_int_coal = - eth_port_set_rx_coal(mp, 133000000, MV643XX_RX_COAL); + eth_port_set_rx_coal(mp, MV643XX_RX_COAL); #endif mp->tx_int_coal = - eth_port_set_tx_coal(mp, 133000000, MV643XX_TX_COAL); + eth_port_set_tx_coal(mp, MV643XX_TX_COAL); /* Unmask phy and link status changes interrupts */ wrl(mp, INTERRUPT_EXTEND_MASK_REG(port_num), ETH_INT_UNMASK_ALL_EXT); @@ -1827,6 +1839,11 @@ static int mv643xx_eth_probe(struct platform_device *pdev) return -ENODEV; } + if (pd->shared == NULL) { + printk(KERN_ERR "No mv643xx_eth_platform_data->shared\n"); + return -ENODEV; + } + dev = alloc_etherdev(sizeof(struct mv643xx_private)); if (!dev) return -ENOMEM; @@ -1877,8 +1894,16 @@ static int mv643xx_eth_probe(struct platform_device *pdev) spin_lock_init(&mp->lock); + mp->shared = platform_get_drvdata(pd->shared); port_num = mp->port_num = pd->port_number; + if (mp->shared->win_protect) + wrl(mp, WINDOW_PROTECT(port_num), mp->shared->win_protect); + + mp->shared_smi = mp->shared; + if (pd->shared_smi != NULL) + mp->shared_smi = platform_get_drvdata(pd->shared_smi); + /* set default config values */ eth_port_uc_addr_get(mp, dev->dev_addr); mp->rx_ring_size = PORT_DEFAULT_RECEIVE_QUEUE_SIZE; @@ -1983,30 +2008,91 @@ static int mv643xx_eth_remove(struct platform_device *pdev) return 0; } +static void mv643xx_eth_conf_mbus_windows(struct mv643xx_shared_private *msp, + struct mbus_dram_target_info *dram) +{ + void __iomem *base = msp->eth_base; + u32 win_enable; + u32 win_protect; + int i; + + for (i = 0; i < 6; i++) { + writel(0, base + WINDOW_BASE(i)); + writel(0, base + WINDOW_SIZE(i)); + if (i < 4) + writel(0, base + WINDOW_REMAP_HIGH(i)); + } + + win_enable = 0x3f; + win_protect = 0; + + for (i = 0; i < dram->num_cs; i++) { + struct mbus_dram_window *cs = dram->cs + i; + + writel((cs->base & 0xffff0000) | + (cs->mbus_attr << 8) | + dram->mbus_dram_target_id, base + WINDOW_BASE(i)); + writel((cs->size - 1) & 0xffff0000, base + WINDOW_SIZE(i)); + + win_enable &= ~(1 << i); + win_protect |= 3 << (2 * i); + } + + writel(win_enable, base + WINDOW_BAR_ENABLE); + msp->win_protect = win_protect; +} + static int mv643xx_eth_shared_probe(struct platform_device *pdev) { static int mv643xx_version_printed = 0; + struct mv643xx_eth_shared_platform_data *pd = pdev->dev.platform_data; + struct mv643xx_shared_private *msp; struct resource *res; + int ret; if (!mv643xx_version_printed++) printk(KERN_NOTICE "MV-643xx 10/100/1000 Ethernet Driver\n"); + ret = -EINVAL; res = platform_get_resource(pdev, IORESOURCE_MEM, 0); if (res == NULL) - return -ENODEV; + goto out; - mv643xx_eth_base = ioremap(res->start, res->end - res->start + 1); - if (mv643xx_eth_base == NULL) - return -ENOMEM; + ret = -ENOMEM; + msp = kmalloc(sizeof(*msp), GFP_KERNEL); + if (msp == NULL) + goto out; + memset(msp, 0, sizeof(*msp)); + + msp->eth_base = ioremap(res->start, res->end - res->start + 1); + if (msp->eth_base == NULL) + goto out_free; + + spin_lock_init(&msp->phy_lock); + msp->t_clk = (pd != NULL && pd->t_clk != 0) ? pd->t_clk : 133000000; + + platform_set_drvdata(pdev, msp); + + /* + * (Re-)program MBUS remapping windows if we are asked to. + */ + if (pd != NULL && pd->dram != NULL) + mv643xx_eth_conf_mbus_windows(msp, pd->dram); return 0; +out_free: + kfree(msp); +out: + return ret; } static int mv643xx_eth_shared_remove(struct platform_device *pdev) { - iounmap(mv643xx_eth_base); - mv643xx_eth_base = NULL; + struct mv643xx_shared_private *msp = platform_get_drvdata(pdev); + + iounmap(msp->eth_base); + kfree(msp); return 0; } @@ -2906,15 +2992,16 @@ static void eth_port_reset(struct mv643xx_private *mp) static void eth_port_read_smi_reg(struct mv643xx_private *mp, unsigned int phy_reg, unsigned int *value) { + void __iomem *smi_reg = mp->shared_smi->eth_base + SMI_REG; int phy_addr = ethernet_phy_get(mp); unsigned long flags; int i; /* the SMI register is a shared resource */ - spin_lock_irqsave(&mv643xx_eth_phy_lock, flags); + spin_lock_irqsave(&mp->shared_smi->phy_lock, flags); /* wait for the SMI register to become available */ - for (i = 0; rdl(mp, SMI_REG) & ETH_SMI_BUSY; i++) { + for (i = 0; readl(smi_reg) & ETH_SMI_BUSY; i++) { if (i == PHY_WAIT_ITERATIONS) { printk("%s: PHY busy timeout\n", mp->dev->name); goto out; @@ -2922,11 +3009,11 @@ static void eth_port_read_smi_reg(struct mv643xx_private *mp, udelay(PHY_WAIT_MICRO_SECONDS); } - wrl(mp, SMI_REG, - (phy_addr << 16) | (phy_reg << 21) | ETH_SMI_OPCODE_READ); + writel((phy_addr << 16) | (phy_reg << 21) | ETH_SMI_OPCODE_READ, + smi_reg); /* now wait for the data to be valid */ - for (i = 0; !(rdl(mp, SMI_REG) & ETH_SMI_READ_VALID); i++) { + for (i = 0; !(readl(smi_reg) & ETH_SMI_READ_VALID); i++) { if (i == PHY_WAIT_ITERATIONS) { printk("%s: PHY read timeout\n", mp->dev->name); goto out; @@ -2934,9 +3021,9 @@ static void eth_port_read_smi_reg(struct mv643xx_private *mp, udelay(PHY_WAIT_MICRO_SECONDS); } - *value = rdl(mp, SMI_REG) & 0xffff; + *value = readl(smi_reg) & 0xffff; out: - spin_unlock_irqrestore(&mv643xx_eth_phy_lock, flags); + spin_unlock_irqrestore(&mp->shared_smi->phy_lock, flags); } /* @@ -2962,17 +3049,16 @@ out: static void eth_port_write_smi_reg(struct mv643xx_private *mp, unsigned int phy_reg, unsigned int value) { - int phy_addr; - int i; + void __iomem *smi_reg = mp->shared_smi->eth_base + SMI_REG; + int phy_addr = ethernet_phy_get(mp); unsigned long flags; - - phy_addr = ethernet_phy_get(mp); + int i; /* the SMI register is a shared resource */ - spin_lock_irqsave(&mv643xx_eth_phy_lock, flags); + spin_lock_irqsave(&mp->shared_smi->phy_lock, flags); /* wait for the SMI register to become available */ - for (i = 0; rdl(mp, SMI_REG) & ETH_SMI_BUSY; i++) { + for (i = 0; readl(smi_reg) & ETH_SMI_BUSY; i++) { if (i == PHY_WAIT_ITERATIONS) { printk("%s: PHY busy timeout\n", mp->dev->name); goto out; @@ -2980,10 +3066,10 @@ static void eth_port_write_smi_reg(struct mv643xx_private *mp, udelay(PHY_WAIT_MICRO_SECONDS); } - wrl(mp, SMI_REG, (phy_addr << 16) | (phy_reg << 21) | - ETH_SMI_OPCODE_WRITE | (value & 0xffff)); + writel((phy_addr << 16) | (phy_reg << 21) | + ETH_SMI_OPCODE_WRITE | (value & 0xffff), smi_reg); out: - spin_unlock_irqrestore(&mv643xx_eth_phy_lock, flags); + spin_unlock_irqrestore(&mp->shared_smi->phy_lock, flags); } /* diff --git a/drivers/net/niu.c b/drivers/net/niu.c index 4009c4ce96b..57cfd72ffdf 100644 --- a/drivers/net/niu.c +++ b/drivers/net/niu.c @@ -1,6 +1,6 @@ /* niu.c: Neptune ethernet driver. * - * Copyright (C) 2007 David S. Miller (davem@davemloft.net) + * Copyright (C) 2007, 2008 David S. Miller (davem@davemloft.net) */ #include <linux/module.h> @@ -33,8 +33,8 @@ #define DRV_MODULE_NAME "niu" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "0.8" -#define DRV_MODULE_RELDATE "April 24, 2008" +#define DRV_MODULE_VERSION "0.9" +#define DRV_MODULE_RELDATE "May 4, 2008" static char version[] __devinitdata = DRV_MODULE_NAME ".c:v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n"; @@ -7264,8 +7264,11 @@ static int __devinit niu_get_and_validate_port(struct niu *np) parent->num_ports = nr64(ESPC_NUM_PORTS_MACS) & ESPC_NUM_PORTS_MACS_VAL; + /* All of the current probing methods fail on + * Maramba on-board parts. + */ if (!parent->num_ports) - return -ENODEV; + parent->num_ports = 4; } } } diff --git a/drivers/net/pcnet32.c b/drivers/net/pcnet32.c index 4eb322e5273..a1c454dbc16 100644 --- a/drivers/net/pcnet32.c +++ b/drivers/net/pcnet32.c @@ -22,12 +22,8 @@ *************************************************************************/ #define DRV_NAME "pcnet32" -#ifdef CONFIG_PCNET32_NAPI -#define DRV_VERSION "1.34-NAPI" -#else -#define DRV_VERSION "1.34" -#endif -#define DRV_RELDATE "14.Aug.2007" +#define DRV_VERSION "1.35" +#define DRV_RELDATE "21.Apr.2008" #define PFX DRV_NAME ": " static const char *const version = @@ -445,30 +441,24 @@ static struct pcnet32_access pcnet32_dwio = { static void pcnet32_netif_stop(struct net_device *dev) { -#ifdef CONFIG_PCNET32_NAPI struct pcnet32_private *lp = netdev_priv(dev); -#endif + dev->trans_start = jiffies; -#ifdef CONFIG_PCNET32_NAPI napi_disable(&lp->napi); -#endif netif_tx_disable(dev); } static void pcnet32_netif_start(struct net_device *dev) { -#ifdef CONFIG_PCNET32_NAPI struct pcnet32_private *lp = netdev_priv(dev); ulong ioaddr = dev->base_addr; u16 val; -#endif + netif_wake_queue(dev); -#ifdef CONFIG_PCNET32_NAPI val = lp->a.read_csr(ioaddr, CSR3); val &= 0x00ff; lp->a.write_csr(ioaddr, CSR3, val); napi_enable(&lp->napi); -#endif } /* @@ -911,11 +901,7 @@ static int pcnet32_loopback_test(struct net_device *dev, uint64_t * data1) rc = 1; /* default to fail */ if (netif_running(dev)) -#ifdef CONFIG_PCNET32_NAPI pcnet32_netif_stop(dev); -#else - pcnet32_close(dev); -#endif spin_lock_irqsave(&lp->lock, flags); lp->a.write_csr(ioaddr, CSR0, CSR0_STOP); /* stop the chip */ @@ -1046,7 +1032,6 @@ static int pcnet32_loopback_test(struct net_device *dev, uint64_t * data1) x = a->read_bcr(ioaddr, 32); /* reset internal loopback */ a->write_bcr(ioaddr, 32, (x & ~0x0002)); -#ifdef CONFIG_PCNET32_NAPI if (netif_running(dev)) { pcnet32_netif_start(dev); pcnet32_restart(dev, CSR0_NORMAL); @@ -1055,16 +1040,6 @@ static int pcnet32_loopback_test(struct net_device *dev, uint64_t * data1) lp->a.write_bcr(ioaddr, 20, 4); /* return to 16bit mode */ } spin_unlock_irqrestore(&lp->lock, flags); -#else - if (netif_running(dev)) { - spin_unlock_irqrestore(&lp->lock, flags); - pcnet32_open(dev); - } else { - pcnet32_purge_rx_ring(dev); - lp->a.write_bcr(ioaddr, 20, 4); /* return to 16bit mode */ - spin_unlock_irqrestore(&lp->lock, flags); - } -#endif return (rc); } /* end pcnet32_loopback_test */ @@ -1270,11 +1245,7 @@ static void pcnet32_rx_entry(struct net_device *dev, } dev->stats.rx_bytes += skb->len; skb->protocol = eth_type_trans(skb, dev); -#ifdef CONFIG_PCNET32_NAPI netif_receive_skb(skb); -#else - netif_rx(skb); -#endif dev->last_rx = jiffies; dev->stats.rx_packets++; return; @@ -1403,7 +1374,6 @@ static int pcnet32_tx(struct net_device *dev) return must_restart; } -#ifdef CONFIG_PCNET32_NAPI static int pcnet32_poll(struct napi_struct *napi, int budget) { struct pcnet32_private *lp = container_of(napi, struct pcnet32_private, napi); @@ -1442,7 +1412,6 @@ static int pcnet32_poll(struct napi_struct *napi, int budget) } return work_done; } -#endif #define PCNET32_REGS_PER_PHY 32 #define PCNET32_MAX_PHYS 32 @@ -1864,9 +1833,7 @@ pcnet32_probe1(unsigned long ioaddr, int shared, struct pci_dev *pdev) /* napi.weight is used in both the napi and non-napi cases */ lp->napi.weight = lp->rx_ring_size / 2; -#ifdef CONFIG_PCNET32_NAPI netif_napi_add(dev, &lp->napi, pcnet32_poll, lp->rx_ring_size / 2); -#endif if (fdx && !(lp->options & PCNET32_PORT_ASEL) && ((cards_found >= MAX_UNITS) || full_duplex[cards_found])) @@ -2297,9 +2264,7 @@ static int pcnet32_open(struct net_device *dev) goto err_free_ring; } -#ifdef CONFIG_PCNET32_NAPI napi_enable(&lp->napi); -#endif /* Re-initialize the PCNET32, and start it when done. */ lp->a.write_csr(ioaddr, 1, (lp->init_dma_addr & 0xffff)); @@ -2623,7 +2588,6 @@ pcnet32_interrupt(int irq, void *dev_id) dev->name, csr0); /* unlike for the lance, there is no restart needed */ } -#ifdef CONFIG_PCNET32_NAPI if (netif_rx_schedule_prep(dev, &lp->napi)) { u16 val; /* set interrupt masks */ @@ -2634,24 +2598,9 @@ pcnet32_interrupt(int irq, void *dev_id) __netif_rx_schedule(dev, &lp->napi); break; } -#else - pcnet32_rx(dev, lp->napi.weight); - if (pcnet32_tx(dev)) { - /* reset the chip to clear the error condition, then restart */ - lp->a.reset(ioaddr); - lp->a.write_csr(ioaddr, CSR4, 0x0915); /* auto tx pad */ - pcnet32_restart(dev, CSR0_START); - netif_wake_queue(dev); - } -#endif csr0 = lp->a.read_csr(ioaddr, CSR0); } -#ifndef CONFIG_PCNET32_NAPI - /* Set interrupt enable. */ - lp->a.write_csr(ioaddr, CSR0, CSR0_INTEN); -#endif - if (netif_msg_intr(lp)) printk(KERN_DEBUG "%s: exiting interrupt, csr0=%#4.4x.\n", dev->name, lp->a.read_csr(ioaddr, CSR0)); @@ -2670,9 +2619,7 @@ static int pcnet32_close(struct net_device *dev) del_timer_sync(&lp->watchdog_timer); netif_stop_queue(dev); -#ifdef CONFIG_PCNET32_NAPI napi_disable(&lp->napi); -#endif spin_lock_irqsave(&lp->lock, flags); diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 3c18bb59495..45cc2914d34 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -547,7 +547,7 @@ static void phy_force_reduction(struct phy_device *phydev) * Must not be called from interrupt context, or while the * phydev->lock is held. */ -void phy_error(struct phy_device *phydev) +static void phy_error(struct phy_device *phydev) { mutex_lock(&phydev->lock); phydev->state = PHY_HALTED; diff --git a/drivers/net/tulip/uli526x.c b/drivers/net/tulip/uli526x.c index a59c1f224aa..2511ca7a12a 100644 --- a/drivers/net/tulip/uli526x.c +++ b/drivers/net/tulip/uli526x.c @@ -434,10 +434,6 @@ static int uli526x_open(struct net_device *dev) ULI526X_DBUG(0, "uli526x_open", 0); - ret = request_irq(dev->irq, &uli526x_interrupt, IRQF_SHARED, dev->name, dev); - if (ret) - return ret; - /* system variable init */ db->cr6_data = CR6_DEFAULT | uli526x_cr6_user_set; db->tx_packet_cnt = 0; @@ -456,6 +452,10 @@ static int uli526x_open(struct net_device *dev) /* Initialize ULI526X board */ uli526x_init(dev); + ret = request_irq(dev->irq, &uli526x_interrupt, IRQF_SHARED, dev->name, dev); + if (ret) + return ret; + /* Active System Interface */ netif_wake_queue(dev); @@ -1368,6 +1368,12 @@ static void update_cr6(u32 cr6_data, unsigned long ioaddr) * This setup frame initialize ULI526X address filter mode */ +#ifdef __BIG_ENDIAN +#define FLT_SHIFT 16 +#else +#define FLT_SHIFT 0 +#endif + static void send_filter_frame(struct net_device *dev, int mc_cnt) { struct uli526x_board_info *db = netdev_priv(dev); @@ -1384,27 +1390,27 @@ static void send_filter_frame(struct net_device *dev, int mc_cnt) /* Node address */ addrptr = (u16 *) dev->dev_addr; - *suptr++ = addrptr[0]; - *suptr++ = addrptr[1]; - *suptr++ = addrptr[2]; + *suptr++ = addrptr[0] << FLT_SHIFT; + *suptr++ = addrptr[1] << FLT_SHIFT; + *suptr++ = addrptr[2] << FLT_SHIFT; /* broadcast address */ - *suptr++ = 0xffff; - *suptr++ = 0xffff; - *suptr++ = 0xffff; + *suptr++ = 0xffff << FLT_SHIFT; + *suptr++ = 0xffff << FLT_SHIFT; + *suptr++ = 0xffff << FLT_SHIFT; /* fit the multicast address */ for (mcptr = dev->mc_list, i = 0; i < mc_cnt; i++, mcptr = mcptr->next) { addrptr = (u16 *) mcptr->dmi_addr; - *suptr++ = addrptr[0]; - *suptr++ = addrptr[1]; - *suptr++ = addrptr[2]; + *suptr++ = addrptr[0] << FLT_SHIFT; + *suptr++ = addrptr[1] << FLT_SHIFT; + *suptr++ = addrptr[2] << FLT_SHIFT; } for (; i<14; i++) { - *suptr++ = 0xffff; - *suptr++ = 0xffff; - *suptr++ = 0xffff; + *suptr++ = 0xffff << FLT_SHIFT; + *suptr++ = 0xffff << FLT_SHIFT; + *suptr++ = 0xffff << FLT_SHIFT; } /* prepare the setup frame */ diff --git a/drivers/net/ucc_geth.c b/drivers/net/ucc_geth.c index 281ce3d3953..ca0bdac07a7 100644 --- a/drivers/net/ucc_geth.c +++ b/drivers/net/ucc_geth.c @@ -62,7 +62,6 @@ #endif /* UGETH_VERBOSE_DEBUG */ #define UGETH_MSG_DEFAULT (NETIF_MSG_IFUP << 1 ) - 1 -void uec_set_ethtool_ops(struct net_device *netdev); static DEFINE_SPINLOCK(ugeth_lock); @@ -216,7 +215,8 @@ static struct list_head *dequeue(struct list_head *lh) } } -static struct sk_buff *get_new_skb(struct ucc_geth_private *ugeth, u8 *bd) +static struct sk_buff *get_new_skb(struct ucc_geth_private *ugeth, + u8 __iomem *bd) { struct sk_buff *skb = NULL; @@ -236,21 +236,22 @@ static struct sk_buff *get_new_skb(struct ucc_geth_private *ugeth, u8 *bd) skb->dev = ugeth->dev; - out_be32(&((struct qe_bd *)bd)->buf, + out_be32(&((struct qe_bd __iomem *)bd)->buf, dma_map_single(NULL, skb->data, ugeth->ug_info->uf_info.max_rx_buf_length + UCC_GETH_RX_DATA_BUF_ALIGNMENT, DMA_FROM_DEVICE)); - out_be32((u32 *)bd, (R_E | R_I | (in_be32((u32 *)bd) & R_W))); + out_be32((u32 __iomem *)bd, + (R_E | R_I | (in_be32((u32 __iomem*)bd) & R_W))); return skb; } static int rx_bd_buffer_set(struct ucc_geth_private *ugeth, u8 rxQ) { - u8 *bd; + u8 __iomem *bd; u32 bd_status; struct sk_buff *skb; int i; @@ -259,7 +260,7 @@ static int rx_bd_buffer_set(struct ucc_geth_private *ugeth, u8 rxQ) i = 0; do { - bd_status = in_be32((u32*)bd); + bd_status = in_be32((u32 __iomem *)bd); skb = get_new_skb(ugeth, bd); if (!skb) /* If can not allocate data buffer, @@ -277,7 +278,7 @@ static int rx_bd_buffer_set(struct ucc_geth_private *ugeth, u8 rxQ) } static int fill_init_enet_entries(struct ucc_geth_private *ugeth, - volatile u32 *p_start, + u32 *p_start, u8 num_entries, u32 thread_size, u32 thread_alignment, @@ -316,7 +317,7 @@ static int fill_init_enet_entries(struct ucc_geth_private *ugeth, } static int return_init_enet_entries(struct ucc_geth_private *ugeth, - volatile u32 *p_start, + u32 *p_start, u8 num_entries, enum qe_risc_allocation risc, int skip_page_for_first_entry) @@ -326,21 +327,22 @@ static int return_init_enet_entries(struct ucc_geth_private *ugeth, int snum; for (i = 0; i < num_entries; i++) { + u32 val = *p_start; + /* Check that this entry was actually valid -- needed in case failed in allocations */ - if ((*p_start & ENET_INIT_PARAM_RISC_MASK) == risc) { + if ((val & ENET_INIT_PARAM_RISC_MASK) == risc) { snum = - (u32) (*p_start & ENET_INIT_PARAM_SNUM_MASK) >> + (u32) (val & ENET_INIT_PARAM_SNUM_MASK) >> ENET_INIT_PARAM_SNUM_SHIFT; qe_put_snum((u8) snum); if (!((i == 0) && skip_page_for_first_entry)) { /* First entry of Rx does not have page */ init_enet_offset = - (in_be32(p_start) & - ENET_INIT_PARAM_PTR_MASK); + (val & ENET_INIT_PARAM_PTR_MASK); qe_muram_free(init_enet_offset); } - *(p_start++) = 0; /* Just for cosmetics */ + *p_start++ = 0; } } @@ -349,7 +351,7 @@ static int return_init_enet_entries(struct ucc_geth_private *ugeth, #ifdef DEBUG static int dump_init_enet_entries(struct ucc_geth_private *ugeth, - volatile u32 *p_start, + u32 __iomem *p_start, u8 num_entries, u32 thread_size, enum qe_risc_allocation risc, @@ -360,11 +362,13 @@ static int dump_init_enet_entries(struct ucc_geth_private *ugeth, int snum; for (i = 0; i < num_entries; i++) { + u32 val = in_be32(p_start); + /* Check that this entry was actually valid -- needed in case failed in allocations */ - if ((*p_start & ENET_INIT_PARAM_RISC_MASK) == risc) { + if ((val & ENET_INIT_PARAM_RISC_MASK) == risc) { snum = - (u32) (*p_start & ENET_INIT_PARAM_SNUM_MASK) >> + (u32) (val & ENET_INIT_PARAM_SNUM_MASK) >> ENET_INIT_PARAM_SNUM_SHIFT; qe_put_snum((u8) snum); if (!((i == 0) && skip_page_for_first_entry)) { @@ -440,7 +444,7 @@ static int hw_add_addr_in_paddr(struct ucc_geth_private *ugeth, static int hw_clear_addr_in_paddr(struct ucc_geth_private *ugeth, u8 paddr_num) { - struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt; + struct ucc_geth_82xx_address_filtering_pram __iomem *p_82xx_addr_filt; if (!(paddr_num < NUM_OF_PADDRS)) { ugeth_warn("%s: Illagel paddr_num.", __FUNCTION__); @@ -448,7 +452,7 @@ static int hw_clear_addr_in_paddr(struct ucc_geth_private *ugeth, u8 paddr_num) } p_82xx_addr_filt = - (struct ucc_geth_82xx_address_filtering_pram *) ugeth->p_rx_glbl_pram-> + (struct ucc_geth_82xx_address_filtering_pram __iomem *) ugeth->p_rx_glbl_pram-> addressfiltering; /* Writing address ff.ff.ff.ff.ff.ff disables address @@ -463,11 +467,11 @@ static int hw_clear_addr_in_paddr(struct ucc_geth_private *ugeth, u8 paddr_num) static void hw_add_addr_in_hash(struct ucc_geth_private *ugeth, u8 *p_enet_addr) { - struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt; + struct ucc_geth_82xx_address_filtering_pram __iomem *p_82xx_addr_filt; u32 cecr_subblock; p_82xx_addr_filt = - (struct ucc_geth_82xx_address_filtering_pram *) ugeth->p_rx_glbl_pram-> + (struct ucc_geth_82xx_address_filtering_pram __iomem *) ugeth->p_rx_glbl_pram-> addressfiltering; cecr_subblock = @@ -487,7 +491,7 @@ static void hw_add_addr_in_hash(struct ucc_geth_private *ugeth, static void magic_packet_detection_enable(struct ucc_geth_private *ugeth) { struct ucc_fast_private *uccf; - struct ucc_geth *ug_regs; + struct ucc_geth __iomem *ug_regs; u32 maccfg2, uccm; uccf = ugeth->uccf; @@ -507,7 +511,7 @@ static void magic_packet_detection_enable(struct ucc_geth_private *ugeth) static void magic_packet_detection_disable(struct ucc_geth_private *ugeth) { struct ucc_fast_private *uccf; - struct ucc_geth *ug_regs; + struct ucc_geth __iomem *ug_regs; u32 maccfg2, uccm; uccf = ugeth->uccf; @@ -538,13 +542,13 @@ static void get_statistics(struct ucc_geth_private *ugeth, rx_firmware_statistics, struct ucc_geth_hardware_statistics *hardware_statistics) { - struct ucc_fast *uf_regs; - struct ucc_geth *ug_regs; + struct ucc_fast __iomem *uf_regs; + struct ucc_geth __iomem *ug_regs; struct ucc_geth_tx_firmware_statistics_pram *p_tx_fw_statistics_pram; struct ucc_geth_rx_firmware_statistics_pram *p_rx_fw_statistics_pram; ug_regs = ugeth->ug_regs; - uf_regs = (struct ucc_fast *) ug_regs; + uf_regs = (struct ucc_fast __iomem *) ug_regs; p_tx_fw_statistics_pram = ugeth->p_tx_fw_statistics_pram; p_rx_fw_statistics_pram = ugeth->p_rx_fw_statistics_pram; @@ -1132,9 +1136,9 @@ static void dump_regs(struct ucc_geth_private *ugeth) } #endif /* DEBUG */ -static void init_default_reg_vals(volatile u32 *upsmr_register, - volatile u32 *maccfg1_register, - volatile u32 *maccfg2_register) +static void init_default_reg_vals(u32 __iomem *upsmr_register, + u32 __iomem *maccfg1_register, + u32 __iomem *maccfg2_register) { out_be32(upsmr_register, UCC_GETH_UPSMR_INIT); out_be32(maccfg1_register, UCC_GETH_MACCFG1_INIT); @@ -1148,7 +1152,7 @@ static int init_half_duplex_params(int alt_beb, u8 alt_beb_truncation, u8 max_retransmissions, u8 collision_window, - volatile u32 *hafdup_register) + u32 __iomem *hafdup_register) { u32 value = 0; @@ -1180,7 +1184,7 @@ static int init_inter_frame_gap_params(u8 non_btb_cs_ipg, u8 non_btb_ipg, u8 min_ifg, u8 btb_ipg, - volatile u32 *ipgifg_register) + u32 __iomem *ipgifg_register) { u32 value = 0; @@ -1215,9 +1219,9 @@ int init_flow_control_params(u32 automatic_flow_control_mode, int tx_flow_control_enable, u16 pause_period, u16 extension_field, - volatile u32 *upsmr_register, - volatile u32 *uempr_register, - volatile u32 *maccfg1_register) + u32 __iomem *upsmr_register, + u32 __iomem *uempr_register, + u32 __iomem *maccfg1_register) { u32 value = 0; @@ -1243,8 +1247,8 @@ int init_flow_control_params(u32 automatic_flow_control_mode, static int init_hw_statistics_gathering_mode(int enable_hardware_statistics, int auto_zero_hardware_statistics, - volatile u32 *upsmr_register, - volatile u16 *uescr_register) + u32 __iomem *upsmr_register, + u16 __iomem *uescr_register) { u32 upsmr_value = 0; u16 uescr_value = 0; @@ -1270,12 +1274,12 @@ static int init_hw_statistics_gathering_mode(int enable_hardware_statistics, static int init_firmware_statistics_gathering_mode(int enable_tx_firmware_statistics, int enable_rx_firmware_statistics, - volatile u32 *tx_rmon_base_ptr, + u32 __iomem *tx_rmon_base_ptr, u32 tx_firmware_statistics_structure_address, - volatile u32 *rx_rmon_base_ptr, + u32 __iomem *rx_rmon_base_ptr, u32 rx_firmware_statistics_structure_address, - volatile u16 *temoder_register, - volatile u32 *remoder_register) + u16 __iomem *temoder_register, + u32 __iomem *remoder_register) { /* Note: this function does not check if */ /* the parameters it receives are NULL */ @@ -1307,8 +1311,8 @@ static int init_mac_station_addr_regs(u8 address_byte_0, u8 address_byte_3, u8 address_byte_4, u8 address_byte_5, - volatile u32 *macstnaddr1_register, - volatile u32 *macstnaddr2_register) + u32 __iomem *macstnaddr1_register, + u32 __iomem *macstnaddr2_register) { u32 value = 0; @@ -1344,7 +1348,7 @@ static int init_mac_station_addr_regs(u8 address_byte_0, } static int init_check_frame_length_mode(int length_check, - volatile u32 *maccfg2_register) + u32 __iomem *maccfg2_register) { u32 value = 0; @@ -1360,7 +1364,7 @@ static int init_check_frame_length_mode(int length_check, } static int init_preamble_length(u8 preamble_length, - volatile u32 *maccfg2_register) + u32 __iomem *maccfg2_register) { u32 value = 0; @@ -1376,7 +1380,7 @@ static int init_preamble_length(u8 preamble_length, static int init_rx_parameters(int reject_broadcast, int receive_short_frames, - int promiscuous, volatile u32 *upsmr_register) + int promiscuous, u32 __iomem *upsmr_register) { u32 value = 0; @@ -1403,7 +1407,7 @@ static int init_rx_parameters(int reject_broadcast, } static int init_max_rx_buff_len(u16 max_rx_buf_len, - volatile u16 *mrblr_register) + u16 __iomem *mrblr_register) { /* max_rx_buf_len value must be a multiple of 128 */ if ((max_rx_buf_len == 0) @@ -1415,8 +1419,8 @@ static int init_max_rx_buff_len(u16 max_rx_buf_len, } static int init_min_frame_len(u16 min_frame_length, - volatile u16 *minflr_register, - volatile u16 *mrblr_register) + u16 __iomem *minflr_register, + u16 __iomem *mrblr_register) { u16 mrblr_value = 0; @@ -1431,8 +1435,8 @@ static int init_min_frame_len(u16 min_frame_length, static int adjust_enet_interface(struct ucc_geth_private *ugeth) { struct ucc_geth_info *ug_info; - struct ucc_geth *ug_regs; - struct ucc_fast *uf_regs; + struct ucc_geth __iomem *ug_regs; + struct ucc_fast __iomem *uf_regs; int ret_val; u32 upsmr, maccfg2, tbiBaseAddress; u16 value; @@ -1517,8 +1521,8 @@ static int adjust_enet_interface(struct ucc_geth_private *ugeth) static void adjust_link(struct net_device *dev) { struct ucc_geth_private *ugeth = netdev_priv(dev); - struct ucc_geth *ug_regs; - struct ucc_fast *uf_regs; + struct ucc_geth __iomem *ug_regs; + struct ucc_fast __iomem *uf_regs; struct phy_device *phydev = ugeth->phydev; unsigned long flags; int new_state = 0; @@ -1678,9 +1682,9 @@ static int ugeth_graceful_stop_rx(struct ucc_geth_private * ugeth) uccf = ugeth->uccf; /* Clear acknowledge bit */ - temp = ugeth->p_rx_glbl_pram->rxgstpack; + temp = in_8(&ugeth->p_rx_glbl_pram->rxgstpack); temp &= ~GRACEFUL_STOP_ACKNOWLEDGE_RX; - ugeth->p_rx_glbl_pram->rxgstpack = temp; + out_8(&ugeth->p_rx_glbl_pram->rxgstpack, temp); /* Keep issuing command and checking acknowledge bit until it is asserted, according to spec */ @@ -1692,7 +1696,7 @@ static int ugeth_graceful_stop_rx(struct ucc_geth_private * ugeth) qe_issue_cmd(QE_GRACEFUL_STOP_RX, cecr_subblock, QE_CR_PROTOCOL_ETHERNET, 0); - temp = ugeth->p_rx_glbl_pram->rxgstpack; + temp = in_8(&ugeth->p_rx_glbl_pram->rxgstpack); } while (!(temp & GRACEFUL_STOP_ACKNOWLEDGE_RX)); uccf->stopped_rx = 1; @@ -1991,19 +1995,20 @@ static int ugeth_82xx_filtering_clear_all_addr_in_hash(struct ucc_geth_private * enum enet_addr_type enet_addr_type) { - struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt; + struct ucc_geth_82xx_address_filtering_pram __iomem *p_82xx_addr_filt; struct ucc_fast_private *uccf; enum comm_dir comm_dir; struct list_head *p_lh; u16 i, num; - u32 *addr_h, *addr_l; + u32 __iomem *addr_h; + u32 __iomem *addr_l; u8 *p_counter; uccf = ugeth->uccf; p_82xx_addr_filt = - (struct ucc_geth_82xx_address_filtering_pram *) ugeth->p_rx_glbl_pram-> - addressfiltering; + (struct ucc_geth_82xx_address_filtering_pram __iomem *) + ugeth->p_rx_glbl_pram->addressfiltering; if (enet_addr_type == ENET_ADDR_TYPE_GROUP) { addr_h = &(p_82xx_addr_filt->gaddr_h); @@ -2079,7 +2084,7 @@ static int ugeth_82xx_filtering_clear_addr_in_paddr(struct ucc_geth_private *uge static void ucc_geth_memclean(struct ucc_geth_private *ugeth) { u16 i, j; - u8 *bd; + u8 __iomem *bd; if (!ugeth) return; @@ -2154,8 +2159,8 @@ static void ucc_geth_memclean(struct ucc_geth_private *ugeth) for (j = 0; j < ugeth->ug_info->bdRingLenTx[i]; j++) { if (ugeth->tx_skbuff[i][j]) { dma_unmap_single(NULL, - ((struct qe_bd *)bd)->buf, - (in_be32((u32 *)bd) & + in_be32(&((struct qe_bd __iomem *)bd)->buf), + (in_be32((u32 __iomem *)bd) & BD_LENGTH_MASK), DMA_TO_DEVICE); dev_kfree_skb_any(ugeth->tx_skbuff[i][j]); @@ -2182,7 +2187,7 @@ static void ucc_geth_memclean(struct ucc_geth_private *ugeth) for (j = 0; j < ugeth->ug_info->bdRingLenRx[i]; j++) { if (ugeth->rx_skbuff[i][j]) { dma_unmap_single(NULL, - ((struct qe_bd *)bd)->buf, + in_be32(&((struct qe_bd __iomem *)bd)->buf), ugeth->ug_info-> uf_info.max_rx_buf_length + UCC_GETH_RX_DATA_BUF_ALIGNMENT, @@ -2218,8 +2223,8 @@ static void ucc_geth_set_multi(struct net_device *dev) { struct ucc_geth_private *ugeth; struct dev_mc_list *dmi; - struct ucc_fast *uf_regs; - struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt; + struct ucc_fast __iomem *uf_regs; + struct ucc_geth_82xx_address_filtering_pram __iomem *p_82xx_addr_filt; int i; ugeth = netdev_priv(dev); @@ -2228,14 +2233,14 @@ static void ucc_geth_set_multi(struct net_device *dev) if (dev->flags & IFF_PROMISC) { - uf_regs->upsmr |= UPSMR_PRO; + out_be32(&uf_regs->upsmr, in_be32(&uf_regs->upsmr) | UPSMR_PRO); } else { - uf_regs->upsmr &= ~UPSMR_PRO; + out_be32(&uf_regs->upsmr, in_be32(&uf_regs->upsmr)&~UPSMR_PRO); p_82xx_addr_filt = - (struct ucc_geth_82xx_address_filtering_pram *) ugeth-> + (struct ucc_geth_82xx_address_filtering_pram __iomem *) ugeth-> p_rx_glbl_pram->addressfiltering; if (dev->flags & IFF_ALLMULTI) { @@ -2270,7 +2275,7 @@ static void ucc_geth_set_multi(struct net_device *dev) static void ucc_geth_stop(struct ucc_geth_private *ugeth) { - struct ucc_geth *ug_regs = ugeth->ug_regs; + struct ucc_geth __iomem *ug_regs = ugeth->ug_regs; struct phy_device *phydev = ugeth->phydev; u32 tempval; @@ -2419,20 +2424,20 @@ static int ucc_struct_init(struct ucc_geth_private *ugeth) return -ENOMEM; } - ugeth->ug_regs = (struct ucc_geth *) ioremap(uf_info->regs, sizeof(struct ucc_geth)); + ugeth->ug_regs = (struct ucc_geth __iomem *) ioremap(uf_info->regs, sizeof(struct ucc_geth)); return 0; } static int ucc_geth_startup(struct ucc_geth_private *ugeth) { - struct ucc_geth_82xx_address_filtering_pram *p_82xx_addr_filt; - struct ucc_geth_init_pram *p_init_enet_pram; + struct ucc_geth_82xx_address_filtering_pram __iomem *p_82xx_addr_filt; + struct ucc_geth_init_pram __iomem *p_init_enet_pram; struct ucc_fast_private *uccf; struct ucc_geth_info *ug_info; struct ucc_fast_info *uf_info; - struct ucc_fast *uf_regs; - struct ucc_geth *ug_regs; + struct ucc_fast __iomem *uf_regs; + struct ucc_geth __iomem *ug_regs; int ret_val = -EINVAL; u32 remoder = UCC_GETH_REMODER_INIT; u32 init_enet_pram_offset, cecr_subblock, command, maccfg1; @@ -2440,7 +2445,8 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) u16 temoder = UCC_GETH_TEMODER_INIT; u16 test; u8 function_code = 0; - u8 *bd, *endOfRing; + u8 __iomem *bd; + u8 __iomem *endOfRing; u8 numThreadsRxNumerical, numThreadsTxNumerical; ugeth_vdbg("%s: IN", __FUNCTION__); @@ -2602,11 +2608,11 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) if (UCC_GETH_TX_BD_RING_ALIGNMENT > 4) align = UCC_GETH_TX_BD_RING_ALIGNMENT; ugeth->tx_bd_ring_offset[j] = - kmalloc((u32) (length + align), GFP_KERNEL); + (u32) kmalloc((u32) (length + align), GFP_KERNEL); if (ugeth->tx_bd_ring_offset[j] != 0) ugeth->p_tx_bd_ring[j] = - (void*)((ugeth->tx_bd_ring_offset[j] + + (u8 __iomem *)((ugeth->tx_bd_ring_offset[j] + align) & ~(align - 1)); } else if (uf_info->bd_mem_part == MEM_PART_MURAM) { ugeth->tx_bd_ring_offset[j] = @@ -2614,7 +2620,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) UCC_GETH_TX_BD_RING_ALIGNMENT); if (!IS_ERR_VALUE(ugeth->tx_bd_ring_offset[j])) ugeth->p_tx_bd_ring[j] = - (u8 *) qe_muram_addr(ugeth-> + (u8 __iomem *) qe_muram_addr(ugeth-> tx_bd_ring_offset[j]); } if (!ugeth->p_tx_bd_ring[j]) { @@ -2626,8 +2632,8 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) return -ENOMEM; } /* Zero unused end of bd ring, according to spec */ - memset(ugeth->p_tx_bd_ring[j] + - ug_info->bdRingLenTx[j] * sizeof(struct qe_bd), 0, + memset_io((void __iomem *)(ugeth->p_tx_bd_ring[j] + + ug_info->bdRingLenTx[j] * sizeof(struct qe_bd)), 0, length - ug_info->bdRingLenTx[j] * sizeof(struct qe_bd)); } @@ -2639,10 +2645,10 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) if (UCC_GETH_RX_BD_RING_ALIGNMENT > 4) align = UCC_GETH_RX_BD_RING_ALIGNMENT; ugeth->rx_bd_ring_offset[j] = - kmalloc((u32) (length + align), GFP_KERNEL); + (u32) kmalloc((u32) (length + align), GFP_KERNEL); if (ugeth->rx_bd_ring_offset[j] != 0) ugeth->p_rx_bd_ring[j] = - (void*)((ugeth->rx_bd_ring_offset[j] + + (u8 __iomem *)((ugeth->rx_bd_ring_offset[j] + align) & ~(align - 1)); } else if (uf_info->bd_mem_part == MEM_PART_MURAM) { ugeth->rx_bd_ring_offset[j] = @@ -2650,7 +2656,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) UCC_GETH_RX_BD_RING_ALIGNMENT); if (!IS_ERR_VALUE(ugeth->rx_bd_ring_offset[j])) ugeth->p_rx_bd_ring[j] = - (u8 *) qe_muram_addr(ugeth-> + (u8 __iomem *) qe_muram_addr(ugeth-> rx_bd_ring_offset[j]); } if (!ugeth->p_rx_bd_ring[j]) { @@ -2685,14 +2691,14 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) bd = ugeth->confBd[j] = ugeth->txBd[j] = ugeth->p_tx_bd_ring[j]; for (i = 0; i < ug_info->bdRingLenTx[j]; i++) { /* clear bd buffer */ - out_be32(&((struct qe_bd *)bd)->buf, 0); + out_be32(&((struct qe_bd __iomem *)bd)->buf, 0); /* set bd status and length */ - out_be32((u32 *)bd, 0); + out_be32((u32 __iomem *)bd, 0); bd += sizeof(struct qe_bd); } bd -= sizeof(struct qe_bd); /* set bd status and length */ - out_be32((u32 *)bd, T_W); /* for last BD set Wrap bit */ + out_be32((u32 __iomem *)bd, T_W); /* for last BD set Wrap bit */ } /* Init Rx bds */ @@ -2717,14 +2723,14 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) bd = ugeth->rxBd[j] = ugeth->p_rx_bd_ring[j]; for (i = 0; i < ug_info->bdRingLenRx[j]; i++) { /* set bd status and length */ - out_be32((u32 *)bd, R_I); + out_be32((u32 __iomem *)bd, R_I); /* clear bd buffer */ - out_be32(&((struct qe_bd *)bd)->buf, 0); + out_be32(&((struct qe_bd __iomem *)bd)->buf, 0); bd += sizeof(struct qe_bd); } bd -= sizeof(struct qe_bd); /* set bd status and length */ - out_be32((u32 *)bd, R_W); /* for last BD set Wrap bit */ + out_be32((u32 __iomem *)bd, R_W); /* for last BD set Wrap bit */ } /* @@ -2744,10 +2750,10 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) return -ENOMEM; } ugeth->p_tx_glbl_pram = - (struct ucc_geth_tx_global_pram *) qe_muram_addr(ugeth-> + (struct ucc_geth_tx_global_pram __iomem *) qe_muram_addr(ugeth-> tx_glbl_pram_offset); /* Zero out p_tx_glbl_pram */ - memset(ugeth->p_tx_glbl_pram, 0, sizeof(struct ucc_geth_tx_global_pram)); + memset_io((void __iomem *)ugeth->p_tx_glbl_pram, 0, sizeof(struct ucc_geth_tx_global_pram)); /* Fill global PRAM */ @@ -2768,7 +2774,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) } ugeth->p_thread_data_tx = - (struct ucc_geth_thread_data_tx *) qe_muram_addr(ugeth-> + (struct ucc_geth_thread_data_tx __iomem *) qe_muram_addr(ugeth-> thread_dat_tx_offset); out_be32(&ugeth->p_tx_glbl_pram->tqptr, ugeth->thread_dat_tx_offset); @@ -2779,7 +2785,8 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) /* iphoffset */ for (i = 0; i < TX_IP_OFFSET_ENTRY_MAX; i++) - ugeth->p_tx_glbl_pram->iphoffset[i] = ug_info->iphoffset[i]; + out_8(&ugeth->p_tx_glbl_pram->iphoffset[i], + ug_info->iphoffset[i]); /* SQPTR */ /* Size varies with number of Tx queues */ @@ -2797,7 +2804,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) } ugeth->p_send_q_mem_reg = - (struct ucc_geth_send_queue_mem_region *) qe_muram_addr(ugeth-> + (struct ucc_geth_send_queue_mem_region __iomem *) qe_muram_addr(ugeth-> send_q_mem_reg_offset); out_be32(&ugeth->p_tx_glbl_pram->sqptr, ugeth->send_q_mem_reg_offset); @@ -2841,25 +2848,26 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) } ugeth->p_scheduler = - (struct ucc_geth_scheduler *) qe_muram_addr(ugeth-> + (struct ucc_geth_scheduler __iomem *) qe_muram_addr(ugeth-> scheduler_offset); out_be32(&ugeth->p_tx_glbl_pram->schedulerbasepointer, ugeth->scheduler_offset); /* Zero out p_scheduler */ - memset(ugeth->p_scheduler, 0, sizeof(struct ucc_geth_scheduler)); + memset_io((void __iomem *)ugeth->p_scheduler, 0, sizeof(struct ucc_geth_scheduler)); /* Set values in scheduler */ out_be32(&ugeth->p_scheduler->mblinterval, ug_info->mblinterval); out_be16(&ugeth->p_scheduler->nortsrbytetime, ug_info->nortsrbytetime); - ugeth->p_scheduler->fracsiz = ug_info->fracsiz; - ugeth->p_scheduler->strictpriorityq = ug_info->strictpriorityq; - ugeth->p_scheduler->txasap = ug_info->txasap; - ugeth->p_scheduler->extrabw = ug_info->extrabw; + out_8(&ugeth->p_scheduler->fracsiz, ug_info->fracsiz); + out_8(&ugeth->p_scheduler->strictpriorityq, + ug_info->strictpriorityq); + out_8(&ugeth->p_scheduler->txasap, ug_info->txasap); + out_8(&ugeth->p_scheduler->extrabw, ug_info->extrabw); for (i = 0; i < NUM_TX_QUEUES; i++) - ugeth->p_scheduler->weightfactor[i] = - ug_info->weightfactor[i]; + out_8(&ugeth->p_scheduler->weightfactor[i], + ug_info->weightfactor[i]); /* Set pointers to cpucount registers in scheduler */ ugeth->p_cpucount[0] = &(ugeth->p_scheduler->cpucount0); @@ -2890,10 +2898,10 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) return -ENOMEM; } ugeth->p_tx_fw_statistics_pram = - (struct ucc_geth_tx_firmware_statistics_pram *) + (struct ucc_geth_tx_firmware_statistics_pram __iomem *) qe_muram_addr(ugeth->tx_fw_statistics_pram_offset); /* Zero out p_tx_fw_statistics_pram */ - memset(ugeth->p_tx_fw_statistics_pram, + memset_io((void __iomem *)ugeth->p_tx_fw_statistics_pram, 0, sizeof(struct ucc_geth_tx_firmware_statistics_pram)); } @@ -2930,10 +2938,10 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) return -ENOMEM; } ugeth->p_rx_glbl_pram = - (struct ucc_geth_rx_global_pram *) qe_muram_addr(ugeth-> + (struct ucc_geth_rx_global_pram __iomem *) qe_muram_addr(ugeth-> rx_glbl_pram_offset); /* Zero out p_rx_glbl_pram */ - memset(ugeth->p_rx_glbl_pram, 0, sizeof(struct ucc_geth_rx_global_pram)); + memset_io((void __iomem *)ugeth->p_rx_glbl_pram, 0, sizeof(struct ucc_geth_rx_global_pram)); /* Fill global PRAM */ @@ -2953,7 +2961,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) } ugeth->p_thread_data_rx = - (struct ucc_geth_thread_data_rx *) qe_muram_addr(ugeth-> + (struct ucc_geth_thread_data_rx __iomem *) qe_muram_addr(ugeth-> thread_dat_rx_offset); out_be32(&ugeth->p_rx_glbl_pram->rqptr, ugeth->thread_dat_rx_offset); @@ -2976,10 +2984,10 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) return -ENOMEM; } ugeth->p_rx_fw_statistics_pram = - (struct ucc_geth_rx_firmware_statistics_pram *) + (struct ucc_geth_rx_firmware_statistics_pram __iomem *) qe_muram_addr(ugeth->rx_fw_statistics_pram_offset); /* Zero out p_rx_fw_statistics_pram */ - memset(ugeth->p_rx_fw_statistics_pram, 0, + memset_io((void __iomem *)ugeth->p_rx_fw_statistics_pram, 0, sizeof(struct ucc_geth_rx_firmware_statistics_pram)); } @@ -3000,7 +3008,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) } ugeth->p_rx_irq_coalescing_tbl = - (struct ucc_geth_rx_interrupt_coalescing_table *) + (struct ucc_geth_rx_interrupt_coalescing_table __iomem *) qe_muram_addr(ugeth->rx_irq_coalescing_tbl_offset); out_be32(&ugeth->p_rx_glbl_pram->intcoalescingptr, ugeth->rx_irq_coalescing_tbl_offset); @@ -3069,11 +3077,11 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) } ugeth->p_rx_bd_qs_tbl = - (struct ucc_geth_rx_bd_queues_entry *) qe_muram_addr(ugeth-> + (struct ucc_geth_rx_bd_queues_entry __iomem *) qe_muram_addr(ugeth-> rx_bd_qs_tbl_offset); out_be32(&ugeth->p_rx_glbl_pram->rbdqptr, ugeth->rx_bd_qs_tbl_offset); /* Zero out p_rx_bd_qs_tbl */ - memset(ugeth->p_rx_bd_qs_tbl, + memset_io((void __iomem *)ugeth->p_rx_bd_qs_tbl, 0, ug_info->numQueuesRx * (sizeof(struct ucc_geth_rx_bd_queues_entry) + sizeof(struct ucc_geth_rx_prefetched_bds))); @@ -3133,7 +3141,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) &ugeth->p_rx_glbl_pram->remoder); /* function code register */ - ugeth->p_rx_glbl_pram->rstate = function_code; + out_8(&ugeth->p_rx_glbl_pram->rstate, function_code); /* initialize extended filtering */ if (ug_info->rxExtendedFiltering) { @@ -3160,7 +3168,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) } ugeth->p_exf_glbl_param = - (struct ucc_geth_exf_global_pram *) qe_muram_addr(ugeth-> + (struct ucc_geth_exf_global_pram __iomem *) qe_muram_addr(ugeth-> exf_glbl_param_offset); out_be32(&ugeth->p_rx_glbl_pram->exfGlobalParam, ugeth->exf_glbl_param_offset); @@ -3175,7 +3183,7 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) ugeth_82xx_filtering_clear_addr_in_paddr(ugeth, (u8) j); p_82xx_addr_filt = - (struct ucc_geth_82xx_address_filtering_pram *) ugeth-> + (struct ucc_geth_82xx_address_filtering_pram __iomem *) ugeth-> p_rx_glbl_pram->addressfiltering; ugeth_82xx_filtering_clear_all_addr_in_hash(ugeth, @@ -3307,17 +3315,21 @@ static int ucc_geth_startup(struct ucc_geth_private *ugeth) return -ENOMEM; } p_init_enet_pram = - (struct ucc_geth_init_pram *) qe_muram_addr(init_enet_pram_offset); + (struct ucc_geth_init_pram __iomem *) qe_muram_addr(init_enet_pram_offset); /* Copy shadow InitEnet command parameter structure into PRAM */ - p_init_enet_pram->resinit1 = ugeth->p_init_enet_param_shadow->resinit1; - p_init_enet_pram->resinit2 = ugeth->p_init_enet_param_shadow->resinit2; - p_init_enet_pram->resinit3 = ugeth->p_init_enet_param_shadow->resinit3; - p_init_enet_pram->resinit4 = ugeth->p_init_enet_param_shadow->resinit4; + out_8(&p_init_enet_pram->resinit1, + ugeth->p_init_enet_param_shadow->resinit1); + out_8(&p_init_enet_pram->resinit2, + ugeth->p_init_enet_param_shadow->resinit2); + out_8(&p_init_enet_pram->resinit3, + ugeth->p_init_enet_param_shadow->resinit3); + out_8(&p_init_enet_pram->resinit4, + ugeth->p_init_enet_param_shadow->resinit4); out_be16(&p_init_enet_pram->resinit5, ugeth->p_init_enet_param_shadow->resinit5); - p_init_enet_pram->largestexternallookupkeysize = - ugeth->p_init_enet_param_shadow->largestexternallookupkeysize; + out_8(&p_init_enet_pram->largestexternallookupkeysize, + ugeth->p_init_enet_param_shadow->largestexternallookupkeysize); out_be32(&p_init_enet_pram->rgftgfrxglobal, ugeth->p_init_enet_param_shadow->rgftgfrxglobal); for (i = 0; i < ENET_INIT_PARAM_MAX_ENTRIES_RX; i++) @@ -3371,7 +3383,7 @@ static int ucc_geth_start_xmit(struct sk_buff *skb, struct net_device *dev) #ifdef CONFIG_UGETH_TX_ON_DEMAND struct ucc_fast_private *uccf; #endif - u8 *bd; /* BD pointer */ + u8 __iomem *bd; /* BD pointer */ u32 bd_status; u8 txQ = 0; @@ -3383,7 +3395,7 @@ static int ucc_geth_start_xmit(struct sk_buff *skb, struct net_device *dev) /* Start from the next BD that should be filled */ bd = ugeth->txBd[txQ]; - bd_status = in_be32((u32 *)bd); + bd_status = in_be32((u32 __iomem *)bd); /* Save the skb pointer so we can free it later */ ugeth->tx_skbuff[txQ][ugeth->skb_curtx[txQ]] = skb; @@ -3393,7 +3405,7 @@ static int ucc_geth_start_xmit(struct sk_buff *skb, struct net_device *dev) 1) & TX_RING_MOD_MASK(ugeth->ug_info->bdRingLenTx[txQ]); /* set up the buffer descriptor */ - out_be32(&((struct qe_bd *)bd)->buf, + out_be32(&((struct qe_bd __iomem *)bd)->buf, dma_map_single(NULL, skb->data, skb->len, DMA_TO_DEVICE)); /* printk(KERN_DEBUG"skb->data is 0x%x\n",skb->data); */ @@ -3401,7 +3413,7 @@ static int ucc_geth_start_xmit(struct sk_buff *skb, struct net_device *dev) bd_status = (bd_status & T_W) | T_R | T_I | T_L | skb->len; /* set bd status and length */ - out_be32((u32 *)bd, bd_status); + out_be32((u32 __iomem *)bd, bd_status); dev->trans_start = jiffies; @@ -3441,7 +3453,7 @@ static int ucc_geth_start_xmit(struct sk_buff *skb, struct net_device *dev) static int ucc_geth_rx(struct ucc_geth_private *ugeth, u8 rxQ, int rx_work_limit) { struct sk_buff *skb; - u8 *bd; + u8 __iomem *bd; u16 length, howmany = 0; u32 bd_status; u8 *bdBuffer; @@ -3454,11 +3466,11 @@ static int ucc_geth_rx(struct ucc_geth_private *ugeth, u8 rxQ, int rx_work_limit /* collect received buffers */ bd = ugeth->rxBd[rxQ]; - bd_status = in_be32((u32 *)bd); + bd_status = in_be32((u32 __iomem *)bd); /* while there are received buffers and BD is full (~R_E) */ while (!((bd_status & (R_E)) || (--rx_work_limit < 0))) { - bdBuffer = (u8 *) in_be32(&((struct qe_bd *)bd)->buf); + bdBuffer = (u8 *) in_be32(&((struct qe_bd __iomem *)bd)->buf); length = (u16) ((bd_status & BD_LENGTH_MASK) - 4); skb = ugeth->rx_skbuff[rxQ][ugeth->skb_currx[rxQ]]; @@ -3516,7 +3528,7 @@ static int ucc_geth_rx(struct ucc_geth_private *ugeth, u8 rxQ, int rx_work_limit else bd += sizeof(struct qe_bd); - bd_status = in_be32((u32 *)bd); + bd_status = in_be32((u32 __iomem *)bd); } ugeth->rxBd[rxQ] = bd; @@ -3527,11 +3539,11 @@ static int ucc_geth_tx(struct net_device *dev, u8 txQ) { /* Start from the next BD that should be filled */ struct ucc_geth_private *ugeth = netdev_priv(dev); - u8 *bd; /* BD pointer */ + u8 __iomem *bd; /* BD pointer */ u32 bd_status; bd = ugeth->confBd[txQ]; - bd_status = in_be32((u32 *)bd); + bd_status = in_be32((u32 __iomem *)bd); /* Normal processing. */ while ((bd_status & T_R) == 0) { @@ -3561,7 +3573,7 @@ static int ucc_geth_tx(struct net_device *dev, u8 txQ) bd += sizeof(struct qe_bd); else bd = ugeth->p_tx_bd_ring[txQ]; - bd_status = in_be32((u32 *)bd); + bd_status = in_be32((u32 __iomem *)bd); } ugeth->confBd[txQ] = bd; return 0; @@ -3910,7 +3922,7 @@ static int ucc_geth_probe(struct of_device* ofdev, const struct of_device_id *ma return -EINVAL; } } else { - prop = of_get_property(np, "rx-clock", NULL); + prop = of_get_property(np, "tx-clock", NULL); if (!prop) { printk(KERN_ERR "ucc_geth: mising tx-clock-name property\n"); diff --git a/drivers/net/ucc_geth.h b/drivers/net/ucc_geth.h index 9f8b7580a3a..abc0e224263 100644 --- a/drivers/net/ucc_geth.h +++ b/drivers/net/ucc_geth.h @@ -700,8 +700,8 @@ struct ucc_geth_82xx_address_filtering_pram { u32 iaddr_l; /* individual address filter, low */ u32 gaddr_h; /* group address filter, high */ u32 gaddr_l; /* group address filter, low */ - struct ucc_geth_82xx_enet_address taddr; - struct ucc_geth_82xx_enet_address paddr[NUM_OF_PADDRS]; + struct ucc_geth_82xx_enet_address __iomem taddr; + struct ucc_geth_82xx_enet_address __iomem paddr[NUM_OF_PADDRS]; u8 res0[0x40 - 0x38]; } __attribute__ ((packed)); @@ -1186,40 +1186,40 @@ struct ucc_geth_private { struct ucc_fast_private *uccf; struct net_device *dev; struct napi_struct napi; - struct ucc_geth *ug_regs; + struct ucc_geth __iomem *ug_regs; struct ucc_geth_init_pram *p_init_enet_param_shadow; - struct ucc_geth_exf_global_pram *p_exf_glbl_param; + struct ucc_geth_exf_global_pram __iomem *p_exf_glbl_param; u32 exf_glbl_param_offset; - struct ucc_geth_rx_global_pram *p_rx_glbl_pram; + struct ucc_geth_rx_global_pram __iomem *p_rx_glbl_pram; u32 rx_glbl_pram_offset; - struct ucc_geth_tx_global_pram *p_tx_glbl_pram; + struct ucc_geth_tx_global_pram __iomem *p_tx_glbl_pram; u32 tx_glbl_pram_offset; - struct ucc_geth_send_queue_mem_region *p_send_q_mem_reg; + struct ucc_geth_send_queue_mem_region __iomem *p_send_q_mem_reg; u32 send_q_mem_reg_offset; - struct ucc_geth_thread_data_tx *p_thread_data_tx; + struct ucc_geth_thread_data_tx __iomem *p_thread_data_tx; u32 thread_dat_tx_offset; - struct ucc_geth_thread_data_rx *p_thread_data_rx; + struct ucc_geth_thread_data_rx __iomem *p_thread_data_rx; u32 thread_dat_rx_offset; - struct ucc_geth_scheduler *p_scheduler; + struct ucc_geth_scheduler __iomem *p_scheduler; u32 scheduler_offset; - struct ucc_geth_tx_firmware_statistics_pram *p_tx_fw_statistics_pram; + struct ucc_geth_tx_firmware_statistics_pram __iomem *p_tx_fw_statistics_pram; u32 tx_fw_statistics_pram_offset; - struct ucc_geth_rx_firmware_statistics_pram *p_rx_fw_statistics_pram; + struct ucc_geth_rx_firmware_statistics_pram __iomem *p_rx_fw_statistics_pram; u32 rx_fw_statistics_pram_offset; - struct ucc_geth_rx_interrupt_coalescing_table *p_rx_irq_coalescing_tbl; + struct ucc_geth_rx_interrupt_coalescing_table __iomem *p_rx_irq_coalescing_tbl; u32 rx_irq_coalescing_tbl_offset; - struct ucc_geth_rx_bd_queues_entry *p_rx_bd_qs_tbl; + struct ucc_geth_rx_bd_queues_entry __iomem *p_rx_bd_qs_tbl; u32 rx_bd_qs_tbl_offset; - u8 *p_tx_bd_ring[NUM_TX_QUEUES]; + u8 __iomem *p_tx_bd_ring[NUM_TX_QUEUES]; u32 tx_bd_ring_offset[NUM_TX_QUEUES]; - u8 *p_rx_bd_ring[NUM_RX_QUEUES]; + u8 __iomem *p_rx_bd_ring[NUM_RX_QUEUES]; u32 rx_bd_ring_offset[NUM_RX_QUEUES]; - u8 *confBd[NUM_TX_QUEUES]; - u8 *txBd[NUM_TX_QUEUES]; - u8 *rxBd[NUM_RX_QUEUES]; + u8 __iomem *confBd[NUM_TX_QUEUES]; + u8 __iomem *txBd[NUM_TX_QUEUES]; + u8 __iomem *rxBd[NUM_RX_QUEUES]; int badFrame[NUM_RX_QUEUES]; u16 cpucount[NUM_TX_QUEUES]; - volatile u16 *p_cpucount[NUM_TX_QUEUES]; + u16 __iomem *p_cpucount[NUM_TX_QUEUES]; int indAddrRegUsed[NUM_OF_PADDRS]; u8 paddr[NUM_OF_PADDRS][ENET_NUM_OCTETS_PER_ADDRESS]; /* ethernet address */ u8 numGroupAddrInHash; @@ -1251,4 +1251,12 @@ struct ucc_geth_private { int oldlink; }; +void uec_set_ethtool_ops(struct net_device *netdev); +int init_flow_control_params(u32 automatic_flow_control_mode, + int rx_flow_control_enable, int tx_flow_control_enable, + u16 pause_period, u16 extension_field, + u32 __iomem *upsmr_register, u32 __iomem *uempr_register, + u32 __iomem *maccfg1_register); + + #endif /* __UCC_GETH_H__ */ diff --git a/drivers/net/ucc_geth_ethtool.c b/drivers/net/ucc_geth_ethtool.c index 9a9622c13e2..299b7f17695 100644 --- a/drivers/net/ucc_geth_ethtool.c +++ b/drivers/net/ucc_geth_ethtool.c @@ -108,12 +108,6 @@ static char rx_fw_stat_gstrings[][ETH_GSTRING_LEN] = { #define UEC_TX_FW_STATS_LEN ARRAY_SIZE(tx_fw_stat_gstrings) #define UEC_RX_FW_STATS_LEN ARRAY_SIZE(rx_fw_stat_gstrings) -extern int init_flow_control_params(u32 automatic_flow_control_mode, - int rx_flow_control_enable, - int tx_flow_control_enable, u16 pause_period, - u16 extension_field, volatile u32 *upsmr_register, - volatile u32 *uempr_register, volatile u32 *maccfg1_register); - static int uec_get_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) { diff --git a/drivers/net/ucc_geth_mii.c b/drivers/net/ucc_geth_mii.c index 2af49078100..94047473692 100644 --- a/drivers/net/ucc_geth_mii.c +++ b/drivers/net/ucc_geth_mii.c @@ -104,7 +104,7 @@ int uec_mdio_read(struct mii_bus *bus, int mii_id, int regnum) } /* Reset the MIIM registers, and wait for the bus to free */ -int uec_mdio_reset(struct mii_bus *bus) +static int uec_mdio_reset(struct mii_bus *bus) { struct ucc_mii_mng __iomem *regs = (void __iomem *)bus->priv; unsigned int timeout = PHY_INIT_TIMEOUT; @@ -240,7 +240,7 @@ reg_map_fail: return err; } -int uec_mdio_remove(struct of_device *ofdev) +static int uec_mdio_remove(struct of_device *ofdev) { struct device *device = &ofdev->dev; struct mii_bus *bus = dev_get_drvdata(device); diff --git a/drivers/net/usb/asix.c b/drivers/net/usb/asix.c index 6f245cfb662..dc6f097062d 100644 --- a/drivers/net/usb/asix.c +++ b/drivers/net/usb/asix.c @@ -1381,6 +1381,10 @@ static const struct usb_device_id products [] = { USB_DEVICE (0x0411, 0x003d), .driver_info = (unsigned long) &ax8817x_info, }, { + // Buffalo LUA-U2-GT 10/100/1000 + USB_DEVICE (0x0411, 0x006e), + .driver_info = (unsigned long) &ax88178_info, +}, { // Sitecom LN-029 "USB 2.0 10/100 Ethernet adapter" USB_DEVICE (0x6189, 0x182d), .driver_info = (unsigned long) &ax8817x_info, diff --git a/drivers/net/wan/lapbether.c b/drivers/net/wan/lapbether.c index b5860b97a93..24fd613466b 100644 --- a/drivers/net/wan/lapbether.c +++ b/drivers/net/wan/lapbether.c @@ -459,6 +459,7 @@ static void __exit lapbeth_cleanup_driver(void) list_for_each_safe(entry, tmp, &lapbeth_devices) { lapbeth = list_entry(entry, struct lapbethdev, node); + dev_put(lapbeth->ethdev); unregister_netdevice(lapbeth->axdev); } rtnl_unlock(); diff --git a/drivers/net/wireless/iwlwifi/Kconfig b/drivers/net/wireless/iwlwifi/Kconfig index d5b7a76fcaa..62fb89d8231 100644 --- a/drivers/net/wireless/iwlwifi/Kconfig +++ b/drivers/net/wireless/iwlwifi/Kconfig @@ -1,6 +1,5 @@ config IWLWIFI - bool - default n + tristate config IWLCORE tristate "Intel Wireless Wifi Core" diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 4a55bf38095..3706ce7972d 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -842,13 +842,25 @@ static void set_pcie_port_type(struct pci_dev *pdev) * reading the dword at 0x100 which must either be 0 or a valid extended * capability header. */ -int pci_cfg_space_size_ext(struct pci_dev *dev, unsigned check_exp_pcix) +int pci_cfg_space_size_ext(struct pci_dev *dev) { - int pos; u32 status; - if (!check_exp_pcix) - goto skip; + if (pci_read_config_dword(dev, 256, &status) != PCIBIOS_SUCCESSFUL) + goto fail; + if (status == 0xffffffff) + goto fail; + + return PCI_CFG_SPACE_EXP_SIZE; + + fail: + return PCI_CFG_SPACE_SIZE; +} + +int pci_cfg_space_size(struct pci_dev *dev) +{ + int pos; + u32 status; pos = pci_find_capability(dev, PCI_CAP_ID_EXP); if (!pos) { @@ -861,23 +873,12 @@ int pci_cfg_space_size_ext(struct pci_dev *dev, unsigned check_exp_pcix) goto fail; } - skip: - if (pci_read_config_dword(dev, 256, &status) != PCIBIOS_SUCCESSFUL) - goto fail; - if (status == 0xffffffff) - goto fail; - - return PCI_CFG_SPACE_EXP_SIZE; + return pci_cfg_space_size_ext(dev); fail: return PCI_CFG_SPACE_SIZE; } -int pci_cfg_space_size(struct pci_dev *dev) -{ - return pci_cfg_space_size_ext(dev, 1); -} - static void pci_release_bus_bridge_dev(struct device *dev) { kfree(dev); diff --git a/drivers/rtc/rtc-ds1511.c b/drivers/rtc/rtc-ds1511.c index a83a40b3eba..0f0d27d1c4c 100644 --- a/drivers/rtc/rtc-ds1511.c +++ b/drivers/rtc/rtc-ds1511.c @@ -184,7 +184,7 @@ ds1511_wdog_disable(void) static int ds1511_rtc_set_time(struct device *dev, struct rtc_time *rtc_tm) { u8 mon, day, dow, hrs, min, sec, yrs, cen; - unsigned int flags; + unsigned long flags; /* * won't have to change this for a while @@ -247,7 +247,7 @@ static int ds1511_rtc_set_time(struct device *dev, struct rtc_time *rtc_tm) static int ds1511_rtc_read_time(struct device *dev, struct rtc_time *rtc_tm) { unsigned int century; - unsigned int flags; + unsigned long flags; spin_lock_irqsave(&ds1511_lock, flags); rtc_disable_update(); diff --git a/drivers/s390/char/tty3270.c b/drivers/s390/char/tty3270.c index c1f2adefad4..5043150019a 100644 --- a/drivers/s390/char/tty3270.c +++ b/drivers/s390/char/tty3270.c @@ -965,8 +965,7 @@ tty3270_write_room(struct tty_struct *tty) * Insert character into the screen at the current position with the * current color and highlight. This function does NOT do cursor movement. */ -static int -tty3270_put_character(struct tty3270 *tp, char ch) +static void tty3270_put_character(struct tty3270 *tp, char ch) { struct tty3270_line *line; struct tty3270_cell *cell; @@ -986,7 +985,6 @@ tty3270_put_character(struct tty3270 *tp, char ch) cell->character = tp->view.ascebc[(unsigned int) ch]; cell->highlight = tp->highlight; cell->f_color = tp->f_color; - return 1; } /* @@ -1612,16 +1610,15 @@ tty3270_write(struct tty_struct * tty, /* * Put single characters to the ttys character buffer */ -static void -tty3270_put_char(struct tty_struct *tty, unsigned char ch) +static int tty3270_put_char(struct tty_struct *tty, unsigned char ch) { struct tty3270 *tp; tp = tty->driver_data; - if (!tp) - return; - if (tp->char_count < TTY3270_CHAR_BUF_SIZE) - tp->char_buf[tp->char_count++] = ch; + if (!tp || tp->char_count >= TTY3270_CHAR_BUF_SIZE) + return 0; + tp->char_buf[tp->char_count++] = ch; + return 1; } /* diff --git a/drivers/s390/cio/blacklist.c b/drivers/s390/cio/blacklist.c index 40ef948fcb3..9c21b8f43f9 100644 --- a/drivers/s390/cio/blacklist.c +++ b/drivers/s390/cio/blacklist.c @@ -19,6 +19,7 @@ #include <asm/cio.h> #include <asm/uaccess.h> +#include <asm/cio.h> #include "blacklist.h" #include "cio.h" @@ -43,164 +44,169 @@ typedef enum {add, free} range_action; * Function: blacklist_range * (Un-)blacklist the devices from-to */ -static void -blacklist_range (range_action action, unsigned int from, unsigned int to, - unsigned int ssid) +static int blacklist_range(range_action action, unsigned int from_ssid, + unsigned int to_ssid, unsigned int from, + unsigned int to, int msgtrigger) { - if (!to) - to = from; - - if (from > to || to > __MAX_SUBCHANNEL || ssid > __MAX_SSID) { - printk (KERN_WARNING "cio: Invalid blacklist range " - "0.%x.%04x to 0.%x.%04x, skipping\n", - ssid, from, ssid, to); - return; + if ((from_ssid > to_ssid) || ((from_ssid == to_ssid) && (from > to))) { + if (msgtrigger) + printk(KERN_WARNING "cio: Invalid cio_ignore range " + "0.%x.%04x-0.%x.%04x\n", from_ssid, from, + to_ssid, to); + return 1; } - for (; from <= to; from++) { + + while ((from_ssid < to_ssid) || ((from_ssid == to_ssid) && + (from <= to))) { if (action == add) - set_bit (from, bl_dev[ssid]); + set_bit(from, bl_dev[from_ssid]); else - clear_bit (from, bl_dev[ssid]); + clear_bit(from, bl_dev[from_ssid]); + from++; + if (from > __MAX_SUBCHANNEL) { + from_ssid++; + from = 0; + } } + + return 0; } -/* - * Function: blacklist_busid - * Get devno/busid from given string. - * Shamelessly grabbed from dasd_devmap.c. - */ -static int -blacklist_busid(char **str, int *id0, int *ssid, int *devno) +static int pure_hex(char **cp, unsigned int *val, int min_digit, + int max_digit, int max_val) { - int val, old_style; - char *sav; + int diff; + unsigned int value; - sav = *str; + diff = 0; + *val = 0; - /* check for leading '0x' */ - old_style = 0; - if ((*str)[0] == '0' && (*str)[1] == 'x') { - *str += 2; - old_style = 1; - } - if (!isxdigit((*str)[0])) /* We require at least one hex digit */ - goto confused; - val = simple_strtoul(*str, str, 16); - if (old_style || (*str)[0] != '.') { - *id0 = *ssid = 0; - if (val < 0 || val > 0xffff) - goto confused; - *devno = val; - if ((*str)[0] != ',' && (*str)[0] != '-' && - (*str)[0] != '\n' && (*str)[0] != '\0') - goto confused; - return 0; + while (isxdigit(**cp) && (diff <= max_digit)) { + + if (isdigit(**cp)) + value = **cp - '0'; + else + value = tolower(**cp) - 'a' + 10; + *val = *val * 16 + value; + (*cp)++; + diff++; } - /* New style x.y.z busid */ - if (val < 0 || val > 0xff) - goto confused; - *id0 = val; - (*str)++; - if (!isxdigit((*str)[0])) /* We require at least one hex digit */ - goto confused; - val = simple_strtoul(*str, str, 16); - if (val < 0 || val > 0xff || (*str)++[0] != '.') - goto confused; - *ssid = val; - if (!isxdigit((*str)[0])) /* We require at least one hex digit */ - goto confused; - val = simple_strtoul(*str, str, 16); - if (val < 0 || val > 0xffff) - goto confused; - *devno = val; - if ((*str)[0] != ',' && (*str)[0] != '-' && - (*str)[0] != '\n' && (*str)[0] != '\0') - goto confused; + + if ((diff < min_digit) || (diff > max_digit) || (*val > max_val)) + return 1; + return 0; -confused: - strsep(str, ",\n"); - printk(KERN_WARNING "cio: Invalid cio_ignore parameter '%s'\n", sav); - return 1; } -static int -blacklist_parse_parameters (char *str, range_action action) +static int parse_busid(char *str, int *cssid, int *ssid, int *devno, + int msgtrigger) { - int from, to, from_id0, to_id0, from_ssid, to_ssid; - - while (*str != 0 && *str != '\n') { - range_action ra = action; - while(*str == ',') - str++; - if (*str == '!') { - ra = !action; - ++str; + char *str_work; + int val, rc, ret; + + rc = 1; + + if (*str == '\0') + goto out; + + /* old style */ + str_work = str; + val = simple_strtoul(str, &str_work, 16); + + if (*str_work == '\0') { + if (val <= __MAX_SUBCHANNEL) { + *devno = val; + *ssid = 0; + *cssid = 0; + rc = 0; } + goto out; + } - /* - * Since we have to parse the proc commands and the - * kernel arguments we have to check four cases - */ - if (strncmp(str,"all,",4) == 0 || strcmp(str,"all") == 0 || - strncmp(str,"all\n",4) == 0 || strncmp(str,"all ",4) == 0) { - int j; - - str += 3; - for (j=0; j <= __MAX_SSID; j++) - blacklist_range(ra, 0, __MAX_SUBCHANNEL, j); - } else { - int rc; + /* new style */ + str_work = str; + ret = pure_hex(&str_work, cssid, 1, 2, __MAX_CSSID); + if (ret || (str_work[0] != '.')) + goto out; + str_work++; + ret = pure_hex(&str_work, ssid, 1, 1, __MAX_SSID); + if (ret || (str_work[0] != '.')) + goto out; + str_work++; + ret = pure_hex(&str_work, devno, 4, 4, __MAX_SUBCHANNEL); + if (ret || (str_work[0] != '\0')) + goto out; + + rc = 0; +out: + if (rc && msgtrigger) + printk(KERN_WARNING "cio: Invalid cio_ignore device '%s'\n", + str); + + return rc; +} - rc = blacklist_busid(&str, &from_id0, - &from_ssid, &from); - if (rc) - continue; - to = from; - to_id0 = from_id0; - to_ssid = from_ssid; - if (*str == '-') { - str++; - rc = blacklist_busid(&str, &to_id0, - &to_ssid, &to); - if (rc) - continue; - } - if (*str == '-') { - printk(KERN_WARNING "cio: invalid cio_ignore " - "parameter '%s'\n", - strsep(&str, ",\n")); - continue; - } - if ((from_id0 != to_id0) || - (from_ssid != to_ssid)) { - printk(KERN_WARNING "cio: invalid cio_ignore " - "range %x.%x.%04x-%x.%x.%04x\n", - from_id0, from_ssid, from, - to_id0, to_ssid, to); - continue; +static int blacklist_parse_parameters(char *str, range_action action, + int msgtrigger) +{ + int from_cssid, to_cssid, from_ssid, to_ssid, from, to; + int rc, totalrc; + char *parm; + range_action ra; + + totalrc = 0; + + while ((parm = strsep(&str, ","))) { + rc = 0; + ra = action; + if (*parm == '!') { + if (ra == add) + ra = free; + else + ra = add; + parm++; + } + if (strcmp(parm, "all") == 0) { + from_cssid = 0; + from_ssid = 0; + from = 0; + to_cssid = __MAX_CSSID; + to_ssid = __MAX_SSID; + to = __MAX_SUBCHANNEL; + } else { + rc = parse_busid(strsep(&parm, "-"), &from_cssid, + &from_ssid, &from, msgtrigger); + if (!rc) { + if (parm != NULL) + rc = parse_busid(parm, &to_cssid, + &to_ssid, &to, + msgtrigger); + else { + to_cssid = from_cssid; + to_ssid = from_ssid; + to = from; + } } - blacklist_range (ra, from, to, to_ssid); } + if (!rc) { + rc = blacklist_range(ra, from_ssid, to_ssid, from, to, + msgtrigger); + if (rc) + totalrc = 1; + } else + totalrc = 1; } - return 1; + + return totalrc; } -/* Parsing the commandline for blacklist parameters, e.g. to blacklist - * bus ids 0.0.1234, 0.0.1235 and 0.0.1236, you could use any of: - * - cio_ignore=1234-1236 - * - cio_ignore=0x1234-0x1235,1236 - * - cio_ignore=0x1234,1235-1236 - * - cio_ignore=1236 cio_ignore=1234-0x1236 - * - cio_ignore=1234 cio_ignore=1236 cio_ignore=0x1235 - * - cio_ignore=0.0.1234-0.0.1236 - * - cio_ignore=0.0.1234,0x1235,1236 - * - ... - */ static int __init blacklist_setup (char *str) { CIO_MSG_EVENT(6, "Reading blacklist parameters\n"); - return blacklist_parse_parameters (str, add); + if (blacklist_parse_parameters(str, add, 1)) + return 0; + return 1; } __setup ("cio_ignore=", blacklist_setup); @@ -224,27 +230,23 @@ is_blacklisted (int ssid, int devno) * Function: blacklist_parse_proc_parameters * parse the stuff which is piped to /proc/cio_ignore */ -static void -blacklist_parse_proc_parameters (char *buf) +static int blacklist_parse_proc_parameters(char *buf) { - if (strncmp (buf, "free ", 5) == 0) { - blacklist_parse_parameters (buf + 5, free); - } else if (strncmp (buf, "add ", 4) == 0) { - /* - * We don't need to check for known devices since - * css_probe_device will handle this correctly. - */ - blacklist_parse_parameters (buf + 4, add); - } else { - printk (KERN_WARNING "cio: cio_ignore: Parse error; \n" - KERN_WARNING "try using 'free all|<devno-range>," - "<devno-range>,...'\n" - KERN_WARNING "or 'add <devno-range>," - "<devno-range>,...'\n"); - return; - } + int rc; + char *parm; + + parm = strsep(&buf, " "); + + if (strcmp("free", parm) == 0) + rc = blacklist_parse_parameters(buf, free, 0); + else if (strcmp("add", parm) == 0) + rc = blacklist_parse_parameters(buf, add, 0); + else + return 1; css_schedule_reprobe(); + + return rc; } /* Iterator struct for all devices. */ @@ -328,6 +330,8 @@ cio_ignore_write(struct file *file, const char __user *user_buf, size_t user_len, loff_t *offset) { char *buf; + size_t i; + ssize_t rc, ret; if (*offset) return -EINVAL; @@ -336,16 +340,27 @@ cio_ignore_write(struct file *file, const char __user *user_buf, buf = vmalloc (user_len + 1); /* maybe better use the stack? */ if (buf == NULL) return -ENOMEM; + memset(buf, 0, user_len + 1); + if (strncpy_from_user (buf, user_buf, user_len) < 0) { - vfree (buf); - return -EFAULT; + rc = -EFAULT; + goto out_free; } - buf[user_len] = '\0'; - blacklist_parse_proc_parameters (buf); + i = user_len - 1; + while ((i >= 0) && (isspace(buf[i]) || (buf[i] == 0))) { + buf[i] = '\0'; + i--; + } + ret = blacklist_parse_proc_parameters(buf); + if (ret) + rc = -EINVAL; + else + rc = user_len; +out_free: vfree (buf); - return user_len; + return rc; } static const struct seq_operations cio_ignore_proc_seq_ops = { diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c index 08a57816130..82c6a2d4512 100644 --- a/drivers/s390/cio/cio.c +++ b/drivers/s390/cio/cio.c @@ -39,23 +39,6 @@ debug_info_t *cio_debug_msg_id; debug_info_t *cio_debug_trace_id; debug_info_t *cio_debug_crw_id; -int cio_show_msg; - -static int __init -cio_setup (char *parm) -{ - if (!strcmp (parm, "yes")) - cio_show_msg = 1; - else if (!strcmp (parm, "no")) - cio_show_msg = 0; - else - printk(KERN_ERR "cio: cio_setup: " - "invalid cio_msg parameter '%s'", parm); - return 1; -} - -__setup ("cio_msg=", cio_setup); - /* * Function: cio_debug_init * Initializes three debug logs for common I/O: @@ -166,7 +149,7 @@ cio_start_handle_notoper(struct subchannel *sch, __u8 lpm) stsch (sch->schid, &sch->schib); - CIO_MSG_EVENT(0, "cio_start: 'not oper' status for " + CIO_MSG_EVENT(2, "cio_start: 'not oper' status for " "subchannel 0.%x.%04x!\n", sch->schid.ssid, sch->schid.sch_no); sprintf(dbf_text, "no%s", sch->dev.bus_id); @@ -567,10 +550,9 @@ cio_validate_subchannel (struct subchannel *sch, struct subchannel_id schid) * ... just being curious we check for non I/O subchannels */ if (sch->st != 0) { - CIO_DEBUG(KERN_INFO, 0, - "Subchannel 0.%x.%04x reports " - "non-I/O subchannel type %04X\n", - sch->schid.ssid, sch->schid.sch_no, sch->st); + CIO_MSG_EVENT(4, "Subchannel 0.%x.%04x reports " + "non-I/O subchannel type %04X\n", + sch->schid.ssid, sch->schid.sch_no, sch->st); /* We stop here for non-io subchannels. */ err = sch->st; goto out; @@ -588,7 +570,7 @@ cio_validate_subchannel (struct subchannel *sch, struct subchannel_id schid) * This device must not be known to Linux. So we simply * say that there is no device and return ENODEV. */ - CIO_MSG_EVENT(4, "Blacklisted device detected " + CIO_MSG_EVENT(6, "Blacklisted device detected " "at devno %04X, subchannel set %x\n", sch->schib.pmcw.dev, sch->schid.ssid); err = -ENODEV; @@ -601,12 +583,11 @@ cio_validate_subchannel (struct subchannel *sch, struct subchannel_id schid) sch->lpm = sch->schib.pmcw.pam & sch->opm; sch->isc = 3; - CIO_DEBUG(KERN_INFO, 0, - "Detected device %04x on subchannel 0.%x.%04X" - " - PIM = %02X, PAM = %02X, POM = %02X\n", - sch->schib.pmcw.dev, sch->schid.ssid, - sch->schid.sch_no, sch->schib.pmcw.pim, - sch->schib.pmcw.pam, sch->schib.pmcw.pom); + CIO_MSG_EVENT(6, "Detected device %04x on subchannel 0.%x.%04X " + "- PIM = %02X, PAM = %02X, POM = %02X\n", + sch->schib.pmcw.dev, sch->schid.ssid, + sch->schid.sch_no, sch->schib.pmcw.pim, + sch->schib.pmcw.pam, sch->schib.pmcw.pom); /* * We now have to initially ... diff --git a/drivers/s390/cio/cio.h b/drivers/s390/cio/cio.h index 3c75412904d..6e933aebe01 100644 --- a/drivers/s390/cio/cio.h +++ b/drivers/s390/cio/cio.h @@ -118,6 +118,4 @@ extern void *cio_get_console_priv(void); #define cio_get_console_priv() NULL #endif -extern int cio_show_msg; - #endif diff --git a/drivers/s390/cio/cio_debug.h b/drivers/s390/cio/cio_debug.h index d7429ef6c66..e64e8278c42 100644 --- a/drivers/s390/cio/cio_debug.h +++ b/drivers/s390/cio/cio_debug.h @@ -31,10 +31,4 @@ static inline void CIO_HEX_EVENT(int level, void *data, int length) } } -#define CIO_DEBUG(printk_level, event_level, msg...) do { \ - if (cio_show_msg) \ - printk(printk_level "cio: " msg); \ - CIO_MSG_EVENT(event_level, msg); \ - } while (0) - #endif diff --git a/drivers/s390/cio/css.c b/drivers/s390/cio/css.c index 595e327d2f7..a76956512b2 100644 --- a/drivers/s390/cio/css.c +++ b/drivers/s390/cio/css.c @@ -570,7 +570,7 @@ static void reprobe_all(struct work_struct *unused) { int ret; - CIO_MSG_EVENT(2, "reprobe start\n"); + CIO_MSG_EVENT(4, "reprobe start\n"); need_reprobe = 0; /* Make sure initial subchannel scan is done. */ @@ -578,7 +578,7 @@ static void reprobe_all(struct work_struct *unused) atomic_read(&ccw_device_init_count) == 0); ret = for_each_subchannel_staged(NULL, reprobe_subchannel, NULL); - CIO_MSG_EVENT(2, "reprobe done (rc=%d, need_reprobe=%d)\n", ret, + CIO_MSG_EVENT(4, "reprobe done (rc=%d, need_reprobe=%d)\n", ret, need_reprobe); } diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c index abfd601d237..e22813db74a 100644 --- a/drivers/s390/cio/device.c +++ b/drivers/s390/cio/device.c @@ -341,7 +341,7 @@ ccw_device_remove_disconnected(struct ccw_device *cdev) rc = device_schedule_callback(&cdev->dev, ccw_device_remove_orphan_cb); if (rc) - CIO_MSG_EVENT(2, "Couldn't unregister orphan " + CIO_MSG_EVENT(0, "Couldn't unregister orphan " "0.%x.%04x\n", cdev->private->dev_id.ssid, cdev->private->dev_id.devno); @@ -351,7 +351,7 @@ ccw_device_remove_disconnected(struct ccw_device *cdev) rc = device_schedule_callback(cdev->dev.parent, ccw_device_remove_sch_cb); if (rc) - CIO_MSG_EVENT(2, "Couldn't unregister disconnected device " + CIO_MSG_EVENT(0, "Couldn't unregister disconnected device " "0.%x.%04x\n", cdev->private->dev_id.ssid, cdev->private->dev_id.devno); @@ -397,7 +397,7 @@ int ccw_device_set_offline(struct ccw_device *cdev) if (ret == 0) wait_event(cdev->private->wait_q, dev_fsm_final_state(cdev)); else { - CIO_MSG_EVENT(2, "ccw_device_offline returned %d, " + CIO_MSG_EVENT(0, "ccw_device_offline returned %d, " "device 0.%x.%04x\n", ret, cdev->private->dev_id.ssid, cdev->private->dev_id.devno); @@ -433,7 +433,7 @@ int ccw_device_set_online(struct ccw_device *cdev) if (ret == 0) wait_event(cdev->private->wait_q, dev_fsm_final_state(cdev)); else { - CIO_MSG_EVENT(2, "ccw_device_online returned %d, " + CIO_MSG_EVENT(0, "ccw_device_online returned %d, " "device 0.%x.%04x\n", ret, cdev->private->dev_id.ssid, cdev->private->dev_id.devno); @@ -451,7 +451,7 @@ int ccw_device_set_online(struct ccw_device *cdev) if (ret == 0) wait_event(cdev->private->wait_q, dev_fsm_final_state(cdev)); else - CIO_MSG_EVENT(2, "ccw_device_offline returned %d, " + CIO_MSG_EVENT(0, "ccw_device_offline returned %d, " "device 0.%x.%04x\n", ret, cdev->private->dev_id.ssid, cdev->private->dev_id.devno); @@ -803,7 +803,7 @@ static void sch_attach_disconnected_device(struct subchannel *sch, other_sch = to_subchannel(get_device(cdev->dev.parent)); ret = device_move(&cdev->dev, &sch->dev); if (ret) { - CIO_MSG_EVENT(2, "Moving disconnected device 0.%x.%04x failed " + CIO_MSG_EVENT(0, "Moving disconnected device 0.%x.%04x failed " "(ret=%d)!\n", cdev->private->dev_id.ssid, cdev->private->dev_id.devno, ret); put_device(&other_sch->dev); @@ -933,7 +933,7 @@ io_subchannel_register(struct work_struct *work) ret = device_reprobe(&cdev->dev); if (ret) /* We can't do much here. */ - CIO_MSG_EVENT(2, "device_reprobe() returned" + CIO_MSG_EVENT(0, "device_reprobe() returned" " %d for 0.%x.%04x\n", ret, cdev->private->dev_id.ssid, cdev->private->dev_id.devno); @@ -1086,7 +1086,7 @@ static void ccw_device_move_to_sch(struct work_struct *work) rc = device_move(&cdev->dev, &sch->dev); mutex_unlock(&sch->reg_mutex); if (rc) { - CIO_MSG_EVENT(2, "Moving device 0.%x.%04x to subchannel " + CIO_MSG_EVENT(0, "Moving device 0.%x.%04x to subchannel " "0.%x.%04x failed (ret=%d)!\n", cdev->private->dev_id.ssid, cdev->private->dev_id.devno, sch->schid.ssid, @@ -1446,8 +1446,7 @@ ccw_device_remove (struct device *dev) wait_event(cdev->private->wait_q, dev_fsm_final_state(cdev)); else - //FIXME: we can't fail! - CIO_MSG_EVENT(2, "ccw_device_offline returned %d, " + CIO_MSG_EVENT(0, "ccw_device_offline returned %d, " "device 0.%x.%04x\n", ret, cdev->private->dev_id.ssid, cdev->private->dev_id.devno); @@ -1524,7 +1523,7 @@ static int recovery_check(struct device *dev, void *data) spin_lock_irq(cdev->ccwlock); switch (cdev->private->state) { case DEV_STATE_DISCONNECTED: - CIO_MSG_EVENT(3, "recovery: trigger 0.%x.%04x\n", + CIO_MSG_EVENT(4, "recovery: trigger 0.%x.%04x\n", cdev->private->dev_id.ssid, cdev->private->dev_id.devno); dev_fsm_event(cdev, DEV_EVENT_VERIFY); @@ -1554,7 +1553,7 @@ static void recovery_work_func(struct work_struct *unused) } spin_unlock_irq(&recovery_lock); } else - CIO_MSG_EVENT(2, "recovery: end\n"); + CIO_MSG_EVENT(4, "recovery: end\n"); } static DECLARE_WORK(recovery_work, recovery_work_func); @@ -1572,7 +1571,7 @@ void ccw_device_schedule_recovery(void) { unsigned long flags; - CIO_MSG_EVENT(2, "recovery: schedule\n"); + CIO_MSG_EVENT(4, "recovery: schedule\n"); spin_lock_irqsave(&recovery_lock, flags); if (!timer_pending(&recovery_timer) || (recovery_phase != 0)) { recovery_phase = 0; diff --git a/drivers/s390/cio/device_fsm.c b/drivers/s390/cio/device_fsm.c index 99403b0a97a..e268d5a77c1 100644 --- a/drivers/s390/cio/device_fsm.c +++ b/drivers/s390/cio/device_fsm.c @@ -322,10 +322,10 @@ ccw_device_recog_done(struct ccw_device *cdev, int state) same_dev = 0; /* Keep the compiler quiet... */ switch (state) { case DEV_STATE_NOT_OPER: - CIO_DEBUG(KERN_WARNING, 2, - "SenseID : unknown device %04x on subchannel " - "0.%x.%04x\n", cdev->private->dev_id.devno, - sch->schid.ssid, sch->schid.sch_no); + CIO_MSG_EVENT(2, "SenseID : unknown device %04x on " + "subchannel 0.%x.%04x\n", + cdev->private->dev_id.devno, + sch->schid.ssid, sch->schid.sch_no); break; case DEV_STATE_OFFLINE: if (cdev->private->state == DEV_STATE_DISCONNECTED_SENSE_ID) { @@ -348,20 +348,19 @@ ccw_device_recog_done(struct ccw_device *cdev, int state) return; } /* Issue device info message. */ - CIO_DEBUG(KERN_INFO, 2, - "SenseID : device 0.%x.%04x reports: " - "CU Type/Mod = %04X/%02X, Dev Type/Mod = " - "%04X/%02X\n", - cdev->private->dev_id.ssid, - cdev->private->dev_id.devno, - cdev->id.cu_type, cdev->id.cu_model, - cdev->id.dev_type, cdev->id.dev_model); + CIO_MSG_EVENT(4, "SenseID : device 0.%x.%04x reports: " + "CU Type/Mod = %04X/%02X, Dev Type/Mod = " + "%04X/%02X\n", + cdev->private->dev_id.ssid, + cdev->private->dev_id.devno, + cdev->id.cu_type, cdev->id.cu_model, + cdev->id.dev_type, cdev->id.dev_model); break; case DEV_STATE_BOXED: - CIO_DEBUG(KERN_WARNING, 2, - "SenseID : boxed device %04x on subchannel " - "0.%x.%04x\n", cdev->private->dev_id.devno, - sch->schid.ssid, sch->schid.sch_no); + CIO_MSG_EVENT(0, "SenseID : boxed device %04x on " + " subchannel 0.%x.%04x\n", + cdev->private->dev_id.devno, + sch->schid.ssid, sch->schid.sch_no); break; } cdev->private->state = state; @@ -443,9 +442,8 @@ ccw_device_done(struct ccw_device *cdev, int state) if (state == DEV_STATE_BOXED) - CIO_DEBUG(KERN_WARNING, 2, - "Boxed device %04x on subchannel %04x\n", - cdev->private->dev_id.devno, sch->schid.sch_no); + CIO_MSG_EVENT(0, "Boxed device %04x on subchannel %04x\n", + cdev->private->dev_id.devno, sch->schid.sch_no); if (cdev->private->flags.donotify) { cdev->private->flags.donotify = 0; @@ -900,7 +898,7 @@ ccw_device_w4sense(struct ccw_device *cdev, enum dev_event dev_event) /* Basic sense hasn't started. Try again. */ ccw_device_do_sense(cdev, irb); else { - CIO_MSG_EVENT(2, "Huh? 0.%x.%04x: unsolicited " + CIO_MSG_EVENT(0, "0.%x.%04x: unsolicited " "interrupt during w4sense...\n", cdev->private->dev_id.ssid, cdev->private->dev_id.devno); @@ -1169,8 +1167,10 @@ ccw_device_nop(struct ccw_device *cdev, enum dev_event dev_event) static void ccw_device_bug(struct ccw_device *cdev, enum dev_event dev_event) { - CIO_MSG_EVENT(0, "dev_jumptable[%i][%i] == NULL\n", - cdev->private->state, dev_event); + CIO_MSG_EVENT(0, "Internal state [%i][%i] not handled for device " + "0.%x.%04x\n", cdev->private->state, dev_event, + cdev->private->dev_id.ssid, + cdev->private->dev_id.devno); BUG(); } diff --git a/drivers/s390/cio/device_id.c b/drivers/s390/cio/device_id.c index dc4d87f77f6..cba7020517e 100644 --- a/drivers/s390/cio/device_id.c +++ b/drivers/s390/cio/device_id.c @@ -214,7 +214,7 @@ ccw_device_check_sense_id(struct ccw_device *cdev) * sense id information. So, for intervention required, * we use the "whack it until it talks" strategy... */ - CIO_MSG_EVENT(2, "SenseID : device %04x on Subchannel " + CIO_MSG_EVENT(0, "SenseID : device %04x on Subchannel " "0.%x.%04x reports cmd reject\n", cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no); @@ -239,7 +239,7 @@ ccw_device_check_sense_id(struct ccw_device *cdev) lpm = to_io_private(sch)->orb.lpm; if ((lpm & sch->schib.pmcw.pim & sch->schib.pmcw.pam) != 0) - CIO_MSG_EVENT(2, "SenseID : path %02X for device %04x " + CIO_MSG_EVENT(4, "SenseID : path %02X for device %04x " "on subchannel 0.%x.%04x is " "'not operational'\n", lpm, cdev->private->dev_id.devno, diff --git a/drivers/s390/cio/device_pgid.c b/drivers/s390/cio/device_pgid.c index c52449a1f9f..ba559053402 100644 --- a/drivers/s390/cio/device_pgid.c +++ b/drivers/s390/cio/device_pgid.c @@ -79,7 +79,7 @@ __ccw_device_sense_pgid_start(struct ccw_device *cdev) /* ret is 0, -EBUSY, -EACCES or -ENODEV */ if (ret != -EACCES) return ret; - CIO_MSG_EVENT(2, "SNID - Device %04x on Subchannel " + CIO_MSG_EVENT(3, "SNID - Device %04x on Subchannel " "0.%x.%04x, lpm %02X, became 'not " "operational'\n", cdev->private->dev_id.devno, @@ -159,7 +159,7 @@ __ccw_device_check_sense_pgid(struct ccw_device *cdev) u8 lpm; lpm = to_io_private(sch)->orb.lpm; - CIO_MSG_EVENT(2, "SNID - Device %04x on Subchannel 0.%x.%04x," + CIO_MSG_EVENT(3, "SNID - Device %04x on Subchannel 0.%x.%04x," " lpm %02X, became 'not operational'\n", cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no, lpm); @@ -275,7 +275,7 @@ __ccw_device_do_pgid(struct ccw_device *cdev, __u8 func) return ret; } /* PGID command failed on this path. */ - CIO_MSG_EVENT(2, "SPID - Device %04x on Subchannel " + CIO_MSG_EVENT(3, "SPID - Device %04x on Subchannel " "0.%x.%04x, lpm %02X, became 'not operational'\n", cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no, cdev->private->imask); @@ -317,7 +317,7 @@ static int __ccw_device_do_nop(struct ccw_device *cdev) return ret; } /* nop command failed on this path. */ - CIO_MSG_EVENT(2, "NOP - Device %04x on Subchannel " + CIO_MSG_EVENT(3, "NOP - Device %04x on Subchannel " "0.%x.%04x, lpm %02X, became 'not operational'\n", cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no, cdev->private->imask); @@ -362,7 +362,7 @@ __ccw_device_check_pgid(struct ccw_device *cdev) return -EAGAIN; } if (irb->scsw.cc == 3) { - CIO_MSG_EVENT(2, "SPID - Device %04x on Subchannel 0.%x.%04x," + CIO_MSG_EVENT(3, "SPID - Device %04x on Subchannel 0.%x.%04x," " lpm %02X, became 'not operational'\n", cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no, cdev->private->imask); @@ -391,7 +391,7 @@ static int __ccw_device_check_nop(struct ccw_device *cdev) return -ETIME; } if (irb->scsw.cc == 3) { - CIO_MSG_EVENT(2, "NOP - Device %04x on Subchannel 0.%x.%04x," + CIO_MSG_EVENT(3, "NOP - Device %04x on Subchannel 0.%x.%04x," " lpm %02X, became 'not operational'\n", cdev->private->dev_id.devno, sch->schid.ssid, sch->schid.sch_no, cdev->private->imask); diff --git a/drivers/s390/s390mach.c b/drivers/s390/s390mach.c index 4d4b54277c4..5080f343ad7 100644 --- a/drivers/s390/s390mach.c +++ b/drivers/s390/s390mach.c @@ -48,10 +48,11 @@ s390_collect_crw_info(void *param) int ccode; struct semaphore *sem; unsigned int chain; + int ignore; sem = (struct semaphore *)param; repeat: - down_interruptible(sem); + ignore = down_interruptible(sem); chain = 0; while (1) { if (unlikely(chain > 1)) { diff --git a/drivers/sbus/char/bpp.c b/drivers/sbus/char/bpp.c index 4fab0c23814..b87037ec980 100644 --- a/drivers/sbus/char/bpp.c +++ b/drivers/sbus/char/bpp.c @@ -41,7 +41,7 @@ #define BPP_DELAY 100 static const unsigned BPP_MAJOR = LP_MAJOR; -static const char* dev_name = "bpp"; +static const char *bpp_dev_name = "bpp"; /* When switching from compatibility to a mode where I can read, try the following mode first. */ diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig index 46d7e400c8b..81ccbd7f9e3 100644 --- a/drivers/scsi/Kconfig +++ b/drivers/scsi/Kconfig @@ -1679,6 +1679,7 @@ config MAC_SCSI config SCSI_MAC_ESP tristate "Macintosh NCR53c9[46] SCSI" depends on MAC && SCSI + select SCSI_SPI_ATTRS help This is the NCR 53c9x SCSI controller found on most of the 68040 based Macintoshes. diff --git a/drivers/scsi/dpt_i2o.c b/drivers/scsi/dpt_i2o.c index 0fb5bf4c43a..8508816f303 100644 --- a/drivers/scsi/dpt_i2o.c +++ b/drivers/scsi/dpt_i2o.c @@ -1967,45 +1967,6 @@ cleanup: return rcode; } - -/* - * This routine returns information about the system. This does not effect - * any logic and if the info is wrong - it doesn't matter. - */ - -/* Get all the info we can not get from kernel services */ -static int adpt_system_info(void __user *buffer) -{ - sysInfo_S si; - - memset(&si, 0, sizeof(si)); - - si.osType = OS_LINUX; - si.osMajorVersion = 0; - si.osMinorVersion = 0; - si.osRevision = 0; - si.busType = SI_PCI_BUS; - si.processorFamily = DPTI_sig.dsProcessorFamily; - -#if defined __i386__ - adpt_i386_info(&si); -#elif defined (__ia64__) - adpt_ia64_info(&si); -#elif defined(__sparc__) - adpt_sparc_info(&si); -#elif defined (__alpha__) - adpt_alpha_info(&si); -#else - si.processorType = 0xff ; -#endif - if(copy_to_user(buffer, &si, sizeof(si))){ - printk(KERN_WARNING"dpti: Could not copy buffer TO user\n"); - return -EFAULT; - } - - return 0; -} - #if defined __ia64__ static void adpt_ia64_info(sysInfo_S* si) { @@ -2016,7 +1977,6 @@ static void adpt_ia64_info(sysInfo_S* si) } #endif - #if defined __sparc__ static void adpt_sparc_info(sysInfo_S* si) { @@ -2026,7 +1986,6 @@ static void adpt_sparc_info(sysInfo_S* si) si->processorType = PROC_ULTRASPARC; } #endif - #if defined __alpha__ static void adpt_alpha_info(sysInfo_S* si) { @@ -2038,7 +1997,6 @@ static void adpt_alpha_info(sysInfo_S* si) #endif #if defined __i386__ - static void adpt_i386_info(sysInfo_S* si) { // This is all the info we need for now @@ -2059,9 +2017,45 @@ static void adpt_i386_info(sysInfo_S* si) break; } } +#endif + +/* + * This routine returns information about the system. This does not effect + * any logic and if the info is wrong - it doesn't matter. + */ +/* Get all the info we can not get from kernel services */ +static int adpt_system_info(void __user *buffer) +{ + sysInfo_S si; + + memset(&si, 0, sizeof(si)); + + si.osType = OS_LINUX; + si.osMajorVersion = 0; + si.osMinorVersion = 0; + si.osRevision = 0; + si.busType = SI_PCI_BUS; + si.processorFamily = DPTI_sig.dsProcessorFamily; + +#if defined __i386__ + adpt_i386_info(&si); +#elif defined (__ia64__) + adpt_ia64_info(&si); +#elif defined(__sparc__) + adpt_sparc_info(&si); +#elif defined (__alpha__) + adpt_alpha_info(&si); +#else + si.processorType = 0xff ; #endif + if (copy_to_user(buffer, &si, sizeof(si))){ + printk(KERN_WARNING"dpti: Could not copy buffer TO user\n"); + return -EFAULT; + } + return 0; +} static int adpt_ioctl(struct inode *inode, struct file *file, uint cmd, ulong arg) diff --git a/drivers/scsi/dpti.h b/drivers/scsi/dpti.h index 924cd5a5167..337746d4604 100644 --- a/drivers/scsi/dpti.h +++ b/drivers/scsi/dpti.h @@ -316,19 +316,6 @@ static int adpt_close(struct inode *inode, struct file *file); static void adpt_delay(int millisec); #endif -#if defined __ia64__ -static void adpt_ia64_info(sysInfo_S* si); -#endif -#if defined __sparc__ -static void adpt_sparc_info(sysInfo_S* si); -#endif -#if defined __alpha__ -static void adpt_sparc_info(sysInfo_S* si); -#endif -#if defined __i386__ -static void adpt_i386_info(sysInfo_S* si); -#endif - #define PRINT_BUFFER_SIZE 512 #define HBA_FLAGS_DBG_FLAGS_MASK 0xffff0000 // Mask for debug flags diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c index 1e2b9d826f6..eab03273379 100644 --- a/drivers/serial/serial_core.c +++ b/drivers/serial/serial_core.c @@ -556,7 +556,7 @@ static int uart_chars_in_buffer(struct tty_struct *tty) static void uart_flush_buffer(struct tty_struct *tty) { struct uart_state *state = tty->driver_data; - struct uart_port *port = state->port; + struct uart_port *port; unsigned long flags; /* @@ -568,6 +568,7 @@ static void uart_flush_buffer(struct tty_struct *tty) return; } + port = state->port; pr_debug("uart_flush_buffer(%d) called\n", tty->index); spin_lock_irqsave(&port->lock, flags); diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig index 33b467a8352..1ef6df395e0 100644 --- a/drivers/usb/host/Kconfig +++ b/drivers/usb/host/Kconfig @@ -129,7 +129,7 @@ config USB_ISP1760_PCI config USB_ISP1760_OF bool "Support for the OF platform bus" - depends on USB_ISP1760_HCD && OF + depends on USB_ISP1760_HCD && PPC_OF ---help--- Enables support for the device present on the PowerPC OpenFirmware platform bus. diff --git a/drivers/usb/serial/iuu_phoenix.c b/drivers/usb/serial/iuu_phoenix.c index 8a217648b25..a01e987c7d3 100644 --- a/drivers/usb/serial/iuu_phoenix.c +++ b/drivers/usb/serial/iuu_phoenix.c @@ -643,7 +643,7 @@ static void read_buf_callback(struct urb *urb) static int iuu_bulk_write(struct usb_serial_port *port) { struct iuu_private *priv = usb_get_serial_port_data(port); - unsigned int flags; + unsigned long flags; int result; int i; char *buf_ptr = port->write_urb->transfer_buffer; @@ -694,7 +694,7 @@ static void iuu_uart_read_callback(struct urb *urb) { struct usb_serial_port *port = urb->context; struct iuu_private *priv = usb_get_serial_port_data(port); - unsigned int flags; + unsigned long flags; int status; int error = 0; int len = 0; @@ -759,7 +759,7 @@ static int iuu_uart_write(struct usb_serial_port *port, const u8 *buf, int count) { struct iuu_private *priv = usb_get_serial_port_data(port); - unsigned int flags; + unsigned long flags; dbg("%s - enter", __func__); if (count > 256) diff --git a/drivers/video/bw2.c b/drivers/video/bw2.c index 275d9dab0c6..79f85dc402d 100644 --- a/drivers/video/bw2.c +++ b/drivers/video/bw2.c @@ -329,7 +329,7 @@ static int __devinit bw2_probe(struct of_device *op, const struct of_device_id * if (!info->screen_base) goto out_unmap_regs; - bw2_blank(0, info); + bw2_blank(FB_BLANK_UNBLANK, info); bw2_init_fix(info, linebytes); diff --git a/drivers/video/cg3.c b/drivers/video/cg3.c index 010ea53978f..e31e26a6bb7 100644 --- a/drivers/video/cg3.c +++ b/drivers/video/cg3.c @@ -398,7 +398,7 @@ static int __devinit cg3_probe(struct of_device *op, if (!info->screen_base) goto out_unmap_regs; - cg3_blank(0, info); + cg3_blank(FB_BLANK_UNBLANK, info); if (!of_find_property(dp, "width", NULL)) { err = cg3_do_default_mode(par); diff --git a/drivers/video/cg6.c b/drivers/video/cg6.c index fc90db6da65..8000bccecdc 100644 --- a/drivers/video/cg6.c +++ b/drivers/video/cg6.c @@ -767,7 +767,7 @@ static int __devinit cg6_probe(struct of_device *op, cg6_bt_init(par); cg6_chip_init(info); - cg6_blank(0, info); + cg6_blank(FB_BLANK_UNBLANK, info); if (fb_alloc_cmap(&info->cmap, 256, 0)) goto out_unmap_regs; diff --git a/drivers/video/ffb.c b/drivers/video/ffb.c index 93dca3e2aa5..0f42a696d17 100644 --- a/drivers/video/ffb.c +++ b/drivers/video/ffb.c @@ -987,7 +987,7 @@ static int __devinit ffb_probe(struct of_device *op, * chosen console, it will have video outputs off in * the DAC. */ - ffb_blank(0, info); + ffb_blank(FB_BLANK_UNBLANK, info); if (fb_alloc_cmap(&info->cmap, 256, 0)) goto out_unmap_dac; diff --git a/drivers/video/leo.c b/drivers/video/leo.c index f3160fc2979..fb129928d5d 100644 --- a/drivers/video/leo.c +++ b/drivers/video/leo.c @@ -601,7 +601,7 @@ static int __devinit leo_probe(struct of_device *op, const struct of_device_id * leo_init_wids(info); leo_init_hw(info); - leo_blank(0, info); + leo_blank(FB_BLANK_UNBLANK, info); if (fb_alloc_cmap(&info->cmap, 256, 0)) goto out_unmap_regs; diff --git a/drivers/video/p9100.c b/drivers/video/p9100.c index c95874fe907..676ffb06d1c 100644 --- a/drivers/video/p9100.c +++ b/drivers/video/p9100.c @@ -295,7 +295,7 @@ static int __devinit p9100_probe(struct of_device *op, const struct of_device_id if (!info->screen_base) goto out_unmap_regs; - p9100_blank(0, info); + p9100_blank(FB_BLANK_UNBLANK, info); if (fb_alloc_cmap(&info->cmap, 256, 0)) goto out_unmap_screen; diff --git a/drivers/video/tcx.c b/drivers/video/tcx.c index a7177430577..44e8c27ed0f 100644 --- a/drivers/video/tcx.c +++ b/drivers/video/tcx.c @@ -84,7 +84,7 @@ struct tcx_tec { struct tcx_thc { u32 thc_rev; - u32 thc_pad0[511]; + u32 thc_pad0[511]; u32 thc_hs; /* hsync timing */ u32 thc_hsdvs; u32 thc_hd; @@ -126,10 +126,10 @@ struct tcx_par { }; /* Reset control plane so that WID is 8-bit plane. */ -static void __tcx_set_control_plane (struct tcx_par *par) +static void __tcx_set_control_plane(struct tcx_par *par) { u32 __iomem *p, *pend; - + if (par->lowdepth) return; @@ -143,8 +143,8 @@ static void __tcx_set_control_plane (struct tcx_par *par) sbus_writel(tmp, p); } } - -static void tcx_reset (struct fb_info *info) + +static void tcx_reset(struct fb_info *info) { struct tcx_par *par = (struct tcx_par *) info->par; unsigned long flags; @@ -365,7 +365,8 @@ static void tcx_unmap_regs(struct of_device *op, struct fb_info *info, info->screen_base, par->fbsize); } -static int __devinit tcx_init_one(struct of_device *op) +static int __devinit tcx_probe(struct of_device *op, + const struct of_device_id *match) { struct device_node *dp = op->node; struct fb_info *info; @@ -488,13 +489,6 @@ out_err: return err; } -static int __devinit tcx_probe(struct of_device *dev, const struct of_device_id *match) -{ - struct of_device *op = to_of_device(&dev->dev); - - return tcx_init_one(op); -} - static int __devexit tcx_remove(struct of_device *op) { struct fb_info *info = dev_get_drvdata(&op->dev); diff --git a/fs/affs/affs.h b/fs/affs/affs.h index d5bd497ab9c..223b1917093 100644 --- a/fs/affs/affs.h +++ b/fs/affs/affs.h @@ -48,7 +48,7 @@ struct affs_ext_key { * affs fs inode data in memory */ struct affs_inode_info { - u32 i_opencnt; + atomic_t i_opencnt; struct semaphore i_link_lock; /* Protects internal inode access. */ struct semaphore i_ext_lock; /* Protects internal inode access. */ #define i_hash_lock i_ext_lock @@ -170,8 +170,6 @@ extern int affs_rename(struct inode *old_dir, struct dentry *old_dentry, extern unsigned long affs_parent_ino(struct inode *dir); extern struct inode *affs_new_inode(struct inode *dir); extern int affs_notify_change(struct dentry *dentry, struct iattr *attr); -extern void affs_put_inode(struct inode *inode); -extern void affs_drop_inode(struct inode *inode); extern void affs_delete_inode(struct inode *inode); extern void affs_clear_inode(struct inode *inode); extern struct inode *affs_iget(struct super_block *sb, diff --git a/fs/affs/file.c b/fs/affs/file.c index 1a4f092f24e..6eac7bdeec9 100644 --- a/fs/affs/file.c +++ b/fs/affs/file.c @@ -48,8 +48,9 @@ affs_file_open(struct inode *inode, struct file *filp) { if (atomic_read(&filp->f_count) != 1) return 0; - pr_debug("AFFS: open(%d)\n", AFFS_I(inode)->i_opencnt); - AFFS_I(inode)->i_opencnt++; + pr_debug("AFFS: open(%lu,%d)\n", + inode->i_ino, atomic_read(&AFFS_I(inode)->i_opencnt)); + atomic_inc(&AFFS_I(inode)->i_opencnt); return 0; } @@ -58,10 +59,16 @@ affs_file_release(struct inode *inode, struct file *filp) { if (atomic_read(&filp->f_count) != 0) return 0; - pr_debug("AFFS: release(%d)\n", AFFS_I(inode)->i_opencnt); - AFFS_I(inode)->i_opencnt--; - if (!AFFS_I(inode)->i_opencnt) + pr_debug("AFFS: release(%lu, %d)\n", + inode->i_ino, atomic_read(&AFFS_I(inode)->i_opencnt)); + + if (atomic_dec_and_test(&AFFS_I(inode)->i_opencnt)) { + mutex_lock(&inode->i_mutex); + if (inode->i_size != AFFS_I(inode)->mmu_private) + affs_truncate(inode); affs_free_prealloc(inode); + mutex_unlock(&inode->i_mutex); + } return 0; } @@ -180,7 +187,7 @@ affs_get_extblock(struct inode *inode, u32 ext) /* inline the simplest case: same extended block as last time */ struct buffer_head *bh = AFFS_I(inode)->i_ext_bh; if (ext == AFFS_I(inode)->i_ext_last) - atomic_inc(&bh->b_count); + get_bh(bh); else /* we have to do more (not inlined) */ bh = affs_get_extblock_slow(inode, ext); @@ -306,7 +313,7 @@ store_ext: affs_brelse(AFFS_I(inode)->i_ext_bh); AFFS_I(inode)->i_ext_last = ext; AFFS_I(inode)->i_ext_bh = bh; - atomic_inc(&bh->b_count); + get_bh(bh); return bh; @@ -324,7 +331,6 @@ affs_get_block(struct inode *inode, sector_t block, struct buffer_head *bh_resul pr_debug("AFFS: get_block(%u, %lu)\n", (u32)inode->i_ino, (unsigned long)block); - BUG_ON(block > (sector_t)0x7fffffffUL); if (block >= AFFS_I(inode)->i_blkcnt) { @@ -827,6 +833,8 @@ affs_truncate(struct inode *inode) res = mapping->a_ops->write_begin(NULL, mapping, size, 0, 0, &page, &fsdata); if (!res) res = mapping->a_ops->write_end(NULL, mapping, size, 0, 0, page, fsdata); + else + inode->i_size = AFFS_I(inode)->mmu_private; mark_inode_dirty(inode); return; } else if (inode->i_size == AFFS_I(inode)->mmu_private) @@ -862,6 +870,7 @@ affs_truncate(struct inode *inode) blk++; } else AFFS_HEAD(ext_bh)->first_data = 0; + AFFS_HEAD(ext_bh)->block_count = cpu_to_be32(i); size = AFFS_SB(sb)->s_hashsize; if (size > blkcnt - blk + i) size = blkcnt - blk + i; diff --git a/fs/affs/inode.c b/fs/affs/inode.c index 27fe6cbe43a..a13b334a391 100644 --- a/fs/affs/inode.c +++ b/fs/affs/inode.c @@ -58,7 +58,7 @@ struct inode *affs_iget(struct super_block *sb, unsigned long ino) AFFS_I(inode)->i_extcnt = 1; AFFS_I(inode)->i_ext_last = ~1; AFFS_I(inode)->i_protect = prot; - AFFS_I(inode)->i_opencnt = 0; + atomic_set(&AFFS_I(inode)->i_opencnt, 0); AFFS_I(inode)->i_blkcnt = 0; AFFS_I(inode)->i_lc = NULL; AFFS_I(inode)->i_lc_size = 0; @@ -108,8 +108,6 @@ struct inode *affs_iget(struct super_block *sb, unsigned long ino) inode->i_mode |= S_IFDIR; } else inode->i_mode = S_IRUGO | S_IXUGO | S_IWUSR | S_IFDIR; - if (tail->link_chain) - inode->i_nlink = 2; /* Maybe it should be controlled by mount parameter? */ //inode->i_mode |= S_ISVTX; inode->i_op = &affs_dir_inode_operations; @@ -245,31 +243,12 @@ out: } void -affs_put_inode(struct inode *inode) -{ - pr_debug("AFFS: put_inode(ino=%lu, nlink=%u)\n", inode->i_ino, inode->i_nlink); - affs_free_prealloc(inode); -} - -void -affs_drop_inode(struct inode *inode) -{ - mutex_lock(&inode->i_mutex); - if (inode->i_size != AFFS_I(inode)->mmu_private) - affs_truncate(inode); - mutex_unlock(&inode->i_mutex); - - generic_drop_inode(inode); -} - -void affs_delete_inode(struct inode *inode) { pr_debug("AFFS: delete_inode(ino=%lu, nlink=%u)\n", inode->i_ino, inode->i_nlink); truncate_inode_pages(&inode->i_data, 0); inode->i_size = 0; - if (S_ISREG(inode->i_mode)) - affs_truncate(inode); + affs_truncate(inode); clear_inode(inode); affs_free_block(inode->i_sb, inode->i_ino); } @@ -277,9 +256,12 @@ affs_delete_inode(struct inode *inode) void affs_clear_inode(struct inode *inode) { - unsigned long cache_page = (unsigned long) AFFS_I(inode)->i_lc; + unsigned long cache_page; pr_debug("AFFS: clear_inode(ino=%lu, nlink=%u)\n", inode->i_ino, inode->i_nlink); + + affs_free_prealloc(inode); + cache_page = (unsigned long)AFFS_I(inode)->i_lc; if (cache_page) { pr_debug("AFFS: freeing ext cache\n"); AFFS_I(inode)->i_lc = NULL; @@ -316,7 +298,7 @@ affs_new_inode(struct inode *dir) inode->i_ino = block; inode->i_nlink = 1; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; - AFFS_I(inode)->i_opencnt = 0; + atomic_set(&AFFS_I(inode)->i_opencnt, 0); AFFS_I(inode)->i_blkcnt = 0; AFFS_I(inode)->i_lc = NULL; AFFS_I(inode)->i_lc_size = 0; @@ -369,12 +351,12 @@ affs_add_entry(struct inode *dir, struct inode *inode, struct dentry *dentry, s3 switch (type) { case ST_LINKFILE: case ST_LINKDIR: - inode_bh = bh; retval = -ENOSPC; block = affs_alloc_block(dir, dir->i_ino); if (!block) goto err; retval = -EIO; + inode_bh = bh; bh = affs_getzeroblk(sb, block); if (!bh) goto err; diff --git a/fs/affs/namei.c b/fs/affs/namei.c index 2218f1ee71c..cfcf1b6cf82 100644 --- a/fs/affs/namei.c +++ b/fs/affs/namei.c @@ -234,7 +234,8 @@ affs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) int affs_unlink(struct inode *dir, struct dentry *dentry) { - pr_debug("AFFS: unlink(dir=%d, \"%.*s\")\n", (u32)dir->i_ino, + pr_debug("AFFS: unlink(dir=%d, %lu \"%.*s\")\n", (u32)dir->i_ino, + dentry->d_inode->i_ino, (int)dentry->d_name.len, dentry->d_name.name); return affs_remove_header(dentry); @@ -302,7 +303,8 @@ affs_mkdir(struct inode *dir, struct dentry *dentry, int mode) int affs_rmdir(struct inode *dir, struct dentry *dentry) { - pr_debug("AFFS: rmdir(dir=%u, \"%.*s\")\n", (u32)dir->i_ino, + pr_debug("AFFS: rmdir(dir=%u, %lu \"%.*s\")\n", (u32)dir->i_ino, + dentry->d_inode->i_ino, (int)dentry->d_name.len, dentry->d_name.name); return affs_remove_header(dentry); diff --git a/fs/affs/super.c b/fs/affs/super.c index 01d25d53254..d214837d5e4 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -71,12 +71,18 @@ static struct kmem_cache * affs_inode_cachep; static struct inode *affs_alloc_inode(struct super_block *sb) { - struct affs_inode_info *ei; - ei = (struct affs_inode_info *)kmem_cache_alloc(affs_inode_cachep, GFP_KERNEL); - if (!ei) + struct affs_inode_info *i; + + i = kmem_cache_alloc(affs_inode_cachep, GFP_KERNEL); + if (!i) return NULL; - ei->vfs_inode.i_version = 1; - return &ei->vfs_inode; + + i->vfs_inode.i_version = 1; + i->i_lc = NULL; + i->i_ext_bh = NULL; + i->i_pa_cnt = 0; + + return &i->vfs_inode; } static void affs_destroy_inode(struct inode *inode) @@ -114,8 +120,6 @@ static const struct super_operations affs_sops = { .alloc_inode = affs_alloc_inode, .destroy_inode = affs_destroy_inode, .write_inode = affs_write_inode, - .put_inode = affs_put_inode, - .drop_inode = affs_drop_inode, .delete_inode = affs_delete_inode, .clear_inode = affs_clear_inode, .put_super = affs_put_super, @@ -158,7 +158,7 @@ struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs) bio_init(bio); if (likely(nr_iovecs)) { - unsigned long idx = 0; /* shut up gcc */ + unsigned long uninitialized_var(idx); bvl = bvec_alloc_bs(gfp_mask, nr_iovecs, &idx, bs); if (unlikely(!bvl)) { @@ -963,6 +963,7 @@ static void bio_copy_kern_endio(struct bio *bio, int err) * @data: pointer to buffer to copy * @len: length in bytes * @gfp_mask: allocation flags for bio and page allocation + * @reading: data direction is READ * * copy the kernel address into a bio suitable for io to a block * device. Returns an error pointer in case of error. diff --git a/fs/inode.c b/fs/inode.c index bf647813042..c36d9480335 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1149,13 +1149,8 @@ static inline void iput_final(struct inode *inode) void iput(struct inode *inode) { if (inode) { - const struct super_operations *op = inode->i_sb->s_op; - BUG_ON(inode->i_state == I_CLEAR); - if (op && op->put_inode) - op->put_inode(inode); - if (atomic_dec_and_lock(&inode->i_count, &inode_lock)) iput_final(inode); } diff --git a/fs/locks.c b/fs/locks.c index 663c069b59b..0ac6b92cb0b 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -1753,6 +1753,7 @@ int fcntl_setlk(unsigned int fd, struct file *filp, unsigned int cmd, struct file_lock *file_lock = locks_alloc_lock(); struct flock flock; struct inode *inode; + struct file *f; int error; if (file_lock == NULL) @@ -1825,7 +1826,15 @@ again: * Attempt to detect a close/fcntl race and recover by * releasing the lock that was just acquired. */ - if (!error && fcheck(fd) != filp && flock.l_type != F_UNLCK) { + /* + * we need that spin_lock here - it prevents reordering between + * update of inode->i_flock and check for it done in close(). + * rcu_read_lock() wouldn't do. + */ + spin_lock(¤t->files->file_lock); + f = fcheck(fd); + spin_unlock(¤t->files->file_lock); + if (!error && f != filp && flock.l_type != F_UNLCK) { flock.l_type = F_UNLCK; goto again; } @@ -1881,6 +1890,7 @@ int fcntl_setlk64(unsigned int fd, struct file *filp, unsigned int cmd, struct file_lock *file_lock = locks_alloc_lock(); struct flock64 flock; struct inode *inode; + struct file *f; int error; if (file_lock == NULL) @@ -1953,7 +1963,10 @@ again: * Attempt to detect a close/fcntl race and recover by * releasing the lock that was just acquired. */ - if (!error && fcheck(fd) != filp && flock.l_type != F_UNLCK) { + spin_lock(¤t->files->file_lock); + f = fcheck(fd); + spin_unlock(¤t->files->file_lock); + if (!error && f != filp && flock.l_type != F_UNLCK) { flock.l_type = F_UNLCK; goto again; } diff --git a/fs/pipe.c b/fs/pipe.c index f73492b6817..ec228bc9f88 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -17,6 +17,7 @@ #include <linux/highmem.h> #include <linux/pagemap.h> #include <linux/audit.h> +#include <linux/syscalls.h> #include <asm/uaccess.h> #include <asm/ioctls.h> @@ -1076,6 +1077,26 @@ int do_pipe(int *fd) } /* + * sys_pipe() is the normal C calling standard for creating + * a pipe. It's not the way Unix traditionally does this, though. + */ +asmlinkage long __weak sys_pipe(int __user *fildes) +{ + int fd[2]; + int error; + + error = do_pipe(fd); + if (!error) { + if (copy_to_user(fildes, fd, sizeof(fd))) { + sys_close(fd[0]); + sys_close(fd[1]); + error = -EFAULT; + } + } + return error; +} + +/* * pipefs should _never_ be mounted by userland - too much of security hassle, * no real gain from having the whole whorehouse mounted. So we don't need * any operations on the root directory. However, we need a non-trivial diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index e2b8e769f51..88717c0f941 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -5,11 +5,9 @@ #include <linux/highmem.h> #include <linux/ptrace.h> #include <linux/pagemap.h> -#include <linux/ptrace.h> #include <linux/mempolicy.h> #include <linux/swap.h> #include <linux/swapops.h> -#include <linux/seq_file.h> #include <asm/elf.h> #include <asm/uaccess.h> diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c index 4b733f10845..4b4f9cc2f18 100644 --- a/fs/proc/task_nommu.c +++ b/fs/proc/task_nommu.c @@ -1,6 +1,7 @@ #include <linux/mm.h> #include <linux/file.h> +#include <linux/fdtable.h> #include <linux/mount.h> #include <linux/ptrace.h> #include <linux/seq_file.h> diff --git a/fs/splice.c b/fs/splice.c index 633f58ebfb7..78150038b58 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -811,24 +811,19 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, { struct address_space *mapping = out->f_mapping; struct inode *inode = mapping->host; - int killsuid, killpriv; + struct splice_desc sd = { + .total_len = len, + .flags = flags, + .pos = *ppos, + .u.file = out, + }; ssize_t ret; - int err = 0; - - killpriv = security_inode_need_killpriv(out->f_path.dentry); - killsuid = should_remove_suid(out->f_path.dentry); - if (unlikely(killsuid || killpriv)) { - mutex_lock(&inode->i_mutex); - if (killpriv) - err = security_inode_killpriv(out->f_path.dentry); - if (!err && killsuid) - err = __remove_suid(out->f_path.dentry, killsuid); - mutex_unlock(&inode->i_mutex); - if (err) - return err; - } - ret = splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_file); + inode_double_lock(inode, pipe->inode); + ret = remove_suid(out->f_path.dentry); + if (likely(!ret)) + ret = __splice_from_pipe(pipe, &sd, pipe_to_file); + inode_double_unlock(inode, pipe->inode); if (ret > 0) { unsigned long nr_pages; @@ -840,6 +835,8 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, * sync it. */ if (unlikely((out->f_flags & O_SYNC) || IS_SYNC(inode))) { + int err; + mutex_lock(&inode->i_mutex); err = generic_osync_inode(inode, mapping, OSYNC_METADATA|OSYNC_DATA); @@ -1075,7 +1072,7 @@ long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, ret = splice_direct_to_actor(in, &sd, direct_splice_actor); if (ret > 0) - *ppos = sd.pos; + *ppos += ret; return ret; } diff --git a/fs/udf/namei.c b/fs/udf/namei.c index 2b34c8ca6c8..d3231947db1 100644 --- a/fs/udf/namei.c +++ b/fs/udf/namei.c @@ -32,6 +32,7 @@ #include <linux/buffer_head.h> #include <linux/sched.h> #include <linux/crc-itu-t.h> +#include <linux/exportfs.h> static inline int udf_match(int len1, const char *name1, int len2, const char *name2) @@ -158,6 +159,8 @@ static struct fileIdentDesc *udf_find_entry(struct inode *dir, sector_t offset; struct extent_position epos = {}; struct udf_inode_info *dinfo = UDF_I(dir); + int isdotdot = dentry->d_name.len == 2 && + dentry->d_name.name[0] == '.' && dentry->d_name.name[1] == '.'; size = udf_ext0_offset(dir) + dir->i_size; f_pos = udf_ext0_offset(dir); @@ -225,6 +228,12 @@ static struct fileIdentDesc *udf_find_entry(struct inode *dir, continue; } + if ((cfi->fileCharacteristics & FID_FILE_CHAR_PARENT) && + isdotdot) { + brelse(epos.bh); + return fi; + } + if (!lfi) continue; @@ -286,9 +295,8 @@ static struct dentry *udf_lookup(struct inode *dir, struct dentry *dentry, } } unlock_kernel(); - d_add(dentry, inode); - return NULL; + return d_splice_alias(inode, dentry); } static struct fileIdentDesc *udf_add_entry(struct inode *dir, @@ -307,7 +315,7 @@ static struct fileIdentDesc *udf_add_entry(struct inode *dir, uint16_t liu; int block; kernel_lb_addr eloc; - uint32_t elen; + uint32_t elen = 0; sector_t offset; struct extent_position epos = {}; struct udf_inode_info *dinfo; @@ -398,7 +406,8 @@ static struct fileIdentDesc *udf_add_entry(struct inode *dir, } add: - if (dinfo->i_alloc_type != ICBTAG_FLAG_AD_IN_ICB) { + /* Is there any extent whose size we need to round up? */ + if (dinfo->i_alloc_type != ICBTAG_FLAG_AD_IN_ICB && elen) { elen = (elen + sb->s_blocksize - 1) & ~(sb->s_blocksize - 1); if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT) epos.offset -= sizeof(short_ad); @@ -1232,6 +1241,134 @@ end_rename: return retval; } +static struct dentry *udf_get_parent(struct dentry *child) +{ + struct dentry *parent; + struct inode *inode = NULL; + struct dentry dotdot; + struct fileIdentDesc cfi; + struct udf_fileident_bh fibh; + + dotdot.d_name.name = ".."; + dotdot.d_name.len = 2; + + lock_kernel(); + if (!udf_find_entry(child->d_inode, &dotdot, &fibh, &cfi)) + goto out_unlock; + + if (fibh.sbh != fibh.ebh) + brelse(fibh.ebh); + brelse(fibh.sbh); + + inode = udf_iget(child->d_inode->i_sb, + lelb_to_cpu(cfi.icb.extLocation)); + if (!inode) + goto out_unlock; + unlock_kernel(); + + parent = d_alloc_anon(inode); + if (!parent) { + iput(inode); + parent = ERR_PTR(-ENOMEM); + } + + return parent; +out_unlock: + unlock_kernel(); + return ERR_PTR(-EACCES); +} + + +static struct dentry *udf_nfs_get_inode(struct super_block *sb, u32 block, + u16 partref, __u32 generation) +{ + struct inode *inode; + struct dentry *result; + kernel_lb_addr loc; + + if (block == 0) + return ERR_PTR(-ESTALE); + + loc.logicalBlockNum = block; + loc.partitionReferenceNum = partref; + inode = udf_iget(sb, loc); + + if (inode == NULL) + return ERR_PTR(-ENOMEM); + + if (generation && inode->i_generation != generation) { + iput(inode); + return ERR_PTR(-ESTALE); + } + result = d_alloc_anon(inode); + if (!result) { + iput(inode); + return ERR_PTR(-ENOMEM); + } + return result; +} + +static struct dentry *udf_fh_to_dentry(struct super_block *sb, + struct fid *fid, int fh_len, int fh_type) +{ + if ((fh_len != 3 && fh_len != 5) || + (fh_type != FILEID_UDF_WITH_PARENT && + fh_type != FILEID_UDF_WITHOUT_PARENT)) + return NULL; + + return udf_nfs_get_inode(sb, fid->udf.block, fid->udf.partref, + fid->udf.generation); +} + +static struct dentry *udf_fh_to_parent(struct super_block *sb, + struct fid *fid, int fh_len, int fh_type) +{ + if (fh_len != 5 || fh_type != FILEID_UDF_WITH_PARENT) + return NULL; + + return udf_nfs_get_inode(sb, fid->udf.parent_block, + fid->udf.parent_partref, + fid->udf.parent_generation); +} +static int udf_encode_fh(struct dentry *de, __u32 *fh, int *lenp, + int connectable) +{ + int len = *lenp; + struct inode *inode = de->d_inode; + kernel_lb_addr location = UDF_I(inode)->i_location; + struct fid *fid = (struct fid *)fh; + int type = FILEID_UDF_WITHOUT_PARENT; + + if (len < 3 || (connectable && len < 5)) + return 255; + + *lenp = 3; + fid->udf.block = location.logicalBlockNum; + fid->udf.partref = location.partitionReferenceNum; + fid->udf.generation = inode->i_generation; + + if (connectable && !S_ISDIR(inode->i_mode)) { + spin_lock(&de->d_lock); + inode = de->d_parent->d_inode; + location = UDF_I(inode)->i_location; + fid->udf.parent_block = location.logicalBlockNum; + fid->udf.parent_partref = location.partitionReferenceNum; + fid->udf.parent_generation = inode->i_generation; + spin_unlock(&de->d_lock); + *lenp = 5; + type = FILEID_UDF_WITH_PARENT; + } + + return type; +} + +const struct export_operations udf_export_ops = { + .encode_fh = udf_encode_fh, + .fh_to_dentry = udf_fh_to_dentry, + .fh_to_parent = udf_fh_to_parent, + .get_parent = udf_get_parent, +}; + const struct inode_operations udf_dir_inode_operations = { .lookup = udf_lookup, .create = udf_create, diff --git a/fs/udf/partition.c b/fs/udf/partition.c index 63610f026ae..96dfd207c3d 100644 --- a/fs/udf/partition.c +++ b/fs/udf/partition.c @@ -27,8 +27,8 @@ #include <linux/slab.h> #include <linux/buffer_head.h> -inline uint32_t udf_get_pblock(struct super_block *sb, uint32_t block, - uint16_t partition, uint32_t offset) +uint32_t udf_get_pblock(struct super_block *sb, uint32_t block, + uint16_t partition, uint32_t offset) { struct udf_sb_info *sbi = UDF_SB(sb); struct udf_part_map *map; diff --git a/fs/udf/super.c b/fs/udf/super.c index 9fb18a340fc..7a5f69be6ac 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -1933,6 +1933,7 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) /* Fill in the rest of the superblock */ sb->s_op = &udf_sb_ops; + sb->s_export_op = &udf_export_ops; sb->dq_op = NULL; sb->s_dirt = 0; sb->s_magic = UDF_SUPER_MAGIC; diff --git a/fs/udf/udfdecl.h b/fs/udf/udfdecl.h index f3f45d02927..8fa9c2d7091 100644 --- a/fs/udf/udfdecl.h +++ b/fs/udf/udfdecl.h @@ -73,6 +73,7 @@ struct task_struct; struct buffer_head; struct super_block; +extern const struct export_operations udf_export_ops; extern const struct inode_operations udf_dir_inode_operations; extern const struct file_operations udf_dir_operations; extern const struct inode_operations udf_file_inode_operations; diff --git a/include/asm-alpha/types.h b/include/asm-alpha/types.h index a9e34ca4d46..c1541353cce 100644 --- a/include/asm-alpha/types.h +++ b/include/asm-alpha/types.h @@ -23,5 +23,11 @@ typedef unsigned int umode_t; #define BITS_PER_LONG 64 +#ifndef __ASSEMBLY__ + +typedef u64 dma_addr_t; +typedef u64 dma64_addr_t; + +#endif /* __ASSEMBLY__ */ #endif /* __KERNEL__ */ #endif /* _ALPHA_TYPES_H */ diff --git a/include/asm-m68k/machw.h b/include/asm-m68k/machw.h index d2e0e25d5c9..35624998291 100644 --- a/include/asm-m68k/machw.h +++ b/include/asm-m68k/machw.h @@ -66,36 +66,6 @@ struct MAC_SCC # define mac_scc ((*(volatile struct SCC*)MAC_SCC_BAS)) #endif -/* hardware stuff */ - -#define MACHW_DECLARE(name) unsigned name : 1 -#define MACHW_SET(name) (mac_hw_present.name = 1) -#define MACHW_PRESENT(name) (mac_hw_present.name) - -struct mac_hw_present { - /* video hardware */ - /* sound hardware */ - /* disk storage interfaces */ - MACHW_DECLARE(MAC_SCSI_80); /* Directly mapped NCR5380 */ - MACHW_DECLARE(MAC_SCSI_96); /* 53c9[46] */ - MACHW_DECLARE(MAC_SCSI_96_2); /* 2nd 53c9[46] Q900 and Q950 */ - MACHW_DECLARE(IDE); /* IDE Interface */ - /* other I/O hardware */ - MACHW_DECLARE(SCC); /* Serial Communications Contr. */ - /* DMA */ - MACHW_DECLARE(SCSI_DMA); /* DMA for the NCR5380 */ - /* real time clocks */ - MACHW_DECLARE(RTC_CLK); /* clock chip */ - /* supporting hardware */ - MACHW_DECLARE(VIA1); /* Versatile Interface Ad. 1 */ - MACHW_DECLARE(VIA2); /* Versatile Interface Ad. 2 */ - MACHW_DECLARE(RBV); /* Versatile Interface Ad. 2+ */ - /* NUBUS */ - MACHW_DECLARE(NUBUS); /* NUBUS */ -}; - -extern struct mac_hw_present mac_hw_present; - #endif /* __ASSEMBLY__ */ #endif /* linux/machw.h */ diff --git a/include/asm-mips/types.h b/include/asm-mips/types.h index 7a2ee4f4013..bcbb8d675af 100644 --- a/include/asm-mips/types.h +++ b/include/asm-mips/types.h @@ -19,8 +19,6 @@ typedef unsigned short umode_t; -#endif - #endif /* __ASSEMBLY__ */ /* diff --git a/include/asm-mn10300/processor.h b/include/asm-mn10300/processor.h index f1b081f5346..73239271873 100644 --- a/include/asm-mn10300/processor.h +++ b/include/asm-mn10300/processor.h @@ -58,7 +58,7 @@ extern struct mn10300_cpuinfo boot_cpu_data; extern void identify_cpu(struct mn10300_cpuinfo *); extern void print_cpu_info(struct mn10300_cpuinfo *); extern void dodgy_tsc(void); -#define cpu_relax() do {} while (0) +#define cpu_relax() barrier() /* * User space process size: 1.75GB (default). diff --git a/include/asm-powerpc/io.h b/include/asm-powerpc/io.h index afae0697e8c..e0062d73db1 100644 --- a/include/asm-powerpc/io.h +++ b/include/asm-powerpc/io.h @@ -2,7 +2,7 @@ #define _ASM_POWERPC_IO_H #ifdef __KERNEL__ -/* +/* * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License * as published by the Free Software Foundation; either version @@ -18,6 +18,9 @@ extern int check_legacy_ioport(unsigned long base_port); #define _PNPWRP 0xa79 #define PNPBIOS_BASE 0xf000 +#include <linux/device.h> +#include <linux/io.h> + #include <linux/compiler.h> #include <asm/page.h> #include <asm/byteorder.h> @@ -744,6 +747,9 @@ static inline void * bus_to_virt(unsigned long address) #define clrsetbits_8(addr, clear, set) clrsetbits(8, addr, clear, set) +void __iomem *devm_ioremap_prot(struct device *dev, resource_size_t offset, + size_t size, unsigned long flags); + #endif /* __KERNEL__ */ #endif /* _ASM_POWERPC_IO_H */ diff --git a/include/asm-powerpc/kvm_host.h b/include/asm-powerpc/kvm_host.h index 04ffbb8e0a3..81a69d71101 100644 --- a/include/asm-powerpc/kvm_host.h +++ b/include/asm-powerpc/kvm_host.h @@ -59,6 +59,7 @@ struct kvm_vcpu_stat { u32 emulated_inst_exits; u32 dec_exits; u32 ext_intr_exits; + u32 halt_wakeup; }; struct tlbe { diff --git a/include/asm-powerpc/kvm_ppc.h b/include/asm-powerpc/kvm_ppc.h index 7ac820308a7..b35a7e3ef97 100644 --- a/include/asm-powerpc/kvm_ppc.h +++ b/include/asm-powerpc/kvm_ppc.h @@ -77,12 +77,17 @@ static inline void kvmppc_clear_exception(struct kvm_vcpu *vcpu, int exception) clear_bit(priority, &vcpu->arch.pending_exceptions); } +/* Helper function for "full" MSR writes. No need to call this if only EE is + * changing. */ static inline void kvmppc_set_msr(struct kvm_vcpu *vcpu, u32 new_msr) { if ((new_msr & MSR_PR) != (vcpu->arch.msr & MSR_PR)) kvmppc_mmu_priv_switch(vcpu, new_msr & MSR_PR); vcpu->arch.msr = new_msr; + + if (vcpu->arch.msr & MSR_WE) + kvm_vcpu_block(vcpu); } #endif /* __POWERPC_KVM_PPC_H__ */ diff --git a/include/asm-powerpc/syscalls.h b/include/asm-powerpc/syscalls.h index b3ca41fc8bb..2b8a458f990 100644 --- a/include/asm-powerpc/syscalls.h +++ b/include/asm-powerpc/syscalls.h @@ -30,7 +30,7 @@ asmlinkage int sys_fork(unsigned long p1, unsigned long p2, asmlinkage int sys_vfork(unsigned long p1, unsigned long p2, unsigned long p3, unsigned long p4, unsigned long p5, unsigned long p6, struct pt_regs *regs); -asmlinkage int sys_pipe(int __user *fildes); +asmlinkage long sys_pipe(int __user *fildes); asmlinkage long sys_rt_sigaction(int sig, const struct sigaction __user *act, struct sigaction __user *oact, size_t sigsetsize); diff --git a/include/asm-s390/kvm_host.h b/include/asm-s390/kvm_host.h index f8204a4f2e0..18cbd8a3979 100644 --- a/include/asm-s390/kvm_host.h +++ b/include/asm-s390/kvm_host.h @@ -104,6 +104,7 @@ struct sie_block { struct kvm_vcpu_stat { u32 exit_userspace; + u32 exit_null; u32 exit_external_request; u32 exit_external_interrupt; u32 exit_stop_request; diff --git a/include/asm-s390/page.h b/include/asm-s390/page.h index f0f4579eac1..12fd9c4f0f1 100644 --- a/include/asm-s390/page.h +++ b/include/asm-s390/page.h @@ -125,6 +125,17 @@ page_get_storage_key(unsigned long addr) return skey; } +#ifdef CONFIG_PAGE_STATES + +struct page; +void arch_free_page(struct page *page, int order); +void arch_alloc_page(struct page *page, int order); + +#define HAVE_ARCH_FREE_PAGE +#define HAVE_ARCH_ALLOC_PAGE + +#endif + #endif /* !__ASSEMBLY__ */ /* to align the pointer to the (next) page boundary */ diff --git a/include/asm-s390/ptrace.h b/include/asm-s390/ptrace.h index 441d7c26085..d7d4e2eb3e6 100644 --- a/include/asm-s390/ptrace.h +++ b/include/asm-s390/ptrace.h @@ -471,6 +471,8 @@ struct task_struct; extern void user_enable_single_step(struct task_struct *); extern void user_disable_single_step(struct task_struct *); +#define __ARCH_WANT_COMPAT_SYS_PTRACE + #define user_mode(regs) (((regs)->psw.mask & PSW_MASK_PSTATE) != 0) #define instruction_pointer(regs) ((regs)->psw.addr & PSW_ADDR_INSN) #define regs_return_value(regs)((regs)->gprs[2]) diff --git a/include/asm-s390/system.h b/include/asm-s390/system.h index c819ae25a84..e0d4500d5f9 100644 --- a/include/asm-s390/system.h +++ b/include/asm-s390/system.h @@ -116,6 +116,12 @@ extern void pfault_fini(void); #define pfault_fini() do { } while (0) #endif /* CONFIG_PFAULT */ +#ifdef CONFIG_PAGE_STATES +extern void cmma_init(void); +#else +static inline void cmma_init(void) { } +#endif + #define finish_arch_switch(prev) do { \ set_fs(current->thread.mm_segment); \ account_vtime(prev); \ diff --git a/include/asm-x86/bootparam.h b/include/asm-x86/bootparam.h index e8659909e5f..f62f4733606 100644 --- a/include/asm-x86/bootparam.h +++ b/include/asm-x86/bootparam.h @@ -14,10 +14,10 @@ /* extensible setup data list node */ struct setup_data { - u64 next; - u32 type; - u32 len; - u8 data[0]; + __u64 next; + __u32 type; + __u32 len; + __u8 data[0]; }; struct setup_header { diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 9d963cd6533..1d8cd01fa51 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -314,6 +314,9 @@ struct kvm_arch{ struct page *apic_access_page; gpa_t wall_clock; + + struct page *ept_identity_pagetable; + bool ept_identity_pagetable_done; }; struct kvm_vm_stat { @@ -422,6 +425,7 @@ struct kvm_x86_ops { struct kvm_run *run); int (*set_tss_addr)(struct kvm *kvm, unsigned int addr); + int (*get_tdp_level)(void); }; extern struct kvm_x86_ops *kvm_x86_ops; @@ -433,6 +437,9 @@ void kvm_mmu_destroy(struct kvm_vcpu *vcpu); int kvm_mmu_create(struct kvm_vcpu *vcpu); int kvm_mmu_setup(struct kvm_vcpu *vcpu); void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte); +void kvm_mmu_set_base_ptes(u64 base_pte); +void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask, + u64 dirty_mask, u64 nx_mask, u64 x_mask); int kvm_mmu_reset_context(struct kvm_vcpu *vcpu); void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot); @@ -620,7 +627,7 @@ static inline void fx_restore(struct i387_fxsave_struct *image) asm("fxrstor (%0)":: "r" (image)); } -static inline void fpu_init(void) +static inline void fx_finit(void) { asm("finit"); } @@ -644,6 +651,7 @@ static inline void kvm_inject_gp(struct kvm_vcpu *vcpu, u32 error_code) #define ASM_VMX_VMWRITE_RSP_RDX ".byte 0x0f, 0x79, 0xd4" #define ASM_VMX_VMXOFF ".byte 0x0f, 0x01, 0xc4" #define ASM_VMX_VMXON_RAX ".byte 0xf3, 0x0f, 0xc7, 0x30" +#define ASM_VMX_INVEPT ".byte 0x66, 0x0f, 0x38, 0x80, 0x08" #define ASM_VMX_INVVPID ".byte 0x66, 0x0f, 0x38, 0x81, 0x08" #define MSR_IA32_TIME_STAMP_COUNTER 0x010 diff --git a/include/asm-x86/pgtable_32.h b/include/asm-x86/pgtable_32.h index 577ab79c4c2..d7f0403bbec 100644 --- a/include/asm-x86/pgtable_32.h +++ b/include/asm-x86/pgtable_32.h @@ -88,14 +88,7 @@ extern unsigned long pg0[]; /* To avoid harmful races, pmd_none(x) should check only the lower when PAE */ #define pmd_none(x) (!(unsigned long)pmd_val((x))) #define pmd_present(x) (pmd_val((x)) & _PAGE_PRESENT) - -extern int pmd_bad(pmd_t pmd); - -#define pmd_bad_v1(x) \ - (_KERNPG_TABLE != (pmd_val((x)) & ~(PAGE_MASK | _PAGE_USER))) -#define pmd_bad_v2(x) \ - (_KERNPG_TABLE != (pmd_val((x)) & ~(PAGE_MASK | _PAGE_USER | \ - _PAGE_PSE | _PAGE_NX))) +#define pmd_bad(x) ((pmd_val(x) & (~PAGE_MASK & ~_PAGE_USER)) != _KERNPG_TABLE) #define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT)) diff --git a/include/asm-x86/pgtable_64.h b/include/asm-x86/pgtable_64.h index a3bbf8766c1..efe83dcbd41 100644 --- a/include/asm-x86/pgtable_64.h +++ b/include/asm-x86/pgtable_64.h @@ -158,14 +158,12 @@ static inline unsigned long pgd_bad(pgd_t pgd) static inline unsigned long pud_bad(pud_t pud) { - return pud_val(pud) & - ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER | _PAGE_PSE | _PAGE_NX); + return pud_val(pud) & ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER); } static inline unsigned long pmd_bad(pmd_t pmd) { - return pmd_val(pmd) & - ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER | _PAGE_PSE | _PAGE_NX); + return pmd_val(pmd) & ~(PTE_MASK | _KERNPG_TABLE | _PAGE_USER); } #define pte_none(x) (!pte_val((x))) diff --git a/include/linux/exportfs.h b/include/linux/exportfs.h index de8387b7ceb..f5abd130663 100644 --- a/include/linux/exportfs.h +++ b/include/linux/exportfs.h @@ -33,6 +33,19 @@ enum fid_type { * 32 bit parent directory inode number. */ FILEID_INO32_GEN_PARENT = 2, + + /* + * 32 bit block number, 16 bit partition reference, + * 16 bit unused, 32 bit generation number. + */ + FILEID_UDF_WITHOUT_PARENT = 0x51, + + /* + * 32 bit block number, 16 bit partition reference, + * 16 bit unused, 32 bit generation number, + * 32 bit parent block number, 32 bit parent generation number + */ + FILEID_UDF_WITH_PARENT = 0x52, }; struct fid { @@ -43,6 +56,14 @@ struct fid { u32 parent_ino; u32 parent_gen; } i32; + struct { + u32 block; + u16 partref; + u16 parent_partref; + u32 generation; + u32 parent_block; + u32 parent_generation; + } udf; __u32 raw[0]; }; }; diff --git a/include/linux/fs.h b/include/linux/fs.h index a1ba005d08e..f413085f748 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1289,17 +1289,12 @@ extern ssize_t vfs_readv(struct file *, const struct iovec __user *, extern ssize_t vfs_writev(struct file *, const struct iovec __user *, unsigned long, loff_t *); -/* - * NOTE: write_inode, delete_inode, clear_inode, put_inode can be called - * without the big kernel lock held in all filesystems. - */ struct super_operations { struct inode *(*alloc_inode)(struct super_block *sb); void (*destroy_inode)(struct inode *); void (*dirty_inode) (struct inode *); int (*write_inode) (struct inode *, int); - void (*put_inode) (struct inode *); void (*drop_inode) (struct inode *); void (*delete_inode) (struct inode *); void (*put_super) (struct super_block *); @@ -1821,7 +1816,6 @@ extern void iget_failed(struct inode *); extern void clear_inode(struct inode *); extern void destroy_inode(struct inode *); extern struct inode *new_inode(struct super_block *); -extern int __remove_suid(struct dentry *, int); extern int should_remove_suid(struct dentry *); extern int remove_suid(struct dentry *); diff --git a/include/linux/genhd.h b/include/linux/genhd.h index ecd2bf63fc8..e9874e7fcdf 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -178,17 +178,17 @@ static inline struct hd_struct *get_part(struct gendisk *gendiskp, static inline void disk_stat_set_all(struct gendisk *gendiskp, int value) { int i; + for_each_possible_cpu(i) memset(per_cpu_ptr(gendiskp->dkstats, i), value, - sizeof (struct disk_stats)); + sizeof(struct disk_stats)); } #define __part_stat_add(part, field, addnd) \ (per_cpu_ptr(part->dkstats, smp_processor_id())->field += addnd) -#define __all_stat_add(gendiskp, field, addnd, sector) \ +#define __all_stat_add(gendiskp, part, field, addnd, sector) \ ({ \ - struct hd_struct *part = get_part(gendiskp, sector); \ if (part) \ __part_stat_add(part, field, addnd); \ __disk_stat_add(gendiskp, field, addnd); \ @@ -203,11 +203,13 @@ static inline void disk_stat_set_all(struct gendisk *gendiskp, int value) { res; \ }) -static inline void part_stat_set_all(struct hd_struct *part, int value) { +static inline void part_stat_set_all(struct hd_struct *part, int value) +{ int i; + for_each_possible_cpu(i) memset(per_cpu_ptr(part->dkstats, i), value, - sizeof(struct disk_stats)); + sizeof(struct disk_stats)); } #else /* !CONFIG_SMP */ @@ -223,9 +225,8 @@ static inline void disk_stat_set_all(struct gendisk *gendiskp, int value) #define __part_stat_add(part, field, addnd) \ (part->dkstats.field += addnd) -#define __all_stat_add(gendiskp, field, addnd, sector) \ +#define __all_stat_add(gendiskp, part, field, addnd, sector) \ ({ \ - struct hd_struct *part = get_part(gendiskp, sector); \ if (part) \ part->dkstats.field += addnd; \ __disk_stat_add(gendiskp, field, addnd); \ @@ -276,10 +277,10 @@ static inline void part_stat_set_all(struct hd_struct *part, int value) #define part_stat_sub(gendiskp, field, subnd) \ part_stat_add(gendiskp, field, -subnd) -#define all_stat_add(gendiskp, field, addnd, sector) \ +#define all_stat_add(gendiskp, part, field, addnd, sector) \ do { \ preempt_disable(); \ - __all_stat_add(gendiskp, field, addnd, sector); \ + __all_stat_add(gendiskp, part, field, addnd, sector); \ preempt_enable(); \ } while (0) @@ -288,15 +289,15 @@ static inline void part_stat_set_all(struct hd_struct *part, int value) #define all_stat_dec(gendiskp, field, sector) \ all_stat_add(gendiskp, field, -1, sector) -#define __all_stat_inc(gendiskp, field, sector) \ - __all_stat_add(gendiskp, field, 1, sector) -#define all_stat_inc(gendiskp, field, sector) \ - all_stat_add(gendiskp, field, 1, sector) +#define __all_stat_inc(gendiskp, part, field, sector) \ + __all_stat_add(gendiskp, part, field, 1, sector) +#define all_stat_inc(gendiskp, part, field, sector) \ + all_stat_add(gendiskp, part, field, 1, sector) -#define __all_stat_sub(gendiskp, field, subnd, sector) \ - __all_stat_add(gendiskp, field, -subnd, sector) -#define all_stat_sub(gendiskp, field, subnd, sector) \ - all_stat_add(gendiskp, field, -subnd, sector) +#define __all_stat_sub(gendiskp, part, field, subnd, sector) \ + __all_stat_add(gendiskp, part, field, -subnd, sector) +#define all_stat_sub(gendiskp, part, field, subnd, sector) \ + all_stat_add(gendiskp, part, field, -subnd, sector) /* Inlines to alloc and free disk stats in struct gendisk */ #ifdef CONFIG_SMP diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h index 31a4d653389..6d93dce61cb 100644 --- a/include/linux/hrtimer.h +++ b/include/linux/hrtimer.h @@ -316,6 +316,15 @@ static inline int hrtimer_is_queued(struct hrtimer *timer) (HRTIMER_STATE_ENQUEUED | HRTIMER_STATE_PENDING); } +/* + * Helper function to check, whether the timer is running the callback + * function + */ +static inline int hrtimer_callback_running(struct hrtimer *timer) +{ + return timer->state & HRTIMER_STATE_CALLBACK; +} + /* Forward a hrtimer so it expires after now: */ extern u64 hrtimer_forward(struct hrtimer *timer, ktime_t now, ktime_t interval); diff --git a/include/linux/io.h b/include/linux/io.h index 3a03a3604cc..6c7f0ba0d5f 100644 --- a/include/linux/io.h +++ b/include/linux/io.h @@ -65,5 +65,6 @@ void __iomem *devm_ioremap_nocache(struct device *dev, resource_size_t offset, void devm_iounmap(struct device *dev, void __iomem *addr); int check_signature(const volatile void __iomem *io_addr, const unsigned char *signature, int length); +void devm_ioremap_release(struct device *dev, void *res); #endif /* _LINUX_IO_H */ diff --git a/include/linux/ioprio.h b/include/linux/ioprio.h index 2a3bb1bb743..f98a656b17e 100644 --- a/include/linux/ioprio.h +++ b/include/linux/ioprio.h @@ -68,6 +68,20 @@ static inline int task_nice_ioprio(struct task_struct *task) } /* + * This is for the case where the task hasn't asked for a specific IO class. + * Check for idle and rt task process, and return appropriate IO class. + */ +static inline int task_nice_ioclass(struct task_struct *task) +{ + if (task->policy == SCHED_IDLE) + return IOPRIO_CLASS_IDLE; + else if (task->policy == SCHED_FIFO || task->policy == SCHED_RR) + return IOPRIO_CLASS_RT; + else + return IOPRIO_CLASS_BE; +} + +/* * For inheritance, return the highest of the two given priorities */ extern int ioprio_best(unsigned short aprio, unsigned short bprio); diff --git a/include/linux/kgdb.h b/include/linux/kgdb.h index 9757b1a6d9d..6adcc297e35 100644 --- a/include/linux/kgdb.h +++ b/include/linux/kgdb.h @@ -261,10 +261,12 @@ struct kgdb_io { extern struct kgdb_arch arch_kgdb_ops; +extern unsigned long __weak kgdb_arch_pc(int exception, struct pt_regs *regs); + extern int kgdb_register_io_module(struct kgdb_io *local_kgdb_io_ops); extern void kgdb_unregister_io_module(struct kgdb_io *local_kgdb_io_ops); -extern int kgdb_hex2long(char **ptr, long *long_val); +extern int kgdb_hex2long(char **ptr, unsigned long *long_val); extern int kgdb_mem2hex(char *mem, char *buf, int count); extern int kgdb_hex2mem(char *buf, char *mem, int count); diff --git a/include/linux/libata.h b/include/linux/libata.h index d1dfe872ee3..7e206da1fbf 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -1039,6 +1039,7 @@ extern void ata_eh_thaw_port(struct ata_port *ap); extern void ata_eh_qc_complete(struct ata_queued_cmd *qc); extern void ata_eh_qc_retry(struct ata_queued_cmd *qc); +extern void ata_eh_analyze_ncq_error(struct ata_link *link); extern void ata_do_eh(struct ata_port *ap, ata_prereset_fn_t prereset, ata_reset_fn_t softreset, ata_reset_fn_t hardreset, @@ -1381,6 +1382,21 @@ static inline struct ata_port *ata_shost_to_port(struct Scsi_Host *host) return *(struct ata_port **)&host->hostdata[0]; } +static inline int ata_check_ready(u8 status) +{ + /* Some controllers report 0x77 or 0x7f during intermediate + * not-ready stages. + */ + if (status == 0x77 || status == 0x7f) + return 0; + + /* 0xff indicates either no device or device not ready */ + if (status == 0xff) + return -ENODEV; + + return !(status & ATA_BUSY); +} + /************************************************************************** * PMP - drivers/ata/libata-pmp.c diff --git a/include/linux/mv643xx_eth.h b/include/linux/mv643xx_eth.h index 30e11aa3c1c..a15cdd4a8e5 100644 --- a/include/linux/mv643xx_eth.h +++ b/include/linux/mv643xx_eth.h @@ -1,19 +1,31 @@ /* * MV-643XX ethernet platform device data definition file. */ + #ifndef __LINUX_MV643XX_ETH_H #define __LINUX_MV643XX_ETH_H -#define MV643XX_ETH_SHARED_NAME "mv643xx_eth_shared" -#define MV643XX_ETH_NAME "mv643xx_eth" +#include <linux/mbus.h> + +#define MV643XX_ETH_SHARED_NAME "mv643xx_eth" +#define MV643XX_ETH_NAME "mv643xx_eth_port" #define MV643XX_ETH_SHARED_REGS 0x2000 #define MV643XX_ETH_SHARED_REGS_SIZE 0x2000 #define MV643XX_ETH_BAR_4 0x2220 #define MV643XX_ETH_SIZE_REG_4 0x2224 #define MV643XX_ETH_BASE_ADDR_ENABLE_REG 0x2290 +struct mv643xx_eth_shared_platform_data { + struct mbus_dram_target_info *dram; + unsigned int t_clk; +}; + struct mv643xx_eth_platform_data { + struct platform_device *shared; int port_number; + + struct platform_device *shared_smi; + u16 force_phy_addr; /* force override if phy_addr == 0 */ u16 phy_addr; diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 5da04e586a3..23aa2ec6b7b 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -7,6 +7,7 @@ struct nf_ct_sip_master { unsigned int register_cseq; + unsigned int invite_cseq; }; enum sip_expectation_classes { diff --git a/include/linux/pci.h b/include/linux/pci.h index 96acd0dae24..509159bcd4e 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -44,6 +44,7 @@ #include <linux/mod_devicetable.h> #include <linux/types.h> +#include <linux/init.h> #include <linux/ioport.h> #include <linux/list.h> #include <linux/compiler.h> @@ -474,7 +475,7 @@ extern struct pci_bus *pci_find_bus(int domain, int busnr); void pci_bus_add_devices(struct pci_bus *bus); struct pci_bus *pci_scan_bus_parented(struct device *parent, int bus, struct pci_ops *ops, void *sysdata); -static inline struct pci_bus *pci_scan_bus(int bus, struct pci_ops *ops, +static inline struct pci_bus * __devinit pci_scan_bus(int bus, struct pci_ops *ops, void *sysdata) { struct pci_bus *root_bus; @@ -666,7 +667,7 @@ int pci_scan_bridge(struct pci_bus *bus, struct pci_dev *dev, int max, void pci_walk_bus(struct pci_bus *top, void (*cb)(struct pci_dev *, void *), void *userdata); -int pci_cfg_space_size_ext(struct pci_dev *dev, unsigned check_exp_pcix); +int pci_cfg_space_size_ext(struct pci_dev *dev); int pci_cfg_space_size(struct pci_dev *dev); unsigned char pci_bus_max_busnr(struct pci_bus *bus); diff --git a/include/linux/phy.h b/include/linux/phy.h index 02df20f085f..7224c4099a2 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -412,6 +412,8 @@ int mdiobus_register(struct mii_bus *bus); void mdiobus_unregister(struct mii_bus *bus); void phy_sanitize_settings(struct phy_device *phydev); int phy_stop_interrupts(struct phy_device *phydev); +int phy_enable_interrupts(struct phy_device *phydev); +int phy_disable_interrupts(struct phy_device *phydev); static inline int phy_read_status(struct phy_device *phydev) { return phydev->drv->read_status(phydev); @@ -447,5 +449,8 @@ int phy_register_fixup_for_uid(u32 phy_uid, u32 phy_uid_mask, int (*run)(struct phy_device *)); int phy_scan_fixups(struct phy_device *phydev); +int __init mdio_bus_init(void); +void mdio_bus_exit(void); + extern struct bus_type mdio_bus_type; #endif /* __PHY_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 03c238088ae..0c35b0343a7 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -158,6 +158,8 @@ print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) } #endif +extern unsigned long long time_sync_thresh; + /* * Task state bitmask. NOTE! These bits are also * encoded in fs/proc/array.c: get_task_state(). @@ -1551,6 +1553,35 @@ static inline int set_cpus_allowed(struct task_struct *p, cpumask_t new_mask) extern unsigned long long sched_clock(void); +#ifndef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK +static inline void sched_clock_init(void) +{ +} + +static inline u64 sched_clock_cpu(int cpu) +{ + return sched_clock(); +} + +static inline void sched_clock_tick(void) +{ +} + +static inline void sched_clock_idle_sleep_event(void) +{ +} + +static inline void sched_clock_idle_wakeup_event(u64 delta_ns) +{ +} +#else +extern void sched_clock_init(void); +extern u64 sched_clock_cpu(int cpu); +extern void sched_clock_tick(void); +extern void sched_clock_idle_sleep_event(void); +extern void sched_clock_idle_wakeup_event(u64 delta_ns); +#endif + /* * For kernel-internal use: high-speed (but slightly incorrect) per-cpu * clock constructed from sched_clock(): @@ -1977,6 +2008,11 @@ static inline void clear_tsk_need_resched(struct task_struct *tsk) clear_tsk_thread_flag(tsk,TIF_NEED_RESCHED); } +static inline int test_tsk_need_resched(struct task_struct *tsk) +{ + return unlikely(test_tsk_thread_flag(tsk,TIF_NEED_RESCHED)); +} + static inline int signal_pending(struct task_struct *p) { return unlikely(test_tsk_thread_flag(p,TIF_SIGPENDING)); @@ -1991,7 +2027,7 @@ static inline int fatal_signal_pending(struct task_struct *p) static inline int need_resched(void) { - return unlikely(test_thread_flag(TIF_NEED_RESCHED)); + return unlikely(test_tsk_need_resched(current)); } /* diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h index 27bad59dae7..7858eac40aa 100644 --- a/include/linux/sysfs.h +++ b/include/linux/sysfs.h @@ -196,12 +196,6 @@ static inline int sysfs_update_group(struct kobject *kobj, return 0; } -static inline int sysfs_update_group(struct kobject *kobj, - const struct attribute_group *grp) -{ - return 0; -} - static inline void sysfs_remove_group(struct kobject *kobj, const struct attribute_group *grp) { diff --git a/include/net/ip.h b/include/net/ip.h index 6d7bcd5e62d..3b40bc2234b 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -210,7 +210,7 @@ int ip_dont_fragment(struct sock *sk, struct dst_entry *dst) { return (inet_sk(sk)->pmtudisc == IP_PMTUDISC_DO || (inet_sk(sk)->pmtudisc == IP_PMTUDISC_WANT && - !(dst_metric(dst, RTAX_LOCK)&(1<<RTAX_MTU)))); + !(dst_metric_locked(dst, RTAX_MTU)))); } extern void __ip_select_ident(struct iphdr *iph, struct dst_entry *dst, int more); diff --git a/include/net/xfrm.h b/include/net/xfrm.h index d1350bcccb0..2933d7474a7 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -648,14 +648,46 @@ extern void xfrm_audit_state_notfound(struct sk_buff *skb, u16 family, extern void xfrm_audit_state_icvfail(struct xfrm_state *x, struct sk_buff *skb, u8 proto); #else -#define xfrm_audit_policy_add(x, r, a, se, s) do { ; } while (0) -#define xfrm_audit_policy_delete(x, r, a, se, s) do { ; } while (0) -#define xfrm_audit_state_add(x, r, a, se, s) do { ; } while (0) -#define xfrm_audit_state_delete(x, r, a, se, s) do { ; } while (0) -#define xfrm_audit_state_replay_overflow(x, s) do { ; } while (0) -#define xfrm_audit_state_notfound_simple(s, f) do { ; } while (0) -#define xfrm_audit_state_notfound(s, f, sp, sq) do { ; } while (0) -#define xfrm_audit_state_icvfail(x, s, p) do { ; } while (0) + +static inline void xfrm_audit_policy_add(struct xfrm_policy *xp, int result, + u32 auid, u32 ses, u32 secid) +{ +} + +static inline void xfrm_audit_policy_delete(struct xfrm_policy *xp, int result, + u32 auid, u32 ses, u32 secid) +{ +} + +static inline void xfrm_audit_state_add(struct xfrm_state *x, int result, + u32 auid, u32 ses, u32 secid) +{ +} + +static inline void xfrm_audit_state_delete(struct xfrm_state *x, int result, + u32 auid, u32 ses, u32 secid) +{ +} + +static inline void xfrm_audit_state_replay_overflow(struct xfrm_state *x, + struct sk_buff *skb) +{ +} + +static inline void xfrm_audit_state_notfound_simple(struct sk_buff *skb, + u16 family) +{ +} + +static inline void xfrm_audit_state_notfound(struct sk_buff *skb, u16 family, + __be32 net_spi, __be32 net_seq) +{ +} + +static inline void xfrm_audit_state_icvfail(struct xfrm_state *x, + struct sk_buff *skb, u8 proto) +{ +} #endif /* CONFIG_AUDITSYSCALL */ static inline void xfrm_pol_hold(struct xfrm_policy *policy) diff --git a/init/Kconfig b/init/Kconfig index 6a44defac3e..3b5adbf228c 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -316,9 +316,16 @@ config CPUSETS Say N if unsure. +# +# Architectures with an unreliable sched_clock() should select this: +# +config HAVE_UNSTABLE_SCHED_CLOCK + bool + config GROUP_SCHED bool "Group CPU scheduler" - default y + depends on EXPERIMENTAL + default n help This feature lets CPU scheduler recognize task groups and control CPU bandwidth allocation to such task groups. @@ -326,7 +333,7 @@ config GROUP_SCHED config FAIR_GROUP_SCHED bool "Group scheduling for SCHED_OTHER" depends on GROUP_SCHED - default y + default GROUP_SCHED config RT_GROUP_SCHED bool "Group scheduling for SCHED_RR/FIFO" @@ -627,6 +634,14 @@ config ELF_CORE help Enable support for generating core dumps. Disabling saves about 4k. +config PCSPKR_PLATFORM + bool "Enable PC-Speaker support" if EMBEDDED + depends on ALPHA || X86 || MIPS || PPC_PREP || PPC_CHRP || PPC_PSERIES + default y + help + This option allows to disable the internal PC-Speaker + support, saving some memory. + config COMPAT_BRK bool "Disable heap randomization" default y @@ -825,6 +840,15 @@ menuconfig MODULES If unsure, say Y. +config MODULE_FORCE_LOAD + bool "Forced module loading" + depends on MODULES + default n + help + This option allows loading of modules even if that would set the + 'F' (forced) taint, due to lack of version info. Which is + usually a really bad idea. + config MODULE_UNLOAD bool "Module unloading" depends on MODULES diff --git a/init/main.c b/init/main.c index a87d4ca5c36..ddada7acf36 100644 --- a/init/main.c +++ b/init/main.c @@ -602,6 +602,7 @@ asmlinkage void __init start_kernel(void) softirq_init(); timekeeping_init(); time_init(); + sched_clock_init(); profile_init(); if (!irqs_disabled()) printk("start_kernel(): bug: interrupts were enabled early\n"); diff --git a/ipc/mqueue.c b/ipc/mqueue.c index 94fd3b08fb7..b3b69fd5133 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -673,7 +673,7 @@ asmlinkage long sys_mq_open(const char __user *u_name, int oflag, mode_t mode, if (IS_ERR(name = getname(u_name))) return PTR_ERR(name); - fd = get_unused_fd(); + fd = get_unused_fd_flags(O_CLOEXEC); if (fd < 0) goto out_putname; @@ -709,7 +709,6 @@ asmlinkage long sys_mq_open(const char __user *u_name, int oflag, mode_t mode, goto out_putfd; } - set_close_on_exec(fd, 1); fd_install(fd, filp); goto out_upsem; diff --git a/kernel/Makefile b/kernel/Makefile index 188c43223f5..1c9938addb9 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -9,7 +9,7 @@ obj-y = sched.o fork.o exec_domain.o panic.o printk.o profile.o \ rcupdate.o extable.o params.o posix-timers.o \ kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \ hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \ - notifier.o ksysfs.o pm_qos_params.o + notifier.o ksysfs.o pm_qos_params.o sched_clock.o obj-$(CONFIG_SYSCTL_SYSCALL_CHECK) += sysctl_check.o obj-$(CONFIG_STACKTRACE) += stacktrace.o diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 8da627d3380..86ea9e34e32 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -1031,11 +1031,9 @@ int current_cpuset_is_being_rebound(void) return task_cs(current) == cpuset_being_rebound; } -static int update_relax_domain_level(struct cpuset *cs, char *buf) +static int update_relax_domain_level(struct cpuset *cs, s64 val) { - int val = simple_strtol(buf, NULL, 10); - - if (val < 0) + if ((int)val < 0) val = -1; if (val != cs->relax_domain_level) { @@ -1280,9 +1278,6 @@ static ssize_t cpuset_common_file_write(struct cgroup *cont, case FILE_MEMLIST: retval = update_nodemask(cs, buffer); break; - case FILE_SCHED_RELAX_DOMAIN_LEVEL: - retval = update_relax_domain_level(cs, buffer); - break; default: retval = -EINVAL; goto out2; @@ -1348,6 +1343,30 @@ static int cpuset_write_u64(struct cgroup *cgrp, struct cftype *cft, u64 val) return retval; } +static int cpuset_write_s64(struct cgroup *cgrp, struct cftype *cft, s64 val) +{ + int retval = 0; + struct cpuset *cs = cgroup_cs(cgrp); + cpuset_filetype_t type = cft->private; + + cgroup_lock(); + + if (cgroup_is_removed(cgrp)) { + cgroup_unlock(); + return -ENODEV; + } + switch (type) { + case FILE_SCHED_RELAX_DOMAIN_LEVEL: + retval = update_relax_domain_level(cs, val); + break; + default: + retval = -EINVAL; + break; + } + cgroup_unlock(); + return retval; +} + /* * These ascii lists should be read in a single call, by using a user * buffer large enough to hold the entire map. If read in smaller @@ -1406,9 +1425,6 @@ static ssize_t cpuset_common_file_read(struct cgroup *cont, case FILE_MEMLIST: s += cpuset_sprintf_memlist(s, cs); break; - case FILE_SCHED_RELAX_DOMAIN_LEVEL: - s += sprintf(s, "%d", cs->relax_domain_level); - break; default: retval = -EINVAL; goto out; @@ -1449,6 +1465,18 @@ static u64 cpuset_read_u64(struct cgroup *cont, struct cftype *cft) } } +static s64 cpuset_read_s64(struct cgroup *cont, struct cftype *cft) +{ + struct cpuset *cs = cgroup_cs(cont); + cpuset_filetype_t type = cft->private; + switch (type) { + case FILE_SCHED_RELAX_DOMAIN_LEVEL: + return cs->relax_domain_level; + default: + BUG(); + } +} + /* * for the common functions, 'private' gives the type of file @@ -1499,8 +1527,8 @@ static struct cftype files[] = { { .name = "sched_relax_domain_level", - .read_u64 = cpuset_read_u64, - .write_u64 = cpuset_write_u64, + .read_s64 = cpuset_read_s64, + .write_s64 = cpuset_write_s64, .private = FILE_SCHED_RELAX_DOMAIN_LEVEL, }, diff --git a/kernel/futex.c b/kernel/futex.c index 98092c9817f..449def8074f 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -104,10 +104,6 @@ struct futex_q { /* Key which the futex is hashed on: */ union futex_key key; - /* For fd, sigio sent using these: */ - int fd; - struct file *filp; - /* Optional priority inheritance state: */ struct futex_pi_state *pi_state; struct task_struct *task; @@ -126,9 +122,6 @@ struct futex_hash_bucket { static struct futex_hash_bucket futex_queues[1<<FUTEX_HASHBITS]; -/* Futex-fs vfsmount entry: */ -static struct vfsmount *futex_mnt; - /* * Take mm->mmap_sem, when futex is shared */ @@ -610,8 +603,6 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, static void wake_futex(struct futex_q *q) { plist_del(&q->list, &q->list.plist); - if (q->filp) - send_sigio(&q->filp->f_owner, q->fd, POLL_IN); /* * The lock in wake_up_all() is a crucial memory barrier after the * plist_del() and also before assigning to q->lock_ptr. @@ -988,14 +979,10 @@ out: } /* The key must be already stored in q->key. */ -static inline struct futex_hash_bucket * -queue_lock(struct futex_q *q, int fd, struct file *filp) +static inline struct futex_hash_bucket *queue_lock(struct futex_q *q) { struct futex_hash_bucket *hb; - q->fd = fd; - q->filp = filp; - init_waitqueue_head(&q->waiters); get_futex_key_refs(&q->key); @@ -1006,7 +993,7 @@ queue_lock(struct futex_q *q, int fd, struct file *filp) return hb; } -static inline void __queue_me(struct futex_q *q, struct futex_hash_bucket *hb) +static inline void queue_me(struct futex_q *q, struct futex_hash_bucket *hb) { int prio; @@ -1041,15 +1028,6 @@ queue_unlock(struct futex_q *q, struct futex_hash_bucket *hb) * exactly once. They are called with the hashed spinlock held. */ -/* The key must be already stored in q->key. */ -static void queue_me(struct futex_q *q, int fd, struct file *filp) -{ - struct futex_hash_bucket *hb; - - hb = queue_lock(q, fd, filp); - __queue_me(q, hb); -} - /* Return 1 if we were still queued (ie. 0 means we were woken) */ static int unqueue_me(struct futex_q *q) { @@ -1194,7 +1172,7 @@ static int futex_wait(u32 __user *uaddr, struct rw_semaphore *fshared, if (unlikely(ret != 0)) goto out_release_sem; - hb = queue_lock(&q, -1, NULL); + hb = queue_lock(&q); /* * Access the page AFTER the futex is queued. @@ -1238,7 +1216,7 @@ static int futex_wait(u32 __user *uaddr, struct rw_semaphore *fshared, goto out_unlock_release_sem; /* Only actually queue if *uaddr contained val. */ - __queue_me(&q, hb); + queue_me(&q, hb); /* * Now the futex is queued and we have checked the data, we @@ -1386,7 +1364,7 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, goto out_release_sem; retry_unlocked: - hb = queue_lock(&q, -1, NULL); + hb = queue_lock(&q); retry_locked: ret = lock_taken = 0; @@ -1499,7 +1477,7 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, /* * Only actually queue now that the atomic ops are done: */ - __queue_me(&q, hb); + queue_me(&q, hb); /* * Now the futex is queued and we have checked the data, we @@ -1746,121 +1724,6 @@ pi_faulted: return ret; } -static int futex_close(struct inode *inode, struct file *filp) -{ - struct futex_q *q = filp->private_data; - - unqueue_me(q); - kfree(q); - - return 0; -} - -/* This is one-shot: once it's gone off you need a new fd */ -static unsigned int futex_poll(struct file *filp, - struct poll_table_struct *wait) -{ - struct futex_q *q = filp->private_data; - int ret = 0; - - poll_wait(filp, &q->waiters, wait); - - /* - * plist_node_empty() is safe here without any lock. - * q->lock_ptr != 0 is not safe, because of ordering against wakeup. - */ - if (plist_node_empty(&q->list)) - ret = POLLIN | POLLRDNORM; - - return ret; -} - -static const struct file_operations futex_fops = { - .release = futex_close, - .poll = futex_poll, -}; - -/* - * Signal allows caller to avoid the race which would occur if they - * set the sigio stuff up afterwards. - */ -static int futex_fd(u32 __user *uaddr, int signal) -{ - struct futex_q *q; - struct file *filp; - int ret, err; - struct rw_semaphore *fshared; - static unsigned long printk_interval; - - if (printk_timed_ratelimit(&printk_interval, 60 * 60 * 1000)) { - printk(KERN_WARNING "Process `%s' used FUTEX_FD, which " - "will be removed from the kernel in June 2007\n", - current->comm); - } - - ret = -EINVAL; - if (!valid_signal(signal)) - goto out; - - ret = get_unused_fd(); - if (ret < 0) - goto out; - filp = get_empty_filp(); - if (!filp) { - put_unused_fd(ret); - ret = -ENFILE; - goto out; - } - filp->f_op = &futex_fops; - filp->f_path.mnt = mntget(futex_mnt); - filp->f_path.dentry = dget(futex_mnt->mnt_root); - filp->f_mapping = filp->f_path.dentry->d_inode->i_mapping; - - if (signal) { - err = __f_setown(filp, task_pid(current), PIDTYPE_PID, 1); - if (err < 0) { - goto error; - } - filp->f_owner.signum = signal; - } - - q = kmalloc(sizeof(*q), GFP_KERNEL); - if (!q) { - err = -ENOMEM; - goto error; - } - q->pi_state = NULL; - - fshared = ¤t->mm->mmap_sem; - down_read(fshared); - err = get_futex_key(uaddr, fshared, &q->key); - - if (unlikely(err != 0)) { - up_read(fshared); - kfree(q); - goto error; - } - - /* - * queue_me() must be called before releasing mmap_sem, because - * key->shared.inode needs to be referenced while holding it. - */ - filp->private_data = q; - - queue_me(q, ret, filp); - up_read(fshared); - - /* Now we map fd to filp, so userspace can access it */ - fd_install(ret, filp); -out: - return ret; -error: - put_unused_fd(ret); - put_filp(filp); - ret = err; - goto out; -} - /* * Support for robust futexes: the kernel cleans up held futexes at * thread exit time. @@ -2092,10 +1955,6 @@ long do_futex(u32 __user *uaddr, int op, u32 val, ktime_t *timeout, case FUTEX_WAKE_BITSET: ret = futex_wake(uaddr, fshared, val, val3); break; - case FUTEX_FD: - /* non-zero val means F_SETOWN(getpid()) & F_SETSIG(val) */ - ret = futex_fd(uaddr, val); - break; case FUTEX_REQUEUE: ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, NULL); break; @@ -2156,19 +2015,6 @@ asmlinkage long sys_futex(u32 __user *uaddr, int op, u32 val, return do_futex(uaddr, op, val, tp, uaddr2, val2, val3); } -static int futexfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, - struct vfsmount *mnt) -{ - return get_sb_pseudo(fs_type, "futex", NULL, FUTEXFS_SUPER_MAGIC, mnt); -} - -static struct file_system_type futex_fs_type = { - .name = "futexfs", - .get_sb = futexfs_get_sb, - .kill_sb = kill_anon_super, -}; - static int __init futex_init(void) { u32 curval; @@ -2193,16 +2039,6 @@ static int __init futex_init(void) spin_lock_init(&futex_queues[i].lock); } - i = register_filesystem(&futex_fs_type); - if (i) - return i; - - futex_mnt = kern_mount(&futex_fs_type); - if (IS_ERR(futex_mnt)) { - unregister_filesystem(&futex_fs_type); - return PTR_ERR(futex_mnt); - } - return 0; } __initcall(futex_init); diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index 9af1d6a8095..421be5fe5cc 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -154,15 +154,6 @@ static void hrtimer_get_softirq_time(struct hrtimer_cpu_base *base) } /* - * Helper function to check, whether the timer is running the callback - * function - */ -static inline int hrtimer_callback_running(struct hrtimer *timer) -{ - return timer->state & HRTIMER_STATE_CALLBACK; -} - -/* * Functions and macros which are different for UP/SMP systems are kept in a * single place */ diff --git a/kernel/kgdb.c b/kernel/kgdb.c index 1bd0ec1c80b..39e31a036f5 100644 --- a/kernel/kgdb.c +++ b/kernel/kgdb.c @@ -61,7 +61,7 @@ struct kgdb_state { int err_code; int cpu; int pass_exception; - long threadid; + unsigned long threadid; long kgdb_usethreadid; struct pt_regs *linux_regs; }; @@ -146,7 +146,7 @@ atomic_t kgdb_cpu_doing_single_step = ATOMIC_INIT(-1); * the other CPUs might interfere with your debugging context, so * use this with care: */ -int kgdb_do_roundup = 1; +static int kgdb_do_roundup = 1; static int __init opt_nokgdbroundup(char *str) { @@ -438,7 +438,7 @@ int kgdb_hex2mem(char *buf, char *mem, int count) * While we find nice hex chars, build a long_val. * Return number of chars processed. */ -int kgdb_hex2long(char **ptr, long *long_val) +int kgdb_hex2long(char **ptr, unsigned long *long_val) { int hex_val; int num = 0; @@ -709,7 +709,7 @@ int kgdb_isremovedbreak(unsigned long addr) return 0; } -int remove_all_break(void) +static int remove_all_break(void) { unsigned long addr; int error; diff --git a/kernel/module.c b/kernel/module.c index 8674a390a2e..8e4528c9909 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -890,6 +890,19 @@ static struct module_attribute *modinfo_attrs[] = { static const char vermagic[] = VERMAGIC_STRING; +static int try_to_force_load(struct module *mod, const char *symname) +{ +#ifdef CONFIG_MODULE_FORCE_LOAD + if (!(tainted & TAINT_FORCED_MODULE)) + printk("%s: no version for \"%s\" found: kernel tainted.\n", + mod->name, symname); + add_taint_module(mod, TAINT_FORCED_MODULE); + return 0; +#else + return -ENOEXEC; +#endif +} + #ifdef CONFIG_MODVERSIONS static int check_version(Elf_Shdr *sechdrs, unsigned int versindex, @@ -914,18 +927,18 @@ static int check_version(Elf_Shdr *sechdrs, if (versions[i].crc == *crc) return 1; - printk("%s: disagrees about version of symbol %s\n", - mod->name, symname); DEBUGP("Found checksum %lX vs module %lX\n", *crc, versions[i].crc); - return 0; + goto bad_version; } - /* Not in module's version table. OK, but that taints the kernel. */ - if (!(tainted & TAINT_FORCED_MODULE)) - printk("%s: no version for \"%s\" found: kernel tainted.\n", - mod->name, symname); - add_taint_module(mod, TAINT_FORCED_MODULE); - return 1; + + if (!try_to_force_load(mod, symname)) + return 1; + +bad_version: + printk("%s: disagrees about version of symbol %s\n", + mod->name, symname); + return 0; } static inline int check_modstruct_version(Elf_Shdr *sechdrs, @@ -1853,9 +1866,9 @@ static struct module *load_module(void __user *umod, modmagic = get_modinfo(sechdrs, infoindex, "vermagic"); /* This is allowed: modprobe --force will invalidate it. */ if (!modmagic) { - add_taint_module(mod, TAINT_FORCED_MODULE); - printk(KERN_WARNING "%s: no version magic, tainting kernel.\n", - mod->name); + err = try_to_force_load(mod, "magic"); + if (err) + goto free_hdr; } else if (!same_magic(modmagic, vermagic)) { printk(KERN_ERR "%s: version magic '%s' should be '%s'\n", mod->name, modmagic, vermagic); @@ -2006,9 +2019,10 @@ static struct module *load_module(void __user *umod, (mod->num_gpl_future_syms && !gplfuturecrcindex) || (mod->num_unused_syms && !unusedcrcindex) || (mod->num_unused_gpl_syms && !unusedgplcrcindex)) { - printk(KERN_WARNING "%s: No versions for exported symbols." - " Tainting kernel.\n", mod->name); - add_taint_module(mod, TAINT_FORCED_MODULE); + printk(KERN_WARNING "%s: No versions for exported symbols.\n", mod->name); + err = try_to_force_load(mod, "nocrc"); + if (err) + goto cleanup; } #endif markersindex = find_sec(hdr, sechdrs, secstrings, "__markers"); diff --git a/kernel/relay.c b/kernel/relay.c index 7de644cdec4..bc24dcdc570 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -1191,7 +1191,7 @@ static ssize_t relay_file_splice_read(struct file *in, ret = 0; spliced = 0; - while (len && !spliced) { + while (len) { ret = subbuf_splice_actor(in, ppos, pipe, len, flags, &nonpad_ret); if (ret < 0) break; diff --git a/kernel/sched.c b/kernel/sched.c index 34bcc5bc120..58fb8af1577 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -75,16 +75,6 @@ #include <asm/irq_regs.h> /* - * Scheduler clock - returns current time in nanosec units. - * This is default implementation. - * Architectures and sub-architectures can override this. - */ -unsigned long long __attribute__((weak)) sched_clock(void) -{ - return (unsigned long long)jiffies * (NSEC_PER_SEC / HZ); -} - -/* * Convert user-nice values [ -20 ... 0 ... 19 ] * to static priority [ MAX_RT_PRIO..MAX_PRIO-1 ], * and back. @@ -242,6 +232,12 @@ static void destroy_rt_bandwidth(struct rt_bandwidth *rt_b) } #endif +/* + * sched_domains_mutex serializes calls to arch_init_sched_domains, + * detach_destroy_domains and partition_sched_domains. + */ +static DEFINE_MUTEX(sched_domains_mutex); + #ifdef CONFIG_GROUP_SCHED #include <linux/cgroup.h> @@ -308,9 +304,6 @@ static DEFINE_PER_CPU(struct rt_rq, init_rt_rq) ____cacheline_aligned_in_smp; */ static DEFINE_SPINLOCK(task_group_lock); -/* doms_cur_mutex serializes access to doms_cur[] array */ -static DEFINE_MUTEX(doms_cur_mutex); - #ifdef CONFIG_FAIR_GROUP_SCHED #ifdef CONFIG_USER_SCHED # define INIT_TASK_GROUP_LOAD (2*NICE_0_LOAD) @@ -318,7 +311,13 @@ static DEFINE_MUTEX(doms_cur_mutex); # define INIT_TASK_GROUP_LOAD NICE_0_LOAD #endif +/* + * A weight of 0, 1 or ULONG_MAX can cause arithmetics problems. + * (The default weight is 1024 - so there's no practical + * limitation from this.) + */ #define MIN_SHARES 2 +#define MAX_SHARES (ULONG_MAX - 1) static int init_task_group_load = INIT_TASK_GROUP_LOAD; #endif @@ -358,21 +357,9 @@ static inline void set_task_rq(struct task_struct *p, unsigned int cpu) #endif } -static inline void lock_doms_cur(void) -{ - mutex_lock(&doms_cur_mutex); -} - -static inline void unlock_doms_cur(void) -{ - mutex_unlock(&doms_cur_mutex); -} - #else static inline void set_task_rq(struct task_struct *p, unsigned int cpu) { } -static inline void lock_doms_cur(void) { } -static inline void unlock_doms_cur(void) { } #endif /* CONFIG_GROUP_SCHED */ @@ -560,13 +547,7 @@ struct rq { unsigned long next_balance; struct mm_struct *prev_mm; - u64 clock, prev_clock_raw; - s64 clock_max_delta; - - unsigned int clock_warps, clock_overflows, clock_underflows; - u64 idle_clock; - unsigned int clock_deep_idle_events; - u64 tick_timestamp; + u64 clock; atomic_t nr_iowait; @@ -631,82 +612,6 @@ static inline int cpu_of(struct rq *rq) #endif } -#ifdef CONFIG_NO_HZ -static inline bool nohz_on(int cpu) -{ - return tick_get_tick_sched(cpu)->nohz_mode != NOHZ_MODE_INACTIVE; -} - -static inline u64 max_skipped_ticks(struct rq *rq) -{ - return nohz_on(cpu_of(rq)) ? jiffies - rq->last_tick_seen + 2 : 1; -} - -static inline void update_last_tick_seen(struct rq *rq) -{ - rq->last_tick_seen = jiffies; -} -#else -static inline u64 max_skipped_ticks(struct rq *rq) -{ - return 1; -} - -static inline void update_last_tick_seen(struct rq *rq) -{ -} -#endif - -/* - * Update the per-runqueue clock, as finegrained as the platform can give - * us, but without assuming monotonicity, etc.: - */ -static void __update_rq_clock(struct rq *rq) -{ - u64 prev_raw = rq->prev_clock_raw; - u64 now = sched_clock(); - s64 delta = now - prev_raw; - u64 clock = rq->clock; - -#ifdef CONFIG_SCHED_DEBUG - WARN_ON_ONCE(cpu_of(rq) != smp_processor_id()); -#endif - /* - * Protect against sched_clock() occasionally going backwards: - */ - if (unlikely(delta < 0)) { - clock++; - rq->clock_warps++; - } else { - /* - * Catch too large forward jumps too: - */ - u64 max_jump = max_skipped_ticks(rq) * TICK_NSEC; - u64 max_time = rq->tick_timestamp + max_jump; - - if (unlikely(clock + delta > max_time)) { - if (clock < max_time) - clock = max_time; - else - clock++; - rq->clock_overflows++; - } else { - if (unlikely(delta > rq->clock_max_delta)) - rq->clock_max_delta = delta; - clock += delta; - } - } - - rq->prev_clock_raw = now; - rq->clock = clock; -} - -static void update_rq_clock(struct rq *rq) -{ - if (likely(smp_processor_id() == cpu_of(rq))) - __update_rq_clock(rq); -} - /* * The domain tree (rq->sd) is protected by RCU's quiescent state transition. * See detach_destroy_domains: synchronize_sched for details. @@ -722,6 +627,11 @@ static void update_rq_clock(struct rq *rq) #define task_rq(p) cpu_rq(task_cpu(p)) #define cpu_curr(cpu) (cpu_rq(cpu)->curr) +static inline void update_rq_clock(struct rq *rq) +{ + rq->clock = sched_clock_cpu(cpu_of(rq)); +} + /* * Tunables that become constants when CONFIG_SCHED_DEBUG is off: */ @@ -757,14 +667,14 @@ const_debug unsigned int sysctl_sched_features = #define SCHED_FEAT(name, enabled) \ #name , -__read_mostly char *sched_feat_names[] = { +static __read_mostly char *sched_feat_names[] = { #include "sched_features.h" NULL }; #undef SCHED_FEAT -int sched_feat_open(struct inode *inode, struct file *filp) +static int sched_feat_open(struct inode *inode, struct file *filp) { filp->private_data = inode->i_private; return 0; @@ -899,7 +809,7 @@ static inline u64 global_rt_runtime(void) return (u64)sysctl_sched_rt_runtime * NSEC_PER_USEC; } -static const unsigned long long time_sync_thresh = 100000; +unsigned long long time_sync_thresh = 100000; static DEFINE_PER_CPU(unsigned long long, time_offset); static DEFINE_PER_CPU(unsigned long long, prev_cpu_time); @@ -913,11 +823,14 @@ static DEFINE_PER_CPU(unsigned long long, prev_cpu_time); static DEFINE_SPINLOCK(time_sync_lock); static unsigned long long prev_global_time; -static unsigned long long __sync_cpu_clock(cycles_t time, int cpu) +static unsigned long long __sync_cpu_clock(unsigned long long time, int cpu) { - unsigned long flags; - - spin_lock_irqsave(&time_sync_lock, flags); + /* + * We want this inlined, to not get tracer function calls + * in this critical section: + */ + spin_acquire(&time_sync_lock.dep_map, 0, 0, _THIS_IP_); + __raw_spin_lock(&time_sync_lock.raw_lock); if (time < prev_global_time) { per_cpu(time_offset, cpu) += prev_global_time - time; @@ -926,7 +839,8 @@ static unsigned long long __sync_cpu_clock(cycles_t time, int cpu) prev_global_time = time; } - spin_unlock_irqrestore(&time_sync_lock, flags); + __raw_spin_unlock(&time_sync_lock.raw_lock); + spin_release(&time_sync_lock.dep_map, 1, _THIS_IP_); return time; } @@ -934,8 +848,6 @@ static unsigned long long __sync_cpu_clock(cycles_t time, int cpu) static unsigned long long __cpu_clock(int cpu) { unsigned long long now; - unsigned long flags; - struct rq *rq; /* * Only call sched_clock() if the scheduler has already been @@ -944,11 +856,7 @@ static unsigned long long __cpu_clock(int cpu) if (unlikely(!scheduler_running)) return 0; - local_irq_save(flags); - rq = cpu_rq(cpu); - update_rq_clock(rq); - now = rq->clock; - local_irq_restore(flags); + now = sched_clock_cpu(cpu); return now; } @@ -960,13 +868,18 @@ static unsigned long long __cpu_clock(int cpu) unsigned long long cpu_clock(int cpu) { unsigned long long prev_cpu_time, time, delta_time; + unsigned long flags; + local_irq_save(flags); prev_cpu_time = per_cpu(prev_cpu_time, cpu); time = __cpu_clock(cpu) + per_cpu(time_offset, cpu); delta_time = time-prev_cpu_time; - if (unlikely(delta_time > time_sync_thresh)) + if (unlikely(delta_time > time_sync_thresh)) { time = __sync_cpu_clock(time, cpu); + per_cpu(prev_cpu_time, cpu) = time; + } + local_irq_restore(flags); return time; } @@ -1117,43 +1030,6 @@ static struct rq *this_rq_lock(void) return rq; } -/* - * We are going deep-idle (irqs are disabled): - */ -void sched_clock_idle_sleep_event(void) -{ - struct rq *rq = cpu_rq(smp_processor_id()); - - spin_lock(&rq->lock); - __update_rq_clock(rq); - spin_unlock(&rq->lock); - rq->clock_deep_idle_events++; -} -EXPORT_SYMBOL_GPL(sched_clock_idle_sleep_event); - -/* - * We just idled delta nanoseconds (called with irqs disabled): - */ -void sched_clock_idle_wakeup_event(u64 delta_ns) -{ - struct rq *rq = cpu_rq(smp_processor_id()); - u64 now = sched_clock(); - - rq->idle_clock += delta_ns; - /* - * Override the previous timestamp and ignore all - * sched_clock() deltas that occured while we idled, - * and use the PM-provided delta_ns to advance the - * rq clock: - */ - spin_lock(&rq->lock); - rq->prev_clock_raw = now; - rq->clock += delta_ns; - spin_unlock(&rq->lock); - touch_softlockup_watchdog(); -} -EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event); - static void __resched_task(struct task_struct *p, int tif_bit); static inline void resched_task(struct task_struct *p) @@ -1189,6 +1065,7 @@ static inline void resched_rq(struct rq *rq) enum { HRTICK_SET, /* re-programm hrtick_timer */ HRTICK_RESET, /* not a new slice */ + HRTICK_BLOCK, /* stop hrtick operations */ }; /* @@ -1200,6 +1077,8 @@ static inline int hrtick_enabled(struct rq *rq) { if (!sched_feat(HRTICK)) return 0; + if (unlikely(test_bit(HRTICK_BLOCK, &rq->hrtick_flags))) + return 0; return hrtimer_is_hres_active(&rq->hrtick_timer); } @@ -1275,14 +1154,70 @@ static enum hrtimer_restart hrtick(struct hrtimer *timer) WARN_ON_ONCE(cpu_of(rq) != smp_processor_id()); spin_lock(&rq->lock); - __update_rq_clock(rq); + update_rq_clock(rq); rq->curr->sched_class->task_tick(rq, rq->curr, 1); spin_unlock(&rq->lock); return HRTIMER_NORESTART; } -static inline void init_rq_hrtick(struct rq *rq) +static void hotplug_hrtick_disable(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + unsigned long flags; + + spin_lock_irqsave(&rq->lock, flags); + rq->hrtick_flags = 0; + __set_bit(HRTICK_BLOCK, &rq->hrtick_flags); + spin_unlock_irqrestore(&rq->lock, flags); + + hrtick_clear(rq); +} + +static void hotplug_hrtick_enable(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + unsigned long flags; + + spin_lock_irqsave(&rq->lock, flags); + __clear_bit(HRTICK_BLOCK, &rq->hrtick_flags); + spin_unlock_irqrestore(&rq->lock, flags); +} + +static int +hotplug_hrtick(struct notifier_block *nfb, unsigned long action, void *hcpu) +{ + int cpu = (int)(long)hcpu; + + switch (action) { + case CPU_UP_CANCELED: + case CPU_UP_CANCELED_FROZEN: + case CPU_DOWN_PREPARE: + case CPU_DOWN_PREPARE_FROZEN: + case CPU_DEAD: + case CPU_DEAD_FROZEN: + hotplug_hrtick_disable(cpu); + return NOTIFY_OK; + + case CPU_UP_PREPARE: + case CPU_UP_PREPARE_FROZEN: + case CPU_DOWN_FAILED: + case CPU_DOWN_FAILED_FROZEN: + case CPU_ONLINE: + case CPU_ONLINE_FROZEN: + hotplug_hrtick_enable(cpu); + return NOTIFY_OK; + } + + return NOTIFY_DONE; +} + +static void init_hrtick(void) +{ + hotcpu_notifier(hotplug_hrtick, 0); +} + +static void init_rq_hrtick(struct rq *rq) { rq->hrtick_flags = 0; hrtimer_init(&rq->hrtick_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); @@ -1319,6 +1254,10 @@ static inline void init_rq_hrtick(struct rq *rq) void hrtick_resched(void) { } + +static inline void init_hrtick(void) +{ +} #endif /* @@ -1438,8 +1377,8 @@ calc_delta_mine(unsigned long delta_exec, unsigned long weight, { u64 tmp; - if (unlikely(!lw->inv_weight)) - lw->inv_weight = (WMULT_CONST-lw->weight/2) / (lw->weight+1); + if (!lw->inv_weight) + lw->inv_weight = 1 + (WMULT_CONST-lw->weight/2)/(lw->weight+1); tmp = (u64)delta_exec * weight; /* @@ -1748,6 +1687,8 @@ __update_group_shares_cpu(struct task_group *tg, struct sched_domain *sd, if (shares < MIN_SHARES) shares = MIN_SHARES; + else if (shares > MAX_SHARES) + shares = MAX_SHARES; __set_se_shares(tg->se[tcpu], shares); } @@ -4339,8 +4280,10 @@ void account_system_time(struct task_struct *p, int hardirq_offset, struct rq *rq = this_rq(); cputime64_t tmp; - if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) - return account_guest_time(p, cputime); + if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) { + account_guest_time(p, cputime); + return; + } p->stime = cputime_add(p->stime, cputime); @@ -4404,19 +4347,11 @@ void scheduler_tick(void) int cpu = smp_processor_id(); struct rq *rq = cpu_rq(cpu); struct task_struct *curr = rq->curr; - u64 next_tick = rq->tick_timestamp + TICK_NSEC; + + sched_clock_tick(); spin_lock(&rq->lock); - __update_rq_clock(rq); - /* - * Let rq->clock advance by at least TICK_NSEC: - */ - if (unlikely(rq->clock < next_tick)) { - rq->clock = next_tick; - rq->clock_underflows++; - } - rq->tick_timestamp = rq->clock; - update_last_tick_seen(rq); + update_rq_clock(rq); update_cpu_load(rq); curr->sched_class->task_tick(rq, curr, 0); spin_unlock(&rq->lock); @@ -4570,7 +4505,7 @@ need_resched_nonpreemptible: * Do the rq-clock update outside the rq lock: */ local_irq_disable(); - __update_rq_clock(rq); + update_rq_clock(rq); spin_lock(&rq->lock); clear_tsk_need_resched(prev); @@ -4595,9 +4530,9 @@ need_resched_nonpreemptible: prev->sched_class->put_prev_task(rq, prev); next = pick_next_task(rq, prev); - sched_info_switch(prev, next); - if (likely(prev != next)) { + sched_info_switch(prev, next); + rq->nr_switches++; rq->curr = next; ++*switch_count; @@ -7755,7 +7690,7 @@ void partition_sched_domains(int ndoms_new, cpumask_t *doms_new, { int i, j; - lock_doms_cur(); + mutex_lock(&sched_domains_mutex); /* always unregister in case we don't destroy any domains */ unregister_sched_domain_sysctl(); @@ -7804,7 +7739,7 @@ match2: register_sched_domain_sysctl(); - unlock_doms_cur(); + mutex_unlock(&sched_domains_mutex); } #if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) @@ -7813,8 +7748,10 @@ int arch_reinit_sched_domains(void) int err; get_online_cpus(); + mutex_lock(&sched_domains_mutex); detach_destroy_domains(&cpu_online_map); err = arch_init_sched_domains(&cpu_online_map); + mutex_unlock(&sched_domains_mutex); put_online_cpus(); return err; @@ -7932,13 +7869,16 @@ void __init sched_init_smp(void) BUG_ON(sched_group_nodes_bycpu == NULL); #endif get_online_cpus(); + mutex_lock(&sched_domains_mutex); arch_init_sched_domains(&cpu_online_map); cpus_andnot(non_isolated_cpus, cpu_possible_map, cpu_isolated_map); if (cpus_empty(non_isolated_cpus)) cpu_set(smp_processor_id(), non_isolated_cpus); + mutex_unlock(&sched_domains_mutex); put_online_cpus(); /* XXX: Theoretical race here - CPU may be hotplugged now */ hotcpu_notifier(update_sched_domains, 0); + init_hrtick(); /* Move init over to a non-isolated CPU */ if (set_cpus_allowed_ptr(current, &non_isolated_cpus) < 0) @@ -8025,7 +7965,7 @@ static void init_tg_cfs_entry(struct task_group *tg, struct cfs_rq *cfs_rq, se->my_q = cfs_rq; se->load.weight = tg->shares; - se->load.inv_weight = div64_u64(1ULL<<32, se->load.weight); + se->load.inv_weight = 0; se->parent = parent; } #endif @@ -8149,8 +8089,6 @@ void __init sched_init(void) spin_lock_init(&rq->lock); lockdep_set_class(&rq->lock, &rq->rq_lock_key); rq->nr_running = 0; - rq->clock = 1; - update_last_tick_seen(rq); init_cfs_rq(&rq->cfs, rq); init_rt_rq(&rq->rt, rq); #ifdef CONFIG_FAIR_GROUP_SCHED @@ -8294,6 +8232,7 @@ EXPORT_SYMBOL(__might_sleep); static void normalize_task(struct rq *rq, struct task_struct *p) { int on_rq; + update_rq_clock(rq); on_rq = p->se.on_rq; if (on_rq) @@ -8325,7 +8264,6 @@ void normalize_rt_tasks(void) p->se.sleep_start = 0; p->se.block_start = 0; #endif - task_rq(p)->clock = 0; if (!rt_task(p)) { /* @@ -8692,7 +8630,7 @@ static void __set_se_shares(struct sched_entity *se, unsigned long shares) dequeue_entity(cfs_rq, se, 0); se->load.weight = shares; - se->load.inv_weight = div64_u64((1ULL<<32), shares); + se->load.inv_weight = 0; if (on_rq) enqueue_entity(cfs_rq, se, 0); @@ -8722,13 +8660,10 @@ int sched_group_set_shares(struct task_group *tg, unsigned long shares) if (!tg->se[0]) return -EINVAL; - /* - * A weight of 0 or 1 can cause arithmetics problems. - * (The default weight is 1024 - so there's no practical - * limitation from this.) - */ if (shares < MIN_SHARES) shares = MIN_SHARES; + else if (shares > MAX_SHARES) + shares = MAX_SHARES; mutex_lock(&shares_mutex); if (tg->shares == shares) @@ -8753,7 +8688,7 @@ int sched_group_set_shares(struct task_group *tg, unsigned long shares) * force a rebalance */ cfs_rq_set_shares(tg->cfs_rq[i], 0); - set_se_shares(tg->se[i], shares/nr_cpu_ids); + set_se_shares(tg->se[i], shares); } /* diff --git a/kernel/sched_clock.c b/kernel/sched_clock.c new file mode 100644 index 00000000000..9c597e37f7d --- /dev/null +++ b/kernel/sched_clock.c @@ -0,0 +1,236 @@ +/* + * sched_clock for unstable cpu clocks + * + * Copyright (C) 2008 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> + * + * Based on code by: + * Ingo Molnar <mingo@redhat.com> + * Guillaume Chazarain <guichaz@gmail.com> + * + * Create a semi stable clock from a mixture of other events, including: + * - gtod + * - jiffies + * - sched_clock() + * - explicit idle events + * + * We use gtod as base and the unstable clock deltas. The deltas are filtered, + * making it monotonic and keeping it within an expected window. This window + * is set up using jiffies. + * + * Furthermore, explicit sleep and wakeup hooks allow us to account for time + * that is otherwise invisible (TSC gets stopped). + * + * The clock: sched_clock_cpu() is monotonic per cpu, and should be somewhat + * consistent between cpus (never more than 1 jiffies difference). + */ +#include <linux/sched.h> +#include <linux/percpu.h> +#include <linux/spinlock.h> +#include <linux/ktime.h> +#include <linux/module.h> + + +#ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK + +struct sched_clock_data { + /* + * Raw spinlock - this is a special case: this might be called + * from within instrumentation code so we dont want to do any + * instrumentation ourselves. + */ + raw_spinlock_t lock; + + unsigned long prev_jiffies; + u64 prev_raw; + u64 tick_raw; + u64 tick_gtod; + u64 clock; +}; + +static DEFINE_PER_CPU_SHARED_ALIGNED(struct sched_clock_data, sched_clock_data); + +static inline struct sched_clock_data *this_scd(void) +{ + return &__get_cpu_var(sched_clock_data); +} + +static inline struct sched_clock_data *cpu_sdc(int cpu) +{ + return &per_cpu(sched_clock_data, cpu); +} + +void sched_clock_init(void) +{ + u64 ktime_now = ktime_to_ns(ktime_get()); + u64 now = 0; + int cpu; + + for_each_possible_cpu(cpu) { + struct sched_clock_data *scd = cpu_sdc(cpu); + + scd->lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + scd->prev_jiffies = jiffies; + scd->prev_raw = now; + scd->tick_raw = now; + scd->tick_gtod = ktime_now; + scd->clock = ktime_now; + } +} + +/* + * update the percpu scd from the raw @now value + * + * - filter out backward motion + * - use jiffies to generate a min,max window to clip the raw values + */ +static void __update_sched_clock(struct sched_clock_data *scd, u64 now) +{ + unsigned long now_jiffies = jiffies; + long delta_jiffies = now_jiffies - scd->prev_jiffies; + u64 clock = scd->clock; + u64 min_clock, max_clock; + s64 delta = now - scd->prev_raw; + + WARN_ON_ONCE(!irqs_disabled()); + min_clock = scd->tick_gtod + delta_jiffies * TICK_NSEC; + + if (unlikely(delta < 0)) { + clock++; + goto out; + } + + max_clock = min_clock + TICK_NSEC; + + if (unlikely(clock + delta > max_clock)) { + if (clock < max_clock) + clock = max_clock; + else + clock++; + } else { + clock += delta; + } + + out: + if (unlikely(clock < min_clock)) + clock = min_clock; + + scd->prev_raw = now; + scd->prev_jiffies = now_jiffies; + scd->clock = clock; +} + +static void lock_double_clock(struct sched_clock_data *data1, + struct sched_clock_data *data2) +{ + if (data1 < data2) { + __raw_spin_lock(&data1->lock); + __raw_spin_lock(&data2->lock); + } else { + __raw_spin_lock(&data2->lock); + __raw_spin_lock(&data1->lock); + } +} + +u64 sched_clock_cpu(int cpu) +{ + struct sched_clock_data *scd = cpu_sdc(cpu); + u64 now, clock; + + WARN_ON_ONCE(!irqs_disabled()); + now = sched_clock(); + + if (cpu != raw_smp_processor_id()) { + /* + * in order to update a remote cpu's clock based on our + * unstable raw time rebase it against: + * tick_raw (offset between raw counters) + * tick_gotd (tick offset between cpus) + */ + struct sched_clock_data *my_scd = this_scd(); + + lock_double_clock(scd, my_scd); + + now -= my_scd->tick_raw; + now += scd->tick_raw; + + now -= my_scd->tick_gtod; + now += scd->tick_gtod; + + __raw_spin_unlock(&my_scd->lock); + } else { + __raw_spin_lock(&scd->lock); + } + + __update_sched_clock(scd, now); + clock = scd->clock; + + __raw_spin_unlock(&scd->lock); + + return clock; +} + +void sched_clock_tick(void) +{ + struct sched_clock_data *scd = this_scd(); + u64 now, now_gtod; + + WARN_ON_ONCE(!irqs_disabled()); + + now = sched_clock(); + now_gtod = ktime_to_ns(ktime_get()); + + __raw_spin_lock(&scd->lock); + __update_sched_clock(scd, now); + /* + * update tick_gtod after __update_sched_clock() because that will + * already observe 1 new jiffy; adding a new tick_gtod to that would + * increase the clock 2 jiffies. + */ + scd->tick_raw = now; + scd->tick_gtod = now_gtod; + __raw_spin_unlock(&scd->lock); +} + +/* + * We are going deep-idle (irqs are disabled): + */ +void sched_clock_idle_sleep_event(void) +{ + sched_clock_cpu(smp_processor_id()); +} +EXPORT_SYMBOL_GPL(sched_clock_idle_sleep_event); + +/* + * We just idled delta nanoseconds (called with irqs disabled): + */ +void sched_clock_idle_wakeup_event(u64 delta_ns) +{ + struct sched_clock_data *scd = this_scd(); + u64 now = sched_clock(); + + /* + * Override the previous timestamp and ignore all + * sched_clock() deltas that occured while we idled, + * and use the PM-provided delta_ns to advance the + * rq clock: + */ + __raw_spin_lock(&scd->lock); + scd->prev_raw = now; + scd->clock += delta_ns; + __raw_spin_unlock(&scd->lock); + + touch_softlockup_watchdog(); +} +EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event); + +#endif + +/* + * Scheduler clock - returns current time in nanosec units. + * This is default implementation. + * Architectures and sub-architectures can override this. + */ +unsigned long long __attribute__((weak)) sched_clock(void) +{ + return (unsigned long long)jiffies * (NSEC_PER_SEC / HZ); +} diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 6b4a12558e8..5f06118fbc3 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -204,13 +204,6 @@ static void print_cpu(struct seq_file *m, int cpu) PN(next_balance); P(curr->pid); PN(clock); - PN(idle_clock); - PN(prev_clock_raw); - P(clock_warps); - P(clock_overflows); - P(clock_underflows); - P(clock_deep_idle_events); - PN(clock_max_delta); P(cpu_load[0]); P(cpu_load[1]); P(cpu_load[2]); diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 89fa32b4edf..e24ecd39c4b 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -662,10 +662,15 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial) if (!initial) { /* sleeps upto a single latency don't count. */ if (sched_feat(NEW_FAIR_SLEEPERS)) { + unsigned long thresh = sysctl_sched_latency; + + /* + * convert the sleeper threshold into virtual time + */ if (sched_feat(NORMALIZED_SLEEPER)) - vruntime -= calc_delta_weight(sysctl_sched_latency, se); - else - vruntime -= sysctl_sched_latency; + thresh = calc_delta_fair(thresh, se); + + vruntime -= thresh; } /* ensure we never gain time by being placed backwards. */ @@ -682,6 +687,7 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int wakeup) * Update run-time statistics of the 'current'. */ update_curr(cfs_rq); + account_entity_enqueue(cfs_rq, se); if (wakeup) { place_entity(cfs_rq, se, 0); @@ -692,7 +698,6 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int wakeup) check_spread(cfs_rq, se); if (se != cfs_rq->curr) __enqueue_entity(cfs_rq, se); - account_entity_enqueue(cfs_rq, se); } static void update_avg(u64 *avg, u64 sample) @@ -841,8 +846,10 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued) * queued ticks are scheduled to match the slice, so don't bother * validating it and just reschedule. */ - if (queued) - return resched_task(rq_of(cfs_rq)->curr); + if (queued) { + resched_task(rq_of(cfs_rq)->curr); + return; + } /* * don't let the period tick interfere with the hrtick preemption */ @@ -957,7 +964,7 @@ static void yield_task_fair(struct rq *rq) return; if (likely(!sysctl_sched_compat_yield) && curr->policy != SCHED_BATCH) { - __update_rq_clock(rq); + update_rq_clock(rq); /* * Update run-time statistics of the 'current'. */ @@ -1007,7 +1014,7 @@ static int wake_idle(int cpu, struct task_struct *p) * sibling runqueue info. This will avoid the checks and cache miss * penalities associated with that. */ - if (idle_cpu(cpu) || cpu_rq(cpu)->nr_running > 1) + if (idle_cpu(cpu) || cpu_rq(cpu)->cfs.nr_running > 1) return cpu; for_each_domain(cpu, sd) { @@ -1611,30 +1618,6 @@ static const struct sched_class fair_sched_class = { }; #ifdef CONFIG_SCHED_DEBUG -static void -print_cfs_rq_tasks(struct seq_file *m, struct cfs_rq *cfs_rq, int depth) -{ - struct sched_entity *se; - - if (!cfs_rq) - return; - - list_for_each_entry_rcu(se, &cfs_rq->tasks, group_node) { - int i; - - for (i = depth; i; i--) - seq_puts(m, " "); - - seq_printf(m, "%lu %s %lu\n", - se->load.weight, - entity_is_task(se) ? "T" : "G", - calc_delta_weight(SCHED_LOAD_SCALE, se) - ); - if (!entity_is_task(se)) - print_cfs_rq_tasks(m, group_cfs_rq(se), depth + 1); - } -} - static void print_cfs_stats(struct seq_file *m, int cpu) { struct cfs_rq *cfs_rq; @@ -1642,9 +1625,6 @@ static void print_cfs_stats(struct seq_file *m, int cpu) rcu_read_lock(); for_each_leaf_cfs_rq(cpu_rq(cpu), cfs_rq) print_cfs_rq(m, cpu, cfs_rq); - - seq_printf(m, "\nWeight tree:\n"); - print_cfs_rq_tasks(m, &cpu_rq(cpu)->cfs, 1); rcu_read_unlock(); } #endif diff --git a/kernel/sched_idletask.c b/kernel/sched_idletask.c index 2bcafa37563..3a4f92dbbe6 100644 --- a/kernel/sched_idletask.c +++ b/kernel/sched_idletask.c @@ -99,7 +99,7 @@ static void prio_changed_idle(struct rq *rq, struct task_struct *p, /* * Simple, special scheduling class for the per-CPU idle tasks: */ -const struct sched_class idle_sched_class = { +static const struct sched_class idle_sched_class = { /* .next is NULL */ /* no enqueue/yield_task for idle tasks */ diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index c2730a5a4f0..060e87b0cb1 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -1098,11 +1098,14 @@ static void post_schedule_rt(struct rq *rq) } } - +/* + * If we are not running and we are not going to reschedule soon, we should + * try to push tasks away now + */ static void task_wake_up_rt(struct rq *rq, struct task_struct *p) { if (!task_running(rq, p) && - (p->prio >= rq->rt.highest_prio) && + !test_tsk_need_resched(rq->curr) && rq->rt.overloaded) push_rt_tasks(rq); } @@ -1309,7 +1312,7 @@ static void set_curr_task_rt(struct rq *rq) p->se.exec_start = rq->clock; } -const struct sched_class rt_sched_class = { +static const struct sched_class rt_sched_class = { .next = &fair_sched_class, .enqueue_task = enqueue_task_rt, .dequeue_task = dequeue_task_rt, diff --git a/kernel/semaphore.c b/kernel/semaphore.c index 5c2942e768c..5e41217239e 100644 --- a/kernel/semaphore.c +++ b/kernel/semaphore.c @@ -54,10 +54,9 @@ void down(struct semaphore *sem) unsigned long flags; spin_lock_irqsave(&sem->lock, flags); - if (likely(sem->count > 0)) - sem->count--; - else + if (unlikely(!sem->count)) __down(sem); + sem->count--; spin_unlock_irqrestore(&sem->lock, flags); } EXPORT_SYMBOL(down); @@ -77,10 +76,10 @@ int down_interruptible(struct semaphore *sem) int result = 0; spin_lock_irqsave(&sem->lock, flags); - if (likely(sem->count > 0)) - sem->count--; - else + if (unlikely(!sem->count)) result = __down_interruptible(sem); + if (!result) + sem->count--; spin_unlock_irqrestore(&sem->lock, flags); return result; @@ -103,10 +102,10 @@ int down_killable(struct semaphore *sem) int result = 0; spin_lock_irqsave(&sem->lock, flags); - if (likely(sem->count > 0)) - sem->count--; - else + if (unlikely(!sem->count)) result = __down_killable(sem); + if (!result) + sem->count--; spin_unlock_irqrestore(&sem->lock, flags); return result; @@ -157,10 +156,10 @@ int down_timeout(struct semaphore *sem, long jiffies) int result = 0; spin_lock_irqsave(&sem->lock, flags); - if (likely(sem->count > 0)) - sem->count--; - else + if (unlikely(!sem->count)) result = __down_timeout(sem, jiffies); + if (!result) + sem->count--; spin_unlock_irqrestore(&sem->lock, flags); return result; @@ -179,9 +178,8 @@ void up(struct semaphore *sem) unsigned long flags; spin_lock_irqsave(&sem->lock, flags); - if (likely(list_empty(&sem->wait_list))) - sem->count++; - else + sem->count++; + if (unlikely(!list_empty(&sem->wait_list))) __up(sem); spin_unlock_irqrestore(&sem->lock, flags); } @@ -192,7 +190,6 @@ EXPORT_SYMBOL(up); struct semaphore_waiter { struct list_head list; struct task_struct *task; - int up; }; /* @@ -205,33 +202,34 @@ static inline int __sched __down_common(struct semaphore *sem, long state, { struct task_struct *task = current; struct semaphore_waiter waiter; + int ret = 0; - list_add_tail(&waiter.list, &sem->wait_list); waiter.task = task; - waiter.up = 0; + list_add_tail(&waiter.list, &sem->wait_list); for (;;) { - if (state == TASK_INTERRUPTIBLE && signal_pending(task)) - goto interrupted; - if (state == TASK_KILLABLE && fatal_signal_pending(task)) - goto interrupted; - if (timeout <= 0) - goto timed_out; + if (state == TASK_INTERRUPTIBLE && signal_pending(task)) { + ret = -EINTR; + break; + } + if (state == TASK_KILLABLE && fatal_signal_pending(task)) { + ret = -EINTR; + break; + } + if (timeout <= 0) { + ret = -ETIME; + break; + } __set_task_state(task, state); spin_unlock_irq(&sem->lock); timeout = schedule_timeout(timeout); spin_lock_irq(&sem->lock); - if (waiter.up) - return 0; + if (sem->count > 0) + break; } - timed_out: - list_del(&waiter.list); - return -ETIME; - - interrupted: list_del(&waiter.list); - return -EINTR; + return ret; } static noinline void __sched __down(struct semaphore *sem) @@ -258,7 +256,5 @@ static noinline void __sched __up(struct semaphore *sem) { struct semaphore_waiter *waiter = list_first_entry(&sem->wait_list, struct semaphore_waiter, list); - list_del(&waiter->list); - waiter->up = 1; wake_up_process(waiter->task); } diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 73961f35fdc..dadde5361f3 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -471,10 +471,10 @@ sysfs_show_available_clocksources(struct sys_device *dev, char *buf) /* * Sysfs setup bits: */ -static SYSDEV_ATTR(current_clocksource, 0600, sysfs_show_current_clocksources, +static SYSDEV_ATTR(current_clocksource, 0644, sysfs_show_current_clocksources, sysfs_override_clocksource); -static SYSDEV_ATTR(available_clocksource, 0600, +static SYSDEV_ATTR(available_clocksource, 0444, sysfs_show_available_clocksources, NULL); static struct sysdev_class clocksource_sysclass = { diff --git a/lib/Kconfig.kgdb b/lib/Kconfig.kgdb index f2e01ac5ab0..a5d4b1dac2a 100644 --- a/lib/Kconfig.kgdb +++ b/lib/Kconfig.kgdb @@ -1,4 +1,10 @@ +config HAVE_ARCH_KGDB_SHADOW_INFO + bool + +config HAVE_ARCH_KGDB + bool + menuconfig KGDB bool "KGDB: kernel debugging with remote gdb" select FRAME_POINTER @@ -10,15 +16,10 @@ menuconfig KGDB at http://kgdb.sourceforge.net as well as in DocBook form in Documentation/DocBook/. If unsure, say N. -config HAVE_ARCH_KGDB_SHADOW_INFO - bool - -config HAVE_ARCH_KGDB - bool +if KGDB config KGDB_SERIAL_CONSOLE tristate "KGDB: use kgdb over the serial console" - depends on KGDB select CONSOLE_POLL select MAGIC_SYSRQ default y @@ -28,7 +29,6 @@ config KGDB_SERIAL_CONSOLE config KGDB_TESTS bool "KGDB: internal test suite" - depends on KGDB default n help This is a kgdb I/O module specifically designed to test @@ -56,3 +56,5 @@ config KGDB_TESTS_BOOT_STRING boot. See the drivers/misc/kgdbts.c for detailed information about other strings you could use beyond the default of V1F100. + +endif # KGDB diff --git a/lib/devres.c b/lib/devres.c index 26c87c49d77..72c8909006d 100644 --- a/lib/devres.c +++ b/lib/devres.c @@ -2,7 +2,7 @@ #include <linux/io.h> #include <linux/module.h> -static void devm_ioremap_release(struct device *dev, void *res) +void devm_ioremap_release(struct device *dev, void *res) { iounmap(*(void __iomem **)res); } diff --git a/mm/filemap.c b/mm/filemap.c index 239d36163bb..2dead9adf8b 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1655,7 +1655,7 @@ int should_remove_suid(struct dentry *dentry) } EXPORT_SYMBOL(should_remove_suid); -int __remove_suid(struct dentry *dentry, int kill) +static int __remove_suid(struct dentry *dentry, int kill) { struct iattr newattrs; diff --git a/mm/memory.c b/mm/memory.c index bbab1e37055..48c122d42ed 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -969,7 +969,7 @@ struct page *follow_page(struct vm_area_struct *vma, unsigned long address, goto no_page_table; pmd = pmd_offset(pud, address); - if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd))) + if (pmd_none(*pmd)) goto no_page_table; if (pmd_huge(*pmd)) { @@ -978,6 +978,9 @@ struct page *follow_page(struct vm_area_struct *vma, unsigned long address, goto out; } + if (unlikely(pmd_bad(*pmd))) + goto no_page_table; + ptep = pte_offset_map_lock(mm, pmd, address, &ptl); if (!ptep) goto out; diff --git a/mm/slub.c b/mm/slub.c index d379b782fc8..a505a828ef4 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -3762,7 +3762,7 @@ static int any_slab_objects(struct kmem_cache *s) if (!n) continue; - if (atomic_read(&n->total_objects)) + if (atomic_long_read(&n->total_objects)) return 1; } return 0; diff --git a/net/atm/br2684.c b/net/atm/br2684.c index 1b228065e74..9d52ebfc196 100644 --- a/net/atm/br2684.c +++ b/net/atm/br2684.c @@ -346,9 +346,9 @@ static void br2684_push(struct atm_vcc *atmvcc, struct sk_buff *skb) /* skb==NULL means VCC is being destroyed */ br2684_close_vcc(brvcc); if (list_empty(&brdev->brvccs)) { - read_lock(&devs_lock); + write_lock_irq(&devs_lock); list_del(&brdev->br2684_devs); - read_unlock(&devs_lock); + write_unlock_irq(&devs_lock); unregister_netdev(net_dev); free_netdev(net_dev); } diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index 77a981a1ee5..c2397f503b0 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -273,15 +273,13 @@ int br_add_bridge(const char *name) rtnl_lock(); if (strchr(dev->name, '%')) { ret = dev_alloc_name(dev, dev->name); - if (ret < 0) { - free_netdev(dev); - goto out; - } + if (ret < 0) + goto out_free; } ret = register_netdevice(dev); if (ret) - goto out; + goto out_free; ret = br_sysfs_addbr(dev); if (ret) @@ -289,6 +287,10 @@ int br_add_bridge(const char *name) out: rtnl_unlock(); return ret; + +out_free: + free_netdev(dev); + goto out; } int br_del_bridge(const char *name) diff --git a/net/can/af_can.c b/net/can/af_can.c index 2759b76f731..7e8ca283645 100644 --- a/net/can/af_can.c +++ b/net/can/af_can.c @@ -208,6 +208,7 @@ static int can_create(struct net *net, struct socket *sock, int protocol) */ int can_send(struct sk_buff *skb, int loop) { + struct sk_buff *newskb = NULL; int err; if (skb->dev->type != ARPHRD_CAN) { @@ -244,8 +245,7 @@ int can_send(struct sk_buff *skb, int loop) * If the interface is not capable to do loopback * itself, we do it here. */ - struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC); - + newskb = skb_clone(skb, GFP_ATOMIC); if (!newskb) { kfree_skb(skb); return -ENOMEM; @@ -254,7 +254,6 @@ int can_send(struct sk_buff *skb, int loop) newskb->sk = skb->sk; newskb->ip_summed = CHECKSUM_UNNECESSARY; newskb->pkt_type = PACKET_BROADCAST; - netif_rx(newskb); } } else { /* indication for the CAN driver: no loopback required */ @@ -266,11 +265,20 @@ int can_send(struct sk_buff *skb, int loop) if (err > 0) err = net_xmit_errno(err); + if (err) { + if (newskb) + kfree_skb(newskb); + return err; + } + + if (newskb) + netif_rx(newskb); + /* update statistics */ can_stats.tx_frames++; can_stats.tx_frames_delta++; - return err; + return 0; } EXPORT_SYMBOL(can_send); diff --git a/net/can/bcm.c b/net/can/bcm.c index 74fd2d33aff..d9a3a9d13be 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -412,12 +412,6 @@ static void bcm_rx_changed(struct bcm_op *op, struct can_frame *data) bcm_send_to_user(op, &head, data, 1); } -/* TODO: move to linux/hrtimer.h */ -static inline int hrtimer_callback_running(struct hrtimer *timer) -{ - return timer->state & HRTIMER_STATE_CALLBACK; -} - /* * bcm_rx_update_and_send - process a detected relevant receive content change * 1. update the last received data diff --git a/net/core/dev.c b/net/core/dev.c index d334446a8ea..a1607bc0cd4 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -994,6 +994,8 @@ int dev_open(struct net_device *dev) { int ret = 0; + ASSERT_RTNL(); + /* * Is it already up? */ @@ -1060,6 +1062,8 @@ int dev_open(struct net_device *dev) */ int dev_close(struct net_device *dev) { + ASSERT_RTNL(); + might_sleep(); if (!(dev->flags & IFF_UP)) @@ -4480,17 +4484,19 @@ static void __net_exit default_device_exit(struct net *net) rtnl_lock(); for_each_netdev_safe(net, dev, next) { int err; + char fb_name[IFNAMSIZ]; /* Ignore unmoveable devices (i.e. loopback) */ if (dev->features & NETIF_F_NETNS_LOCAL) continue; /* Push remaing network devices to init_net */ - err = dev_change_net_namespace(dev, &init_net, "dev%d"); + snprintf(fb_name, IFNAMSIZ, "dev%d", dev->ifindex); + err = dev_change_net_namespace(dev, &init_net, fb_name); if (err) { - printk(KERN_WARNING "%s: failed to move %s to init_net: %d\n", + printk(KERN_EMERG "%s: failed to move %s to init_net: %d\n", __func__, dev->name, err); - unregister_netdevice(dev); + BUG(); } } rtnl_unlock(); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 4fe605fa6f8..5c459f2b798 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -200,7 +200,9 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask, goto nodata; /* - * See comment in sk_buff definition, just before the 'tail' member + * Only clear those fields we need to clear, not those that we will + * actually initialise below. Hence, don't put any more fields after + * the tail pointer in struct sk_buff! */ memset(skb, 0, offsetof(struct sk_buff, tail)); skb->truesize = size + sizeof(struct sk_buff); diff --git a/net/dccp/feat.c b/net/dccp/feat.c index 4a4f6ce4498..933a0ecf8d4 100644 --- a/net/dccp/feat.c +++ b/net/dccp/feat.c @@ -32,7 +32,7 @@ int dccp_feat_change(struct dccp_minisock *dmsk, u8 type, u8 feature, if (len > 3) { DCCP_WARN("invalid length %d\n", len); - return 1; + return -EINVAL; } /* XXX add further sanity checks */ diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 2f665a51647..f50e88bf266 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -235,14 +235,14 @@ static void dn_dst_update_pmtu(struct dst_entry *dst, u32 mtu) else min_mtu -= 21; - if (dst->metrics[RTAX_MTU-1] > mtu && mtu >= min_mtu) { + if (dst_metric(dst, RTAX_MTU) > mtu && mtu >= min_mtu) { if (!(dst_metric_locked(dst, RTAX_MTU))) { dst->metrics[RTAX_MTU-1] = mtu; dst_set_expires(dst, dn_rt_mtu_expires); } if (!(dst_metric_locked(dst, RTAX_ADVMSS))) { u32 mss = mtu - DN_MAX_NSP_DATA_HEADER; - if (dst->metrics[RTAX_ADVMSS-1] > mss) + if (dst_metric(dst, RTAX_ADVMSS) > mss) dst->metrics[RTAX_ADVMSS-1] = mss; } } @@ -805,12 +805,12 @@ static int dn_rt_set_next_hop(struct dn_route *rt, struct dn_fib_res *res) rt->u.dst.neighbour = n; } - if (rt->u.dst.metrics[RTAX_MTU-1] == 0 || - rt->u.dst.metrics[RTAX_MTU-1] > rt->u.dst.dev->mtu) + if (dst_metric(&rt->u.dst, RTAX_MTU) == 0 || + dst_metric(&rt->u.dst, RTAX_MTU) > rt->u.dst.dev->mtu) rt->u.dst.metrics[RTAX_MTU-1] = rt->u.dst.dev->mtu; mss = dn_mss_from_pmtu(dev, dst_mtu(&rt->u.dst)); - if (rt->u.dst.metrics[RTAX_ADVMSS-1] == 0 || - rt->u.dst.metrics[RTAX_ADVMSS-1] > mss) + if (dst_metric(&rt->u.dst, RTAX_ADVMSS) == 0 || + dst_metric(&rt->u.dst, RTAX_ADVMSS) > mss) rt->u.dst.metrics[RTAX_ADVMSS-1] = mss; return 0; } diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index 7b4bad6d572..ff77a4a7f9e 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -397,7 +397,7 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, iph = ip_hdr(skb); /* - * RFC1122: 3.1.2.2 MUST silently discard any IP frame that fails the checksum. + * RFC1122: 3.2.1.2 MUST silently discard any IP frame that fails the checksum. * * Is the datagram acceptable? * diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 5e3685c5c40..92f90ae46f4 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1468,14 +1468,14 @@ unsigned short ip_rt_frag_needed(struct net *net, struct iphdr *iph, /* BSD 4.2 compatibility hack :-( */ if (mtu == 0 && - old_mtu >= rth->u.dst.metrics[RTAX_MTU-1] && + old_mtu >= dst_metric(&rth->u.dst, RTAX_MTU) && old_mtu >= 68 + (iph->ihl << 2)) old_mtu -= iph->ihl << 2; mtu = guess_mtu(old_mtu); } - if (mtu <= rth->u.dst.metrics[RTAX_MTU-1]) { - if (mtu < rth->u.dst.metrics[RTAX_MTU-1]) { + if (mtu <= dst_metric(&rth->u.dst, RTAX_MTU)) { + if (mtu < dst_metric(&rth->u.dst, RTAX_MTU)) { dst_confirm(&rth->u.dst); if (mtu < ip_rt_min_pmtu) { mtu = ip_rt_min_pmtu; @@ -1497,7 +1497,7 @@ unsigned short ip_rt_frag_needed(struct net *net, struct iphdr *iph, static void ip_rt_update_pmtu(struct dst_entry *dst, u32 mtu) { - if (dst->metrics[RTAX_MTU-1] > mtu && mtu >= 68 && + if (dst_metric(dst, RTAX_MTU) > mtu && mtu >= 68 && !(dst_metric_locked(dst, RTAX_MTU))) { if (mtu < ip_rt_min_pmtu) { mtu = ip_rt_min_pmtu; @@ -1613,7 +1613,7 @@ static void rt_set_nexthop(struct rtable *rt, struct fib_result *res, u32 itag) sizeof(rt->u.dst.metrics)); if (fi->fib_mtu == 0) { rt->u.dst.metrics[RTAX_MTU-1] = rt->u.dst.dev->mtu; - if (rt->u.dst.metrics[RTAX_LOCK-1] & (1 << RTAX_MTU) && + if (dst_metric_locked(&rt->u.dst, RTAX_MTU) && rt->rt_gateway != rt->rt_dst && rt->u.dst.dev->mtu > 576) rt->u.dst.metrics[RTAX_MTU-1] = 576; @@ -1624,14 +1624,14 @@ static void rt_set_nexthop(struct rtable *rt, struct fib_result *res, u32 itag) } else rt->u.dst.metrics[RTAX_MTU-1]= rt->u.dst.dev->mtu; - if (rt->u.dst.metrics[RTAX_HOPLIMIT-1] == 0) + if (dst_metric(&rt->u.dst, RTAX_HOPLIMIT) == 0) rt->u.dst.metrics[RTAX_HOPLIMIT-1] = sysctl_ip_default_ttl; - if (rt->u.dst.metrics[RTAX_MTU-1] > IP_MAX_MTU) + if (dst_metric(&rt->u.dst, RTAX_MTU) > IP_MAX_MTU) rt->u.dst.metrics[RTAX_MTU-1] = IP_MAX_MTU; - if (rt->u.dst.metrics[RTAX_ADVMSS-1] == 0) + if (dst_metric(&rt->u.dst, RTAX_ADVMSS) == 0) rt->u.dst.metrics[RTAX_ADVMSS-1] = max_t(unsigned int, rt->u.dst.dev->mtu - 40, ip_rt_min_advmss); - if (rt->u.dst.metrics[RTAX_ADVMSS-1] > 65535 - 40) + if (dst_metric(&rt->u.dst, RTAX_ADVMSS) > 65535 - 40) rt->u.dst.metrics[RTAX_ADVMSS-1] = 65535 - 40; #ifdef CONFIG_NET_CLS_ROUTE diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index eda4f4a233f..26c936930e9 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -66,6 +66,7 @@ #include <linux/mm.h> #include <linux/module.h> #include <linux/sysctl.h> +#include <net/dst.h> #include <net/tcp.h> #include <net/inet_common.h> #include <linux/ipsec.h> @@ -113,8 +114,6 @@ int sysctl_tcp_abc __read_mostly; #define FLAG_FORWARD_PROGRESS (FLAG_ACKED|FLAG_DATA_SACKED) #define FLAG_ANY_PROGRESS (FLAG_FORWARD_PROGRESS|FLAG_SND_UNA_ADVANCED) -#define IsSackFrto() (sysctl_tcp_frto == 0x2) - #define TCP_REMNANT (TCP_FLAG_FIN|TCP_FLAG_URG|TCP_FLAG_SYN|TCP_FLAG_PSH) #define TCP_HP_BITS (~(TCP_RESERVED_BITS|TCP_FLAG_PSH)) @@ -605,7 +604,7 @@ static u32 tcp_rto_min(struct sock *sk) u32 rto_min = TCP_RTO_MIN; if (dst && dst_metric_locked(dst, RTAX_RTO_MIN)) - rto_min = dst->metrics[RTAX_RTO_MIN - 1]; + rto_min = dst_metric(dst, RTAX_RTO_MIN); return rto_min; } @@ -769,7 +768,7 @@ void tcp_update_metrics(struct sock *sk) dst->metrics[RTAX_RTTVAR - 1] = m; else dst->metrics[RTAX_RTTVAR-1] -= - (dst->metrics[RTAX_RTTVAR-1] - m)>>2; + (dst_metric(dst, RTAX_RTTVAR) - m)>>2; } if (tp->snd_ssthresh >= 0xFFFF) { @@ -788,21 +787,21 @@ void tcp_update_metrics(struct sock *sk) dst->metrics[RTAX_SSTHRESH-1] = max(tp->snd_cwnd >> 1, tp->snd_ssthresh); if (!dst_metric_locked(dst, RTAX_CWND)) - dst->metrics[RTAX_CWND-1] = (dst->metrics[RTAX_CWND-1] + tp->snd_cwnd) >> 1; + dst->metrics[RTAX_CWND-1] = (dst_metric(dst, RTAX_CWND) + tp->snd_cwnd) >> 1; } else { /* Else slow start did not finish, cwnd is non-sense, ssthresh may be also invalid. */ if (!dst_metric_locked(dst, RTAX_CWND)) - dst->metrics[RTAX_CWND-1] = (dst->metrics[RTAX_CWND-1] + tp->snd_ssthresh) >> 1; - if (dst->metrics[RTAX_SSTHRESH-1] && + dst->metrics[RTAX_CWND-1] = (dst_metric(dst, RTAX_CWND) + tp->snd_ssthresh) >> 1; + if (dst_metric(dst, RTAX_SSTHRESH) && !dst_metric_locked(dst, RTAX_SSTHRESH) && - tp->snd_ssthresh > dst->metrics[RTAX_SSTHRESH-1]) + tp->snd_ssthresh > dst_metric(dst, RTAX_SSTHRESH)) dst->metrics[RTAX_SSTHRESH-1] = tp->snd_ssthresh; } if (!dst_metric_locked(dst, RTAX_REORDERING)) { - if (dst->metrics[RTAX_REORDERING-1] < tp->reordering && + if (dst_metric(dst, RTAX_REORDERING) < tp->reordering && tp->reordering != sysctl_tcp_reordering) dst->metrics[RTAX_REORDERING-1] = tp->reordering; } @@ -1685,6 +1684,11 @@ static inline void tcp_reset_reno_sack(struct tcp_sock *tp) tp->sacked_out = 0; } +static int tcp_is_sackfrto(const struct tcp_sock *tp) +{ + return (sysctl_tcp_frto == 0x2) && !tcp_is_reno(tp); +} + /* F-RTO can only be used if TCP has never retransmitted anything other than * head (SACK enhanced variant from Appendix B of RFC4138 is more robust here) */ @@ -1701,7 +1705,7 @@ int tcp_use_frto(struct sock *sk) if (icsk->icsk_mtup.probe_size) return 0; - if (IsSackFrto()) + if (tcp_is_sackfrto(tp)) return 1; /* Avoid expensive walking of rexmit queue if possible */ @@ -1791,7 +1795,7 @@ void tcp_enter_frto(struct sock *sk) /* Earlier loss recovery underway (see RFC4138; Appendix B). * The last condition is necessary at least in tp->frto_counter case. */ - if (IsSackFrto() && (tp->frto_counter || + if (tcp_is_sackfrto(tp) && (tp->frto_counter || ((1 << icsk->icsk_ca_state) & (TCPF_CA_Recovery|TCPF_CA_Loss))) && after(tp->high_seq, tp->snd_una)) { tp->frto_highmark = tp->high_seq; @@ -3123,7 +3127,7 @@ static int tcp_process_frto(struct sock *sk, int flag) return 1; } - if (!IsSackFrto() || tcp_is_reno(tp)) { + if (!tcp_is_sackfrto(tp)) { /* RFC4138 shortcoming in step 2; should also have case c): * ACK isn't duplicate nor advances window, e.g., opposite dir * data, winupdate diff --git a/net/ipv6/route.c b/net/ipv6/route.c index a493ad9b891..12bba088034 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1243,11 +1243,11 @@ install_route: } } - if (rt->u.dst.metrics[RTAX_HOPLIMIT-1] == 0) + if (dst_metric(&rt->u.dst, RTAX_HOPLIMIT) == 0) rt->u.dst.metrics[RTAX_HOPLIMIT-1] = -1; - if (!rt->u.dst.metrics[RTAX_MTU-1]) + if (!dst_metric(&rt->u.dst, RTAX_MTU)) rt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(dev); - if (!rt->u.dst.metrics[RTAX_ADVMSS-1]) + if (!dst_metric(&rt->u.dst, RTAX_ADVMSS)) rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->u.dst)); rt->u.dst.dev = dev; rt->rt6i_idev = idev; diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 9ad4e3631b6..915afadb060 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -1766,6 +1766,7 @@ fail_wep: fail_rate: ieee80211_debugfs_remove_netdev(IEEE80211_DEV_TO_SUB_IF(local->mdev)); unregister_netdevice(local->mdev); + local->mdev = NULL; fail_dev: rtnl_unlock(); sta_info_stop(local); @@ -1773,8 +1774,10 @@ fail_sta_info: debugfs_hw_del(local); destroy_workqueue(local->hw.workqueue); fail_workqueue: - ieee80211_if_free(local->mdev); - local->mdev = NULL; + if (local->mdev != NULL) { + ieee80211_if_free(local->mdev); + local->mdev = NULL; + } fail_mdev_alloc: wiphy_unregister(local->hw.wiphy); return result; diff --git a/net/mac80211/rc80211_pid_debugfs.c b/net/mac80211/rc80211_pid_debugfs.c index ae75d417873..ff5c380f3c1 100644 --- a/net/mac80211/rc80211_pid_debugfs.c +++ b/net/mac80211/rc80211_pid_debugfs.c @@ -85,7 +85,7 @@ static int rate_control_pid_events_open(struct inode *inode, struct file *file) struct rc_pid_sta_info *sinfo = inode->i_private; struct rc_pid_event_buffer *events = &sinfo->events; struct rc_pid_events_file_info *file_info; - unsigned int status; + unsigned long status; /* Allocate a state struct */ file_info = kmalloc(sizeof(*file_info), GFP_KERNEL); @@ -135,7 +135,7 @@ static ssize_t rate_control_pid_events_read(struct file *file, char __user *buf, char pb[RC_PID_PRINT_BUF_SIZE]; int ret; int p; - unsigned int status; + unsigned long status; /* Check if there is something to read. */ if (events->next_entry == file_info->next_entry) { diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index c1fc0f1a641..aa8d80c35e2 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -90,6 +90,7 @@ config NF_CT_PROTO_DCCP tristate 'DCCP protocol connection tracking support (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK depends on NETFILTER_ADVANCED + default IP_DCCP help With this option enabled, the layer 3 independent connection tracking code will be able to do state tracking on DCCP connections. @@ -104,6 +105,7 @@ config NF_CT_PROTO_SCTP tristate 'SCTP protocol connection tracking support (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK depends on NETFILTER_ADVANCED + default IP_SCTP help With this option enabled, the layer 3 independent connection tracking code will be able to do state tracking on SCTP connections. @@ -532,6 +534,7 @@ config NETFILTER_XT_MATCH_DCCP tristate '"dccp" protocol match support' depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED + default IP_DCCP help With this option enabled, you will be able to use the iptables `dccp' match in order to match on DCCP source/destination ports @@ -725,6 +728,7 @@ config NETFILTER_XT_MATCH_SCTP tristate '"sctp" protocol match support (EXPERIMENTAL)' depends on NETFILTER_XTABLES && EXPERIMENTAL depends on NETFILTER_ADVANCED + default IP_SCTP help With this option enabled, you will be able to use the `sctp' match in order to match on SCTP source/destination ports diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 9f490006956..2f9bbc058b4 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -870,6 +870,7 @@ static int process_sdp(struct sk_buff *skb, { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + struct nf_conn_help *help = nfct_help(ct); unsigned int matchoff, matchlen; unsigned int mediaoff, medialen; unsigned int sdpoff; @@ -959,6 +960,9 @@ static int process_sdp(struct sk_buff *skb, if (nf_nat_sdp_session && ct->status & IPS_NAT_MASK) ret = nf_nat_sdp_session(skb, dptr, sdpoff, datalen, &rtp_addr); + if (ret == NF_ACCEPT && i > 0) + help->help.ct_sip_info.invite_cseq = cseq; + return ret; } static int process_invite_response(struct sk_buff *skb, @@ -967,14 +971,14 @@ static int process_invite_response(struct sk_buff *skb, { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + struct nf_conn_help *help = nfct_help(ct); if ((code >= 100 && code <= 199) || (code >= 200 && code <= 299)) return process_sdp(skb, dptr, datalen, cseq); - else { + else if (help->help.ct_sip_info.invite_cseq == cseq) flush_expectations(ct, true); - return NF_ACCEPT; - } + return NF_ACCEPT; } static int process_update_response(struct sk_buff *skb, @@ -983,14 +987,14 @@ static int process_update_response(struct sk_buff *skb, { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + struct nf_conn_help *help = nfct_help(ct); if ((code >= 100 && code <= 199) || (code >= 200 && code <= 299)) return process_sdp(skb, dptr, datalen, cseq); - else { + else if (help->help.ct_sip_info.invite_cseq == cseq) flush_expectations(ct, true); - return NF_ACCEPT; - } + return NF_ACCEPT; } static int process_prack_response(struct sk_buff *skb, @@ -999,14 +1003,14 @@ static int process_prack_response(struct sk_buff *skb, { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + struct nf_conn_help *help = nfct_help(ct); if ((code >= 100 && code <= 199) || (code >= 200 && code <= 299)) return process_sdp(skb, dptr, datalen, cseq); - else { + else if (help->help.ct_sip_info.invite_cseq == cseq) flush_expectations(ct, true); - return NF_ACCEPT; - } + return NF_ACCEPT; } static int process_bye_request(struct sk_buff *skb, diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c index 64b2d136c78..1d421d059ca 100644 --- a/net/sched/act_simple.c +++ b/net/sched/act_simple.c @@ -6,7 +6,7 @@ * as published by the Free Software Foundation; either version * 2 of the License, or (at your option) any later version. * - * Authors: Jamal Hadi Salim (2005) + * Authors: Jamal Hadi Salim (2005-8) * */ @@ -34,6 +34,7 @@ static struct tcf_hashinfo simp_hash_info = { .lock = &simp_lock, }; +#define SIMP_MAX_DATA 32 static int tcf_simp(struct sk_buff *skb, struct tc_action *a, struct tcf_result *res) { struct tcf_defact *d = a->priv; @@ -69,23 +70,28 @@ static int tcf_simp_release(struct tcf_defact *d, int bind) return ret; } -static int alloc_defdata(struct tcf_defact *d, u32 datalen, void *defdata) +static int alloc_defdata(struct tcf_defact *d, char *defdata) { - d->tcfd_defdata = kmemdup(defdata, datalen, GFP_KERNEL); + d->tcfd_defdata = kstrndup(defdata, SIMP_MAX_DATA, GFP_KERNEL); if (unlikely(!d->tcfd_defdata)) return -ENOMEM; - d->tcfd_datalen = datalen; + return 0; } -static int realloc_defdata(struct tcf_defact *d, u32 datalen, void *defdata) +static void reset_policy(struct tcf_defact *d, char *defdata, + struct tc_defact *p) { - kfree(d->tcfd_defdata); - return alloc_defdata(d, datalen, defdata); + spin_lock_bh(&d->tcf_lock); + d->tcf_action = p->action; + memset(d->tcfd_defdata, 0, SIMP_MAX_DATA); + strlcpy(d->tcfd_defdata, defdata, SIMP_MAX_DATA); + spin_unlock_bh(&d->tcf_lock); } static const struct nla_policy simple_policy[TCA_DEF_MAX + 1] = { [TCA_DEF_PARMS] = { .len = sizeof(struct tc_defact) }, + [TCA_DEF_DATA] = { .type = NLA_STRING, .len = SIMP_MAX_DATA }, }; static int tcf_simp_init(struct nlattr *nla, struct nlattr *est, @@ -95,28 +101,24 @@ static int tcf_simp_init(struct nlattr *nla, struct nlattr *est, struct tc_defact *parm; struct tcf_defact *d; struct tcf_common *pc; - void *defdata; - u32 datalen = 0; + char *defdata; int ret = 0, err; if (nla == NULL) return -EINVAL; - err = nla_parse_nested(tb, TCA_DEF_MAX, nla, NULL); + err = nla_parse_nested(tb, TCA_DEF_MAX, nla, simple_policy); if (err < 0) return err; if (tb[TCA_DEF_PARMS] == NULL) return -EINVAL; - parm = nla_data(tb[TCA_DEF_PARMS]); - defdata = nla_data(tb[TCA_DEF_DATA]); - if (defdata == NULL) + if (tb[TCA_DEF_DATA] == NULL) return -EINVAL; - datalen = nla_len(tb[TCA_DEF_DATA]); - if (datalen == 0) - return -EINVAL; + parm = nla_data(tb[TCA_DEF_PARMS]); + defdata = nla_data(tb[TCA_DEF_DATA]); pc = tcf_hash_check(parm->index, a, bind, &simp_hash_info); if (!pc) { @@ -126,11 +128,12 @@ static int tcf_simp_init(struct nlattr *nla, struct nlattr *est, return -ENOMEM; d = to_defact(pc); - ret = alloc_defdata(d, datalen, defdata); + ret = alloc_defdata(d, defdata); if (ret < 0) { kfree(pc); return ret; } + d->tcf_action = parm->action; ret = ACT_P_CREATED; } else { d = to_defact(pc); @@ -138,13 +141,9 @@ static int tcf_simp_init(struct nlattr *nla, struct nlattr *est, tcf_simp_release(d, bind); return -EEXIST; } - realloc_defdata(d, datalen, defdata); + reset_policy(d, defdata, parm); } - spin_lock_bh(&d->tcf_lock); - d->tcf_action = parm->action; - spin_unlock_bh(&d->tcf_lock); - if (ret == ACT_P_CREATED) tcf_hash_insert(pc, &simp_hash_info); return ret; @@ -172,7 +171,7 @@ static inline int tcf_simp_dump(struct sk_buff *skb, struct tc_action *a, opt.bindcnt = d->tcf_bindcnt - bind; opt.action = d->tcf_action; NLA_PUT(skb, TCA_DEF_PARMS, sizeof(opt), &opt); - NLA_PUT(skb, TCA_DEF_DATA, d->tcfd_datalen, d->tcfd_defdata); + NLA_PUT_STRING(skb, TCA_DEF_DATA, d->tcfd_defdata); t.install = jiffies_to_clock_t(jiffies - d->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - d->tcf_tm.lastuse); t.expires = jiffies_to_clock_t(d->tcf_tm.expires); diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 66148cc4759..5bc1ed49018 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -1197,12 +1197,16 @@ static inline int htb_parent_last_child(struct htb_class *cl) return 1; } -static void htb_parent_to_leaf(struct htb_class *cl, struct Qdisc *new_q) +static void htb_parent_to_leaf(struct htb_sched *q, struct htb_class *cl, + struct Qdisc *new_q) { struct htb_class *parent = cl->parent; BUG_TRAP(!cl->level && cl->un.leaf.q && !cl->prio_activity); + if (parent->cmode != HTB_CAN_SEND) + htb_safe_rb_erase(&parent->pq_node, q->wait_pq + parent->level); + parent->level = 0; memset(&parent->un.inner, 0, sizeof(parent->un.inner)); INIT_LIST_HEAD(&parent->un.leaf.drop_list); @@ -1300,7 +1304,7 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg) htb_deactivate(q, cl); if (last_child) - htb_parent_to_leaf(cl, new_q); + htb_parent_to_leaf(q, cl, new_q); if (--cl->refcnt == 0) htb_destroy_class(sch, cl); diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index d74c2d26953..01c7e311b90 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -18,7 +18,6 @@ #include <linux/mm.h> #include <linux/interrupt.h> #include <linux/module.h> -#include <linux/sched.h> #include <linux/sunrpc/types.h> #include <linux/sunrpc/xdr.h> diff --git a/scripts/kconfig/lkc.h b/scripts/kconfig/lkc.h index 4bc68f20a73..96521cb087e 100644 --- a/scripts/kconfig/lkc.h +++ b/scripts/kconfig/lkc.h @@ -11,9 +11,9 @@ #ifndef KBUILD_NO_NLS # include <libintl.h> #else -# define gettext(Msgid) ((const char *) (Msgid)) -# define textdomain(Domainname) ((const char *) (Domainname)) -# define bindtextdomain(Domainname, Dirname) ((const char *) (Dirname)) +static inline const char *gettext(const char *txt) { return txt; } +static inline void textdomain(const char *domainname) {} +static inline void bindtextdomain(const char *name, const char *dir) {} #endif #ifdef __cplusplus diff --git a/scripts/kconfig/mconf.c b/scripts/kconfig/mconf.c index 734cf4f3131..6841e95c098 100644 --- a/scripts/kconfig/mconf.c +++ b/scripts/kconfig/mconf.c @@ -773,7 +773,7 @@ static void conf_string(struct menu *menu) while (1) { int res; - char *heading; + const char *heading; switch (sym_get_type(menu->sym)) { case S_INT: @@ -925,3 +925,4 @@ int main(int ac, char **av) return 0; } + diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c index e04c4218cb5..cea4a790e1e 100644 --- a/scripts/mod/file2alias.c +++ b/scripts/mod/file2alias.c @@ -51,6 +51,15 @@ do { \ sprintf(str + strlen(str), "*"); \ } while(0) +/* Always end in a wildcard, for future extension */ +static inline void add_wildcard(char *str) +{ + int len = strlen(str); + + if (str[len - 1] != '*') + strcat(str + len, "*"); +} + unsigned int cross_build = 0; /** * Check that sizeof(device_id type) are consistent with size of section @@ -133,9 +142,7 @@ static void do_usb_entry(struct usb_device_id *id, id->match_flags&USB_DEVICE_ID_MATCH_INT_PROTOCOL, id->bInterfaceProtocol); - /* Always end in a wildcard, for future extension */ - if (alias[strlen(alias)-1] != '*') - strcat(alias, "*"); + add_wildcard(alias); buf_printf(&mod->dev_table_buf, "MODULE_ALIAS(\"%s\");\n", alias); } @@ -219,6 +226,7 @@ static int do_ieee1394_entry(const char *filename, ADD(alias, "ver", id->match_flags & IEEE1394_MATCH_VERSION, id->version); + add_wildcard(alias); return 1; } @@ -261,6 +269,7 @@ static int do_pci_entry(const char *filename, ADD(alias, "bc", baseclass_mask == 0xFF, baseclass); ADD(alias, "sc", subclass_mask == 0xFF, subclass); ADD(alias, "i", interface_mask == 0xFF, interface); + add_wildcard(alias); return 1; } @@ -283,6 +292,7 @@ static int do_ccw_entry(const char *filename, id->dev_type); ADD(alias, "dm", id->match_flags&CCW_DEVICE_ID_MATCH_DEVICE_MODEL, id->dev_model); + add_wildcard(alias); return 1; } @@ -290,7 +300,7 @@ static int do_ccw_entry(const char *filename, static int do_ap_entry(const char *filename, struct ap_device_id *id, char *alias) { - sprintf(alias, "ap:t%02X", id->dev_type); + sprintf(alias, "ap:t%02X*", id->dev_type); return 1; } @@ -309,6 +319,7 @@ static int do_serio_entry(const char *filename, ADD(alias, "id", id->id != SERIO_ANY, id->id); ADD(alias, "ex", id->extra != SERIO_ANY, id->extra); + add_wildcard(alias); return 1; } @@ -316,7 +327,7 @@ static int do_serio_entry(const char *filename, static int do_acpi_entry(const char *filename, struct acpi_device_id *id, char *alias) { - sprintf(alias, "acpi*:%s:", id->id); + sprintf(alias, "acpi*:%s:*", id->id); return 1; } @@ -324,7 +335,7 @@ static int do_acpi_entry(const char *filename, static int do_pnp_entry(const char *filename, struct pnp_device_id *id, char *alias) { - sprintf(alias, "pnp:d%s", id->id); + sprintf(alias, "pnp:d%s*", id->id); return 1; } @@ -409,6 +420,7 @@ static int do_pcmcia_entry(const char *filename, ADD(alias, "pc", id->match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID3, id->prod_id_hash[2]); ADD(alias, "pd", id->match_flags & PCMCIA_DEV_ID_MATCH_PROD_ID4, id->prod_id_hash[3]); + add_wildcard(alias); return 1; } @@ -432,6 +444,7 @@ static int do_of_entry (const char *filename, struct of_device_id *of, char *ali if (isspace (*tmp)) *tmp = '_'; + add_wildcard(alias); return 1; } @@ -448,6 +461,7 @@ static int do_vio_entry(const char *filename, struct vio_device_id *vio, if (isspace (*tmp)) *tmp = '_'; + add_wildcard(alias); return 1; } @@ -511,6 +525,8 @@ static int do_eisa_entry(const char *filename, struct eisa_device_id *eisa, { if (eisa->sig[0]) sprintf(alias, EISA_DEVICE_MODALIAS_FMT "*", eisa->sig); + else + strcat(alias, "*"); return 1; } @@ -529,6 +545,7 @@ static int do_parisc_entry(const char *filename, struct parisc_device_id *id, ADD(alias, "rev", id->hversion_rev != PA_HVERSION_REV_ANY_ID, id->hversion_rev); ADD(alias, "sv", id->sversion != PA_SVERSION_ANY_ID, id->sversion); + add_wildcard(alias); return 1; } @@ -544,6 +561,7 @@ static int do_sdio_entry(const char *filename, ADD(alias, "c", id->class != (__u8)SDIO_ANY_ID, id->class); ADD(alias, "v", id->vendor != (__u16)SDIO_ANY_ID, id->vendor); ADD(alias, "d", id->device != (__u16)SDIO_ANY_ID, id->device); + add_wildcard(alias); return 1; } @@ -559,6 +577,7 @@ static int do_ssb_entry(const char *filename, ADD(alias, "v", id->vendor != SSB_ANY_VENDOR, id->vendor); ADD(alias, "id", id->coreid != SSB_ANY_ID, id->coreid); ADD(alias, "rev", id->revision != SSB_ANY_REV, id->revision); + add_wildcard(alias); return 1; } @@ -573,6 +592,7 @@ static int do_virtio_entry(const char *filename, struct virtio_device_id *id, ADD(alias, "d", 1, id->device); ADD(alias, "v", id->vendor != VIRTIO_DEV_ANY_ID, id->vendor); + add_wildcard(alias); return 1; } @@ -612,9 +632,6 @@ static void do_table(void *symval, unsigned long size, for (i = 0; i < size; i += id_size) { if (do_entry(mod->name, symval+i, alias)) { - /* Always end in a wildcard, for future extension */ - if (alias[strlen(alias)-1] != '*') - strcat(alias, "*"); buf_printf(&mod->dev_table_buf, "MODULE_ALIAS(\"%s\");\n", alias); } diff --git a/sound/drivers/Kconfig b/sound/drivers/Kconfig index a78a8d04517..379bcb07446 100644 --- a/sound/drivers/Kconfig +++ b/sound/drivers/Kconfig @@ -5,8 +5,8 @@ menu "Generic devices" config SND_PCSP - tristate "Internal PC speaker support" - depends on X86_PC && HIGH_RES_TIMERS + tristate "PC-Speaker support" + depends on PCSPKR_PLATFORM && X86_PC && HIGH_RES_TIMERS depends on INPUT depends on SND select SND_PCM diff --git a/sound/drivers/pcsp/pcsp.c b/sound/drivers/pcsp/pcsp.c index 59203511e77..54a1f9036c6 100644 --- a/sound/drivers/pcsp/pcsp.c +++ b/sound/drivers/pcsp/pcsp.c @@ -194,6 +194,7 @@ static void pcsp_stop_beep(struct snd_pcsp *chip) spin_unlock_irq(&chip->substream_lock); } +#ifdef CONFIG_PM static int pcsp_suspend(struct platform_device *dev, pm_message_t state) { struct snd_pcsp *chip = platform_get_drvdata(dev); @@ -201,6 +202,9 @@ static int pcsp_suspend(struct platform_device *dev, pm_message_t state) snd_pcm_suspend_all(chip->pcm); return 0; } +#else +#define pcsp_suspend NULL +#endif /* CONFIG_PM */ static void pcsp_shutdown(struct platform_device *dev) { diff --git a/sound/oss/kahlua.c b/sound/oss/kahlua.c index dfe670f12e6..eb9bc365530 100644 --- a/sound/oss/kahlua.c +++ b/sound/oss/kahlua.c @@ -67,7 +67,7 @@ static int __devinit probe_one(struct pci_dev *pdev, const struct pci_device_id return 1; mem = ioremap(base, 128); - if(mem == 0UL) + if (!mem) return 1; map = readw(mem + 0x18); /* Read the SMI enables */ iounmap(mem); diff --git a/sound/pci/Kconfig b/sound/pci/Kconfig index 581debf37dc..7e474210957 100644 --- a/sound/pci/Kconfig +++ b/sound/pci/Kconfig @@ -515,19 +515,16 @@ config SND_FM801 config SND_FM801_TEA575X_BOOL bool "ForteMedia FM801 + TEA5757 tuner" depends on SND_FM801 + depends on VIDEO_V4L1=y || VIDEO_V4L1=SND_FM801 help Say Y here to include support for soundcards based on the ForteMedia FM801 chip with a TEA5757 tuner connected to GPIO1-3 pins (Media Forte SF256-PCS-02) into the snd-fm801 driver. - This will enable support for the old V4L1 API. - config SND_FM801_TEA575X tristate depends on SND_FM801_TEA575X_BOOL default SND_FM801 - select VIDEO_V4L1 - select VIDEO_DEV config SND_HDA_INTEL tristate "Intel HD Audio" diff --git a/sound/pci/ac97/ac97_patch.c b/sound/pci/ac97/ac97_patch.c index 39198e505b1..2da89810ca1 100644 --- a/sound/pci/ac97/ac97_patch.c +++ b/sound/pci/ac97/ac97_patch.c @@ -3446,6 +3446,7 @@ static const struct snd_kcontrol_new snd_ac97_controls_vt1617a[] = { int patch_vt1617a(struct snd_ac97 * ac97) { int err = 0; + int val; /* we choose to not fail out at this point, but we tell the caller when we return */ @@ -3456,7 +3457,13 @@ int patch_vt1617a(struct snd_ac97 * ac97) /* bring analog power consumption to normal by turning off the * headphone amplifier, like WinXP driver for EPIA SP */ - snd_ac97_write_cache(ac97, 0x5c, 0x20); + /* We need to check the bit before writing it. + * On some (many?) hardwares, setting bit actually clears it! + */ + val = snd_ac97_read(ac97, 0x5c); + if (!(val & 0x20)) + snd_ac97_write_cache(ac97, 0x5c, 0x20); + ac97->ext_id |= AC97_EI_SPDIF; /* force the detection of spdif */ ac97->rates[AC97_RATES_SPDIF] = SNDRV_PCM_RATE_44100 | SNDRV_PCM_RATE_48000; ac97->build_ops = &patch_vt1616_ops; diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index d9783a4263e..6d4df45e81e 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -11902,7 +11902,10 @@ static void alc861_auto_set_output_and_unmute(struct hda_codec *codec, hda_nid_t nid, int pin_type, int dac_idx) { - alc_set_pin_output(codec, nid, pin_type); + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, + pin_type); + snd_hda_codec_write(codec, dac_idx, 0, AC_VERB_SET_AMP_GAIN_MUTE, + AMP_OUT_UNMUTE); } static void alc861_auto_init_multi_out(struct hda_codec *codec) diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index b3a15d61687..393f7fd2b1b 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -4289,6 +4289,8 @@ struct hda_codec_preset snd_hda_preset_sigmatel[] = { { .id = 0x83847635, .name = "STAC9250D", .patch = patch_stac925x }, { .id = 0x83847636, .name = "STAC9251", .patch = patch_stac925x }, { .id = 0x83847637, .name = "STAC9250D", .patch = patch_stac925x }, + { .id = 0x83847645, .name = "92HD206X", .patch = patch_stac927x }, + { .id = 0x83847646, .name = "92HD206D", .patch = patch_stac927x }, /* The following does not take into account .id=0x83847661 when subsys = * 104D0C00 which is STAC9225s. Because of this, some SZ Notebooks are * currently not fully supported. diff --git a/sound/soc/at91/at91-pcm.c b/sound/soc/at91/at91-pcm.c index 67c88e322fb..ccac6bd2889 100644 --- a/sound/soc/at91/at91-pcm.c +++ b/sound/soc/at91/at91-pcm.c @@ -103,7 +103,8 @@ static void at91_pcm_dma_irq(u32 ssc_sr, if (prtd->period_ptr >= prtd->dma_buffer_end) { prtd->period_ptr = prtd->dma_buffer; } - at91_ssc_write(params->ssc_base + params->pdc->xnpr, prtd->period_ptr); + at91_ssc_write(params->ssc_base + params->pdc->xnpr, + prtd->period_ptr); at91_ssc_write(params->ssc_base + params->pdc->xncr, prtd->period_size / params->pdc_xfer_size); } @@ -191,10 +192,12 @@ static int at91_pcm_trigger(struct snd_pcm_substream *substream, at91_ssc_write(params->ssc_base + AT91_SSC_IER, params->mask->ssc_endx | params->mask->ssc_endbuf); - at91_ssc_write(params->ssc_base + ATMEL_PDC_PTCR, params->mask->pdc_enable); + at91_ssc_write(params->ssc_base + ATMEL_PDC_PTCR, + params->mask->pdc_enable); - DBG("sr=%lx imr=%lx\n", at91_ssc_read(params->ssc_base + AT91_SSC_SR), - at91_ssc_read(params->ssc_base + AT91_SSC_IER)); + DBG("sr=%lx imr=%lx\n", + at91_ssc_read(params->ssc_base + AT91_SSC_SR), + at91_ssc_read(params->ssc_base + AT91_SSC_IMR)); break; case SNDRV_PCM_TRIGGER_STOP: diff --git a/sound/soc/at91/at91-ssc.c b/sound/soc/at91/at91-ssc.c index f642d2dd4ec..bc35d00a38f 100644 --- a/sound/soc/at91/at91-ssc.c +++ b/sound/soc/at91/at91-ssc.c @@ -590,7 +590,7 @@ static int at91_ssc_hw_params(struct snd_pcm_substream *substream, printk(KERN_WARNING "at91-ssc: request_irq failure\n"); DBG("Stopping pid %d clock\n", ssc_p->ssc.pid); - at91_sys_write(AT91_PMC_PCER, 1<<ssc_p->ssc.pid); + at91_sys_write(AT91_PMC_PCDR, 1<<ssc_p->ssc.pid); return ret; } diff --git a/sound/soc/s3c24xx/s3c24xx-i2s.c b/sound/soc/s3c24xx/s3c24xx-i2s.c index 4ebcd6a8bf2..1ed6afd4545 100644 --- a/sound/soc/s3c24xx/s3c24xx-i2s.c +++ b/sound/soc/s3c24xx/s3c24xx-i2s.c @@ -224,6 +224,7 @@ static int s3c24xx_i2s_set_fmt(struct snd_soc_cpu_dai *cpu_dai, iismod |= S3C2410_IISMOD_SLAVE; break; case SND_SOC_DAIFMT_CBS_CFS: + iismod &= ~S3C2410_IISMOD_SLAVE; break; default: return -EINVAL; @@ -234,6 +235,7 @@ static int s3c24xx_i2s_set_fmt(struct snd_soc_cpu_dai *cpu_dai, iismod |= S3C2410_IISMOD_MSB; break; case SND_SOC_DAIFMT_I2S: + iismod &= ~S3C2410_IISMOD_MSB; break; default: return -EINVAL; diff --git a/sound/soc/s3c24xx/s3c24xx-pcm.c b/sound/soc/s3c24xx/s3c24xx-pcm.c index 6c70a81c730..7806ae61461 100644 --- a/sound/soc/s3c24xx/s3c24xx-pcm.c +++ b/sound/soc/s3c24xx/s3c24xx-pcm.c @@ -171,7 +171,7 @@ static int s3c24xx_pcm_hw_params(struct snd_pcm_substream *substream, ret = s3c2410_dma_request(prtd->params->channel, prtd->params->client, NULL); - if (ret) { + if (ret < 0) { DBG(KERN_ERR "failed to get dma channel\n"); return ret; } diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index e89338e2b04..f7ba099049e 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -522,6 +522,7 @@ unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn) return bad_hva(); return (slot->userspace_addr + (gfn - slot->base_gfn) * PAGE_SIZE); } +EXPORT_SYMBOL_GPL(gfn_to_hva); /* * Requires current->mm->mmap_sem to be held |