diff options
123 files changed, 2433 insertions, 1016 deletions
diff --git a/Documentation/ABI/testing/sysfs-bus-pci-devices-cciss b/Documentation/ABI/testing/sysfs-bus-pci-devices-cciss index 0a92a7c93a6..4f29e5f1ebf 100644 --- a/Documentation/ABI/testing/sysfs-bus-pci-devices-cciss +++ b/Documentation/ABI/testing/sysfs-bus-pci-devices-cciss @@ -31,3 +31,31 @@ Date: March 2009 Kernel Version: 2.6.30 Contact: iss_storagedev@hp.com Description: A symbolic link to /sys/block/cciss!cXdY + +Where: /sys/bus/pci/devices/<dev>/ccissX/rescan +Date: August 2009 +Kernel Version: 2.6.31 +Contact: iss_storagedev@hp.com +Description: Kicks of a rescan of the controller to discover logical + drive topology changes. + +Where: /sys/bus/pci/devices/<dev>/ccissX/cXdY/lunid +Date: August 2009 +Kernel Version: 2.6.31 +Contact: iss_storagedev@hp.com +Description: Displays the 8-byte LUN ID used to address logical + drive Y of controller X. + +Where: /sys/bus/pci/devices/<dev>/ccissX/cXdY/raid_level +Date: August 2009 +Kernel Version: 2.6.31 +Contact: iss_storagedev@hp.com +Description: Displays the RAID level of logical drive Y of + controller X. + +Where: /sys/bus/pci/devices/<dev>/ccissX/cXdY/usage_count +Date: August 2009 +Kernel Version: 2.6.31 +Contact: iss_storagedev@hp.com +Description: Displays the usage count (number of opens) of logical drive Y + of controller X. diff --git a/Documentation/hwmon/ltc4215 b/Documentation/hwmon/ltc4215 index 2e6a21eb656..c196a184625 100644 --- a/Documentation/hwmon/ltc4215 +++ b/Documentation/hwmon/ltc4215 @@ -22,12 +22,13 @@ Usage Notes ----------- This driver does not probe for LTC4215 devices, due to the fact that some -of the possible addresses are unfriendly to probing. You will need to use -the "force" parameter to tell the driver where to find the device. +of the possible addresses are unfriendly to probing. You will have to +instantiate the devices explicitly. Example: the following will load the driver for an LTC4215 at address 0x44 on I2C bus #0: -$ modprobe ltc4215 force=0,0x44 +$ modprobe ltc4215 +$ echo ltc4215 0x44 > /sys/bus/i2c/devices/i2c-0/new_device Sysfs entries diff --git a/Documentation/hwmon/ltc4245 b/Documentation/hwmon/ltc4245 index bae7a3adc5d..02838a47d86 100644 --- a/Documentation/hwmon/ltc4245 +++ b/Documentation/hwmon/ltc4245 @@ -23,12 +23,13 @@ Usage Notes ----------- This driver does not probe for LTC4245 devices, due to the fact that some -of the possible addresses are unfriendly to probing. You will need to use -the "force" parameter to tell the driver where to find the device. +of the possible addresses are unfriendly to probing. You will have to +instantiate the devices explicitly. Example: the following will load the driver for an LTC4245 at address 0x23 on I2C bus #1: -$ modprobe ltc4245 force=1,0x23 +$ modprobe ltc4245 +$ echo ltc4245 0x23 > /sys/bus/i2c/devices/i2c-1/new_device Sysfs entries diff --git a/Documentation/i2c/instantiating-devices b/Documentation/i2c/instantiating-devices index c740b7b4108..e89490270ab 100644 --- a/Documentation/i2c/instantiating-devices +++ b/Documentation/i2c/instantiating-devices @@ -188,7 +188,7 @@ segment, the address is sufficient to uniquely identify the device to be deleted. Example: -# echo eeprom 0x50 > /sys/class/i2c-adapter/i2c-3/new_device +# echo eeprom 0x50 > /sys/bus/i2c/devices/i2c-3/new_device While this interface should only be used when in-kernel device declaration can't be done, there is a variety of cases where it can be helpful: diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 6fa7292947e..9107b387e91 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -671,6 +671,7 @@ and is between 256 and 4096 characters. It is defined in the file earlyprintk= [X86,SH,BLACKFIN] earlyprintk=vga earlyprintk=serial[,ttySn[,baudrate]] + earlyprintk=ttySn[,baudrate] earlyprintk=dbgp[debugController#] Append ",keep" to not disable it when the real console diff --git a/Documentation/i2c/chips/eeprom b/Documentation/misc-devices/eeprom index f7e8104b576..f7e8104b576 100644 --- a/Documentation/i2c/chips/eeprom +++ b/Documentation/misc-devices/eeprom diff --git a/Documentation/i2c/chips/max6875 b/Documentation/misc-devices/max6875 index 10ca43cd1a7..1e89ee3ccc1 100644 --- a/Documentation/i2c/chips/max6875 +++ b/Documentation/misc-devices/max6875 @@ -42,10 +42,12 @@ General Remarks Valid addresses for the MAX6875 are 0x50 and 0x52. Valid addresses for the MAX6874 are 0x50, 0x52, 0x54 and 0x56. -The driver does not probe any address, so you must force the address. +The driver does not probe any address, so you explicitly instantiate the +devices. Example: -$ modprobe max6875 force=0,0x50 +$ modprobe max6875 +$ echo max6875 0x50 > /sys/bus/i2c/devices/i2c-0/new_device The MAX6874/MAX6875 ignores address bit 0, so this driver attaches to multiple addresses. For example, for address 0x50, it also reserves 0x51. diff --git a/Documentation/w1/masters/ds2482 b/Documentation/w1/masters/ds2482 index 9210d6fa502..299b91c7609 100644 --- a/Documentation/w1/masters/ds2482 +++ b/Documentation/w1/masters/ds2482 @@ -24,8 +24,8 @@ General Remarks Valid addresses are 0x18, 0x19, 0x1a, and 0x1b. However, the device cannot be detected without writing to the i2c bus, so no -detection is done. -You should force the device address. +detection is done. You should instantiate the device explicitly. -$ modprobe ds2482 force=0,0x18 +$ modprobe ds2482 +$ echo ds2482 0x18 > /sys/bus/i2c/devices/i2c-0/new_device diff --git a/MAINTAINERS b/MAINTAINERS index 737a9b2c532..09a2028bab7 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -257,6 +257,13 @@ W: http://www.lesswatts.org/projects/acpi/ S: Supported F: drivers/acpi/fan.c +ACPI PROCESSOR AGGREGATOR DRIVER +M: Shaohua Li <shaohua.li@intel.com> +L: linux-acpi@vger.kernel.org +W: http://www.lesswatts.org/projects/acpi/ +S: Supported +F: drivers/acpi/acpi_pad.c + ACPI THERMAL DRIVER M: Zhang Rui <rui.zhang@intel.com> L: linux-acpi@vger.kernel.org @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 32 -EXTRAVERSION = -rc2 +EXTRAVERSION = -rc3 NAME = Man-Eating Seals of Antiquity # *DOCUMENTATION* diff --git a/arch/m68knommu/kernel/asm-offsets.c b/arch/m68knommu/kernel/asm-offsets.c index 594ee0e657f..9a8876f715d 100644 --- a/arch/m68knommu/kernel/asm-offsets.c +++ b/arch/m68knommu/kernel/asm-offsets.c @@ -45,25 +45,25 @@ int main(void) DEFINE(THREAD_FPSTATE, offsetof(struct thread_struct, fpstate)); /* offsets into the pt_regs */ - DEFINE(PT_D0, offsetof(struct pt_regs, d0)); - DEFINE(PT_ORIG_D0, offsetof(struct pt_regs, orig_d0)); - DEFINE(PT_D1, offsetof(struct pt_regs, d1)); - DEFINE(PT_D2, offsetof(struct pt_regs, d2)); - DEFINE(PT_D3, offsetof(struct pt_regs, d3)); - DEFINE(PT_D4, offsetof(struct pt_regs, d4)); - DEFINE(PT_D5, offsetof(struct pt_regs, d5)); - DEFINE(PT_A0, offsetof(struct pt_regs, a0)); - DEFINE(PT_A1, offsetof(struct pt_regs, a1)); - DEFINE(PT_A2, offsetof(struct pt_regs, a2)); - DEFINE(PT_PC, offsetof(struct pt_regs, pc)); - DEFINE(PT_SR, offsetof(struct pt_regs, sr)); + DEFINE(PT_OFF_D0, offsetof(struct pt_regs, d0)); + DEFINE(PT_OFF_ORIG_D0, offsetof(struct pt_regs, orig_d0)); + DEFINE(PT_OFF_D1, offsetof(struct pt_regs, d1)); + DEFINE(PT_OFF_D2, offsetof(struct pt_regs, d2)); + DEFINE(PT_OFF_D3, offsetof(struct pt_regs, d3)); + DEFINE(PT_OFF_D4, offsetof(struct pt_regs, d4)); + DEFINE(PT_OFF_D5, offsetof(struct pt_regs, d5)); + DEFINE(PT_OFF_A0, offsetof(struct pt_regs, a0)); + DEFINE(PT_OFF_A1, offsetof(struct pt_regs, a1)); + DEFINE(PT_OFF_A2, offsetof(struct pt_regs, a2)); + DEFINE(PT_OFF_PC, offsetof(struct pt_regs, pc)); + DEFINE(PT_OFF_SR, offsetof(struct pt_regs, sr)); #ifdef CONFIG_COLDFIRE /* bitfields are a bit difficult */ - DEFINE(PT_FORMATVEC, offsetof(struct pt_regs, sr) - 2); + DEFINE(PT_OFF_FORMATVEC, offsetof(struct pt_regs, sr) - 2); #else /* bitfields are a bit difficult */ - DEFINE(PT_VECTOR, offsetof(struct pt_regs, pc) + 4); + DEFINE(PT_OFF_VECTOR, offsetof(struct pt_regs, pc) + 4); #endif /* signal defines */ diff --git a/arch/m68knommu/kernel/entry.S b/arch/m68knommu/kernel/entry.S index f56faa5c9cd..56043ade394 100644 --- a/arch/m68knommu/kernel/entry.S +++ b/arch/m68knommu/kernel/entry.S @@ -46,7 +46,7 @@ ENTRY(buserr) SAVE_ALL moveq #-1,%d0 - movel %d0,%sp@(PT_ORIG_D0) + movel %d0,%sp@(PT_OFF_ORIG_D0) movel %sp,%sp@- /* stack frame pointer argument */ jsr buserr_c addql #4,%sp @@ -55,7 +55,7 @@ ENTRY(buserr) ENTRY(trap) SAVE_ALL moveq #-1,%d0 - movel %d0,%sp@(PT_ORIG_D0) + movel %d0,%sp@(PT_OFF_ORIG_D0) movel %sp,%sp@- /* stack frame pointer argument */ jsr trap_c addql #4,%sp @@ -67,7 +67,7 @@ ENTRY(trap) ENTRY(dbginterrupt) SAVE_ALL moveq #-1,%d0 - movel %d0,%sp@(PT_ORIG_D0) + movel %d0,%sp@(PT_OFF_ORIG_D0) movel %sp,%sp@- /* stack frame pointer argument */ jsr dbginterrupt_c addql #4,%sp diff --git a/arch/m68knommu/mm/init.c b/arch/m68knommu/mm/init.c index b1703c67a4f..f3236d0b522 100644 --- a/arch/m68knommu/mm/init.c +++ b/arch/m68knommu/mm/init.c @@ -162,7 +162,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) totalram_pages++; pages++; } - printk (KERN_NOTICE "Freeing initrd memory: %dk freed\n", pages); + printk (KERN_NOTICE "Freeing initrd memory: %dk freed\n", pages * (PAGE_SIZE / 1024)); } #endif diff --git a/arch/m68knommu/platform/5206e/config.c b/arch/m68knommu/platform/5206e/config.c index 0f41ba82a3b..942397984c6 100644 --- a/arch/m68knommu/platform/5206e/config.c +++ b/arch/m68knommu/platform/5206e/config.c @@ -17,7 +17,6 @@ #include <asm/mcfsim.h> #include <asm/mcfuart.h> #include <asm/mcfdma.h> -#include <asm/mcfuart.h> /***************************************************************************/ diff --git a/arch/m68knommu/platform/68328/entry.S b/arch/m68knommu/platform/68328/entry.S index b1aef72f3ba..9d80d2c4286 100644 --- a/arch/m68knommu/platform/68328/entry.S +++ b/arch/m68knommu/platform/68328/entry.S @@ -39,17 +39,17 @@ .globl inthandler7 badsys: - movel #-ENOSYS,%sp@(PT_D0) + movel #-ENOSYS,%sp@(PT_OFF_D0) jra ret_from_exception do_trace: - movel #-ENOSYS,%sp@(PT_D0) /* needed for strace*/ + movel #-ENOSYS,%sp@(PT_OFF_D0) /* needed for strace*/ subql #4,%sp SAVE_SWITCH_STACK jbsr syscall_trace RESTORE_SWITCH_STACK addql #4,%sp - movel %sp@(PT_ORIG_D0),%d1 + movel %sp@(PT_OFF_ORIG_D0),%d1 movel #-ENOSYS,%d0 cmpl #NR_syscalls,%d1 jcc 1f @@ -57,7 +57,7 @@ do_trace: lea sys_call_table, %a0 jbsr %a0@(%d1) -1: movel %d0,%sp@(PT_D0) /* save the return value */ +1: movel %d0,%sp@(PT_OFF_D0) /* save the return value */ subql #4,%sp /* dummy return address */ SAVE_SWITCH_STACK jbsr syscall_trace @@ -75,7 +75,7 @@ ENTRY(system_call) jbsr set_esp0 addql #4,%sp - movel %sp@(PT_ORIG_D0),%d0 + movel %sp@(PT_OFF_ORIG_D0),%d0 movel %sp,%d1 /* get thread_info pointer */ andl #-THREAD_SIZE,%d1 @@ -88,10 +88,10 @@ ENTRY(system_call) lea sys_call_table,%a0 movel %a0@(%d0), %a0 jbsr %a0@ - movel %d0,%sp@(PT_D0) /* save the return value*/ + movel %d0,%sp@(PT_OFF_D0) /* save the return value*/ ret_from_exception: - btst #5,%sp@(PT_SR) /* check if returning to kernel*/ + btst #5,%sp@(PT_OFF_SR) /* check if returning to kernel*/ jeq Luser_return /* if so, skip resched, signals*/ Lkernel_return: @@ -133,7 +133,7 @@ Lreturn: */ inthandler1: SAVE_ALL - movew %sp@(PT_VECTOR), %d0 + movew %sp@(PT_OFF_VECTOR), %d0 and #0x3ff, %d0 movel %sp,%sp@- @@ -144,7 +144,7 @@ inthandler1: inthandler2: SAVE_ALL - movew %sp@(PT_VECTOR), %d0 + movew %sp@(PT_OFF_VECTOR), %d0 and #0x3ff, %d0 movel %sp,%sp@- @@ -155,7 +155,7 @@ inthandler2: inthandler3: SAVE_ALL - movew %sp@(PT_VECTOR), %d0 + movew %sp@(PT_OFF_VECTOR), %d0 and #0x3ff, %d0 movel %sp,%sp@- @@ -166,7 +166,7 @@ inthandler3: inthandler4: SAVE_ALL - movew %sp@(PT_VECTOR), %d0 + movew %sp@(PT_OFF_VECTOR), %d0 and #0x3ff, %d0 movel %sp,%sp@- @@ -177,7 +177,7 @@ inthandler4: inthandler5: SAVE_ALL - movew %sp@(PT_VECTOR), %d0 + movew %sp@(PT_OFF_VECTOR), %d0 and #0x3ff, %d0 movel %sp,%sp@- @@ -188,7 +188,7 @@ inthandler5: inthandler6: SAVE_ALL - movew %sp@(PT_VECTOR), %d0 + movew %sp@(PT_OFF_VECTOR), %d0 and #0x3ff, %d0 movel %sp,%sp@- @@ -199,7 +199,7 @@ inthandler6: inthandler7: SAVE_ALL - movew %sp@(PT_VECTOR), %d0 + movew %sp@(PT_OFF_VECTOR), %d0 and #0x3ff, %d0 movel %sp,%sp@- @@ -210,7 +210,7 @@ inthandler7: inthandler: SAVE_ALL - movew %sp@(PT_VECTOR), %d0 + movew %sp@(PT_OFF_VECTOR), %d0 and #0x3ff, %d0 movel %sp,%sp@- @@ -224,7 +224,7 @@ ret_from_interrupt: 2: RESTORE_ALL 1: - moveb %sp@(PT_SR), %d0 + moveb %sp@(PT_OFF_SR), %d0 and #7, %d0 jhi 2b diff --git a/arch/m68knommu/platform/68360/entry.S b/arch/m68knommu/platform/68360/entry.S index 55dfefe3864..6d3460a39ca 100644 --- a/arch/m68knommu/platform/68360/entry.S +++ b/arch/m68knommu/platform/68360/entry.S @@ -35,17 +35,17 @@ .globl inthandler badsys: - movel #-ENOSYS,%sp@(PT_D0) + movel #-ENOSYS,%sp@(PT_OFF_D0) jra ret_from_exception do_trace: - movel #-ENOSYS,%sp@(PT_D0) /* needed for strace*/ + movel #-ENOSYS,%sp@(PT_OFF_D0) /* needed for strace*/ subql #4,%sp SAVE_SWITCH_STACK jbsr syscall_trace RESTORE_SWITCH_STACK addql #4,%sp - movel %sp@(PT_ORIG_D0),%d1 + movel %sp@(PT_OFF_ORIG_D0),%d1 movel #-ENOSYS,%d0 cmpl #NR_syscalls,%d1 jcc 1f @@ -53,7 +53,7 @@ do_trace: lea sys_call_table, %a0 jbsr %a0@(%d1) -1: movel %d0,%sp@(PT_D0) /* save the return value */ +1: movel %d0,%sp@(PT_OFF_D0) /* save the return value */ subql #4,%sp /* dummy return address */ SAVE_SWITCH_STACK jbsr syscall_trace @@ -79,10 +79,10 @@ ENTRY(system_call) lea sys_call_table,%a0 movel %a0@(%d0), %a0 jbsr %a0@ - movel %d0,%sp@(PT_D0) /* save the return value*/ + movel %d0,%sp@(PT_OFF_D0) /* save the return value*/ ret_from_exception: - btst #5,%sp@(PT_SR) /* check if returning to kernel*/ + btst #5,%sp@(PT_OFF_SR) /* check if returning to kernel*/ jeq Luser_return /* if so, skip resched, signals*/ Lkernel_return: @@ -124,7 +124,7 @@ Lreturn: */ inthandler: SAVE_ALL - movew %sp@(PT_VECTOR), %d0 + movew %sp@(PT_OFF_VECTOR), %d0 and.l #0x3ff, %d0 lsr.l #0x02, %d0 @@ -139,7 +139,7 @@ ret_from_interrupt: 2: RESTORE_ALL 1: - moveb %sp@(PT_SR), %d0 + moveb %sp@(PT_OFF_SR), %d0 and #7, %d0 jhi 2b /* check if we need to do software interrupts */ diff --git a/arch/m68knommu/platform/coldfire/entry.S b/arch/m68knommu/platform/coldfire/entry.S index 3b471c0da24..dd7d591f70e 100644 --- a/arch/m68knommu/platform/coldfire/entry.S +++ b/arch/m68knommu/platform/coldfire/entry.S @@ -81,11 +81,11 @@ ENTRY(system_call) movel %d3,%a0 jbsr %a0@ - movel %d0,%sp@(PT_D0) /* save the return value */ + movel %d0,%sp@(PT_OFF_D0) /* save the return value */ jra ret_from_exception 1: - movel #-ENOSYS,%d2 /* strace needs -ENOSYS in PT_D0 */ - movel %d2,PT_D0(%sp) /* on syscall entry */ + movel #-ENOSYS,%d2 /* strace needs -ENOSYS in PT_OFF_D0 */ + movel %d2,PT_OFF_D0(%sp) /* on syscall entry */ subql #4,%sp SAVE_SWITCH_STACK jbsr syscall_trace @@ -93,7 +93,7 @@ ENTRY(system_call) addql #4,%sp movel %d3,%a0 jbsr %a0@ - movel %d0,%sp@(PT_D0) /* save the return value */ + movel %d0,%sp@(PT_OFF_D0) /* save the return value */ subql #4,%sp /* dummy return address */ SAVE_SWITCH_STACK jbsr syscall_trace @@ -104,7 +104,7 @@ ret_from_signal: ret_from_exception: move #0x2700,%sr /* disable intrs */ - btst #5,%sp@(PT_SR) /* check if returning to kernel */ + btst #5,%sp@(PT_OFF_SR) /* check if returning to kernel */ jeq Luser_return /* if so, skip resched, signals */ #ifdef CONFIG_PREEMPT @@ -142,8 +142,8 @@ Luser_return: Lreturn: move #0x2700,%sr /* disable intrs */ movel sw_usp,%a0 /* get usp */ - movel %sp@(PT_PC),%a0@- /* copy exception program counter */ - movel %sp@(PT_FORMATVEC),%a0@-/* copy exception format/vector/sr */ + movel %sp@(PT_OFF_PC),%a0@- /* copy exception program counter */ + movel %sp@(PT_OFF_FORMATVEC),%a0@- /* copy exception format/vector/sr */ moveml %sp@,%d1-%d5/%a0-%a2 lea %sp@(32),%sp /* space for 8 regs */ movel %sp@+,%d0 @@ -181,9 +181,9 @@ Lsignal_return: ENTRY(inthandler) SAVE_ALL moveq #-1,%d0 - movel %d0,%sp@(PT_ORIG_D0) + movel %d0,%sp@(PT_OFF_ORIG_D0) - movew %sp@(PT_FORMATVEC),%d0 /* put exception # in d0 */ + movew %sp@(PT_OFF_FORMATVEC),%d0 /* put exception # in d0 */ andl #0x03fc,%d0 /* mask out vector only */ movel %sp,%sp@- /* push regs arg */ @@ -203,7 +203,7 @@ ENTRY(inthandler) ENTRY(fasthandler) SAVE_LOCAL - movew %sp@(PT_FORMATVEC),%d0 + movew %sp@(PT_OFF_FORMATVEC),%d0 andl #0x03fc,%d0 /* mask out vector only */ movel %sp,%sp@- /* push regs arg */ diff --git a/arch/microblaze/kernel/entry.S b/arch/microblaze/kernel/entry.S index acc1f05d1e2..e3ecb36dd55 100644 --- a/arch/microblaze/kernel/entry.S +++ b/arch/microblaze/kernel/entry.S @@ -592,6 +592,8 @@ C_ENTRY(full_exception_trap): nop mfs r7, rfsr; /* save FSR */ nop + mts rfsr, r0; /* Clear sticky fsr */ + nop la r12, r0, full_exception set_vms; rtbd r12, 0; diff --git a/arch/microblaze/kernel/hw_exception_handler.S b/arch/microblaze/kernel/hw_exception_handler.S index 6b0288ebccd..2b86c03aa84 100644 --- a/arch/microblaze/kernel/hw_exception_handler.S +++ b/arch/microblaze/kernel/hw_exception_handler.S @@ -384,7 +384,7 @@ handle_other_ex: /* Handle Other exceptions here */ addk r8, r17, r0; /* Load exception address */ bralid r15, full_exception; /* Branch to the handler */ nop; - mts r0, rfsr; /* Clear sticky fsr */ + mts rfsr, r0; /* Clear sticky fsr */ nop /* diff --git a/arch/microblaze/kernel/process.c b/arch/microblaze/kernel/process.c index 4201c743cc9..c592d475b3d 100644 --- a/arch/microblaze/kernel/process.c +++ b/arch/microblaze/kernel/process.c @@ -235,7 +235,9 @@ void start_thread(struct pt_regs *regs, unsigned long pc, unsigned long usp) regs->pc = pc; regs->r1 = usp; regs->pt_mode = 0; +#ifdef CONFIG_MMU regs->msr |= MSR_UMS; +#endif } #ifdef CONFIG_MMU diff --git a/arch/sparc/include/asm/hardirq_32.h b/arch/sparc/include/asm/hardirq_32.h index 4f63ed8df55..162007643cd 100644 --- a/arch/sparc/include/asm/hardirq_32.h +++ b/arch/sparc/include/asm/hardirq_32.h @@ -7,17 +7,7 @@ #ifndef __SPARC_HARDIRQ_H #define __SPARC_HARDIRQ_H -#include <linux/threads.h> -#include <linux/spinlock.h> -#include <linux/cache.h> - -/* entry.S is sensitive to the offsets of these fields */ /* XXX P3 Is it? */ -typedef struct { - unsigned int __softirq_pending; -} ____cacheline_aligned irq_cpustat_t; - -#include <linux/irq_cpustat.h> /* Standard mappings for irq_cpustat_t above */ - #define HARDIRQ_BITS 8 +#include <asm-generic/hardirq.h> #endif /* __SPARC_HARDIRQ_H */ diff --git a/arch/sparc/include/asm/irq_32.h b/arch/sparc/include/asm/irq_32.h index ea43057d476..cbf4801deaa 100644 --- a/arch/sparc/include/asm/irq_32.h +++ b/arch/sparc/include/asm/irq_32.h @@ -6,10 +6,10 @@ #ifndef _SPARC_IRQ_H #define _SPARC_IRQ_H -#include <linux/interrupt.h> - #define NR_IRQS 16 +#include <linux/interrupt.h> + #define irq_canonicalize(irq) (irq) extern void __init init_IRQ(void); diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index 0ff92fa2206..f3cb790fa2a 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -41,8 +41,8 @@ #define LOW_OBP_ADDRESS _AC(0x00000000f0000000,UL) #define HI_OBP_ADDRESS _AC(0x0000000100000000,UL) #define VMALLOC_START _AC(0x0000000100000000,UL) -#define VMALLOC_END _AC(0x0000000200000000,UL) -#define VMEMMAP_BASE _AC(0x0000000200000000,UL) +#define VMALLOC_END _AC(0x0000010000000000,UL) +#define VMEMMAP_BASE _AC(0x0000010000000000,UL) #define vmemmap ((struct page *)VMEMMAP_BASE) diff --git a/arch/sparc/kernel/ktlb.S b/arch/sparc/kernel/ktlb.S index 3ea6e8cde8c..1d361477d7d 100644 --- a/arch/sparc/kernel/ktlb.S +++ b/arch/sparc/kernel/ktlb.S @@ -280,8 +280,8 @@ kvmap_dtlb_nonlinear: #ifdef CONFIG_SPARSEMEM_VMEMMAP /* Do not use the TSB for vmemmap. */ - mov (VMEMMAP_BASE >> 24), %g5 - sllx %g5, 24, %g5 + mov (VMEMMAP_BASE >> 40), %g5 + sllx %g5, 40, %g5 cmp %g4,%g5 bgeu,pn %xcc, kvmap_vmemmap nop @@ -293,8 +293,8 @@ kvmap_dtlb_tsbmiss: sethi %hi(MODULES_VADDR), %g5 cmp %g4, %g5 blu,pn %xcc, kvmap_dtlb_longpath - mov (VMALLOC_END >> 24), %g5 - sllx %g5, 24, %g5 + mov (VMALLOC_END >> 40), %g5 + sllx %g5, 40, %g5 cmp %g4, %g5 bgeu,pn %xcc, kvmap_dtlb_longpath nop diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index 2d6a1b10c81..04db9274389 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -56,7 +56,8 @@ struct cpu_hw_events { struct perf_event *events[MAX_HWEVENTS]; unsigned long used_mask[BITS_TO_LONGS(MAX_HWEVENTS)]; unsigned long active_mask[BITS_TO_LONGS(MAX_HWEVENTS)]; - int enabled; + u64 pcr; + int enabled; }; DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = { .enabled = 1, }; @@ -68,8 +69,30 @@ struct perf_event_map { #define PIC_LOWER 0x02 }; +static unsigned long perf_event_encode(const struct perf_event_map *pmap) +{ + return ((unsigned long) pmap->encoding << 16) | pmap->pic_mask; +} + +static void perf_event_decode(unsigned long val, u16 *enc, u8 *msk) +{ + *msk = val & 0xff; + *enc = val >> 16; +} + +#define C(x) PERF_COUNT_HW_CACHE_##x + +#define CACHE_OP_UNSUPPORTED 0xfffe +#define CACHE_OP_NONSENSE 0xffff + +typedef struct perf_event_map cache_map_t + [PERF_COUNT_HW_CACHE_MAX] + [PERF_COUNT_HW_CACHE_OP_MAX] + [PERF_COUNT_HW_CACHE_RESULT_MAX]; + struct sparc_pmu { const struct perf_event_map *(*event_map)(int); + const cache_map_t *cache_map; int max_events; int upper_shift; int lower_shift; @@ -80,21 +103,109 @@ struct sparc_pmu { int lower_nop; }; -static const struct perf_event_map ultra3i_perfmon_event_map[] = { +static const struct perf_event_map ultra3_perfmon_event_map[] = { [PERF_COUNT_HW_CPU_CYCLES] = { 0x0000, PIC_UPPER | PIC_LOWER }, [PERF_COUNT_HW_INSTRUCTIONS] = { 0x0001, PIC_UPPER | PIC_LOWER }, [PERF_COUNT_HW_CACHE_REFERENCES] = { 0x0009, PIC_LOWER }, [PERF_COUNT_HW_CACHE_MISSES] = { 0x0009, PIC_UPPER }, }; -static const struct perf_event_map *ultra3i_event_map(int event_id) +static const struct perf_event_map *ultra3_event_map(int event_id) { - return &ultra3i_perfmon_event_map[event_id]; + return &ultra3_perfmon_event_map[event_id]; } -static const struct sparc_pmu ultra3i_pmu = { - .event_map = ultra3i_event_map, - .max_events = ARRAY_SIZE(ultra3i_perfmon_event_map), +static const cache_map_t ultra3_cache_map = { +[C(L1D)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { 0x09, PIC_LOWER, }, + [C(RESULT_MISS)] = { 0x09, PIC_UPPER, }, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = { 0x0a, PIC_LOWER }, + [C(RESULT_MISS)] = { 0x0a, PIC_UPPER }, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(L1I)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { 0x09, PIC_LOWER, }, + [C(RESULT_MISS)] = { 0x09, PIC_UPPER, }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_NONSENSE }, + [ C(RESULT_MISS) ] = { CACHE_OP_NONSENSE }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(LL)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { 0x0c, PIC_LOWER, }, + [C(RESULT_MISS)] = { 0x0c, PIC_UPPER, }, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = { 0x0c, PIC_LOWER }, + [C(RESULT_MISS)] = { 0x0c, PIC_UPPER }, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(DTLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { 0x12, PIC_UPPER, }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(ITLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { 0x11, PIC_UPPER, }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(BPU)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +}; + +static const struct sparc_pmu ultra3_pmu = { + .event_map = ultra3_event_map, + .cache_map = &ultra3_cache_map, + .max_events = ARRAY_SIZE(ultra3_perfmon_event_map), .upper_shift = 11, .lower_shift = 4, .event_mask = 0x3f, @@ -102,6 +213,121 @@ static const struct sparc_pmu ultra3i_pmu = { .lower_nop = 0x14, }; +/* Niagara1 is very limited. The upper PIC is hard-locked to count + * only instructions, so it is free running which creates all kinds of + * problems. Some hardware designs make one wonder if the creator + * even looked at how this stuff gets used by software. + */ +static const struct perf_event_map niagara1_perfmon_event_map[] = { + [PERF_COUNT_HW_CPU_CYCLES] = { 0x00, PIC_UPPER }, + [PERF_COUNT_HW_INSTRUCTIONS] = { 0x00, PIC_UPPER }, + [PERF_COUNT_HW_CACHE_REFERENCES] = { 0, PIC_NONE }, + [PERF_COUNT_HW_CACHE_MISSES] = { 0x03, PIC_LOWER }, +}; + +static const struct perf_event_map *niagara1_event_map(int event_id) +{ + return &niagara1_perfmon_event_map[event_id]; +} + +static const cache_map_t niagara1_cache_map = { +[C(L1D)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { 0x03, PIC_LOWER, }, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { 0x03, PIC_LOWER, }, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(L1I)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { 0x00, PIC_UPPER }, + [C(RESULT_MISS)] = { 0x02, PIC_LOWER, }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_NONSENSE }, + [ C(RESULT_MISS) ] = { CACHE_OP_NONSENSE }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(LL)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { 0x07, PIC_LOWER, }, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { 0x07, PIC_LOWER, }, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(DTLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { 0x05, PIC_LOWER, }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(ITLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { 0x04, PIC_LOWER, }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(BPU)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +}; + +static const struct sparc_pmu niagara1_pmu = { + .event_map = niagara1_event_map, + .cache_map = &niagara1_cache_map, + .max_events = ARRAY_SIZE(niagara1_perfmon_event_map), + .upper_shift = 0, + .lower_shift = 4, + .event_mask = 0x7, + .upper_nop = 0x0, + .lower_nop = 0x0, +}; + static const struct perf_event_map niagara2_perfmon_event_map[] = { [PERF_COUNT_HW_CPU_CYCLES] = { 0x02ff, PIC_UPPER | PIC_LOWER }, [PERF_COUNT_HW_INSTRUCTIONS] = { 0x02ff, PIC_UPPER | PIC_LOWER }, @@ -116,8 +342,96 @@ static const struct perf_event_map *niagara2_event_map(int event_id) return &niagara2_perfmon_event_map[event_id]; } +static const cache_map_t niagara2_cache_map = { +[C(L1D)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { 0x0208, PIC_UPPER | PIC_LOWER, }, + [C(RESULT_MISS)] = { 0x0302, PIC_UPPER | PIC_LOWER, }, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = { 0x0210, PIC_UPPER | PIC_LOWER, }, + [C(RESULT_MISS)] = { 0x0302, PIC_UPPER | PIC_LOWER, }, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(L1I)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { 0x02ff, PIC_UPPER | PIC_LOWER, }, + [C(RESULT_MISS)] = { 0x0301, PIC_UPPER | PIC_LOWER, }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_NONSENSE }, + [ C(RESULT_MISS) ] = { CACHE_OP_NONSENSE }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(LL)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { 0x0208, PIC_UPPER | PIC_LOWER, }, + [C(RESULT_MISS)] = { 0x0330, PIC_UPPER | PIC_LOWER, }, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = { 0x0210, PIC_UPPER | PIC_LOWER, }, + [C(RESULT_MISS)] = { 0x0320, PIC_UPPER | PIC_LOWER, }, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(DTLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { 0x0b08, PIC_UPPER | PIC_LOWER, }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(ITLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { 0xb04, PIC_UPPER | PIC_LOWER, }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(BPU)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +}; + static const struct sparc_pmu niagara2_pmu = { .event_map = niagara2_event_map, + .cache_map = &niagara2_cache_map, .max_events = ARRAY_SIZE(niagara2_perfmon_event_map), .upper_shift = 19, .lower_shift = 6, @@ -151,23 +465,30 @@ static u64 nop_for_index(int idx) sparc_pmu->lower_nop, idx); } -static inline void sparc_pmu_enable_event(struct hw_perf_event *hwc, - int idx) +static inline void sparc_pmu_enable_event(struct cpu_hw_events *cpuc, struct hw_perf_event *hwc, int idx) { u64 val, mask = mask_for_index(idx); - val = pcr_ops->read(); - pcr_ops->write((val & ~mask) | hwc->config); + val = cpuc->pcr; + val &= ~mask; + val |= hwc->config; + cpuc->pcr = val; + + pcr_ops->write(cpuc->pcr); } -static inline void sparc_pmu_disable_event(struct hw_perf_event *hwc, - int idx) +static inline void sparc_pmu_disable_event(struct cpu_hw_events *cpuc, struct hw_perf_event *hwc, int idx) { u64 mask = mask_for_index(idx); u64 nop = nop_for_index(idx); - u64 val = pcr_ops->read(); + u64 val; - pcr_ops->write((val & ~mask) | nop); + val = cpuc->pcr; + val &= ~mask; + val |= nop; + cpuc->pcr = val; + + pcr_ops->write(cpuc->pcr); } void hw_perf_enable(void) @@ -182,7 +503,7 @@ void hw_perf_enable(void) cpuc->enabled = 1; barrier(); - val = pcr_ops->read(); + val = cpuc->pcr; for (i = 0; i < MAX_HWEVENTS; i++) { struct perf_event *cp = cpuc->events[i]; @@ -194,7 +515,9 @@ void hw_perf_enable(void) val |= hwc->config_base; } - pcr_ops->write(val); + cpuc->pcr = val; + + pcr_ops->write(cpuc->pcr); } void hw_perf_disable(void) @@ -207,10 +530,12 @@ void hw_perf_disable(void) cpuc->enabled = 0; - val = pcr_ops->read(); + val = cpuc->pcr; val &= ~(PCR_UTRACE | PCR_STRACE | sparc_pmu->hv_bit | sparc_pmu->irq_bit); - pcr_ops->write(val); + cpuc->pcr = val; + + pcr_ops->write(cpuc->pcr); } static u32 read_pmc(int idx) @@ -242,7 +567,7 @@ static void write_pmc(int idx, u64 val) } static int sparc_perf_event_set_period(struct perf_event *event, - struct hw_perf_event *hwc, int idx) + struct hw_perf_event *hwc, int idx) { s64 left = atomic64_read(&hwc->period_left); s64 period = hwc->sample_period; @@ -282,19 +607,19 @@ static int sparc_pmu_enable(struct perf_event *event) if (test_and_set_bit(idx, cpuc->used_mask)) return -EAGAIN; - sparc_pmu_disable_event(hwc, idx); + sparc_pmu_disable_event(cpuc, hwc, idx); cpuc->events[idx] = event; set_bit(idx, cpuc->active_mask); sparc_perf_event_set_period(event, hwc, idx); - sparc_pmu_enable_event(hwc, idx); + sparc_pmu_enable_event(cpuc, hwc, idx); perf_event_update_userpage(event); return 0; } static u64 sparc_perf_event_update(struct perf_event *event, - struct hw_perf_event *hwc, int idx) + struct hw_perf_event *hwc, int idx) { int shift = 64 - 32; u64 prev_raw_count, new_raw_count; @@ -324,7 +649,7 @@ static void sparc_pmu_disable(struct perf_event *event) int idx = hwc->idx; clear_bit(idx, cpuc->active_mask); - sparc_pmu_disable_event(hwc, idx); + sparc_pmu_disable_event(cpuc, hwc, idx); barrier(); @@ -338,18 +663,29 @@ static void sparc_pmu_disable(struct perf_event *event) static void sparc_pmu_read(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; + sparc_perf_event_update(event, hwc, hwc->idx); } static void sparc_pmu_unthrottle(struct perf_event *event) { + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); struct hw_perf_event *hwc = &event->hw; - sparc_pmu_enable_event(hwc, hwc->idx); + + sparc_pmu_enable_event(cpuc, hwc, hwc->idx); } static atomic_t active_events = ATOMIC_INIT(0); static DEFINE_MUTEX(pmc_grab_mutex); +static void perf_stop_nmi_watchdog(void *unused) +{ + struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + + stop_nmi_watchdog(NULL); + cpuc->pcr = pcr_ops->read(); +} + void perf_event_grab_pmc(void) { if (atomic_inc_not_zero(&active_events)) @@ -358,7 +694,7 @@ void perf_event_grab_pmc(void) mutex_lock(&pmc_grab_mutex); if (atomic_read(&active_events) == 0) { if (atomic_read(&nmi_active) > 0) { - on_each_cpu(stop_nmi_watchdog, NULL, 1); + on_each_cpu(perf_stop_nmi_watchdog, NULL, 1); BUG_ON(atomic_read(&nmi_active) != 0); } atomic_inc(&active_events); @@ -375,30 +711,160 @@ void perf_event_release_pmc(void) } } +static const struct perf_event_map *sparc_map_cache_event(u64 config) +{ + unsigned int cache_type, cache_op, cache_result; + const struct perf_event_map *pmap; + + if (!sparc_pmu->cache_map) + return ERR_PTR(-ENOENT); + + cache_type = (config >> 0) & 0xff; + if (cache_type >= PERF_COUNT_HW_CACHE_MAX) + return ERR_PTR(-EINVAL); + + cache_op = (config >> 8) & 0xff; + if (cache_op >= PERF_COUNT_HW_CACHE_OP_MAX) + return ERR_PTR(-EINVAL); + + cache_result = (config >> 16) & 0xff; + if (cache_result >= PERF_COUNT_HW_CACHE_RESULT_MAX) + return ERR_PTR(-EINVAL); + + pmap = &((*sparc_pmu->cache_map)[cache_type][cache_op][cache_result]); + + if (pmap->encoding == CACHE_OP_UNSUPPORTED) + return ERR_PTR(-ENOENT); + + if (pmap->encoding == CACHE_OP_NONSENSE) + return ERR_PTR(-EINVAL); + + return pmap; +} + static void hw_perf_event_destroy(struct perf_event *event) { perf_event_release_pmc(); } +/* Make sure all events can be scheduled into the hardware at + * the same time. This is simplified by the fact that we only + * need to support 2 simultaneous HW events. + */ +static int sparc_check_constraints(unsigned long *events, int n_ev) +{ + if (n_ev <= perf_max_events) { + u8 msk1, msk2; + u16 dummy; + + if (n_ev == 1) + return 0; + BUG_ON(n_ev != 2); + perf_event_decode(events[0], &dummy, &msk1); + perf_event_decode(events[1], &dummy, &msk2); + + /* If both events can go on any counter, OK. */ + if (msk1 == (PIC_UPPER | PIC_LOWER) && + msk2 == (PIC_UPPER | PIC_LOWER)) + return 0; + + /* If one event is limited to a specific counter, + * and the other can go on both, OK. + */ + if ((msk1 == PIC_UPPER || msk1 == PIC_LOWER) && + msk2 == (PIC_UPPER | PIC_LOWER)) + return 0; + if ((msk2 == PIC_UPPER || msk2 == PIC_LOWER) && + msk1 == (PIC_UPPER | PIC_LOWER)) + return 0; + + /* If the events are fixed to different counters, OK. */ + if ((msk1 == PIC_UPPER && msk2 == PIC_LOWER) || + (msk1 == PIC_LOWER && msk2 == PIC_UPPER)) + return 0; + + /* Otherwise, there is a conflict. */ + } + + return -1; +} + +static int check_excludes(struct perf_event **evts, int n_prev, int n_new) +{ + int eu = 0, ek = 0, eh = 0; + struct perf_event *event; + int i, n, first; + + n = n_prev + n_new; + if (n <= 1) + return 0; + + first = 1; + for (i = 0; i < n; i++) { + event = evts[i]; + if (first) { + eu = event->attr.exclude_user; + ek = event->attr.exclude_kernel; + eh = event->attr.exclude_hv; + first = 0; + } else if (event->attr.exclude_user != eu || + event->attr.exclude_kernel != ek || + event->attr.exclude_hv != eh) { + return -EAGAIN; + } + } + + return 0; +} + +static int collect_events(struct perf_event *group, int max_count, + struct perf_event *evts[], unsigned long *events) +{ + struct perf_event *event; + int n = 0; + + if (!is_software_event(group)) { + if (n >= max_count) + return -1; + evts[n] = group; + events[n++] = group->hw.event_base; + } + list_for_each_entry(event, &group->sibling_list, group_entry) { + if (!is_software_event(event) && + event->state != PERF_EVENT_STATE_OFF) { + if (n >= max_count) + return -1; + evts[n] = event; + events[n++] = event->hw.event_base; + } + } + return n; +} + static int __hw_perf_event_init(struct perf_event *event) { struct perf_event_attr *attr = &event->attr; + struct perf_event *evts[MAX_HWEVENTS]; struct hw_perf_event *hwc = &event->hw; + unsigned long events[MAX_HWEVENTS]; const struct perf_event_map *pmap; u64 enc; + int n; if (atomic_read(&nmi_active) < 0) return -ENODEV; - if (attr->type != PERF_TYPE_HARDWARE) + if (attr->type == PERF_TYPE_HARDWARE) { + if (attr->config >= sparc_pmu->max_events) + return -EINVAL; + pmap = sparc_pmu->event_map(attr->config); + } else if (attr->type == PERF_TYPE_HW_CACHE) { + pmap = sparc_map_cache_event(attr->config); + if (IS_ERR(pmap)) + return PTR_ERR(pmap); + } else return -EOPNOTSUPP; - if (attr->config >= sparc_pmu->max_events) - return -EINVAL; - - perf_event_grab_pmc(); - event->destroy = hw_perf_event_destroy; - /* We save the enable bits in the config_base. So to * turn off sampling just write 'config', and to enable * things write 'config | config_base'. @@ -411,15 +877,39 @@ static int __hw_perf_event_init(struct perf_event *event) if (!attr->exclude_hv) hwc->config_base |= sparc_pmu->hv_bit; + hwc->event_base = perf_event_encode(pmap); + + enc = pmap->encoding; + + n = 0; + if (event->group_leader != event) { + n = collect_events(event->group_leader, + perf_max_events - 1, + evts, events); + if (n < 0) + return -EINVAL; + } + events[n] = hwc->event_base; + evts[n] = event; + + if (check_excludes(evts, n, 1)) + return -EINVAL; + + if (sparc_check_constraints(events, n + 1)) + return -EINVAL; + + /* Try to do all error checking before this point, as unwinding + * state after grabbing the PMC is difficult. + */ + perf_event_grab_pmc(); + event->destroy = hw_perf_event_destroy; + if (!hwc->sample_period) { hwc->sample_period = MAX_PERIOD; hwc->last_period = hwc->sample_period; atomic64_set(&hwc->period_left, hwc->sample_period); } - pmap = sparc_pmu->event_map(attr->config); - - enc = pmap->encoding; if (pmap->pic_mask & PIC_UPPER) { hwc->idx = PIC_UPPER_INDEX; enc <<= sparc_pmu->upper_shift; @@ -472,7 +962,7 @@ void perf_event_print_debug(void) } static int __kprobes perf_event_nmi_handler(struct notifier_block *self, - unsigned long cmd, void *__args) + unsigned long cmd, void *__args) { struct die_args *args = __args; struct perf_sample_data data; @@ -513,7 +1003,7 @@ static int __kprobes perf_event_nmi_handler(struct notifier_block *self, continue; if (perf_event_overflow(event, 1, &data, regs)) - sparc_pmu_disable_event(hwc, idx); + sparc_pmu_disable_event(cpuc, hwc, idx); } return NOTIFY_STOP; @@ -525,8 +1015,15 @@ static __read_mostly struct notifier_block perf_event_nmi_notifier = { static bool __init supported_pmu(void) { - if (!strcmp(sparc_pmu_type, "ultra3i")) { - sparc_pmu = &ultra3i_pmu; + if (!strcmp(sparc_pmu_type, "ultra3") || + !strcmp(sparc_pmu_type, "ultra3+") || + !strcmp(sparc_pmu_type, "ultra3i") || + !strcmp(sparc_pmu_type, "ultra4+")) { + sparc_pmu = &ultra3_pmu; + return true; + } + if (!strcmp(sparc_pmu_type, "niagara")) { + sparc_pmu = &niagara1_pmu; return true; } if (!strcmp(sparc_pmu_type, "niagara2")) { diff --git a/arch/sparc/oprofile/init.c b/arch/sparc/oprofile/init.c index f97cb8b6ee5..f9024bccff1 100644 --- a/arch/sparc/oprofile/init.c +++ b/arch/sparc/oprofile/init.c @@ -11,6 +11,7 @@ #include <linux/oprofile.h> #include <linux/errno.h> #include <linux/init.h> +#include <linux/param.h> /* for HZ */ #ifdef CONFIG_SPARC64 #include <linux/notifier.h> diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 8da93745c08..c876bace8fd 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -86,10 +86,6 @@ config STACKTRACE_SUPPORT config HAVE_LATENCYTOP_SUPPORT def_bool y -config FAST_CMPXCHG_LOCAL - bool - default y - config MMU def_bool y diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu index 527519b8a9f..f2824fb8c79 100644 --- a/arch/x86/Kconfig.cpu +++ b/arch/x86/Kconfig.cpu @@ -400,7 +400,7 @@ config X86_TSC config X86_CMPXCHG64 def_bool y - depends on X86_PAE || X86_64 + depends on X86_PAE || X86_64 || MCORE2 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MATOM # this should be set for all -march=.. options where the compiler # generates cmov. @@ -412,6 +412,7 @@ config X86_MINIMUM_CPU_FAMILY int default "64" if X86_64 default "6" if X86_32 && X86_P6_NOP + default "5" if X86_32 && X86_CMPXCHG64 default "4" if X86_32 && (X86_XADD || X86_CMPXCHG || X86_BSWAP || X86_WP_WORKS_OK) default "3" diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index 74619c4f9fd..1733f9f65e8 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -21,8 +21,8 @@ #define __AUDIT_ARCH_LE 0x40000000 #ifndef CONFIG_AUDITSYSCALL -#define sysexit_audit int_ret_from_sys_call -#define sysretl_audit int_ret_from_sys_call +#define sysexit_audit ia32_ret_from_sys_call +#define sysretl_audit ia32_ret_from_sys_call #endif #define IA32_NR_syscalls ((ia32_syscall_end - ia32_sys_call_table)/8) @@ -39,12 +39,12 @@ .endm /* clobbers %eax */ - .macro CLEAR_RREGS _r9=rax + .macro CLEAR_RREGS offset=0, _r9=rax xorl %eax,%eax - movq %rax,R11(%rsp) - movq %rax,R10(%rsp) - movq %\_r9,R9(%rsp) - movq %rax,R8(%rsp) + movq %rax,\offset+R11(%rsp) + movq %rax,\offset+R10(%rsp) + movq %\_r9,\offset+R9(%rsp) + movq %rax,\offset+R8(%rsp) .endm /* @@ -172,6 +172,10 @@ sysexit_from_sys_call: movl RIP-R11(%rsp),%edx /* User %eip */ CFI_REGISTER rip,rdx RESTORE_ARGS 1,24,1,1,1,1 + xorq %r8,%r8 + xorq %r9,%r9 + xorq %r10,%r10 + xorq %r11,%r11 popfq CFI_ADJUST_CFA_OFFSET -8 /*CFI_RESTORE rflags*/ @@ -202,7 +206,7 @@ sysexit_from_sys_call: .macro auditsys_exit exit,ebpsave=RBP testl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT),TI_flags(%r10) - jnz int_ret_from_sys_call + jnz ia32_ret_from_sys_call TRACE_IRQS_ON sti movl %eax,%esi /* second arg, syscall return value */ @@ -218,8 +222,9 @@ sysexit_from_sys_call: cli TRACE_IRQS_OFF testl %edi,TI_flags(%r10) - jnz int_with_check - jmp \exit + jz \exit + CLEAR_RREGS -ARGOFFSET + jmp int_with_check .endm sysenter_auditsys: @@ -329,6 +334,9 @@ sysretl_from_sys_call: CFI_REGISTER rip,rcx movl EFLAGS-ARGOFFSET(%rsp),%r11d /*CFI_REGISTER rflags,r11*/ + xorq %r10,%r10 + xorq %r9,%r9 + xorq %r8,%r8 TRACE_IRQS_ON movl RSP-ARGOFFSET(%rsp),%esp CFI_RESTORE rsp @@ -353,7 +361,7 @@ cstar_tracesys: #endif xchgl %r9d,%ebp SAVE_REST - CLEAR_RREGS r9 + CLEAR_RREGS 0, r9 movq $-ENOSYS,RAX(%rsp) /* ptrace can change this for a bad syscall */ movq %rsp,%rdi /* &pt_regs -> arg1 */ call syscall_trace_enter @@ -425,6 +433,8 @@ ia32_do_call: call *ia32_sys_call_table(,%rax,8) # xxx: rip relative ia32_sysret: movq %rax,RAX-ARGOFFSET(%rsp) +ia32_ret_from_sys_call: + CLEAR_RREGS -ARGOFFSET jmp int_ret_from_sys_call ia32_tracesys: @@ -442,8 +452,8 @@ END(ia32_syscall) ia32_badsys: movq $0,ORIG_RAX-ARGOFFSET(%rsp) - movq $-ENOSYS,RAX-ARGOFFSET(%rsp) - jmp int_ret_from_sys_call + movq $-ENOSYS,%rax + jmp ia32_sysret quiet_ni_syscall: movq $-ENOSYS,%rax diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c index 41fd965c80c..b9c830c12b4 100644 --- a/arch/x86/kernel/early_printk.c +++ b/arch/x86/kernel/early_printk.c @@ -206,8 +206,11 @@ static int __init setup_early_printk(char *buf) while (*buf != '\0') { if (!strncmp(buf, "serial", 6)) { - early_serial_init(buf + 6); + buf += 6; + early_serial_init(buf); early_console_register(&early_serial_console, keep); + if (!strncmp(buf, ",ttyS", 5)) + buf += 5; } if (!strncmp(buf, "ttyS", 4)) { early_serial_init(buf + 4); diff --git a/arch/x86/kernel/i386_ksyms_32.c b/arch/x86/kernel/i386_ksyms_32.c index 1736c5a725a..9c3bd4a2050 100644 --- a/arch/x86/kernel/i386_ksyms_32.c +++ b/arch/x86/kernel/i386_ksyms_32.c @@ -15,8 +15,10 @@ EXPORT_SYMBOL(mcount); * the export, but dont use it from C code, it is used * by assembly code and is not using C calling convention! */ +#ifndef CONFIG_X86_CMPXCHG64 extern void cmpxchg8b_emu(void); EXPORT_SYMBOL(cmpxchg8b_emu); +#endif /* Networking helper routines. */ EXPORT_SYMBOL(csum_partial_copy_generic); diff --git a/arch/x86/lib/Makefile b/arch/x86/lib/Makefile index 3e549b8ec8c..85f5db95c60 100644 --- a/arch/x86/lib/Makefile +++ b/arch/x86/lib/Makefile @@ -15,8 +15,10 @@ ifeq ($(CONFIG_X86_32),y) obj-y += atomic64_32.o lib-y += checksum_32.o lib-y += strstr_32.o - lib-y += semaphore_32.o string_32.o cmpxchg8b_emu.o - + lib-y += semaphore_32.o string_32.o +ifneq ($(CONFIG_X86_CMPXCHG64),y) + lib-y += cmpxchg8b_emu.o +endif lib-$(CONFIG_X86_USE_3DNOW) += mmx_32.o else obj-y += io_64.o iomap_copy_64.o diff --git a/block/blk-barrier.c b/block/blk-barrier.c index 6593ab39cfe..8873b9b439f 100644 --- a/block/blk-barrier.c +++ b/block/blk-barrier.c @@ -350,6 +350,7 @@ static void blkdev_discard_end_io(struct bio *bio, int err) if (bio->bi_private) complete(bio->bi_private); + __free_page(bio_page(bio)); bio_put(bio); } @@ -372,30 +373,50 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector, struct request_queue *q = bdev_get_queue(bdev); int type = flags & DISCARD_FL_BARRIER ? DISCARD_BARRIER : DISCARD_NOBARRIER; + struct bio *bio; + struct page *page; int ret = 0; if (!q) return -ENXIO; - if (!q->prepare_discard_fn) + if (!blk_queue_discard(q)) return -EOPNOTSUPP; while (nr_sects && !ret) { - struct bio *bio = bio_alloc(gfp_mask, 0); - if (!bio) - return -ENOMEM; + unsigned int sector_size = q->limits.logical_block_size; + unsigned int max_discard_sectors = + min(q->limits.max_discard_sectors, UINT_MAX >> 9); + bio = bio_alloc(gfp_mask, 1); + if (!bio) + goto out; + bio->bi_sector = sector; bio->bi_end_io = blkdev_discard_end_io; bio->bi_bdev = bdev; if (flags & DISCARD_FL_WAIT) bio->bi_private = &wait; - bio->bi_sector = sector; + /* + * Add a zeroed one-sector payload as that's what + * our current implementations need. If we'll ever need + * more the interface will need revisiting. + */ + page = alloc_page(GFP_KERNEL | __GFP_ZERO); + if (!page) + goto out_free_bio; + if (bio_add_pc_page(q, bio, page, sector_size, 0) < sector_size) + goto out_free_page; - if (nr_sects > queue_max_hw_sectors(q)) { - bio->bi_size = queue_max_hw_sectors(q) << 9; - nr_sects -= queue_max_hw_sectors(q); - sector += queue_max_hw_sectors(q); + /* + * And override the bio size - the way discard works we + * touch many more blocks on disk than the actual payload + * length. + */ + if (nr_sects > max_discard_sectors) { + bio->bi_size = max_discard_sectors << 9; + nr_sects -= max_discard_sectors; + sector += max_discard_sectors; } else { bio->bi_size = nr_sects << 9; nr_sects = 0; @@ -414,5 +435,11 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector, bio_put(bio); } return ret; +out_free_page: + __free_page(page); +out_free_bio: + bio_put(bio); +out: + return -ENOMEM; } EXPORT_SYMBOL(blkdev_issue_discard); diff --git a/block/blk-core.c b/block/blk-core.c index 8135228e4b2..81f34311659 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -34,6 +34,7 @@ #include "blk.h" EXPORT_TRACEPOINT_SYMBOL_GPL(block_remap); +EXPORT_TRACEPOINT_SYMBOL_GPL(block_rq_remap); EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_complete); static int __make_request(struct request_queue *q, struct bio *bio); @@ -69,7 +70,7 @@ static void drive_stat_acct(struct request *rq, int new_io) part_stat_inc(cpu, part, merges[rw]); else { part_round_stats(cpu, part); - part_inc_in_flight(part, rw); + part_inc_in_flight(part); } part_stat_unlock(); @@ -1031,7 +1032,7 @@ static void part_round_stats_single(int cpu, struct hd_struct *part, if (part->in_flight) { __part_stat_add(cpu, part, time_in_queue, - part_in_flight(part) * (now - part->stamp)); + part->in_flight * (now - part->stamp)); __part_stat_add(cpu, part, io_ticks, (now - part->stamp)); } part->stamp = now; @@ -1124,7 +1125,6 @@ void init_request_from_bio(struct request *req, struct bio *bio) req->cmd_flags |= REQ_DISCARD; if (bio_rw_flagged(bio, BIO_RW_BARRIER)) req->cmd_flags |= REQ_SOFTBARRIER; - req->q->prepare_discard_fn(req->q, req); } else if (unlikely(bio_rw_flagged(bio, BIO_RW_BARRIER))) req->cmd_flags |= REQ_HARDBARRIER; @@ -1437,7 +1437,8 @@ static inline void __generic_make_request(struct bio *bio) goto end_io; } - if (unlikely(nr_sectors > queue_max_hw_sectors(q))) { + if (unlikely(!bio_rw_flagged(bio, BIO_RW_DISCARD) && + nr_sectors > queue_max_hw_sectors(q))) { printk(KERN_ERR "bio too big device %s (%u > %u)\n", bdevname(bio->bi_bdev, b), bio_sectors(bio), @@ -1470,7 +1471,7 @@ static inline void __generic_make_request(struct bio *bio) goto end_io; if (bio_rw_flagged(bio, BIO_RW_DISCARD) && - !q->prepare_discard_fn) { + !blk_queue_discard(q)) { err = -EOPNOTSUPP; goto end_io; } @@ -1738,7 +1739,7 @@ static void blk_account_io_done(struct request *req) part_stat_inc(cpu, part, ios[rw]); part_stat_add(cpu, part, ticks[rw], duration); part_round_stats(cpu, part); - part_dec_in_flight(part, rw); + part_dec_in_flight(part); part_stat_unlock(); } @@ -2491,6 +2492,14 @@ int kblockd_schedule_work(struct request_queue *q, struct work_struct *work) } EXPORT_SYMBOL(kblockd_schedule_work); +int kblockd_schedule_delayed_work(struct request_queue *q, + struct delayed_work *work, + unsigned long delay) +{ + return queue_delayed_work(kblockd_workqueue, work, delay); +} +EXPORT_SYMBOL(kblockd_schedule_delayed_work); + int __init blk_dev_init(void) { BUILD_BUG_ON(__REQ_NR_BITS > 8 * diff --git a/block/blk-merge.c b/block/blk-merge.c index 99cb5cf1f44..b0de8574fdc 100644 --- a/block/blk-merge.c +++ b/block/blk-merge.c @@ -351,7 +351,7 @@ static void blk_account_io_merge(struct request *req) part = disk_map_sector_rcu(req->rq_disk, blk_rq_pos(req)); part_round_stats(cpu, part); - part_dec_in_flight(part, rq_data_dir(req)); + part_dec_in_flight(part); part_stat_unlock(); } diff --git a/block/blk-settings.c b/block/blk-settings.c index 83413ff8373..e0695bca702 100644 --- a/block/blk-settings.c +++ b/block/blk-settings.c @@ -34,23 +34,6 @@ void blk_queue_prep_rq(struct request_queue *q, prep_rq_fn *pfn) EXPORT_SYMBOL(blk_queue_prep_rq); /** - * blk_queue_set_discard - set a discard_sectors function for queue - * @q: queue - * @dfn: prepare_discard function - * - * It's possible for a queue to register a discard callback which is used - * to transform a discard request into the appropriate type for the - * hardware. If none is registered, then discard requests are failed - * with %EOPNOTSUPP. - * - */ -void blk_queue_set_discard(struct request_queue *q, prepare_discard_fn *dfn) -{ - q->prepare_discard_fn = dfn; -} -EXPORT_SYMBOL(blk_queue_set_discard); - -/** * blk_queue_merge_bvec - set a merge_bvec function for queue * @q: queue * @mbfn: merge_bvec_fn @@ -111,7 +94,9 @@ void blk_set_default_limits(struct queue_limits *lim) lim->max_hw_segments = MAX_HW_SEGMENTS; lim->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK; lim->max_segment_size = MAX_SEGMENT_SIZE; - lim->max_sectors = lim->max_hw_sectors = SAFE_MAX_SECTORS; + lim->max_sectors = BLK_DEF_MAX_SECTORS; + lim->max_hw_sectors = INT_MAX; + lim->max_discard_sectors = SAFE_MAX_SECTORS; lim->logical_block_size = lim->physical_block_size = lim->io_min = 512; lim->bounce_pfn = (unsigned long)(BLK_BOUNCE_ANY >> PAGE_SHIFT); lim->alignment_offset = 0; @@ -164,6 +149,7 @@ void blk_queue_make_request(struct request_queue *q, make_request_fn *mfn) q->unplug_timer.data = (unsigned long)q; blk_set_default_limits(&q->limits); + blk_queue_max_sectors(q, SAFE_MAX_SECTORS); /* * If the caller didn't supply a lock, fall back to our embedded @@ -254,6 +240,18 @@ void blk_queue_max_hw_sectors(struct request_queue *q, unsigned int max_sectors) EXPORT_SYMBOL(blk_queue_max_hw_sectors); /** + * blk_queue_max_discard_sectors - set max sectors for a single discard + * @q: the request queue for the device + * @max_discard: maximum number of sectors to discard + **/ +void blk_queue_max_discard_sectors(struct request_queue *q, + unsigned int max_discard_sectors) +{ + q->limits.max_discard_sectors = max_discard_sectors; +} +EXPORT_SYMBOL(blk_queue_max_discard_sectors); + +/** * blk_queue_max_phys_segments - set max phys segments for a request for this queue * @q: the request queue for the device * @max_segments: max number of segments diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index b78c9c3e267..8a6d81afb28 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -452,6 +452,7 @@ int blk_register_queue(struct gendisk *disk) if (ret) { kobject_uevent(&q->kobj, KOBJ_REMOVE); kobject_del(&q->kobj); + blk_trace_remove_sysfs(disk_to_dev(disk)); return ret; } @@ -465,11 +466,11 @@ void blk_unregister_queue(struct gendisk *disk) if (WARN_ON(!q)) return; - if (q->request_fn) { + if (q->request_fn) elv_unregister_queue(q); - kobject_uevent(&q->kobj, KOBJ_REMOVE); - kobject_del(&q->kobj); - kobject_put(&disk_to_dev(disk)->kobj); - } + kobject_uevent(&q->kobj, KOBJ_REMOVE); + kobject_del(&q->kobj); + blk_trace_remove_sysfs(disk_to_dev(disk)); + kobject_put(&disk_to_dev(disk)->kobj); } diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 1ca813b16e7..9c4b679908f 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -150,7 +150,7 @@ struct cfq_data { * idle window management */ struct timer_list idle_slice_timer; - struct work_struct unplug_work; + struct delayed_work unplug_work; struct cfq_queue *active_queue; struct cfq_io_context *active_cic; @@ -173,6 +173,7 @@ struct cfq_data { unsigned int cfq_slice[2]; unsigned int cfq_slice_async_rq; unsigned int cfq_slice_idle; + unsigned int cfq_latency; struct list_head cic_list; @@ -180,6 +181,8 @@ struct cfq_data { * Fallback dummy cfqq for extreme OOM conditions */ struct cfq_queue oom_cfqq; + + unsigned long last_end_sync_rq; }; enum cfqq_state_flags { @@ -265,11 +268,13 @@ static inline int cfq_bio_sync(struct bio *bio) * scheduler run of queue, if there are requests pending and no one in the * driver that will restart queueing */ -static inline void cfq_schedule_dispatch(struct cfq_data *cfqd) +static inline void cfq_schedule_dispatch(struct cfq_data *cfqd, + unsigned long delay) { if (cfqd->busy_queues) { cfq_log(cfqd, "schedule dispatch"); - kblockd_schedule_work(cfqd->queue, &cfqd->unplug_work); + kblockd_schedule_delayed_work(cfqd->queue, &cfqd->unplug_work, + delay); } } @@ -1326,12 +1331,30 @@ static int cfq_dispatch_requests(struct request_queue *q, int force) return 0; /* - * we are the only queue, allow up to 4 times of 'quantum' + * Sole queue user, allow bigger slice */ - if (cfqq->dispatched >= 4 * max_dispatch) - return 0; + max_dispatch *= 4; + } + + /* + * Async queues must wait a bit before being allowed dispatch. + * We also ramp up the dispatch depth gradually for async IO, + * based on the last sync IO we serviced + */ + if (!cfq_cfqq_sync(cfqq) && cfqd->cfq_latency) { + unsigned long last_sync = jiffies - cfqd->last_end_sync_rq; + unsigned int depth; + + depth = last_sync / cfqd->cfq_slice[1]; + if (!depth && !cfqq->dispatched) + depth = 1; + if (depth < max_dispatch) + max_dispatch = depth; } + if (cfqq->dispatched >= max_dispatch) + return 0; + /* * Dispatch a request from this cfqq */ @@ -1376,7 +1399,7 @@ static void cfq_put_queue(struct cfq_queue *cfqq) if (unlikely(cfqd->active_queue == cfqq)) { __cfq_slice_expired(cfqd, cfqq, 0); - cfq_schedule_dispatch(cfqd); + cfq_schedule_dispatch(cfqd, 0); } kmem_cache_free(cfq_pool, cfqq); @@ -1471,7 +1494,7 @@ static void cfq_exit_cfqq(struct cfq_data *cfqd, struct cfq_queue *cfqq) { if (unlikely(cfqq == cfqd->active_queue)) { __cfq_slice_expired(cfqd, cfqq, 0); - cfq_schedule_dispatch(cfqd); + cfq_schedule_dispatch(cfqd, 0); } cfq_put_queue(cfqq); @@ -1951,7 +1974,7 @@ cfq_update_idle_window(struct cfq_data *cfqd, struct cfq_queue *cfqq, enable_idle = old_idle = cfq_cfqq_idle_window(cfqq); if (!atomic_read(&cic->ioc->nr_tasks) || !cfqd->cfq_slice_idle || - (cfqd->hw_tag && CIC_SEEKY(cic))) + (!cfqd->cfq_latency && cfqd->hw_tag && CIC_SEEKY(cic))) enable_idle = 0; else if (sample_valid(cic->ttime_samples)) { if (cic->ttime_mean > cfqd->cfq_slice_idle) @@ -2157,8 +2180,10 @@ static void cfq_completed_request(struct request_queue *q, struct request *rq) if (cfq_cfqq_sync(cfqq)) cfqd->sync_flight--; - if (sync) + if (sync) { RQ_CIC(rq)->last_end_request = now; + cfqd->last_end_sync_rq = now; + } /* * If this is the active queue, check if it needs to be expired, @@ -2186,7 +2211,7 @@ static void cfq_completed_request(struct request_queue *q, struct request *rq) } if (!rq_in_driver(cfqd)) - cfq_schedule_dispatch(cfqd); + cfq_schedule_dispatch(cfqd, 0); } /* @@ -2316,7 +2341,7 @@ queue_fail: if (cic) put_io_context(cic->ioc); - cfq_schedule_dispatch(cfqd); + cfq_schedule_dispatch(cfqd, 0); spin_unlock_irqrestore(q->queue_lock, flags); cfq_log(cfqd, "set_request fail"); return 1; @@ -2325,7 +2350,7 @@ queue_fail: static void cfq_kick_queue(struct work_struct *work) { struct cfq_data *cfqd = - container_of(work, struct cfq_data, unplug_work); + container_of(work, struct cfq_data, unplug_work.work); struct request_queue *q = cfqd->queue; spin_lock_irq(q->queue_lock); @@ -2379,7 +2404,7 @@ static void cfq_idle_slice_timer(unsigned long data) expire: cfq_slice_expired(cfqd, timed_out); out_kick: - cfq_schedule_dispatch(cfqd); + cfq_schedule_dispatch(cfqd, 0); out_cont: spin_unlock_irqrestore(cfqd->queue->queue_lock, flags); } @@ -2387,7 +2412,7 @@ out_cont: static void cfq_shutdown_timer_wq(struct cfq_data *cfqd) { del_timer_sync(&cfqd->idle_slice_timer); - cancel_work_sync(&cfqd->unplug_work); + cancel_delayed_work_sync(&cfqd->unplug_work); } static void cfq_put_async_queues(struct cfq_data *cfqd) @@ -2469,7 +2494,7 @@ static void *cfq_init_queue(struct request_queue *q) cfqd->idle_slice_timer.function = cfq_idle_slice_timer; cfqd->idle_slice_timer.data = (unsigned long) cfqd; - INIT_WORK(&cfqd->unplug_work, cfq_kick_queue); + INIT_DELAYED_WORK(&cfqd->unplug_work, cfq_kick_queue); cfqd->cfq_quantum = cfq_quantum; cfqd->cfq_fifo_expire[0] = cfq_fifo_expire[0]; @@ -2480,8 +2505,9 @@ static void *cfq_init_queue(struct request_queue *q) cfqd->cfq_slice[1] = cfq_slice_sync; cfqd->cfq_slice_async_rq = cfq_slice_async_rq; cfqd->cfq_slice_idle = cfq_slice_idle; + cfqd->cfq_latency = 1; cfqd->hw_tag = 1; - + cfqd->last_end_sync_rq = jiffies; return cfqd; } @@ -2549,6 +2575,7 @@ SHOW_FUNCTION(cfq_slice_idle_show, cfqd->cfq_slice_idle, 1); SHOW_FUNCTION(cfq_slice_sync_show, cfqd->cfq_slice[1], 1); SHOW_FUNCTION(cfq_slice_async_show, cfqd->cfq_slice[0], 1); SHOW_FUNCTION(cfq_slice_async_rq_show, cfqd->cfq_slice_async_rq, 0); +SHOW_FUNCTION(cfq_low_latency_show, cfqd->cfq_latency, 0); #undef SHOW_FUNCTION #define STORE_FUNCTION(__FUNC, __PTR, MIN, MAX, __CONV) \ @@ -2580,6 +2607,7 @@ STORE_FUNCTION(cfq_slice_sync_store, &cfqd->cfq_slice[1], 1, UINT_MAX, 1); STORE_FUNCTION(cfq_slice_async_store, &cfqd->cfq_slice[0], 1, UINT_MAX, 1); STORE_FUNCTION(cfq_slice_async_rq_store, &cfqd->cfq_slice_async_rq, 1, UINT_MAX, 0); +STORE_FUNCTION(cfq_low_latency_store, &cfqd->cfq_latency, 0, 1, 0); #undef STORE_FUNCTION #define CFQ_ATTR(name) \ @@ -2595,6 +2623,7 @@ static struct elv_fs_entry cfq_attrs[] = { CFQ_ATTR(slice_async), CFQ_ATTR(slice_async_rq), CFQ_ATTR(slice_idle), + CFQ_ATTR(low_latency), __ATTR_NULL }; diff --git a/block/compat_ioctl.c b/block/compat_ioctl.c index 7865a34e0fa..9bd086c1a4d 100644 --- a/block/compat_ioctl.c +++ b/block/compat_ioctl.c @@ -21,6 +21,11 @@ static int compat_put_int(unsigned long arg, int val) return put_user(val, (compat_int_t __user *)compat_ptr(arg)); } +static int compat_put_uint(unsigned long arg, unsigned int val) +{ + return put_user(val, (compat_uint_t __user *)compat_ptr(arg)); +} + static int compat_put_long(unsigned long arg, long val) { return put_user(val, (compat_long_t __user *)compat_ptr(arg)); @@ -734,6 +739,14 @@ long compat_blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg) switch (cmd) { case HDIO_GETGEO: return compat_hdio_getgeo(disk, bdev, compat_ptr(arg)); + case BLKPBSZGET: + return compat_put_uint(arg, bdev_physical_block_size(bdev)); + case BLKIOMIN: + return compat_put_uint(arg, bdev_io_min(bdev)); + case BLKIOOPT: + return compat_put_uint(arg, bdev_io_opt(bdev)); + case BLKALIGNOFF: + return compat_put_int(arg, bdev_alignment_offset(bdev)); case BLKFLSBUF: case BLKROSET: case BLKDISCARD: diff --git a/block/genhd.c b/block/genhd.c index 517e4332cb3..5a0861da324 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -869,7 +869,6 @@ static DEVICE_ATTR(size, S_IRUGO, part_size_show, NULL); static DEVICE_ATTR(alignment_offset, S_IRUGO, disk_alignment_offset_show, NULL); static DEVICE_ATTR(capability, S_IRUGO, disk_capability_show, NULL); static DEVICE_ATTR(stat, S_IRUGO, part_stat_show, NULL); -static DEVICE_ATTR(inflight, S_IRUGO, part_inflight_show, NULL); #ifdef CONFIG_FAIL_MAKE_REQUEST static struct device_attribute dev_attr_fail = __ATTR(make-it-fail, S_IRUGO|S_IWUSR, part_fail_show, part_fail_store); @@ -889,7 +888,6 @@ static struct attribute *disk_attrs[] = { &dev_attr_alignment_offset.attr, &dev_attr_capability.attr, &dev_attr_stat.attr, - &dev_attr_inflight.attr, #ifdef CONFIG_FAIL_MAKE_REQUEST &dev_attr_fail.attr, #endif @@ -1055,7 +1053,7 @@ static int diskstats_show(struct seq_file *seqf, void *v) part_stat_read(hd, merges[1]), (unsigned long long)part_stat_read(hd, sectors[1]), jiffies_to_msecs(part_stat_read(hd, ticks[1])), - part_in_flight(hd), + hd->in_flight, jiffies_to_msecs(part_stat_read(hd, io_ticks)), jiffies_to_msecs(part_stat_read(hd, time_in_queue)) ); diff --git a/block/ioctl.c b/block/ioctl.c index d3e6b5827a3..1f4d1de12b0 100644 --- a/block/ioctl.c +++ b/block/ioctl.c @@ -138,6 +138,11 @@ static int put_int(unsigned long arg, int val) return put_user(val, (int __user *)arg); } +static int put_uint(unsigned long arg, unsigned int val) +{ + return put_user(val, (unsigned int __user *)arg); +} + static int put_long(unsigned long arg, long val) { return put_user(val, (long __user *)arg); @@ -263,10 +268,18 @@ int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd, return put_long(arg, (bdi->ra_pages * PAGE_CACHE_SIZE) / 512); case BLKROGET: return put_int(arg, bdev_read_only(bdev) != 0); - case BLKBSZGET: /* get the logical block size (cf. BLKSSZGET) */ + case BLKBSZGET: /* get block device soft block size (cf. BLKSSZGET) */ return put_int(arg, block_size(bdev)); - case BLKSSZGET: /* get block device hardware sector size */ + case BLKSSZGET: /* get block device logical block size */ return put_int(arg, bdev_logical_block_size(bdev)); + case BLKPBSZGET: /* get block device physical block size */ + return put_uint(arg, bdev_physical_block_size(bdev)); + case BLKIOMIN: + return put_uint(arg, bdev_io_min(bdev)); + case BLKIOOPT: + return put_uint(arg, bdev_io_opt(bdev)); + case BLKALIGNOFF: + return put_int(arg, bdev_alignment_offset(bdev)); case BLKSECTGET: return put_ushort(arg, queue_max_sectors(bdev_get_queue(bdev))); case BLKRASET: diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig index dd8729d674e..0ed42d8870c 100644 --- a/drivers/acpi/Kconfig +++ b/drivers/acpi/Kconfig @@ -211,6 +211,18 @@ config ACPI_HOTPLUG_CPU select ACPI_CONTAINER default y +config ACPI_PROCESSOR_AGGREGATOR + tristate "Processor Aggregator" + depends on ACPI_PROCESSOR + depends on EXPERIMENTAL + depends on X86 + help + ACPI 4.0 defines processor Aggregator, which enables OS to perform + specfic processor configuration and control that applies to all + processors in the platform. Currently only logical processor idling + is defined, which is to reduce power consumption. This driver + support the new device. + config ACPI_THERMAL tristate "Thermal Zone" depends on ACPI_PROCESSOR diff --git a/drivers/acpi/Makefile b/drivers/acpi/Makefile index 82cd49dc603..7702118509a 100644 --- a/drivers/acpi/Makefile +++ b/drivers/acpi/Makefile @@ -62,3 +62,5 @@ obj-$(CONFIG_ACPI_POWER_METER) += power_meter.o processor-y := processor_core.o processor_throttling.o processor-y += processor_idle.o processor_thermal.o processor-$(CONFIG_CPU_FREQ) += processor_perflib.o + +obj-$(CONFIG_ACPI_PROCESSOR_AGGREGATOR) += acpi_pad.o diff --git a/drivers/acpi/acpi_pad.c b/drivers/acpi/acpi_pad.c new file mode 100644 index 00000000000..0d2cdb86158 --- /dev/null +++ b/drivers/acpi/acpi_pad.c @@ -0,0 +1,514 @@ +/* + * acpi_pad.c ACPI Processor Aggregator Driver + * + * Copyright (c) 2009, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope 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; if not, write to the Free Software Foundation, Inc., + * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. + * + */ + +#include <linux/kernel.h> +#include <linux/cpumask.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/types.h> +#include <linux/kthread.h> +#include <linux/freezer.h> +#include <linux/cpu.h> +#include <linux/clockchips.h> +#include <acpi/acpi_bus.h> +#include <acpi/acpi_drivers.h> + +#define ACPI_PROCESSOR_AGGREGATOR_CLASS "processor_aggregator" +#define ACPI_PROCESSOR_AGGREGATOR_DEVICE_NAME "Processor Aggregator" +#define ACPI_PROCESSOR_AGGREGATOR_NOTIFY 0x80 +static DEFINE_MUTEX(isolated_cpus_lock); + +#define MWAIT_SUBSTATE_MASK (0xf) +#define MWAIT_CSTATE_MASK (0xf) +#define MWAIT_SUBSTATE_SIZE (4) +#define CPUID_MWAIT_LEAF (5) +#define CPUID5_ECX_EXTENSIONS_SUPPORTED (0x1) +#define CPUID5_ECX_INTERRUPT_BREAK (0x2) +static unsigned long power_saving_mwait_eax; +static void power_saving_mwait_init(void) +{ + unsigned int eax, ebx, ecx, edx; + unsigned int highest_cstate = 0; + unsigned int highest_subcstate = 0; + int i; + + if (!boot_cpu_has(X86_FEATURE_MWAIT)) + return; + if (boot_cpu_data.cpuid_level < CPUID_MWAIT_LEAF) + return; + + cpuid(CPUID_MWAIT_LEAF, &eax, &ebx, &ecx, &edx); + + if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED) || + !(ecx & CPUID5_ECX_INTERRUPT_BREAK)) + return; + + edx >>= MWAIT_SUBSTATE_SIZE; + for (i = 0; i < 7 && edx; i++, edx >>= MWAIT_SUBSTATE_SIZE) { + if (edx & MWAIT_SUBSTATE_MASK) { + highest_cstate = i; + highest_subcstate = edx & MWAIT_SUBSTATE_MASK; + } + } + power_saving_mwait_eax = (highest_cstate << MWAIT_SUBSTATE_SIZE) | + (highest_subcstate - 1); + + for_each_online_cpu(i) + clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_ON, &i); + +#if defined(CONFIG_GENERIC_TIME) && defined(CONFIG_X86) + switch (boot_cpu_data.x86_vendor) { + case X86_VENDOR_AMD: + case X86_VENDOR_INTEL: + /* + * AMD Fam10h TSC will tick in all + * C/P/S0/S1 states when this bit is set. + */ + if (boot_cpu_has(X86_FEATURE_NONSTOP_TSC)) + return; + + /*FALL THROUGH*/ + default: + /* TSC could halt in idle, so notify users */ + mark_tsc_unstable("TSC halts in idle"); + } +#endif +} + +static unsigned long cpu_weight[NR_CPUS]; +static int tsk_in_cpu[NR_CPUS] = {[0 ... NR_CPUS-1] = -1}; +static DECLARE_BITMAP(pad_busy_cpus_bits, NR_CPUS); +static void round_robin_cpu(unsigned int tsk_index) +{ + struct cpumask *pad_busy_cpus = to_cpumask(pad_busy_cpus_bits); + cpumask_var_t tmp; + int cpu; + unsigned long min_weight = -1, preferred_cpu; + + if (!alloc_cpumask_var(&tmp, GFP_KERNEL)) + return; + + mutex_lock(&isolated_cpus_lock); + cpumask_clear(tmp); + for_each_cpu(cpu, pad_busy_cpus) + cpumask_or(tmp, tmp, topology_thread_cpumask(cpu)); + cpumask_andnot(tmp, cpu_online_mask, tmp); + /* avoid HT sibilings if possible */ + if (cpumask_empty(tmp)) + cpumask_andnot(tmp, cpu_online_mask, pad_busy_cpus); + if (cpumask_empty(tmp)) { + mutex_unlock(&isolated_cpus_lock); + return; + } + for_each_cpu(cpu, tmp) { + if (cpu_weight[cpu] < min_weight) { + min_weight = cpu_weight[cpu]; + preferred_cpu = cpu; + } + } + + if (tsk_in_cpu[tsk_index] != -1) + cpumask_clear_cpu(tsk_in_cpu[tsk_index], pad_busy_cpus); + tsk_in_cpu[tsk_index] = preferred_cpu; + cpumask_set_cpu(preferred_cpu, pad_busy_cpus); + cpu_weight[preferred_cpu]++; + mutex_unlock(&isolated_cpus_lock); + + set_cpus_allowed_ptr(current, cpumask_of(preferred_cpu)); +} + +static void exit_round_robin(unsigned int tsk_index) +{ + struct cpumask *pad_busy_cpus = to_cpumask(pad_busy_cpus_bits); + cpumask_clear_cpu(tsk_in_cpu[tsk_index], pad_busy_cpus); + tsk_in_cpu[tsk_index] = -1; +} + +static unsigned int idle_pct = 5; /* percentage */ +static unsigned int round_robin_time = 10; /* second */ +static int power_saving_thread(void *data) +{ + struct sched_param param = {.sched_priority = 1}; + int do_sleep; + unsigned int tsk_index = (unsigned long)data; + u64 last_jiffies = 0; + + sched_setscheduler(current, SCHED_RR, ¶m); + + while (!kthread_should_stop()) { + int cpu; + u64 expire_time; + + try_to_freeze(); + + /* round robin to cpus */ + if (last_jiffies + round_robin_time * HZ < jiffies) { + last_jiffies = jiffies; + round_robin_cpu(tsk_index); + } + + do_sleep = 0; + + current_thread_info()->status &= ~TS_POLLING; + /* + * TS_POLLING-cleared state must be visible before we test + * NEED_RESCHED: + */ + smp_mb(); + + expire_time = jiffies + HZ * (100 - idle_pct) / 100; + + while (!need_resched()) { + local_irq_disable(); + cpu = smp_processor_id(); + clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_ENTER, + &cpu); + stop_critical_timings(); + + __monitor((void *)¤t_thread_info()->flags, 0, 0); + smp_mb(); + if (!need_resched()) + __mwait(power_saving_mwait_eax, 1); + + start_critical_timings(); + clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_EXIT, + &cpu); + local_irq_enable(); + + if (jiffies > expire_time) { + do_sleep = 1; + break; + } + } + + current_thread_info()->status |= TS_POLLING; + + /* + * current sched_rt has threshold for rt task running time. + * When a rt task uses 95% CPU time, the rt thread will be + * scheduled out for 5% CPU time to not starve other tasks. But + * the mechanism only works when all CPUs have RT task running, + * as if one CPU hasn't RT task, RT task from other CPUs will + * borrow CPU time from this CPU and cause RT task use > 95% + * CPU time. To make 'avoid staration' work, takes a nap here. + */ + if (do_sleep) + schedule_timeout_killable(HZ * idle_pct / 100); + } + + exit_round_robin(tsk_index); + return 0; +} + +static struct task_struct *ps_tsks[NR_CPUS]; +static unsigned int ps_tsk_num; +static int create_power_saving_task(void) +{ + ps_tsks[ps_tsk_num] = kthread_run(power_saving_thread, + (void *)(unsigned long)ps_tsk_num, + "power_saving/%d", ps_tsk_num); + if (ps_tsks[ps_tsk_num]) { + ps_tsk_num++; + return 0; + } + return -EINVAL; +} + +static void destroy_power_saving_task(void) +{ + if (ps_tsk_num > 0) { + ps_tsk_num--; + kthread_stop(ps_tsks[ps_tsk_num]); + } +} + +static void set_power_saving_task_num(unsigned int num) +{ + if (num > ps_tsk_num) { + while (ps_tsk_num < num) { + if (create_power_saving_task()) + return; + } + } else if (num < ps_tsk_num) { + while (ps_tsk_num > num) + destroy_power_saving_task(); + } +} + +static int acpi_pad_idle_cpus(unsigned int num_cpus) +{ + get_online_cpus(); + + num_cpus = min_t(unsigned int, num_cpus, num_online_cpus()); + set_power_saving_task_num(num_cpus); + + put_online_cpus(); + return 0; +} + +static uint32_t acpi_pad_idle_cpus_num(void) +{ + return ps_tsk_num; +} + +static ssize_t acpi_pad_rrtime_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + unsigned long num; + if (strict_strtoul(buf, 0, &num)) + return -EINVAL; + if (num < 1 || num >= 100) + return -EINVAL; + mutex_lock(&isolated_cpus_lock); + round_robin_time = num; + mutex_unlock(&isolated_cpus_lock); + return count; +} + +static ssize_t acpi_pad_rrtime_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return scnprintf(buf, PAGE_SIZE, "%d", round_robin_time); +} +static DEVICE_ATTR(rrtime, S_IRUGO|S_IWUSR, + acpi_pad_rrtime_show, + acpi_pad_rrtime_store); + +static ssize_t acpi_pad_idlepct_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + unsigned long num; + if (strict_strtoul(buf, 0, &num)) + return -EINVAL; + if (num < 1 || num >= 100) + return -EINVAL; + mutex_lock(&isolated_cpus_lock); + idle_pct = num; + mutex_unlock(&isolated_cpus_lock); + return count; +} + +static ssize_t acpi_pad_idlepct_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return scnprintf(buf, PAGE_SIZE, "%d", idle_pct); +} +static DEVICE_ATTR(idlepct, S_IRUGO|S_IWUSR, + acpi_pad_idlepct_show, + acpi_pad_idlepct_store); + +static ssize_t acpi_pad_idlecpus_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + unsigned long num; + if (strict_strtoul(buf, 0, &num)) + return -EINVAL; + mutex_lock(&isolated_cpus_lock); + acpi_pad_idle_cpus(num); + mutex_unlock(&isolated_cpus_lock); + return count; +} + +static ssize_t acpi_pad_idlecpus_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return cpumask_scnprintf(buf, PAGE_SIZE, + to_cpumask(pad_busy_cpus_bits)); +} +static DEVICE_ATTR(idlecpus, S_IRUGO|S_IWUSR, + acpi_pad_idlecpus_show, + acpi_pad_idlecpus_store); + +static int acpi_pad_add_sysfs(struct acpi_device *device) +{ + int result; + + result = device_create_file(&device->dev, &dev_attr_idlecpus); + if (result) + return -ENODEV; + result = device_create_file(&device->dev, &dev_attr_idlepct); + if (result) { + device_remove_file(&device->dev, &dev_attr_idlecpus); + return -ENODEV; + } + result = device_create_file(&device->dev, &dev_attr_rrtime); + if (result) { + device_remove_file(&device->dev, &dev_attr_idlecpus); + device_remove_file(&device->dev, &dev_attr_idlepct); + return -ENODEV; + } + return 0; +} + +static void acpi_pad_remove_sysfs(struct acpi_device *device) +{ + device_remove_file(&device->dev, &dev_attr_idlecpus); + device_remove_file(&device->dev, &dev_attr_idlepct); + device_remove_file(&device->dev, &dev_attr_rrtime); +} + +/* Query firmware how many CPUs should be idle */ +static int acpi_pad_pur(acpi_handle handle, int *num_cpus) +{ + struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL}; + acpi_status status; + union acpi_object *package; + int rev, num, ret = -EINVAL; + + status = acpi_evaluate_object(handle, "_PUR", NULL, &buffer); + if (ACPI_FAILURE(status)) + return -EINVAL; + package = buffer.pointer; + if (package->type != ACPI_TYPE_PACKAGE || package->package.count != 2) + goto out; + rev = package->package.elements[0].integer.value; + num = package->package.elements[1].integer.value; + if (rev != 1) + goto out; + *num_cpus = num; + ret = 0; +out: + kfree(buffer.pointer); + return ret; +} + +/* Notify firmware how many CPUs are idle */ +static void acpi_pad_ost(acpi_handle handle, int stat, + uint32_t idle_cpus) +{ + union acpi_object params[3] = { + {.type = ACPI_TYPE_INTEGER,}, + {.type = ACPI_TYPE_INTEGER,}, + {.type = ACPI_TYPE_BUFFER,}, + }; + struct acpi_object_list arg_list = {3, params}; + + params[0].integer.value = ACPI_PROCESSOR_AGGREGATOR_NOTIFY; + params[1].integer.value = stat; + params[2].buffer.length = 4; + params[2].buffer.pointer = (void *)&idle_cpus; + acpi_evaluate_object(handle, "_OST", &arg_list, NULL); +} + +static void acpi_pad_handle_notify(acpi_handle handle) +{ + int num_cpus, ret; + uint32_t idle_cpus; + + mutex_lock(&isolated_cpus_lock); + if (acpi_pad_pur(handle, &num_cpus)) { + mutex_unlock(&isolated_cpus_lock); + return; + } + ret = acpi_pad_idle_cpus(num_cpus); + idle_cpus = acpi_pad_idle_cpus_num(); + if (!ret) + acpi_pad_ost(handle, 0, idle_cpus); + else + acpi_pad_ost(handle, 1, 0); + mutex_unlock(&isolated_cpus_lock); +} + +static void acpi_pad_notify(acpi_handle handle, u32 event, + void *data) +{ + struct acpi_device *device = data; + + switch (event) { + case ACPI_PROCESSOR_AGGREGATOR_NOTIFY: + acpi_pad_handle_notify(handle); + acpi_bus_generate_proc_event(device, event, 0); + acpi_bus_generate_netlink_event(device->pnp.device_class, + dev_name(&device->dev), event, 0); + break; + default: + printk(KERN_WARNING"Unsupported event [0x%x]\n", event); + break; + } +} + +static int acpi_pad_add(struct acpi_device *device) +{ + acpi_status status; + + strcpy(acpi_device_name(device), ACPI_PROCESSOR_AGGREGATOR_DEVICE_NAME); + strcpy(acpi_device_class(device), ACPI_PROCESSOR_AGGREGATOR_CLASS); + + if (acpi_pad_add_sysfs(device)) + return -ENODEV; + + status = acpi_install_notify_handler(device->handle, + ACPI_DEVICE_NOTIFY, acpi_pad_notify, device); + if (ACPI_FAILURE(status)) { + acpi_pad_remove_sysfs(device); + return -ENODEV; + } + + return 0; +} + +static int acpi_pad_remove(struct acpi_device *device, + int type) +{ + mutex_lock(&isolated_cpus_lock); + acpi_pad_idle_cpus(0); + mutex_unlock(&isolated_cpus_lock); + + acpi_remove_notify_handler(device->handle, + ACPI_DEVICE_NOTIFY, acpi_pad_notify); + acpi_pad_remove_sysfs(device); + return 0; +} + +static const struct acpi_device_id pad_device_ids[] = { + {"ACPI000C", 0}, + {"", 0}, +}; +MODULE_DEVICE_TABLE(acpi, pad_device_ids); + +static struct acpi_driver acpi_pad_driver = { + .name = "processor_aggregator", + .class = ACPI_PROCESSOR_AGGREGATOR_CLASS, + .ids = pad_device_ids, + .ops = { + .add = acpi_pad_add, + .remove = acpi_pad_remove, + }, +}; + +static int __init acpi_pad_init(void) +{ + power_saving_mwait_init(); + if (power_saving_mwait_eax == 0) + return -EINVAL; + + return acpi_bus_register_driver(&acpi_pad_driver); +} + +static void __exit acpi_pad_exit(void) +{ + acpi_bus_unregister_driver(&acpi_pad_driver); +} + +module_init(acpi_pad_init); +module_exit(acpi_pad_exit); +MODULE_AUTHOR("Shaohua Li<shaohua.li@intel.com>"); +MODULE_DESCRIPTION("ACPI Processor Aggregator Driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/acpi/dock.c b/drivers/acpi/dock.c index 3a2cfefc71a..7338b6a3e04 100644 --- a/drivers/acpi/dock.c +++ b/drivers/acpi/dock.c @@ -67,7 +67,7 @@ struct dock_station { struct list_head dependent_devices; struct list_head hotplug_devices; - struct list_head sibiling; + struct list_head sibling; struct platform_device *dock_device; }; static LIST_HEAD(dock_stations); @@ -275,7 +275,7 @@ int is_dock_device(acpi_handle handle) if (is_dock(handle)) return 1; - list_for_each_entry(dock_station, &dock_stations, sibiling) { + list_for_each_entry(dock_station, &dock_stations, sibling) { if (find_dock_dependent_device(dock_station, handle)) return 1; } @@ -619,7 +619,7 @@ register_hotplug_dock_device(acpi_handle handle, struct acpi_dock_ops *ops, * make sure this handle is for a device dependent on the dock, * this would include the dock station itself */ - list_for_each_entry(dock_station, &dock_stations, sibiling) { + list_for_each_entry(dock_station, &dock_stations, sibling) { /* * An ATA bay can be in a dock and itself can be ejected * seperately, so there are two 'dock stations' which need the @@ -651,7 +651,7 @@ void unregister_hotplug_dock_device(acpi_handle handle) if (!dock_station_count) return; - list_for_each_entry(dock_station, &dock_stations, sibiling) { + list_for_each_entry(dock_station, &dock_stations, sibling) { dd = find_dock_dependent_device(dock_station, handle); if (dd) dock_del_hotplug_device(dock_station, dd); @@ -787,7 +787,7 @@ static int acpi_dock_notifier_call(struct notifier_block *this, if (event != ACPI_NOTIFY_BUS_CHECK && event != ACPI_NOTIFY_DEVICE_CHECK && event != ACPI_NOTIFY_EJECT_REQUEST) return 0; - list_for_each_entry(dock_station, &dock_stations, sibiling) { + list_for_each_entry(dock_station, &dock_stations, sibling) { if (dock_station->handle == handle) { struct dock_data *dock_data; @@ -958,7 +958,7 @@ static int dock_add(acpi_handle handle) dock_station->last_dock_time = jiffies - HZ; INIT_LIST_HEAD(&dock_station->dependent_devices); INIT_LIST_HEAD(&dock_station->hotplug_devices); - INIT_LIST_HEAD(&dock_station->sibiling); + INIT_LIST_HEAD(&dock_station->sibling); spin_lock_init(&dock_station->dd_lock); mutex_init(&dock_station->hp_lock); ATOMIC_INIT_NOTIFIER_HEAD(&dock_notifier_list); @@ -1044,7 +1044,7 @@ static int dock_add(acpi_handle handle) add_dock_dependent_device(dock_station, dd); dock_station_count++; - list_add(&dock_station->sibiling, &dock_stations); + list_add(&dock_station->sibling, &dock_stations); return 0; dock_add_err_unregister: @@ -1149,7 +1149,7 @@ static void __exit dock_exit(void) struct dock_station *tmp; unregister_acpi_bus_notifier(&dock_acpi_notifier); - list_for_each_entry_safe(dock_station, tmp, &dock_stations, sibiling) + list_for_each_entry_safe(dock_station, tmp, &dock_stations, sibling) dock_remove(dock_station); } diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index f70796081c4..baef28c1e63 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -119,6 +119,8 @@ static struct acpi_ec { } *boot_ec, *first_ec; static int EC_FLAGS_MSI; /* Out-of-spec MSI controller */ +static int EC_FLAGS_VALIDATE_ECDT; /* ASUStec ECDTs need to be validated */ +static int EC_FLAGS_SKIP_DSDT_SCAN; /* Not all BIOS survive early DSDT scan */ /* -------------------------------------------------------------------------- Transaction Management @@ -232,10 +234,8 @@ static int ec_poll(struct acpi_ec *ec) } advance_transaction(ec, acpi_ec_read_status(ec)); } while (time_before(jiffies, delay)); - if (!ec->curr->irq_count || - (acpi_ec_read_status(ec) & ACPI_EC_FLAG_IBF)) + if (acpi_ec_read_status(ec) & ACPI_EC_FLAG_IBF) break; - /* try restart command if we get any false interrupts */ pr_debug(PREFIX "controller reset, restart transaction\n"); spin_lock_irqsave(&ec->curr_lock, flags); start_transaction(ec); @@ -899,6 +899,44 @@ static const struct acpi_device_id ec_device_ids[] = { {"", 0}, }; +/* Some BIOS do not survive early DSDT scan, skip it */ +static int ec_skip_dsdt_scan(const struct dmi_system_id *id) +{ + EC_FLAGS_SKIP_DSDT_SCAN = 1; + return 0; +} + +/* ASUStek often supplies us with broken ECDT, validate it */ +static int ec_validate_ecdt(const struct dmi_system_id *id) +{ + EC_FLAGS_VALIDATE_ECDT = 1; + return 0; +} + +/* MSI EC needs special treatment, enable it */ +static int ec_flag_msi(const struct dmi_system_id *id) +{ + EC_FLAGS_MSI = 1; + EC_FLAGS_VALIDATE_ECDT = 1; + return 0; +} + +static struct dmi_system_id __initdata ec_dmi_table[] = { + { + ec_skip_dsdt_scan, "Compal JFL92", { + DMI_MATCH(DMI_BIOS_VENDOR, "COMPAL"), + DMI_MATCH(DMI_BOARD_NAME, "JFL92") }, NULL}, + { + ec_flag_msi, "MSI hardware", { + DMI_MATCH(DMI_BIOS_VENDOR, "Micro-Star"), + DMI_MATCH(DMI_CHASSIS_VENDOR, "MICRO-Star") }, NULL}, + { + ec_validate_ecdt, "ASUS hardware", { + DMI_MATCH(DMI_BIOS_VENDOR, "ASUS") }, NULL}, + {}, +}; + + int __init acpi_ec_ecdt_probe(void) { acpi_status status; @@ -911,11 +949,7 @@ int __init acpi_ec_ecdt_probe(void) /* * Generate a boot ec context */ - if (dmi_name_in_vendors("Micro-Star") || - dmi_name_in_vendors("Notebook")) { - pr_info(PREFIX "Enabling special treatment for EC from MSI.\n"); - EC_FLAGS_MSI = 1; - } + dmi_check_system(ec_dmi_table); status = acpi_get_table(ACPI_SIG_ECDT, 1, (struct acpi_table_header **)&ecdt_ptr); if (ACPI_SUCCESS(status)) { @@ -926,7 +960,7 @@ int __init acpi_ec_ecdt_probe(void) boot_ec->handle = ACPI_ROOT_OBJECT; acpi_get_handle(ACPI_ROOT_OBJECT, ecdt_ptr->id, &boot_ec->handle); /* Don't trust ECDT, which comes from ASUSTek */ - if (!dmi_name_in_vendors("ASUS") && EC_FLAGS_MSI == 0) + if (!EC_FLAGS_VALIDATE_ECDT) goto install; saved_ec = kmalloc(sizeof(struct acpi_ec), GFP_KERNEL); if (!saved_ec) @@ -934,6 +968,10 @@ int __init acpi_ec_ecdt_probe(void) memcpy(saved_ec, boot_ec, sizeof(struct acpi_ec)); /* fall through */ } + + if (EC_FLAGS_SKIP_DSDT_SCAN) + return -ENODEV; + /* This workaround is needed only on some broken machines, * which require early EC, but fail to provide ECDT */ printk(KERN_DEBUG PREFIX "Look up EC in DSDT\n"); diff --git a/drivers/acpi/proc.c b/drivers/acpi/proc.c index d0d550d22a6..f8b6f555ba5 100644 --- a/drivers/acpi/proc.c +++ b/drivers/acpi/proc.c @@ -398,6 +398,8 @@ acpi_system_write_wakeup_device(struct file *file, if (len > 4) len = 4; + if (len < 0) + return -EFAULT; if (copy_from_user(strbuf, buffer, len)) return -EFAULT; diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c index c2d4d6e0936..c567b46dfa0 100644 --- a/drivers/acpi/processor_core.c +++ b/drivers/acpi/processor_core.c @@ -863,13 +863,6 @@ static int acpi_processor_add(struct acpi_device *device) goto err_remove_sysfs; } - if (pr->flags.throttling) { - printk(KERN_INFO PREFIX "%s [%s] (supports", - acpi_device_name(device), acpi_device_bid(device)); - printk(" %d throttling states", pr->throttling.state_count); - printk(")\n"); - } - return 0; err_remove_sysfs: diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c index 468921bed22..14a7481c97d 100644 --- a/drivers/acpi/scan.c +++ b/drivers/acpi/scan.c @@ -1052,6 +1052,8 @@ static void acpi_device_set_id(struct acpi_device *device) device->flags.bus_address = 1; } + kfree(info); + /* * Some devices don't reliably have _HIDs & _CIDs, so add * synthetic HIDs to make sure drivers can find them. @@ -1325,13 +1327,8 @@ static int acpi_bus_scan(acpi_handle handle, struct acpi_bus_ops *ops, struct acpi_device **child) { acpi_status status; - struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL }; void *device = NULL; - acpi_get_name(handle, ACPI_FULL_PATHNAME, &buffer); - printk(KERN_INFO PREFIX "Enumerating devices from [%s]\n", - (char *) buffer.pointer); - status = acpi_bus_check_add(handle, 0, ops, &device); if (ACPI_SUCCESS(status)) acpi_walk_namespace(ACPI_TYPE_ANY, handle, ACPI_UINT32_MAX, diff --git a/drivers/block/DAC960.c b/drivers/block/DAC960.c index 6fa7b0fdbdf..eb4fa194394 100644 --- a/drivers/block/DAC960.c +++ b/drivers/block/DAC960.c @@ -38,6 +38,7 @@ #include <linux/slab.h> #include <linux/smp_lock.h> #include <linux/proc_fs.h> +#include <linux/seq_file.h> #include <linux/reboot.h> #include <linux/spinlock.h> #include <linux/timer.h> @@ -6422,16 +6423,10 @@ static bool DAC960_V2_ExecuteUserCommand(DAC960_Controller_T *Controller, return true; } - -/* - DAC960_ProcReadStatus implements reading /proc/rd/status. -*/ - -static int DAC960_ProcReadStatus(char *Page, char **Start, off_t Offset, - int Count, int *EOF, void *Data) +static int dac960_proc_show(struct seq_file *m, void *v) { unsigned char *StatusMessage = "OK\n"; - int ControllerNumber, BytesAvailable; + int ControllerNumber; for (ControllerNumber = 0; ControllerNumber < DAC960_ControllerCount; ControllerNumber++) @@ -6444,52 +6439,49 @@ static int DAC960_ProcReadStatus(char *Page, char **Start, off_t Offset, break; } } - BytesAvailable = strlen(StatusMessage) - Offset; - if (Count >= BytesAvailable) - { - Count = BytesAvailable; - *EOF = true; - } - if (Count <= 0) return 0; - *Start = Page; - memcpy(Page, &StatusMessage[Offset], Count); - return Count; + seq_puts(m, StatusMessage); + return 0; } +static int dac960_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, dac960_proc_show, NULL); +} -/* - DAC960_ProcReadInitialStatus implements reading /proc/rd/cN/initial_status. -*/ +static const struct file_operations dac960_proc_fops = { + .owner = THIS_MODULE, + .open = dac960_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; -static int DAC960_ProcReadInitialStatus(char *Page, char **Start, off_t Offset, - int Count, int *EOF, void *Data) +static int dac960_initial_status_proc_show(struct seq_file *m, void *v) { - DAC960_Controller_T *Controller = (DAC960_Controller_T *) Data; - int BytesAvailable = Controller->InitialStatusLength - Offset; - if (Count >= BytesAvailable) - { - Count = BytesAvailable; - *EOF = true; - } - if (Count <= 0) return 0; - *Start = Page; - memcpy(Page, &Controller->CombinedStatusBuffer[Offset], Count); - return Count; + DAC960_Controller_T *Controller = (DAC960_Controller_T *)m->private; + seq_printf(m, "%.*s", Controller->InitialStatusLength, Controller->CombinedStatusBuffer); + return 0; } +static int dac960_initial_status_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, dac960_initial_status_proc_show, PDE(inode)->data); +} -/* - DAC960_ProcReadCurrentStatus implements reading /proc/rd/cN/current_status. -*/ +static const struct file_operations dac960_initial_status_proc_fops = { + .owner = THIS_MODULE, + .open = dac960_initial_status_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; -static int DAC960_ProcReadCurrentStatus(char *Page, char **Start, off_t Offset, - int Count, int *EOF, void *Data) +static int dac960_current_status_proc_show(struct seq_file *m, void *v) { - DAC960_Controller_T *Controller = (DAC960_Controller_T *) Data; + DAC960_Controller_T *Controller = (DAC960_Controller_T *) m->private; unsigned char *StatusMessage = "No Rebuild or Consistency Check in Progress\n"; int ProgressMessageLength = strlen(StatusMessage); - int BytesAvailable; if (jiffies != Controller->LastCurrentStatusTime) { Controller->CurrentStatusLength = 0; @@ -6513,49 +6505,41 @@ static int DAC960_ProcReadCurrentStatus(char *Page, char **Start, off_t Offset, } Controller->LastCurrentStatusTime = jiffies; } - BytesAvailable = Controller->CurrentStatusLength - Offset; - if (Count >= BytesAvailable) - { - Count = BytesAvailable; - *EOF = true; - } - if (Count <= 0) return 0; - *Start = Page; - memcpy(Page, &Controller->CurrentStatusBuffer[Offset], Count); - return Count; + seq_printf(m, "%.*s", Controller->CurrentStatusLength, Controller->CurrentStatusBuffer); + return 0; } +static int dac960_current_status_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, dac960_current_status_proc_show, PDE(inode)->data); +} -/* - DAC960_ProcReadUserCommand implements reading /proc/rd/cN/user_command. -*/ +static const struct file_operations dac960_current_status_proc_fops = { + .owner = THIS_MODULE, + .open = dac960_current_status_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; -static int DAC960_ProcReadUserCommand(char *Page, char **Start, off_t Offset, - int Count, int *EOF, void *Data) +static int dac960_user_command_proc_show(struct seq_file *m, void *v) { - DAC960_Controller_T *Controller = (DAC960_Controller_T *) Data; - int BytesAvailable = Controller->UserStatusLength - Offset; - if (Count >= BytesAvailable) - { - Count = BytesAvailable; - *EOF = true; - } - if (Count <= 0) return 0; - *Start = Page; - memcpy(Page, &Controller->UserStatusBuffer[Offset], Count); - return Count; -} + DAC960_Controller_T *Controller = (DAC960_Controller_T *)m->private; + seq_printf(m, "%.*s", Controller->UserStatusLength, Controller->UserStatusBuffer); + return 0; +} -/* - DAC960_ProcWriteUserCommand implements writing /proc/rd/cN/user_command. -*/ +static int dac960_user_command_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, dac960_user_command_proc_show, PDE(inode)->data); +} -static int DAC960_ProcWriteUserCommand(struct file *file, +static ssize_t dac960_user_command_proc_write(struct file *file, const char __user *Buffer, - unsigned long Count, void *Data) + size_t Count, loff_t *pos) { - DAC960_Controller_T *Controller = (DAC960_Controller_T *) Data; + DAC960_Controller_T *Controller = (DAC960_Controller_T *) PDE(file->f_path.dentry->d_inode)->data; unsigned char CommandBuffer[80]; int Length; if (Count > sizeof(CommandBuffer)-1) return -EINVAL; @@ -6572,6 +6556,14 @@ static int DAC960_ProcWriteUserCommand(struct file *file, ? Count : -EBUSY); } +static const struct file_operations dac960_user_command_proc_fops = { + .owner = THIS_MODULE, + .open = dac960_user_command_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = dac960_user_command_proc_write, +}; /* DAC960_CreateProcEntries creates the /proc/rd/... entries for the @@ -6586,23 +6578,17 @@ static void DAC960_CreateProcEntries(DAC960_Controller_T *Controller) if (DAC960_ProcDirectoryEntry == NULL) { DAC960_ProcDirectoryEntry = proc_mkdir("rd", NULL); - StatusProcEntry = create_proc_read_entry("status", 0, + StatusProcEntry = proc_create("status", 0, DAC960_ProcDirectoryEntry, - DAC960_ProcReadStatus, NULL); + &dac960_proc_fops); } sprintf(Controller->ControllerName, "c%d", Controller->ControllerNumber); ControllerProcEntry = proc_mkdir(Controller->ControllerName, DAC960_ProcDirectoryEntry); - create_proc_read_entry("initial_status", 0, ControllerProcEntry, - DAC960_ProcReadInitialStatus, Controller); - create_proc_read_entry("current_status", 0, ControllerProcEntry, - DAC960_ProcReadCurrentStatus, Controller); - UserCommandProcEntry = - create_proc_read_entry("user_command", S_IWUSR | S_IRUSR, - ControllerProcEntry, DAC960_ProcReadUserCommand, - Controller); - UserCommandProcEntry->write_proc = DAC960_ProcWriteUserCommand; + proc_create_data("initial_status", 0, ControllerProcEntry, &dac960_initial_status_proc_fops, Controller); + proc_create_data("current_status", 0, ControllerProcEntry, &dac960_current_status_proc_fops, Controller); + UserCommandProcEntry = proc_create_data("user_command", S_IWUSR | S_IRUSR, ControllerProcEntry, &dac960_user_command_proc_fops, Controller); Controller->ControllerProcEntry = ControllerProcEntry; } diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index 1ece0b47b58..fb5be2d95d5 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -36,9 +36,11 @@ #include <linux/proc_fs.h> #include <linux/seq_file.h> #include <linux/init.h> +#include <linux/jiffies.h> #include <linux/hdreg.h> #include <linux/spinlock.h> #include <linux/compat.h> +#include <linux/mutex.h> #include <asm/uaccess.h> #include <asm/io.h> @@ -155,6 +157,10 @@ static struct board_type products[] = { static ctlr_info_t *hba[MAX_CTLR]; +static struct task_struct *cciss_scan_thread; +static DEFINE_MUTEX(scan_mutex); +static LIST_HEAD(scan_q); + static void do_cciss_request(struct request_queue *q); static irqreturn_t do_cciss_intr(int irq, void *dev_id); static int cciss_open(struct block_device *bdev, fmode_t mode); @@ -164,9 +170,9 @@ static int cciss_ioctl(struct block_device *bdev, fmode_t mode, static int cciss_getgeo(struct block_device *bdev, struct hd_geometry *geo); static int cciss_revalidate(struct gendisk *disk); -static int rebuild_lun_table(ctlr_info_t *h, int first_time); +static int rebuild_lun_table(ctlr_info_t *h, int first_time, int via_ioctl); static int deregister_disk(ctlr_info_t *h, int drv_index, - int clear_all); + int clear_all, int via_ioctl); static void cciss_read_capacity(int ctlr, int logvol, int withirq, sector_t *total_size, unsigned int *block_size); @@ -189,8 +195,13 @@ static int sendcmd_withirq_core(ctlr_info_t *h, CommandList_struct *c, static int process_sendcmd_error(ctlr_info_t *h, CommandList_struct *c); static void fail_all_cmds(unsigned long ctlr); +static int add_to_scan_list(struct ctlr_info *h); static int scan_thread(void *data); static int check_for_unit_attention(ctlr_info_t *h, CommandList_struct *c); +static void cciss_hba_release(struct device *dev); +static void cciss_device_release(struct device *dev); +static void cciss_free_gendisk(ctlr_info_t *h, int drv_index); +static void cciss_free_drive_info(ctlr_info_t *h, int drv_index); #ifdef CONFIG_PROC_FS static void cciss_procinit(int i); @@ -245,7 +256,10 @@ static inline void removeQ(CommandList_struct *c) #include "cciss_scsi.c" /* For SCSI tape support */ -#define RAID_UNKNOWN 6 +static const char *raid_label[] = { "0", "4", "1(1+0)", "5", "5+1", "ADG", + "UNKNOWN" +}; +#define RAID_UNKNOWN (sizeof(raid_label) / sizeof(raid_label[0])-1) #ifdef CONFIG_PROC_FS @@ -255,9 +269,6 @@ static inline void removeQ(CommandList_struct *c) #define ENG_GIG 1000000000 #define ENG_GIG_FACTOR (ENG_GIG/512) #define ENGAGE_SCSI "engage scsi" -static const char *raid_label[] = { "0", "4", "1(1+0)", "5", "5+1", "ADG", - "UNKNOWN" -}; static struct proc_dir_entry *proc_cciss; @@ -318,7 +329,7 @@ static int cciss_seq_show(struct seq_file *seq, void *v) ctlr_info_t *h = seq->private; unsigned ctlr = h->ctlr; loff_t *pos = v; - drive_info_struct *drv = &h->drv[*pos]; + drive_info_struct *drv = h->drv[*pos]; if (*pos > h->highest_lun) return 0; @@ -331,7 +342,7 @@ static int cciss_seq_show(struct seq_file *seq, void *v) vol_sz_frac *= 100; sector_div(vol_sz_frac, ENG_GIG_FACTOR); - if (drv->raid_level > 5) + if (drv->raid_level < 0 || drv->raid_level > RAID_UNKNOWN) drv->raid_level = RAID_UNKNOWN; seq_printf(seq, "cciss/c%dd%d:" "\t%4u.%02uGB\tRAID %s\n", @@ -454,9 +465,19 @@ static void __devinit cciss_procinit(int i) #define to_hba(n) container_of(n, struct ctlr_info, dev) #define to_drv(n) container_of(n, drive_info_struct, dev) -static struct device_type cciss_host_type = { - .name = "cciss_host", -}; +static ssize_t host_store_rescan(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct ctlr_info *h = to_hba(dev); + + add_to_scan_list(h); + wake_up_process(cciss_scan_thread); + wait_for_completion_interruptible(&h->scan_wait); + + return count; +} +DEVICE_ATTR(rescan, S_IWUSR, NULL, host_store_rescan); static ssize_t dev_show_unique_id(struct device *dev, struct device_attribute *attr, @@ -560,11 +581,101 @@ static ssize_t dev_show_rev(struct device *dev, } DEVICE_ATTR(rev, S_IRUGO, dev_show_rev, NULL); +static ssize_t cciss_show_lunid(struct device *dev, + struct device_attribute *attr, char *buf) +{ + drive_info_struct *drv = to_drv(dev); + struct ctlr_info *h = to_hba(drv->dev.parent); + unsigned long flags; + unsigned char lunid[8]; + + spin_lock_irqsave(CCISS_LOCK(h->ctlr), flags); + if (h->busy_configuring) { + spin_unlock_irqrestore(CCISS_LOCK(h->ctlr), flags); + return -EBUSY; + } + if (!drv->heads) { + spin_unlock_irqrestore(CCISS_LOCK(h->ctlr), flags); + return -ENOTTY; + } + memcpy(lunid, drv->LunID, sizeof(lunid)); + spin_unlock_irqrestore(CCISS_LOCK(h->ctlr), flags); + return snprintf(buf, 20, "0x%02x%02x%02x%02x%02x%02x%02x%02x\n", + lunid[0], lunid[1], lunid[2], lunid[3], + lunid[4], lunid[5], lunid[6], lunid[7]); +} +DEVICE_ATTR(lunid, S_IRUGO, cciss_show_lunid, NULL); + +static ssize_t cciss_show_raid_level(struct device *dev, + struct device_attribute *attr, char *buf) +{ + drive_info_struct *drv = to_drv(dev); + struct ctlr_info *h = to_hba(drv->dev.parent); + int raid; + unsigned long flags; + + spin_lock_irqsave(CCISS_LOCK(h->ctlr), flags); + if (h->busy_configuring) { + spin_unlock_irqrestore(CCISS_LOCK(h->ctlr), flags); + return -EBUSY; + } + raid = drv->raid_level; + spin_unlock_irqrestore(CCISS_LOCK(h->ctlr), flags); + if (raid < 0 || raid > RAID_UNKNOWN) + raid = RAID_UNKNOWN; + + return snprintf(buf, strlen(raid_label[raid]) + 7, "RAID %s\n", + raid_label[raid]); +} +DEVICE_ATTR(raid_level, S_IRUGO, cciss_show_raid_level, NULL); + +static ssize_t cciss_show_usage_count(struct device *dev, + struct device_attribute *attr, char *buf) +{ + drive_info_struct *drv = to_drv(dev); + struct ctlr_info *h = to_hba(drv->dev.parent); + unsigned long flags; + int count; + + spin_lock_irqsave(CCISS_LOCK(h->ctlr), flags); + if (h->busy_configuring) { + spin_unlock_irqrestore(CCISS_LOCK(h->ctlr), flags); + return -EBUSY; + } + count = drv->usage_count; + spin_unlock_irqrestore(CCISS_LOCK(h->ctlr), flags); + return snprintf(buf, 20, "%d\n", count); +} +DEVICE_ATTR(usage_count, S_IRUGO, cciss_show_usage_count, NULL); + +static struct attribute *cciss_host_attrs[] = { + &dev_attr_rescan.attr, + NULL +}; + +static struct attribute_group cciss_host_attr_group = { + .attrs = cciss_host_attrs, +}; + +static const struct attribute_group *cciss_host_attr_groups[] = { + &cciss_host_attr_group, + NULL +}; + +static struct device_type cciss_host_type = { + .name = "cciss_host", + .groups = cciss_host_attr_groups, + .release = cciss_hba_release, +}; + static struct attribute *cciss_dev_attrs[] = { &dev_attr_unique_id.attr, &dev_attr_model.attr, &dev_attr_vendor.attr, &dev_attr_rev.attr, + &dev_attr_lunid.attr, + &dev_attr_raid_level.attr, + &dev_attr_usage_count.attr, NULL }; @@ -580,12 +691,24 @@ static const struct attribute_group *cciss_dev_attr_groups[] = { static struct device_type cciss_dev_type = { .name = "cciss_device", .groups = cciss_dev_attr_groups, + .release = cciss_device_release, }; static struct bus_type cciss_bus_type = { .name = "cciss", }; +/* + * cciss_hba_release is called when the reference count + * of h->dev goes to zero. + */ +static void cciss_hba_release(struct device *dev) +{ + /* + * nothing to do, but need this to avoid a warning + * about not having a release handler from lib/kref.c. + */ +} /* * Initialize sysfs entry for each controller. This sets up and registers @@ -609,6 +732,16 @@ static int cciss_create_hba_sysfs_entry(struct ctlr_info *h) static void cciss_destroy_hba_sysfs_entry(struct ctlr_info *h) { device_del(&h->dev); + put_device(&h->dev); /* final put. */ +} + +/* cciss_device_release is called when the reference count + * of h->drv[x]dev goes to zero. + */ +static void cciss_device_release(struct device *dev) +{ + drive_info_struct *drv = to_drv(dev); + kfree(drv); } /* @@ -617,24 +750,39 @@ static void cciss_destroy_hba_sysfs_entry(struct ctlr_info *h) * /sys/bus/pci/devices/<dev/ccis#/. We also create a link from * /sys/block/cciss!c#d# to this entry. */ -static int cciss_create_ld_sysfs_entry(struct ctlr_info *h, - drive_info_struct *drv, +static long cciss_create_ld_sysfs_entry(struct ctlr_info *h, int drv_index) { - device_initialize(&drv->dev); - drv->dev.type = &cciss_dev_type; - drv->dev.bus = &cciss_bus_type; - dev_set_name(&drv->dev, "c%dd%d", h->ctlr, drv_index); - drv->dev.parent = &h->dev; - return device_add(&drv->dev); + struct device *dev; + + if (h->drv[drv_index]->device_initialized) + return 0; + + dev = &h->drv[drv_index]->dev; + device_initialize(dev); + dev->type = &cciss_dev_type; + dev->bus = &cciss_bus_type; + dev_set_name(dev, "c%dd%d", h->ctlr, drv_index); + dev->parent = &h->dev; + h->drv[drv_index]->device_initialized = 1; + return device_add(dev); } /* * Remove sysfs entries for a logical drive. */ -static void cciss_destroy_ld_sysfs_entry(drive_info_struct *drv) +static void cciss_destroy_ld_sysfs_entry(struct ctlr_info *h, int drv_index, + int ctlr_exiting) { - device_del(&drv->dev); + struct device *dev = &h->drv[drv_index]->dev; + + /* special case for c*d0, we only destroy it on controller exit */ + if (drv_index == 0 && !ctlr_exiting) + return; + + device_del(dev); + put_device(dev); /* the "final" put. */ + h->drv[drv_index] = NULL; } /* @@ -751,7 +899,7 @@ static int cciss_open(struct block_device *bdev, fmode_t mode) printk(KERN_DEBUG "cciss_open %s\n", bdev->bd_disk->disk_name); #endif /* CCISS_DEBUG */ - if (host->busy_initializing || drv->busy_configuring) + if (drv->busy_configuring) return -EBUSY; /* * Root is allowed to open raw volume zero even if it's not configured @@ -767,7 +915,8 @@ static int cciss_open(struct block_device *bdev, fmode_t mode) if (MINOR(bdev->bd_dev) & 0x0f) { return -ENXIO; /* if it is, make sure we have a LUN ID */ - } else if (drv->LunID == 0) { + } else if (memcmp(drv->LunID, CTLR_LUNID, + sizeof(drv->LunID))) { return -ENXIO; } } @@ -1132,12 +1281,13 @@ static int cciss_ioctl(struct block_device *bdev, fmode_t mode, case CCISS_DEREGDISK: case CCISS_REGNEWD: case CCISS_REVALIDVOLS: - return rebuild_lun_table(host, 0); + return rebuild_lun_table(host, 0, 1); case CCISS_GETLUNINFO:{ LogvolInfo_struct luninfo; - luninfo.LunID = drv->LunID; + memcpy(&luninfo.LunID, drv->LunID, + sizeof(luninfo.LunID)); luninfo.num_opens = drv->usage_count; luninfo.num_parts = 0; if (copy_to_user(argp, &luninfo, @@ -1475,7 +1625,10 @@ static void cciss_check_queues(ctlr_info_t *h) /* make sure the disk has been added and the drive is real * because this can be called from the middle of init_one. */ - if (!(h->drv[curr_queue].queue) || !(h->drv[curr_queue].heads)) + if (!h->drv[curr_queue]) + continue; + if (!(h->drv[curr_queue]->queue) || + !(h->drv[curr_queue]->heads)) continue; blk_start_queue(h->gendisk[curr_queue]->queue); @@ -1532,13 +1685,11 @@ static void cciss_softirq_done(struct request *rq) spin_unlock_irqrestore(&h->lock, flags); } -static void log_unit_to_scsi3addr(ctlr_info_t *h, unsigned char scsi3addr[], - uint32_t log_unit) +static inline void log_unit_to_scsi3addr(ctlr_info_t *h, + unsigned char scsi3addr[], uint32_t log_unit) { - log_unit = h->drv[log_unit].LunID & 0x03fff; - memset(&scsi3addr[4], 0, 4); - memcpy(&scsi3addr[0], &log_unit, 4); - scsi3addr[3] |= 0x40; + memcpy(scsi3addr, h->drv[log_unit]->LunID, + sizeof(h->drv[log_unit]->LunID)); } /* This function gets the SCSI vendor, model, and revision of a logical drive @@ -1615,16 +1766,23 @@ static void cciss_get_serial_no(int ctlr, int logvol, int withirq, return; } -static void cciss_add_disk(ctlr_info_t *h, struct gendisk *disk, +/* + * cciss_add_disk sets up the block device queue for a logical drive + */ +static int cciss_add_disk(ctlr_info_t *h, struct gendisk *disk, int drv_index) { disk->queue = blk_init_queue(do_cciss_request, &h->lock); + if (!disk->queue) + goto init_queue_failure; sprintf(disk->disk_name, "cciss/c%dd%d", h->ctlr, drv_index); disk->major = h->major; disk->first_minor = drv_index << NWD_SHIFT; disk->fops = &cciss_fops; - disk->private_data = &h->drv[drv_index]; - disk->driverfs_dev = &h->drv[drv_index].dev; + if (cciss_create_ld_sysfs_entry(h, drv_index)) + goto cleanup_queue; + disk->private_data = h->drv[drv_index]; + disk->driverfs_dev = &h->drv[drv_index]->dev; /* Set up queue information */ blk_queue_bounce_limit(disk->queue, h->pdev->dma_mask); @@ -1642,14 +1800,21 @@ static void cciss_add_disk(ctlr_info_t *h, struct gendisk *disk, disk->queue->queuedata = h; blk_queue_logical_block_size(disk->queue, - h->drv[drv_index].block_size); + h->drv[drv_index]->block_size); /* Make sure all queue data is written out before */ - /* setting h->drv[drv_index].queue, as setting this */ + /* setting h->drv[drv_index]->queue, as setting this */ /* allows the interrupt handler to start the queue */ wmb(); - h->drv[drv_index].queue = disk->queue; + h->drv[drv_index]->queue = disk->queue; add_disk(disk); + return 0; + +cleanup_queue: + blk_cleanup_queue(disk->queue); + disk->queue = NULL; +init_queue_failure: + return -1; } /* This function will check the usage_count of the drive to be updated/added. @@ -1662,7 +1827,8 @@ static void cciss_add_disk(ctlr_info_t *h, struct gendisk *disk, * is also the controller node. Any changes to disk 0 will show up on * the next reboot. */ -static void cciss_update_drive_info(int ctlr, int drv_index, int first_time) +static void cciss_update_drive_info(int ctlr, int drv_index, int first_time, + int via_ioctl) { ctlr_info_t *h = hba[ctlr]; struct gendisk *disk; @@ -1672,21 +1838,13 @@ static void cciss_update_drive_info(int ctlr, int drv_index, int first_time) unsigned long flags = 0; int ret = 0; drive_info_struct *drvinfo; - int was_only_controller_node; /* Get information about the disk and modify the driver structure */ inq_buff = kmalloc(sizeof(InquiryData_struct), GFP_KERNEL); - drvinfo = kmalloc(sizeof(*drvinfo), GFP_KERNEL); + drvinfo = kzalloc(sizeof(*drvinfo), GFP_KERNEL); if (inq_buff == NULL || drvinfo == NULL) goto mem_msg; - /* See if we're trying to update the "controller node" - * this will happen the when the first logical drive gets - * created by ACU. - */ - was_only_controller_node = (drv_index == 0 && - h->drv[0].raid_level == -1); - /* testing to see if 16-byte CDBs are already being used */ if (h->cciss_read == CCISS_READ_16) { cciss_read_capacity_16(h->ctlr, drv_index, 1, @@ -1719,16 +1877,19 @@ static void cciss_update_drive_info(int ctlr, int drv_index, int first_time) drvinfo->model, drvinfo->rev); cciss_get_serial_no(ctlr, drv_index, 1, drvinfo->serial_no, sizeof(drvinfo->serial_no)); + /* Save the lunid in case we deregister the disk, below. */ + memcpy(drvinfo->LunID, h->drv[drv_index]->LunID, + sizeof(drvinfo->LunID)); /* Is it the same disk we already know, and nothing's changed? */ - if (h->drv[drv_index].raid_level != -1 && + if (h->drv[drv_index]->raid_level != -1 && ((memcmp(drvinfo->serial_no, - h->drv[drv_index].serial_no, 16) == 0) && - drvinfo->block_size == h->drv[drv_index].block_size && - drvinfo->nr_blocks == h->drv[drv_index].nr_blocks && - drvinfo->heads == h->drv[drv_index].heads && - drvinfo->sectors == h->drv[drv_index].sectors && - drvinfo->cylinders == h->drv[drv_index].cylinders)) + h->drv[drv_index]->serial_no, 16) == 0) && + drvinfo->block_size == h->drv[drv_index]->block_size && + drvinfo->nr_blocks == h->drv[drv_index]->nr_blocks && + drvinfo->heads == h->drv[drv_index]->heads && + drvinfo->sectors == h->drv[drv_index]->sectors && + drvinfo->cylinders == h->drv[drv_index]->cylinders)) /* The disk is unchanged, nothing to update */ goto freeret; @@ -1738,18 +1899,17 @@ static void cciss_update_drive_info(int ctlr, int drv_index, int first_time) * If the disk already exists then deregister it before proceeding * (unless it's the first disk (for the controller node). */ - if (h->drv[drv_index].raid_level != -1 && drv_index != 0) { + if (h->drv[drv_index]->raid_level != -1 && drv_index != 0) { printk(KERN_WARNING "disk %d has changed.\n", drv_index); spin_lock_irqsave(CCISS_LOCK(h->ctlr), flags); - h->drv[drv_index].busy_configuring = 1; + h->drv[drv_index]->busy_configuring = 1; spin_unlock_irqrestore(CCISS_LOCK(h->ctlr), flags); - /* deregister_disk sets h->drv[drv_index].queue = NULL + /* deregister_disk sets h->drv[drv_index]->queue = NULL * which keeps the interrupt handler from starting * the queue. */ - ret = deregister_disk(h, drv_index, 0); - h->drv[drv_index].busy_configuring = 0; + ret = deregister_disk(h, drv_index, 0, via_ioctl); } /* If the disk is in use return */ @@ -1757,22 +1917,31 @@ static void cciss_update_drive_info(int ctlr, int drv_index, int first_time) goto freeret; /* Save the new information from cciss_geometry_inquiry - * and serial number inquiry. + * and serial number inquiry. If the disk was deregistered + * above, then h->drv[drv_index] will be NULL. */ - h->drv[drv_index].block_size = drvinfo->block_size; - h->drv[drv_index].nr_blocks = drvinfo->nr_blocks; - h->drv[drv_index].heads = drvinfo->heads; - h->drv[drv_index].sectors = drvinfo->sectors; - h->drv[drv_index].cylinders = drvinfo->cylinders; - h->drv[drv_index].raid_level = drvinfo->raid_level; - memcpy(h->drv[drv_index].serial_no, drvinfo->serial_no, 16); - memcpy(h->drv[drv_index].vendor, drvinfo->vendor, VENDOR_LEN + 1); - memcpy(h->drv[drv_index].model, drvinfo->model, MODEL_LEN + 1); - memcpy(h->drv[drv_index].rev, drvinfo->rev, REV_LEN + 1); + if (h->drv[drv_index] == NULL) { + drvinfo->device_initialized = 0; + h->drv[drv_index] = drvinfo; + drvinfo = NULL; /* so it won't be freed below. */ + } else { + /* special case for cxd0 */ + h->drv[drv_index]->block_size = drvinfo->block_size; + h->drv[drv_index]->nr_blocks = drvinfo->nr_blocks; + h->drv[drv_index]->heads = drvinfo->heads; + h->drv[drv_index]->sectors = drvinfo->sectors; + h->drv[drv_index]->cylinders = drvinfo->cylinders; + h->drv[drv_index]->raid_level = drvinfo->raid_level; + memcpy(h->drv[drv_index]->serial_no, drvinfo->serial_no, 16); + memcpy(h->drv[drv_index]->vendor, drvinfo->vendor, + VENDOR_LEN + 1); + memcpy(h->drv[drv_index]->model, drvinfo->model, MODEL_LEN + 1); + memcpy(h->drv[drv_index]->rev, drvinfo->rev, REV_LEN + 1); + } ++h->num_luns; disk = h->gendisk[drv_index]; - set_capacity(disk, h->drv[drv_index].nr_blocks); + set_capacity(disk, h->drv[drv_index]->nr_blocks); /* If it's not disk 0 (drv_index != 0) * or if it was disk 0, but there was previously @@ -1780,8 +1949,15 @@ static void cciss_update_drive_info(int ctlr, int drv_index, int first_time) * (raid_leve == -1) then we want to update the * logical drive's information. */ - if (drv_index || first_time) - cciss_add_disk(h, disk, drv_index); + if (drv_index || first_time) { + if (cciss_add_disk(h, disk, drv_index) != 0) { + cciss_free_gendisk(h, drv_index); + cciss_free_drive_info(h, drv_index); + printk(KERN_WARNING "cciss:%d could not update " + "disk %d\n", h->ctlr, drv_index); + --h->num_luns; + } + } freeret: kfree(inq_buff); @@ -1793,28 +1969,70 @@ mem_msg: } /* This function will find the first index of the controllers drive array - * that has a -1 for the raid_level and will return that index. This is - * where new drives will be added. If the index to be returned is greater - * than the highest_lun index for the controller then highest_lun is set - * to this new index. If there are no available indexes then -1 is returned. - * "controller_node" is used to know if this is a real logical drive, or just - * the controller node, which determines if this counts towards highest_lun. + * that has a null drv pointer and allocate the drive info struct and + * will return that index This is where new drives will be added. + * If the index to be returned is greater than the highest_lun index for + * the controller then highest_lun is set * to this new index. + * If there are no available indexes or if tha allocation fails, then -1 + * is returned. * "controller_node" is used to know if this is a real + * logical drive, or just the controller node, which determines if this + * counts towards highest_lun. */ -static int cciss_find_free_drive_index(int ctlr, int controller_node) +static int cciss_alloc_drive_info(ctlr_info_t *h, int controller_node) { int i; + drive_info_struct *drv; + /* Search for an empty slot for our drive info */ for (i = 0; i < CISS_MAX_LUN; i++) { - if (hba[ctlr]->drv[i].raid_level == -1) { - if (i > hba[ctlr]->highest_lun) - if (!controller_node) - hba[ctlr]->highest_lun = i; + + /* if not cxd0 case, and it's occupied, skip it. */ + if (h->drv[i] && i != 0) + continue; + /* + * If it's cxd0 case, and drv is alloc'ed already, and a + * disk is configured there, skip it. + */ + if (i == 0 && h->drv[i] && h->drv[i]->raid_level != -1) + continue; + + /* + * We've found an empty slot. Update highest_lun + * provided this isn't just the fake cxd0 controller node. + */ + if (i > h->highest_lun && !controller_node) + h->highest_lun = i; + + /* If adding a real disk at cxd0, and it's already alloc'ed */ + if (i == 0 && h->drv[i] != NULL) return i; - } + + /* + * Found an empty slot, not already alloc'ed. Allocate it. + * Mark it with raid_level == -1, so we know it's new later on. + */ + drv = kzalloc(sizeof(*drv), GFP_KERNEL); + if (!drv) + return -1; + drv->raid_level = -1; /* so we know it's new */ + h->drv[i] = drv; + return i; } return -1; } +static void cciss_free_drive_info(ctlr_info_t *h, int drv_index) +{ + kfree(h->drv[drv_index]); + h->drv[drv_index] = NULL; +} + +static void cciss_free_gendisk(ctlr_info_t *h, int drv_index) +{ + put_disk(h->gendisk[drv_index]); + h->gendisk[drv_index] = NULL; +} + /* cciss_add_gendisk finds a free hba[]->drv structure * and allocates a gendisk if needed, and sets the lunid * in the drvinfo structure. It returns the index into @@ -1824,13 +2042,15 @@ static int cciss_find_free_drive_index(int ctlr, int controller_node) * a means to talk to the controller in case no logical * drives have yet been configured. */ -static int cciss_add_gendisk(ctlr_info_t *h, __u32 lunid, int controller_node) +static int cciss_add_gendisk(ctlr_info_t *h, unsigned char lunid[], + int controller_node) { int drv_index; - drv_index = cciss_find_free_drive_index(h->ctlr, controller_node); + drv_index = cciss_alloc_drive_info(h, controller_node); if (drv_index == -1) return -1; + /*Check if the gendisk needs to be allocated */ if (!h->gendisk[drv_index]) { h->gendisk[drv_index] = @@ -1839,23 +2059,24 @@ static int cciss_add_gendisk(ctlr_info_t *h, __u32 lunid, int controller_node) printk(KERN_ERR "cciss%d: could not " "allocate a new disk %d\n", h->ctlr, drv_index); - return -1; + goto err_free_drive_info; } } - h->drv[drv_index].LunID = lunid; - if (cciss_create_ld_sysfs_entry(h, &h->drv[drv_index], drv_index)) + memcpy(h->drv[drv_index]->LunID, lunid, + sizeof(h->drv[drv_index]->LunID)); + if (cciss_create_ld_sysfs_entry(h, drv_index)) goto err_free_disk; - /* Don't need to mark this busy because nobody */ /* else knows about this disk yet to contend */ /* for access to it. */ - h->drv[drv_index].busy_configuring = 0; + h->drv[drv_index]->busy_configuring = 0; wmb(); return drv_index; err_free_disk: - put_disk(h->gendisk[drv_index]); - h->gendisk[drv_index] = NULL; + cciss_free_gendisk(h, drv_index); +err_free_drive_info: + cciss_free_drive_info(h, drv_index); return -1; } @@ -1872,21 +2093,25 @@ static void cciss_add_controller_node(ctlr_info_t *h) if (h->gendisk[0] != NULL) /* already did this? Then bail. */ return; - drv_index = cciss_add_gendisk(h, 0, 1); - if (drv_index == -1) { - printk(KERN_WARNING "cciss%d: could not " - "add disk 0.\n", h->ctlr); - return; - } - h->drv[drv_index].block_size = 512; - h->drv[drv_index].nr_blocks = 0; - h->drv[drv_index].heads = 0; - h->drv[drv_index].sectors = 0; - h->drv[drv_index].cylinders = 0; - h->drv[drv_index].raid_level = -1; - memset(h->drv[drv_index].serial_no, 0, 16); + drv_index = cciss_add_gendisk(h, CTLR_LUNID, 1); + if (drv_index == -1) + goto error; + h->drv[drv_index]->block_size = 512; + h->drv[drv_index]->nr_blocks = 0; + h->drv[drv_index]->heads = 0; + h->drv[drv_index]->sectors = 0; + h->drv[drv_index]->cylinders = 0; + h->drv[drv_index]->raid_level = -1; + memset(h->drv[drv_index]->serial_no, 0, 16); disk = h->gendisk[drv_index]; - cciss_add_disk(h, disk, drv_index); + if (cciss_add_disk(h, disk, drv_index) == 0) + return; + cciss_free_gendisk(h, drv_index); + cciss_free_drive_info(h, drv_index); +error: + printk(KERN_WARNING "cciss%d: could not " + "add disk 0.\n", h->ctlr); + return; } /* This function will add and remove logical drives from the Logical @@ -1897,7 +2122,8 @@ static void cciss_add_controller_node(ctlr_info_t *h) * INPUT * h = The controller to perform the operations on */ -static int rebuild_lun_table(ctlr_info_t *h, int first_time) +static int rebuild_lun_table(ctlr_info_t *h, int first_time, + int via_ioctl) { int ctlr = h->ctlr; int num_luns; @@ -1907,7 +2133,7 @@ static int rebuild_lun_table(ctlr_info_t *h, int first_time) int i; int drv_found; int drv_index = 0; - __u32 lunid = 0; + unsigned char lunid[8] = CTLR_LUNID; unsigned long flags; if (!capable(CAP_SYS_RAWIO)) @@ -1960,13 +2186,13 @@ static int rebuild_lun_table(ctlr_info_t *h, int first_time) drv_found = 0; /* skip holes in the array from already deleted drives */ - if (h->drv[i].raid_level == -1) + if (h->drv[i] == NULL) continue; for (j = 0; j < num_luns; j++) { - memcpy(&lunid, &ld_buff->LUN[j][0], 4); - lunid = le32_to_cpu(lunid); - if (h->drv[i].LunID == lunid) { + memcpy(lunid, &ld_buff->LUN[j][0], sizeof(lunid)); + if (memcmp(h->drv[i]->LunID, lunid, + sizeof(lunid)) == 0) { drv_found = 1; break; } @@ -1974,11 +2200,11 @@ static int rebuild_lun_table(ctlr_info_t *h, int first_time) if (!drv_found) { /* Deregister it from the OS, it's gone. */ spin_lock_irqsave(CCISS_LOCK(h->ctlr), flags); - h->drv[i].busy_configuring = 1; + h->drv[i]->busy_configuring = 1; spin_unlock_irqrestore(CCISS_LOCK(h->ctlr), flags); - return_code = deregister_disk(h, i, 1); - cciss_destroy_ld_sysfs_entry(&h->drv[i]); - h->drv[i].busy_configuring = 0; + return_code = deregister_disk(h, i, 1, via_ioctl); + if (h->drv[i] != NULL) + h->drv[i]->busy_configuring = 0; } } @@ -1992,17 +2218,16 @@ static int rebuild_lun_table(ctlr_info_t *h, int first_time) drv_found = 0; - memcpy(&lunid, &ld_buff->LUN[i][0], 4); - lunid = le32_to_cpu(lunid); - + memcpy(lunid, &ld_buff->LUN[i][0], sizeof(lunid)); /* Find if the LUN is already in the drive array * of the driver. If so then update its info * if not in use. If it does not exist then find * the first free index and add it. */ for (j = 0; j <= h->highest_lun; j++) { - if (h->drv[j].raid_level != -1 && - h->drv[j].LunID == lunid) { + if (h->drv[j] != NULL && + memcmp(h->drv[j]->LunID, lunid, + sizeof(h->drv[j]->LunID)) == 0) { drv_index = j; drv_found = 1; break; @@ -2015,7 +2240,8 @@ static int rebuild_lun_table(ctlr_info_t *h, int first_time) if (drv_index == -1) goto freeret; } - cciss_update_drive_info(ctlr, drv_index, first_time); + cciss_update_drive_info(ctlr, drv_index, first_time, + via_ioctl); } /* end for */ freeret: @@ -2032,6 +2258,25 @@ mem_msg: goto freeret; } +static void cciss_clear_drive_info(drive_info_struct *drive_info) +{ + /* zero out the disk size info */ + drive_info->nr_blocks = 0; + drive_info->block_size = 0; + drive_info->heads = 0; + drive_info->sectors = 0; + drive_info->cylinders = 0; + drive_info->raid_level = -1; + memset(drive_info->serial_no, 0, sizeof(drive_info->serial_no)); + memset(drive_info->model, 0, sizeof(drive_info->model)); + memset(drive_info->rev, 0, sizeof(drive_info->rev)); + memset(drive_info->vendor, 0, sizeof(drive_info->vendor)); + /* + * don't clear the LUNID though, we need to remember which + * one this one is. + */ +} + /* This function will deregister the disk and it's queue from the * kernel. It must be called with the controller lock held and the * drv structures busy_configuring flag set. It's parameters are: @@ -2046,43 +2291,48 @@ mem_msg: * the disk in preparation for re-adding it. In this case * the highest_lun should be left unchanged and the LunID * should not be cleared. + * via_ioctl + * This indicates whether we've reached this path via ioctl. + * This affects the maximum usage count allowed for c0d0 to be messed with. + * If this path is reached via ioctl(), then the max_usage_count will + * be 1, as the process calling ioctl() has got to have the device open. + * If we get here via sysfs, then the max usage count will be zero. */ static int deregister_disk(ctlr_info_t *h, int drv_index, - int clear_all) + int clear_all, int via_ioctl) { int i; struct gendisk *disk; drive_info_struct *drv; + int recalculate_highest_lun; if (!capable(CAP_SYS_RAWIO)) return -EPERM; - drv = &h->drv[drv_index]; + drv = h->drv[drv_index]; disk = h->gendisk[drv_index]; /* make sure logical volume is NOT is use */ if (clear_all || (h->gendisk[0] == disk)) { - if (drv->usage_count > 1) + if (drv->usage_count > via_ioctl) return -EBUSY; } else if (drv->usage_count > 0) return -EBUSY; + recalculate_highest_lun = (drv == h->drv[h->highest_lun]); + /* invalidate the devices and deregister the disk. If it is disk * zero do not deregister it but just zero out it's values. This * allows us to delete disk zero but keep the controller registered. */ if (h->gendisk[0] != disk) { struct request_queue *q = disk->queue; - if (disk->flags & GENHD_FL_UP) + if (disk->flags & GENHD_FL_UP) { + cciss_destroy_ld_sysfs_entry(h, drv_index, 0); del_gendisk(disk); - if (q) { - blk_cleanup_queue(q); - /* Set drv->queue to NULL so that we do not try - * to call blk_start_queue on this queue in the - * interrupt handler - */ - drv->queue = NULL; } + if (q) + blk_cleanup_queue(q); /* If clear_all is set then we are deleting the logical * drive, not just refreshing its info. For drives * other than disk 0 we will call put_disk. We do not @@ -2105,34 +2355,20 @@ static int deregister_disk(ctlr_info_t *h, int drv_index, } } else { set_capacity(disk, 0); + cciss_clear_drive_info(drv); } --h->num_luns; - /* zero out the disk size info */ - drv->nr_blocks = 0; - drv->block_size = 0; - drv->heads = 0; - drv->sectors = 0; - drv->cylinders = 0; - drv->raid_level = -1; /* This can be used as a flag variable to - * indicate that this element of the drive - * array is free. - */ - - if (clear_all) { - /* check to see if it was the last disk */ - if (drv == h->drv + h->highest_lun) { - /* if so, find the new hightest lun */ - int i, newhighest = -1; - for (i = 0; i <= h->highest_lun; i++) { - /* if the disk has size > 0, it is available */ - if (h->drv[i].heads) - newhighest = i; - } - h->highest_lun = newhighest; - } - drv->LunID = 0; + /* if it was the last disk, find the new hightest lun */ + if (clear_all && recalculate_highest_lun) { + int i, newhighest = -1; + for (i = 0; i <= h->highest_lun; i++) { + /* if the disk has size > 0, it is available */ + if (h->drv[i] && h->drv[i]->heads) + newhighest = i; + } + h->highest_lun = newhighest; } return 0; } @@ -2479,8 +2715,6 @@ static void cciss_geometry_inquiry(int ctlr, int logvol, } else { /* Get geometry failed */ printk(KERN_WARNING "cciss: reading geometry failed\n"); } - printk(KERN_INFO " heads=%d, sectors=%d, cylinders=%d\n\n", - drv->heads, drv->sectors, drv->cylinders); } static void @@ -2514,9 +2748,6 @@ cciss_read_capacity(int ctlr, int logvol, int withirq, sector_t *total_size, *total_size = 0; *block_size = BLOCK_SIZE; } - if (*total_size != 0) - printk(KERN_INFO " blocks= %llu block_size= %d\n", - (unsigned long long)*total_size+1, *block_size); kfree(buf); } @@ -2568,7 +2799,8 @@ static int cciss_revalidate(struct gendisk *disk) InquiryData_struct *inq_buff = NULL; for (logvol = 0; logvol < CISS_MAX_LUN; logvol++) { - if (h->drv[logvol].LunID == drv->LunID) { + if (memcmp(h->drv[logvol]->LunID, drv->LunID, + sizeof(drv->LunID)) == 0) { FOUND = 1; break; } @@ -3053,8 +3285,7 @@ static void do_cciss_request(struct request_queue *q) /* The first 2 bits are reserved for controller error reporting. */ c->Header.Tag.lower = (c->cmdindex << 3); c->Header.Tag.lower |= 0x04; /* flag for direct lookup. */ - c->Header.LUN.LogDev.VolId = drv->LunID; - c->Header.LUN.LogDev.Mode = 1; + memcpy(&c->Header.LUN, drv->LunID, sizeof(drv->LunID)); c->Request.CDBLen = 10; // 12 byte commands not in FW yet; c->Request.Type.Type = TYPE_CMD; // It is a command. c->Request.Type.Attribute = ATTR_SIMPLE; @@ -3232,20 +3463,121 @@ static irqreturn_t do_cciss_intr(int irq, void *dev_id) return IRQ_HANDLED; } +/** + * add_to_scan_list() - add controller to rescan queue + * @h: Pointer to the controller. + * + * Adds the controller to the rescan queue if not already on the queue. + * + * returns 1 if added to the queue, 0 if skipped (could be on the + * queue already, or the controller could be initializing or shutting + * down). + **/ +static int add_to_scan_list(struct ctlr_info *h) +{ + struct ctlr_info *test_h; + int found = 0; + int ret = 0; + + if (h->busy_initializing) + return 0; + + if (!mutex_trylock(&h->busy_shutting_down)) + return 0; + + mutex_lock(&scan_mutex); + list_for_each_entry(test_h, &scan_q, scan_list) { + if (test_h == h) { + found = 1; + break; + } + } + if (!found && !h->busy_scanning) { + INIT_COMPLETION(h->scan_wait); + list_add_tail(&h->scan_list, &scan_q); + ret = 1; + } + mutex_unlock(&scan_mutex); + mutex_unlock(&h->busy_shutting_down); + + return ret; +} + +/** + * remove_from_scan_list() - remove controller from rescan queue + * @h: Pointer to the controller. + * + * Removes the controller from the rescan queue if present. Blocks if + * the controller is currently conducting a rescan. + **/ +static void remove_from_scan_list(struct ctlr_info *h) +{ + struct ctlr_info *test_h, *tmp_h; + int scanning = 0; + + mutex_lock(&scan_mutex); + list_for_each_entry_safe(test_h, tmp_h, &scan_q, scan_list) { + if (test_h == h) { + list_del(&h->scan_list); + complete_all(&h->scan_wait); + mutex_unlock(&scan_mutex); + return; + } + } + if (&h->busy_scanning) + scanning = 0; + mutex_unlock(&scan_mutex); + + if (scanning) + wait_for_completion(&h->scan_wait); +} + +/** + * scan_thread() - kernel thread used to rescan controllers + * @data: Ignored. + * + * A kernel thread used scan for drive topology changes on + * controllers. The thread processes only one controller at a time + * using a queue. Controllers are added to the queue using + * add_to_scan_list() and removed from the queue either after done + * processing or using remove_from_scan_list(). + * + * returns 0. + **/ static int scan_thread(void *data) { - ctlr_info_t *h = data; - int rc; - DECLARE_COMPLETION_ONSTACK(wait); - h->rescan_wait = &wait; + struct ctlr_info *h; - for (;;) { - rc = wait_for_completion_interruptible(&wait); + while (1) { + set_current_state(TASK_INTERRUPTIBLE); + schedule(); if (kthread_should_stop()) break; - if (!rc) - rebuild_lun_table(h, 0); + + while (1) { + mutex_lock(&scan_mutex); + if (list_empty(&scan_q)) { + mutex_unlock(&scan_mutex); + break; + } + + h = list_entry(scan_q.next, + struct ctlr_info, + scan_list); + list_del(&h->scan_list); + h->busy_scanning = 1; + mutex_unlock(&scan_mutex); + + if (h) { + rebuild_lun_table(h, 0, 0); + complete_all(&h->scan_wait); + mutex_lock(&scan_mutex); + h->busy_scanning = 0; + mutex_unlock(&scan_mutex); + } + } } + return 0; } @@ -3268,8 +3600,8 @@ static int check_for_unit_attention(ctlr_info_t *h, CommandList_struct *c) case REPORT_LUNS_CHANGED: printk(KERN_WARNING "cciss%d: report LUN data " "changed\n", h->ctlr); - if (h->rescan_wait) - complete(h->rescan_wait); + add_to_scan_list(h); + wake_up_process(cciss_scan_thread); return 1; break; case POWER_OR_RESET: @@ -3489,7 +3821,7 @@ static int __devinit cciss_pci_init(ctlr_info_t *c, struct pci_dev *pdev) if (scratchpad == CCISS_FIRMWARE_READY) break; set_current_state(TASK_INTERRUPTIBLE); - schedule_timeout(HZ / 10); /* wait 100ms */ + schedule_timeout(msecs_to_jiffies(100)); /* wait 100ms */ } if (scratchpad != CCISS_FIRMWARE_READY) { printk(KERN_WARNING "cciss: Board not ready. Timed out.\n"); @@ -3615,7 +3947,7 @@ static int __devinit cciss_pci_init(ctlr_info_t *c, struct pci_dev *pdev) break; /* delay and try again */ set_current_state(TASK_INTERRUPTIBLE); - schedule_timeout(10); + schedule_timeout(msecs_to_jiffies(1)); } #ifdef CCISS_DEBUG @@ -3669,15 +4001,16 @@ Enomem: return -1; } -static void free_hba(int i) +static void free_hba(int n) { - ctlr_info_t *p = hba[i]; - int n; + ctlr_info_t *h = hba[n]; + int i; - hba[i] = NULL; - for (n = 0; n < CISS_MAX_LUN; n++) - put_disk(p->gendisk[n]); - kfree(p); + hba[n] = NULL; + for (i = 0; i < h->highest_lun + 1; i++) + if (h->gendisk[i] != NULL) + put_disk(h->gendisk[i]); + kfree(h); } /* Send a message CDB to the firmware. */ @@ -3918,6 +4251,7 @@ static int __devinit cciss_init_one(struct pci_dev *pdev, hba[i]->busy_initializing = 1; INIT_HLIST_HEAD(&hba[i]->cmpQ); INIT_HLIST_HEAD(&hba[i]->reqQ); + mutex_init(&hba[i]->busy_shutting_down); if (cciss_pci_init(hba[i], pdev) != 0) goto clean0; @@ -3926,6 +4260,8 @@ static int __devinit cciss_init_one(struct pci_dev *pdev, hba[i]->ctlr = i; hba[i]->pdev = pdev; + init_completion(&hba[i]->scan_wait); + if (cciss_create_hba_sysfs_entry(hba[i])) goto clean0; @@ -4001,8 +4337,7 @@ static int __devinit cciss_init_one(struct pci_dev *pdev, hba[i]->num_luns = 0; hba[i]->highest_lun = -1; for (j = 0; j < CISS_MAX_LUN; j++) { - hba[i]->drv[j].raid_level = -1; - hba[i]->drv[j].queue = NULL; + hba[i]->drv[j] = NULL; hba[i]->gendisk[j] = NULL; } @@ -4035,14 +4370,8 @@ static int __devinit cciss_init_one(struct pci_dev *pdev, hba[i]->cciss_max_sectors = 2048; + rebuild_lun_table(hba[i], 1, 0); hba[i]->busy_initializing = 0; - - rebuild_lun_table(hba[i], 1); - hba[i]->cciss_scan_thread = kthread_run(scan_thread, hba[i], - "cciss_scan%02d", i); - if (IS_ERR(hba[i]->cciss_scan_thread)) - return PTR_ERR(hba[i]->cciss_scan_thread); - return 1; clean4: @@ -4063,12 +4392,7 @@ clean1: cciss_destroy_hba_sysfs_entry(hba[i]); clean0: hba[i]->busy_initializing = 0; - /* cleanup any queues that may have been initialized */ - for (j=0; j <= hba[i]->highest_lun; j++){ - drive_info_struct *drv = &(hba[i]->drv[j]); - if (drv->queue) - blk_cleanup_queue(drv->queue); - } + /* * Deliberately omit pci_disable_device(): it does something nasty to * Smart Array controllers that pci_enable_device does not undo @@ -4125,8 +4449,9 @@ static void __devexit cciss_remove_one(struct pci_dev *pdev) return; } - kthread_stop(hba[i]->cciss_scan_thread); + mutex_lock(&hba[i]->busy_shutting_down); + remove_from_scan_list(hba[i]); remove_proc_entry(hba[i]->devname, proc_cciss); unregister_blkdev(hba[i]->major, hba[i]->devname); @@ -4136,8 +4461,10 @@ static void __devexit cciss_remove_one(struct pci_dev *pdev) if (disk) { struct request_queue *q = disk->queue; - if (disk->flags & GENHD_FL_UP) + if (disk->flags & GENHD_FL_UP) { + cciss_destroy_ld_sysfs_entry(hba[i], j, 1); del_gendisk(disk); + } if (q) blk_cleanup_queue(q); } @@ -4170,6 +4497,7 @@ static void __devexit cciss_remove_one(struct pci_dev *pdev) pci_release_regions(pdev); pci_set_drvdata(pdev, NULL); cciss_destroy_hba_sysfs_entry(hba[i]); + mutex_unlock(&hba[i]->busy_shutting_down); free_hba(i); } @@ -4202,15 +4530,25 @@ static int __init cciss_init(void) if (err) return err; + /* Start the scan thread */ + cciss_scan_thread = kthread_run(scan_thread, NULL, "cciss_scan"); + if (IS_ERR(cciss_scan_thread)) { + err = PTR_ERR(cciss_scan_thread); + goto err_bus_unregister; + } + /* Register for our PCI devices */ err = pci_register_driver(&cciss_pci_driver); if (err) - goto err_bus_register; + goto err_thread_stop; - return 0; + return err; -err_bus_register: +err_thread_stop: + kthread_stop(cciss_scan_thread); +err_bus_unregister: bus_unregister(&cciss_bus_type); + return err; } @@ -4227,6 +4565,7 @@ static void __exit cciss_cleanup(void) cciss_remove_one(hba[i]->pdev); } } + kthread_stop(cciss_scan_thread); remove_proc_entry("driver/cciss", NULL); bus_unregister(&cciss_bus_type); } diff --git a/drivers/block/cciss.h b/drivers/block/cciss.h index 06a5db25b29..31524cf42c7 100644 --- a/drivers/block/cciss.h +++ b/drivers/block/cciss.h @@ -2,6 +2,7 @@ #define CCISS_H #include <linux/genhd.h> +#include <linux/mutex.h> #include "cciss_cmd.h" @@ -29,7 +30,7 @@ struct access_method { }; typedef struct _drive_info_struct { - __u32 LunID; + unsigned char LunID[8]; int usage_count; struct request_queue *queue; sector_t nr_blocks; @@ -51,6 +52,7 @@ typedef struct _drive_info_struct char vendor[VENDOR_LEN + 1]; /* SCSI vendor string */ char model[MODEL_LEN + 1]; /* SCSI model string */ char rev[REV_LEN + 1]; /* SCSI revision string */ + char device_initialized; /* indicates whether dev is initialized */ } drive_info_struct; struct ctlr_info @@ -86,7 +88,7 @@ struct ctlr_info BYTE cciss_read_capacity; // information about each logical volume - drive_info_struct drv[CISS_MAX_LUN]; + drive_info_struct *drv[CISS_MAX_LUN]; struct access_method access; @@ -108,6 +110,8 @@ struct ctlr_info int nr_frees; int busy_configuring; int busy_initializing; + int busy_scanning; + struct mutex busy_shutting_down; /* This element holds the zero based queue number of the last * queue to be started. It is used for fairness. @@ -122,8 +126,8 @@ struct ctlr_info /* and saved for later processing */ #endif unsigned char alive; - struct completion *rescan_wait; - struct task_struct *cciss_scan_thread; + struct list_head scan_list; + struct completion scan_wait; struct device dev; }; diff --git a/drivers/block/cpqarray.c b/drivers/block/cpqarray.c index b82d438e260..6422651ec36 100644 --- a/drivers/block/cpqarray.c +++ b/drivers/block/cpqarray.c @@ -32,6 +32,7 @@ #include <linux/blkpg.h> #include <linux/timer.h> #include <linux/proc_fs.h> +#include <linux/seq_file.h> #include <linux/init.h> #include <linux/hdreg.h> #include <linux/spinlock.h> @@ -177,7 +178,6 @@ static int cpqarray_register_ctlr(int ctlr, struct pci_dev *pdev); #ifdef CONFIG_PROC_FS static void ida_procinit(int i); -static int ida_proc_get_info(char *buffer, char **start, off_t offset, int length, int *eof, void *data); #else static void ida_procinit(int i) {} #endif @@ -206,6 +206,7 @@ static const struct block_device_operations ida_fops = { #ifdef CONFIG_PROC_FS static struct proc_dir_entry *proc_array; +static const struct file_operations ida_proc_fops; /* * Get us a file in /proc/array that says something about each controller. @@ -218,19 +219,16 @@ static void __init ida_procinit(int i) if (!proc_array) return; } - create_proc_read_entry(hba[i]->devname, 0, proc_array, - ida_proc_get_info, hba[i]); + proc_create_data(hba[i]->devname, 0, proc_array, &ida_proc_fops, hba[i]); } /* * Report information about this controller. */ -static int ida_proc_get_info(char *buffer, char **start, off_t offset, int length, int *eof, void *data) +static int ida_proc_show(struct seq_file *m, void *v) { - off_t pos = 0; - off_t len = 0; - int size, i, ctlr; - ctlr_info_t *h = (ctlr_info_t*)data; + int i, ctlr; + ctlr_info_t *h = (ctlr_info_t*)m->private; drv_info_t *drv; #ifdef CPQ_PROC_PRINT_QUEUES cmdlist_t *c; @@ -238,7 +236,7 @@ static int ida_proc_get_info(char *buffer, char **start, off_t offset, int lengt #endif ctlr = h->ctlr; - size = sprintf(buffer, "%s: Compaq %s Controller\n" + seq_printf(m, "%s: Compaq %s Controller\n" " Board ID: 0x%08lx\n" " Firmware Revision: %c%c%c%c\n" " Controller Sig: 0x%08lx\n" @@ -258,55 +256,54 @@ static int ida_proc_get_info(char *buffer, char **start, off_t offset, int lengt h->log_drives, h->phys_drives, h->Qdepth, h->maxQsinceinit); - pos += size; len += size; - - size = sprintf(buffer+len, "Logical Drive Info:\n"); - pos += size; len += size; + seq_puts(m, "Logical Drive Info:\n"); for(i=0; i<h->log_drives; i++) { drv = &h->drv[i]; - size = sprintf(buffer+len, "ida/c%dd%d: blksz=%d nr_blks=%d\n", + seq_printf(m, "ida/c%dd%d: blksz=%d nr_blks=%d\n", ctlr, i, drv->blk_size, drv->nr_blks); - pos += size; len += size; } #ifdef CPQ_PROC_PRINT_QUEUES spin_lock_irqsave(IDA_LOCK(h->ctlr), flags); - size = sprintf(buffer+len, "\nCurrent Queues:\n"); - pos += size; len += size; + seq_puts(m, "\nCurrent Queues:\n"); c = h->reqQ; - size = sprintf(buffer+len, "reqQ = %p", c); pos += size; len += size; + seq_printf(m, "reqQ = %p", c); if (c) c=c->next; while(c && c != h->reqQ) { - size = sprintf(buffer+len, "->%p", c); - pos += size; len += size; + seq_printf(m, "->%p", c); c=c->next; } c = h->cmpQ; - size = sprintf(buffer+len, "\ncmpQ = %p", c); pos += size; len += size; + seq_printf(m, "\ncmpQ = %p", c); if (c) c=c->next; while(c && c != h->cmpQ) { - size = sprintf(buffer+len, "->%p", c); - pos += size; len += size; + seq_printf(m, "->%p", c); c=c->next; } - size = sprintf(buffer+len, "\n"); pos += size; len += size; + seq_putc(m, '\n'); spin_unlock_irqrestore(IDA_LOCK(h->ctlr), flags); #endif - size = sprintf(buffer+len, "nr_allocs = %d\nnr_frees = %d\n", + seq_printf(m, "nr_allocs = %d\nnr_frees = %d\n", h->nr_allocs, h->nr_frees); - pos += size; len += size; - - *eof = 1; - *start = buffer+offset; - len -= offset; - if (len>length) - len = length; - return len; + return 0; +} + +static int ida_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, ida_proc_show, PDE(inode)->data); } + +static const struct file_operations ida_proc_fops = { + .owner = THIS_MODULE, + .open = ida_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; #endif /* CONFIG_PROC_FS */ module_param_array(eisa, int, NULL, 0); diff --git a/drivers/char/dtlk.c b/drivers/char/dtlk.c index 52e06589821..045c930e632 100644 --- a/drivers/char/dtlk.c +++ b/drivers/char/dtlk.c @@ -56,6 +56,7 @@ #include <linux/errno.h> /* for -EBUSY */ #include <linux/ioport.h> /* for request_region */ #include <linux/delay.h> /* for loops_per_jiffy */ +#include <linux/sched.h> #include <linux/smp_lock.h> /* cycle_kernel_lock() */ #include <asm/io.h> /* for inb_p, outb_p, inb, outb, etc. */ #include <asm/uaccess.h> /* for get_user, etc. */ diff --git a/drivers/char/ipmi/ipmi_devintf.c b/drivers/char/ipmi/ipmi_devintf.c index 41fc11dc921..65545de3dbf 100644 --- a/drivers/char/ipmi/ipmi_devintf.c +++ b/drivers/char/ipmi/ipmi_devintf.c @@ -36,6 +36,7 @@ #include <linux/errno.h> #include <asm/system.h> #include <linux/poll.h> +#include <linux/sched.h> #include <linux/spinlock.h> #include <linux/slab.h> #include <linux/ipmi.h> diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index 09050797c76..ec5e3f8df64 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -35,6 +35,7 @@ #include <linux/errno.h> #include <asm/system.h> #include <linux/poll.h> +#include <linux/sched.h> #include <linux/spinlock.h> #include <linux/mutex.h> #include <linux/slab.h> diff --git a/drivers/firewire/core-cdev.c b/drivers/firewire/core-cdev.c index ced186d7e9a..5089331544e 100644 --- a/drivers/firewire/core-cdev.c +++ b/drivers/firewire/core-cdev.c @@ -33,6 +33,7 @@ #include <linux/mutex.h> #include <linux/poll.h> #include <linux/preempt.h> +#include <linux/sched.h> #include <linux/spinlock.h> #include <linux/time.h> #include <linux/uaccess.h> diff --git a/drivers/firmware/iscsi_ibft.c b/drivers/firmware/iscsi_ibft.c index 420a96e7f2d..051d1ebbd28 100644 --- a/drivers/firmware/iscsi_ibft.c +++ b/drivers/firmware/iscsi_ibft.c @@ -939,7 +939,7 @@ static int __init ibft_init(void) if (ibft_addr) { printk(KERN_INFO "iBFT detected at 0x%llx.\n", - (u64)virt_to_phys((void *)ibft_addr)); + (u64)isa_virt_to_bus(ibft_addr)); rc = ibft_check_device(); if (rc) diff --git a/drivers/firmware/iscsi_ibft_find.c b/drivers/firmware/iscsi_ibft_find.c index d53fbbfefa3..dfb15c06c88 100644 --- a/drivers/firmware/iscsi_ibft_find.c +++ b/drivers/firmware/iscsi_ibft_find.c @@ -65,10 +65,10 @@ void __init reserve_ibft_region(void) * so skip that area */ if (pos == VGA_MEM) pos += VGA_SIZE; - virt = phys_to_virt(pos); + virt = isa_bus_to_virt(pos); if (memcmp(virt, IBFT_SIGN, IBFT_SIGN_LEN) == 0) { unsigned long *addr = - (unsigned long *)phys_to_virt(pos + 4); + (unsigned long *)isa_bus_to_virt(pos + 4); len = *addr; /* if the length of the table extends past 1M, * the table cannot be valid. */ diff --git a/drivers/hid/hidraw.c b/drivers/hid/hidraw.c index 0c6639ea03d..ba05275e510 100644 --- a/drivers/hid/hidraw.c +++ b/drivers/hid/hidraw.c @@ -30,6 +30,7 @@ #include <linux/major.h> #include <linux/hid.h> #include <linux/mutex.h> +#include <linux/sched.h> #include <linux/smp_lock.h> #include <linux/hidraw.h> diff --git a/drivers/hwmon/ltc4215.c b/drivers/hwmon/ltc4215.c index 6c9a04136e0..00d975eb5b8 100644 --- a/drivers/hwmon/ltc4215.c +++ b/drivers/hwmon/ltc4215.c @@ -20,11 +20,6 @@ #include <linux/hwmon.h> #include <linux/hwmon-sysfs.h> -static const unsigned short normal_i2c[] = { I2C_CLIENT_END }; - -/* Insmod parameters */ -I2C_CLIENT_INSMOD_1(ltc4215); - /* Here are names of the chip's registers (a.k.a. commands) */ enum ltc4215_cmd { LTC4215_CONTROL = 0x00, /* rw */ @@ -246,9 +241,13 @@ static const struct attribute_group ltc4215_group = { static int ltc4215_probe(struct i2c_client *client, const struct i2c_device_id *id) { + struct i2c_adapter *adapter = client->adapter; struct ltc4215_data *data; int ret; + if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA)) + return -ENODEV; + data = kzalloc(sizeof(*data), GFP_KERNEL); if (!data) { ret = -ENOMEM; @@ -294,56 +293,20 @@ static int ltc4215_remove(struct i2c_client *client) return 0; } -static int ltc4215_detect(struct i2c_client *client, - int kind, - struct i2c_board_info *info) -{ - struct i2c_adapter *adapter = client->adapter; - - if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA)) - return -ENODEV; - - if (kind < 0) { /* probed detection - check the chip type */ - s32 v; /* 8 bits from the chip, or -ERRNO */ - - /* - * Register 0x01 bit b7 is reserved, expect 0 - * Register 0x03 bit b6 and b7 are reserved, expect 0 - */ - v = i2c_smbus_read_byte_data(client, LTC4215_ALERT); - if (v < 0 || (v & (1 << 7)) != 0) - return -ENODEV; - - v = i2c_smbus_read_byte_data(client, LTC4215_FAULT); - if (v < 0 || (v & ((1 << 6) | (1 << 7))) != 0) - return -ENODEV; - } - - strlcpy(info->type, "ltc4215", I2C_NAME_SIZE); - dev_info(&adapter->dev, "ltc4215 %s at address 0x%02x\n", - kind < 0 ? "probed" : "forced", - client->addr); - - return 0; -} - static const struct i2c_device_id ltc4215_id[] = { - { "ltc4215", ltc4215 }, + { "ltc4215", 0 }, { } }; MODULE_DEVICE_TABLE(i2c, ltc4215_id); /* This is the driver that will be inserted */ static struct i2c_driver ltc4215_driver = { - .class = I2C_CLASS_HWMON, .driver = { .name = "ltc4215", }, .probe = ltc4215_probe, .remove = ltc4215_remove, .id_table = ltc4215_id, - .detect = ltc4215_detect, - .address_data = &addr_data, }; static int __init ltc4215_init(void) diff --git a/drivers/hwmon/ltc4245.c b/drivers/hwmon/ltc4245.c index e3896433361..65c232a9d0c 100644 --- a/drivers/hwmon/ltc4245.c +++ b/drivers/hwmon/ltc4245.c @@ -22,15 +22,6 @@ #include <linux/hwmon.h> #include <linux/hwmon-sysfs.h> -/* Valid addresses are 0x20 - 0x3f - * - * For now, we do not probe, since some of these addresses - * are known to be unfriendly to probing */ -static const unsigned short normal_i2c[] = { I2C_CLIENT_END }; - -/* Insmod parameters */ -I2C_CLIENT_INSMOD_1(ltc4245); - /* Here are names of the chip's registers (a.k.a. commands) */ enum ltc4245_cmd { LTC4245_STATUS = 0x00, /* readonly */ @@ -369,9 +360,13 @@ static const struct attribute_group ltc4245_group = { static int ltc4245_probe(struct i2c_client *client, const struct i2c_device_id *id) { + struct i2c_adapter *adapter = client->adapter; struct ltc4245_data *data; int ret; + if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA)) + return -ENODEV; + data = kzalloc(sizeof(*data), GFP_KERNEL); if (!data) { ret = -ENOMEM; @@ -418,136 +413,20 @@ static int ltc4245_remove(struct i2c_client *client) return 0; } -/* Check that some bits in a control register appear at all possible - * locations without changing value - * - * @client: the i2c client to use - * @reg: the register to read - * @bits: the bits to check (0xff checks all bits, - * 0x03 checks only the last two bits) - * - * return -ERRNO if the register read failed - * return -ENODEV if the register value doesn't stay constant at all - * possible addresses - * - * return 0 for success - */ -static int ltc4245_check_control_reg(struct i2c_client *client, u8 reg, u8 bits) -{ - int i; - s32 v, voff1, voff2; - - /* Read register and check for error */ - v = i2c_smbus_read_byte_data(client, reg); - if (v < 0) - return v; - - v &= bits; - - for (i = 0x00; i < 0xff; i += 0x20) { - - voff1 = i2c_smbus_read_byte_data(client, reg + i); - if (voff1 < 0) - return voff1; - - voff2 = i2c_smbus_read_byte_data(client, reg + i + 0x08); - if (voff2 < 0) - return voff2; - - voff1 &= bits; - voff2 &= bits; - - if (v != voff1 || v != voff2) - return -ENODEV; - } - - return 0; -} - -static int ltc4245_detect(struct i2c_client *client, - int kind, - struct i2c_board_info *info) -{ - struct i2c_adapter *adapter = client->adapter; - - if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE_DATA)) - return -ENODEV; - - if (kind < 0) { /* probed detection - check the chip type */ - s32 v; /* 8 bits from the chip, or -ERRNO */ - - /* Chip registers 0x00-0x07 are control registers - * Chip registers 0x10-0x1f are data registers - * - * Address bits b7-b5 are ignored. This makes the chip "repeat" - * in steps of 0x20. Any control registers should appear with - * the same values across all duplicated addresses. - * - * Register 0x02 bit b2 is reserved, expect 0 - * Register 0x07 bits b7 to b4 are reserved, expect 0 - * - * Registers 0x01, 0x02 are control registers and should not - * change on their own. - * - * Register 0x06 bits b6 and b7 are control bits, and should - * not change on their own. - * - * Register 0x07 bits b3 to b0 are control bits, and should - * not change on their own. - */ - - /* read register 0x02 reserved bit, expect 0 */ - v = i2c_smbus_read_byte_data(client, LTC4245_CONTROL); - if (v < 0 || (v & 0x04) != 0) - return -ENODEV; - - /* read register 0x07 reserved bits, expect 0 */ - v = i2c_smbus_read_byte_data(client, LTC4245_ADCADR); - if (v < 0 || (v & 0xf0) != 0) - return -ENODEV; - - /* check that the alert register appears at all locations */ - if (ltc4245_check_control_reg(client, LTC4245_ALERT, 0xff)) - return -ENODEV; - - /* check that the control register appears at all locations */ - if (ltc4245_check_control_reg(client, LTC4245_CONTROL, 0xff)) - return -ENODEV; - - /* check that register 0x06 bits b6 and b7 stay constant */ - if (ltc4245_check_control_reg(client, LTC4245_GPIO, 0xc0)) - return -ENODEV; - - /* check that register 0x07 bits b3-b0 stay constant */ - if (ltc4245_check_control_reg(client, LTC4245_ADCADR, 0x0f)) - return -ENODEV; - } - - strlcpy(info->type, "ltc4245", I2C_NAME_SIZE); - dev_info(&adapter->dev, "ltc4245 %s at address 0x%02x\n", - kind < 0 ? "probed" : "forced", - client->addr); - - return 0; -} - static const struct i2c_device_id ltc4245_id[] = { - { "ltc4245", ltc4245 }, + { "ltc4245", 0 }, { } }; MODULE_DEVICE_TABLE(i2c, ltc4245_id); /* This is the driver that will be inserted */ static struct i2c_driver ltc4245_driver = { - .class = I2C_CLASS_HWMON, .driver = { .name = "ltc4245", }, .probe = ltc4245_probe, .remove = ltc4245_remove, .id_table = ltc4245_id, - .detect = ltc4245_detect, - .address_data = &addr_data, }; static int __init ltc4245_init(void) diff --git a/drivers/i2c/busses/i2c-amd756.c b/drivers/i2c/busses/i2c-amd756.c index f7d6fe9c49b..8f0b90ef8c7 100644 --- a/drivers/i2c/busses/i2c-amd756.c +++ b/drivers/i2c/busses/i2c-amd756.c @@ -364,7 +364,7 @@ static int __devinit amd756_probe(struct pci_dev *pdev, error = acpi_check_region(amd756_ioport, SMB_IOSIZE, amd756_driver.name); if (error) - return error; + return -ENODEV; if (!request_region(amd756_ioport, SMB_IOSIZE, amd756_driver.name)) { dev_err(&pdev->dev, "SMB region 0x%x already in use!\n", diff --git a/drivers/i2c/busses/i2c-amd8111.c b/drivers/i2c/busses/i2c-amd8111.c index a7c59908c45..5b4ad86ca16 100644 --- a/drivers/i2c/busses/i2c-amd8111.c +++ b/drivers/i2c/busses/i2c-amd8111.c @@ -376,8 +376,10 @@ static int __devinit amd8111_probe(struct pci_dev *dev, smbus->size = pci_resource_len(dev, 0); error = acpi_check_resource_conflict(&dev->resource[0]); - if (error) + if (error) { + error = -ENODEV; goto out_kfree; + } if (!request_region(smbus->base, smbus->size, amd8111_driver.name)) { error = -EBUSY; diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c index 9d2c5adf5d4..55edcfe5b85 100644 --- a/drivers/i2c/busses/i2c-i801.c +++ b/drivers/i2c/busses/i2c-i801.c @@ -732,8 +732,10 @@ static int __devinit i801_probe(struct pci_dev *dev, const struct pci_device_id } err = acpi_check_resource_conflict(&dev->resource[SMBBAR]); - if (err) + if (err) { + err = -ENODEV; goto exit; + } err = pci_request_region(dev, SMBBAR, i801_driver.name); if (err) { diff --git a/drivers/i2c/busses/i2c-isch.c b/drivers/i2c/busses/i2c-isch.c index 9f6b8e0f863..dba6eb053e2 100644 --- a/drivers/i2c/busses/i2c-isch.c +++ b/drivers/i2c/busses/i2c-isch.c @@ -281,7 +281,7 @@ static int __devinit sch_probe(struct pci_dev *dev, return -ENODEV; } if (acpi_check_region(sch_smba, SMBIOSIZE, sch_driver.name)) - return -EBUSY; + return -ENODEV; if (!request_region(sch_smba, SMBIOSIZE, sch_driver.name)) { dev_err(&dev->dev, "SMBus region 0x%x already in use!\n", sch_smba); diff --git a/drivers/i2c/busses/i2c-piix4.c b/drivers/i2c/busses/i2c-piix4.c index a782c7a08f9..d26a972aaca 100644 --- a/drivers/i2c/busses/i2c-piix4.c +++ b/drivers/i2c/busses/i2c-piix4.c @@ -169,7 +169,7 @@ static int __devinit piix4_setup(struct pci_dev *PIIX4_dev, } if (acpi_check_region(piix4_smba, SMBIOSIZE, piix4_driver.name)) - return -EBUSY; + return -ENODEV; if (!request_region(piix4_smba, SMBIOSIZE, piix4_driver.name)) { dev_err(&PIIX4_dev->dev, "SMBus region 0x%x already in use!\n", @@ -260,7 +260,7 @@ static int __devinit piix4_setup_sb800(struct pci_dev *PIIX4_dev, piix4_smba = ((smba_en_hi << 8) | smba_en_lo) & 0xffe0; if (acpi_check_region(piix4_smba, SMBIOSIZE, piix4_driver.name)) - return -EBUSY; + return -ENODEV; if (!request_region(piix4_smba, SMBIOSIZE, piix4_driver.name)) { dev_err(&PIIX4_dev->dev, "SMBus region 0x%x already in use!\n", diff --git a/drivers/i2c/busses/i2c-sis96x.c b/drivers/i2c/busses/i2c-sis96x.c index 8295885b2fd..1649963b00d 100644 --- a/drivers/i2c/busses/i2c-sis96x.c +++ b/drivers/i2c/busses/i2c-sis96x.c @@ -280,7 +280,7 @@ static int __devinit sis96x_probe(struct pci_dev *dev, retval = acpi_check_resource_conflict(&dev->resource[SIS96x_BAR]); if (retval) - return retval; + return -ENODEV; /* Everything is happy, let's grab the memory and set things up. */ if (!request_region(sis96x_smbus_base, SMB_IOSIZE, diff --git a/drivers/i2c/busses/i2c-viapro.c b/drivers/i2c/busses/i2c-viapro.c index 54d810a4d00..e4b1543015a 100644 --- a/drivers/i2c/busses/i2c-viapro.c +++ b/drivers/i2c/busses/i2c-viapro.c @@ -365,7 +365,7 @@ static int __devinit vt596_probe(struct pci_dev *pdev, found: error = acpi_check_region(vt596_smba, 8, vt596_driver.name); if (error) - return error; + return -ENODEV; if (!request_region(vt596_smba, 8, vt596_driver.name)) { dev_err(&pdev->dev, "SMBus region 0x%x already in use!\n", diff --git a/drivers/infiniband/core/ucm.c b/drivers/infiniband/core/ucm.c index 51bd9669cb1..f504c9b00c1 100644 --- a/drivers/infiniband/core/ucm.c +++ b/drivers/infiniband/core/ucm.c @@ -38,6 +38,7 @@ #include <linux/device.h> #include <linux/err.h> #include <linux/poll.h> +#include <linux/sched.h> #include <linux/file.h> #include <linux/mount.h> #include <linux/cdev.h> diff --git a/drivers/infiniband/core/user_mad.c b/drivers/infiniband/core/user_mad.c index 8c46f225709..7de02969ed7 100644 --- a/drivers/infiniband/core/user_mad.c +++ b/drivers/infiniband/core/user_mad.c @@ -44,6 +44,7 @@ #include <linux/mutex.h> #include <linux/kref.h> #include <linux/compat.h> +#include <linux/sched.h> #include <linux/semaphore.h> #include <asm/uaccess.h> diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c index d3fff9e008a..aec0fbdfe7f 100644 --- a/drivers/infiniband/core/uverbs_main.c +++ b/drivers/infiniband/core/uverbs_main.c @@ -40,6 +40,7 @@ #include <linux/err.h> #include <linux/fs.h> #include <linux/poll.h> +#include <linux/sched.h> #include <linux/file.h> #include <linux/mount.h> #include <linux/cdev.h> diff --git a/drivers/input/evdev.c b/drivers/input/evdev.c index 1148140d08a..dee6706038a 100644 --- a/drivers/input/evdev.c +++ b/drivers/input/evdev.c @@ -13,6 +13,7 @@ #define EVDEV_BUFFER_SIZE 64 #include <linux/poll.h> +#include <linux/sched.h> #include <linux/slab.h> #include <linux/module.h> #include <linux/init.h> diff --git a/drivers/input/input.c b/drivers/input/input.c index 16ec33f27c5..c6f88ebb40c 100644 --- a/drivers/input/input.c +++ b/drivers/input/input.c @@ -17,6 +17,7 @@ #include <linux/random.h> #include <linux/major.h> #include <linux/proc_fs.h> +#include <linux/sched.h> #include <linux/seq_file.h> #include <linux/poll.h> #include <linux/device.h> diff --git a/drivers/input/joydev.c b/drivers/input/joydev.c index 901b2525993..b1bd6dd3228 100644 --- a/drivers/input/joydev.c +++ b/drivers/input/joydev.c @@ -18,6 +18,7 @@ #include <linux/input.h> #include <linux/kernel.h> #include <linux/major.h> +#include <linux/sched.h> #include <linux/slab.h> #include <linux/mm.h> #include <linux/miscdevice.h> diff --git a/drivers/input/misc/uinput.c b/drivers/input/misc/uinput.c index c5a49aba418..d3f57245420 100644 --- a/drivers/input/misc/uinput.c +++ b/drivers/input/misc/uinput.c @@ -30,6 +30,7 @@ * - first public version */ #include <linux/poll.h> +#include <linux/sched.h> #include <linux/slab.h> #include <linux/module.h> #include <linux/init.h> diff --git a/drivers/input/mousedev.c b/drivers/input/mousedev.c index 966b8868f79..a13d80f7da1 100644 --- a/drivers/input/mousedev.c +++ b/drivers/input/mousedev.c @@ -13,6 +13,7 @@ #define MOUSEDEV_MINORS 32 #define MOUSEDEV_MIX 31 +#include <linux/sched.h> #include <linux/slab.h> #include <linux/smp_lock.h> #include <linux/poll.h> diff --git a/drivers/isdn/divert/divert_procfs.c b/drivers/isdn/divert/divert_procfs.c index 8b256a617c8..3697c409bec 100644 --- a/drivers/isdn/divert/divert_procfs.c +++ b/drivers/isdn/divert/divert_procfs.c @@ -16,6 +16,7 @@ #else #include <linux/fs.h> #endif +#include <linux/sched.h> #include <linux/isdnif.h> #include <net/net_namespace.h> #include "isdn_divert.h" diff --git a/drivers/leds/leds-pca9532.c b/drivers/leds/leds-pca9532.c index 708a8017c21..adc561eb59d 100644 --- a/drivers/leds/leds-pca9532.c +++ b/drivers/leds/leds-pca9532.c @@ -19,9 +19,6 @@ #include <linux/workqueue.h> #include <linux/leds-pca9532.h> -static const unsigned short normal_i2c[] = { /*0x60,*/ I2C_CLIENT_END}; -I2C_CLIENT_INSMOD_1(pca9532); - #define PCA9532_REG_PSC(i) (0x2+(i)*2) #define PCA9532_REG_PWM(i) (0x3+(i)*2) #define PCA9532_REG_LS0 0x6 diff --git a/drivers/macintosh/therm_adt746x.c b/drivers/macintosh/therm_adt746x.c index fde377c60cc..556f0feaa4d 100644 --- a/drivers/macintosh/therm_adt746x.c +++ b/drivers/macintosh/therm_adt746x.c @@ -124,6 +124,8 @@ read_reg(struct thermostat* th, int reg) return data; } +static struct i2c_driver thermostat_driver; + static int attach_thermostat(struct i2c_adapter *adapter) { @@ -148,7 +150,7 @@ attach_thermostat(struct i2c_adapter *adapter) * Let i2c-core delete that device on driver removal. * This is safe because i2c-core holds the core_lock mutex for us. */ - list_add_tail(&client->detected, &client->driver->clients); + list_add_tail(&client->detected, &thermostat_driver.clients); return 0; } diff --git a/drivers/macintosh/therm_pm72.c b/drivers/macintosh/therm_pm72.c index a028598af2d..ea32c7e5a9a 100644 --- a/drivers/macintosh/therm_pm72.c +++ b/drivers/macintosh/therm_pm72.c @@ -286,6 +286,8 @@ struct fcu_fan_table fcu_fans[] = { }, }; +static struct i2c_driver therm_pm72_driver; + /* * Utility function to create an i2c_client structure and * attach it to one of u3 adapters @@ -318,7 +320,7 @@ static struct i2c_client *attach_i2c_chip(int id, const char *name) * Let i2c-core delete that device on driver removal. * This is safe because i2c-core holds the core_lock mutex for us. */ - list_add_tail(&clt->detected, &clt->driver->clients); + list_add_tail(&clt->detected, &therm_pm72_driver.clients); return clt; } diff --git a/drivers/macintosh/windfarm_lm75_sensor.c b/drivers/macintosh/windfarm_lm75_sensor.c index 529886c7a82..ed6426a1077 100644 --- a/drivers/macintosh/windfarm_lm75_sensor.c +++ b/drivers/macintosh/windfarm_lm75_sensor.c @@ -115,6 +115,8 @@ static int wf_lm75_probe(struct i2c_client *client, return rc; } +static struct i2c_driver wf_lm75_driver; + static struct i2c_client *wf_lm75_create(struct i2c_adapter *adapter, u8 addr, int ds1775, const char *loc) @@ -157,7 +159,7 @@ static struct i2c_client *wf_lm75_create(struct i2c_adapter *adapter, * Let i2c-core delete that device on driver removal. * This is safe because i2c-core holds the core_lock mutex for us. */ - list_add_tail(&client->detected, &client->driver->clients); + list_add_tail(&client->detected, &wf_lm75_driver.clients); return client; fail: return NULL; diff --git a/drivers/macintosh/windfarm_max6690_sensor.c b/drivers/macintosh/windfarm_max6690_sensor.c index e2a55ecda2b..a67b349319e 100644 --- a/drivers/macintosh/windfarm_max6690_sensor.c +++ b/drivers/macintosh/windfarm_max6690_sensor.c @@ -88,6 +88,8 @@ static int wf_max6690_probe(struct i2c_client *client, return rc; } +static struct i2c_driver wf_max6690_driver; + static struct i2c_client *wf_max6690_create(struct i2c_adapter *adapter, u8 addr, const char *loc) { @@ -119,7 +121,7 @@ static struct i2c_client *wf_max6690_create(struct i2c_adapter *adapter, * Let i2c-core delete that device on driver removal. * This is safe because i2c-core holds the core_lock mutex for us. */ - list_add_tail(&client->detected, &client->driver->clients); + list_add_tail(&client->detected, &wf_max6690_driver.clients); return client; fail: diff --git a/drivers/macintosh/windfarm_smu_sat.c b/drivers/macintosh/windfarm_smu_sat.c index 5da729e58f9..e20330a2895 100644 --- a/drivers/macintosh/windfarm_smu_sat.c +++ b/drivers/macintosh/windfarm_smu_sat.c @@ -194,6 +194,8 @@ static struct wf_sensor_ops wf_sat_ops = { .owner = THIS_MODULE, }; +static struct i2c_driver wf_sat_driver; + static void wf_sat_create(struct i2c_adapter *adapter, struct device_node *dev) { struct i2c_board_info info; @@ -222,7 +224,7 @@ static void wf_sat_create(struct i2c_adapter *adapter, struct device_node *dev) * Let i2c-core delete that device on driver removal. * This is safe because i2c-core holds the core_lock mutex for us. */ - list_add_tail(&client->detected, &client->driver->clients); + list_add_tail(&client->detected, &wf_sat_driver.clients); } static int wf_sat_probe(struct i2c_client *client, diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 376f1ab48a2..23e76fe0d35 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -130,7 +130,7 @@ struct mapped_device { /* * A list of ios that arrived while we were suspended. */ - atomic_t pending[2]; + atomic_t pending; wait_queue_head_t wait; struct work_struct work; struct bio_list deferred; @@ -453,14 +453,13 @@ static void start_io_acct(struct dm_io *io) { struct mapped_device *md = io->md; int cpu; - int rw = bio_data_dir(io->bio); io->start_time = jiffies; cpu = part_stat_lock(); part_round_stats(cpu, &dm_disk(md)->part0); part_stat_unlock(); - dm_disk(md)->part0.in_flight[rw] = atomic_inc_return(&md->pending[rw]); + dm_disk(md)->part0.in_flight = atomic_inc_return(&md->pending); } static void end_io_acct(struct dm_io *io) @@ -480,9 +479,8 @@ static void end_io_acct(struct dm_io *io) * After this is decremented the bio must not be touched if it is * a barrier. */ - dm_disk(md)->part0.in_flight[rw] = pending = - atomic_dec_return(&md->pending[rw]); - pending += atomic_read(&md->pending[rw^0x1]); + dm_disk(md)->part0.in_flight = pending = + atomic_dec_return(&md->pending); /* nudge anyone waiting on suspend queue */ if (!pending) @@ -1787,8 +1785,7 @@ static struct mapped_device *alloc_dev(int minor) if (!md->disk) goto bad_disk; - atomic_set(&md->pending[0], 0); - atomic_set(&md->pending[1], 0); + atomic_set(&md->pending, 0); init_waitqueue_head(&md->wait); INIT_WORK(&md->work, dm_wq_work); init_waitqueue_head(&md->eventq); @@ -2091,8 +2088,7 @@ static int dm_wait_for_completion(struct mapped_device *md, int interruptible) break; } spin_unlock_irqrestore(q->queue_lock, flags); - } else if (!atomic_read(&md->pending[0]) && - !atomic_read(&md->pending[1])) + } else if (!atomic_read(&md->pending)) break; if (interruptible == TASK_INTERRUPTIBLE && diff --git a/drivers/media/dvb/dvb-core/dmxdev.c b/drivers/media/dvb/dvb-core/dmxdev.c index 51641498359..c37790ad92d 100644 --- a/drivers/media/dvb/dvb-core/dmxdev.c +++ b/drivers/media/dvb/dvb-core/dmxdev.c @@ -20,6 +20,7 @@ * */ +#include <linux/sched.h> #include <linux/spinlock.h> #include <linux/slab.h> #include <linux/vmalloc.h> diff --git a/drivers/media/dvb/dvb-core/dvb_demux.c b/drivers/media/dvb/dvb-core/dvb_demux.c index eef6d361662..91c537bca8a 100644 --- a/drivers/media/dvb/dvb-core/dvb_demux.c +++ b/drivers/media/dvb/dvb-core/dvb_demux.c @@ -21,6 +21,7 @@ * */ +#include <linux/sched.h> #include <linux/spinlock.h> #include <linux/slab.h> #include <linux/vmalloc.h> diff --git a/drivers/media/radio/radio-cadet.c b/drivers/media/radio/radio-cadet.c index 8b1440136c4..482d0f3be5f 100644 --- a/drivers/media/radio/radio-cadet.c +++ b/drivers/media/radio/radio-cadet.c @@ -38,6 +38,7 @@ #include <linux/videodev2.h> /* V4L2 API defs */ #include <linux/param.h> #include <linux/pnp.h> +#include <linux/sched.h> #include <linux/io.h> /* outb, outb_p */ #include <media/v4l2-device.h> #include <media/v4l2-ioctl.h> diff --git a/drivers/media/video/cpia.c b/drivers/media/video/cpia.c index 43ab0adf3b6..2377313c041 100644 --- a/drivers/media/video/cpia.c +++ b/drivers/media/video/cpia.c @@ -31,6 +31,7 @@ #include <linux/init.h> #include <linux/fs.h> #include <linux/vmalloc.h> +#include <linux/sched.h> #include <linux/slab.h> #include <linux/proc_fs.h> #include <linux/ctype.h> diff --git a/drivers/mfd/ab3100-core.c b/drivers/mfd/ab3100-core.c index 5447da16a17..61348102827 100644 --- a/drivers/mfd/ab3100-core.c +++ b/drivers/mfd/ab3100-core.c @@ -57,8 +57,6 @@ * The AB3100 is usually assigned address 0x48 (7-bit) * The chip is defined in the platform i2c_board_data section. */ -static unsigned short normal_i2c[] = { 0x48, I2C_CLIENT_END }; -I2C_CLIENT_INSMOD_1(ab3100); u8 ab3100_get_chip_type(struct ab3100 *ab3100) { @@ -966,7 +964,7 @@ static int __exit ab3100_remove(struct i2c_client *client) } static const struct i2c_device_id ab3100_id[] = { - { "ab3100", ab3100 }, + { "ab3100", 0 }, { } }; MODULE_DEVICE_TABLE(i2c, ab3100_id); diff --git a/drivers/mfd/ucb1400_core.c b/drivers/mfd/ucb1400_core.c index 2afc08006e6..fa294b6d600 100644 --- a/drivers/mfd/ucb1400_core.c +++ b/drivers/mfd/ucb1400_core.c @@ -21,6 +21,7 @@ */ #include <linux/module.h> +#include <linux/sched.h> #include <linux/ucb1400.h> unsigned int ucb1400_adc_read(struct snd_ac97 *ac97, u16 adc_channel, diff --git a/drivers/misc/eeprom/max6875.c b/drivers/misc/eeprom/max6875.c index 3c0c58eed34..5a6b2bce8ad 100644 --- a/drivers/misc/eeprom/max6875.c +++ b/drivers/misc/eeprom/max6875.c @@ -33,12 +33,6 @@ #include <linux/i2c.h> #include <linux/mutex.h> -/* Do not scan - the MAX6875 access method will write to some EEPROM chips */ -static const unsigned short normal_i2c[] = { I2C_CLIENT_END }; - -/* Insmod parameters */ -I2C_CLIENT_INSMOD_1(max6875); - /* The MAX6875 can only read/write 16 bytes at a time */ #define SLICE_SIZE 16 #define SLICE_BITS 4 @@ -146,31 +140,21 @@ static struct bin_attribute user_eeprom_attr = { .read = max6875_read, }; -/* Return 0 if detection is successful, -ENODEV otherwise */ -static int max6875_detect(struct i2c_client *client, int kind, - struct i2c_board_info *info) +static int max6875_probe(struct i2c_client *client, + const struct i2c_device_id *id) { struct i2c_adapter *adapter = client->adapter; + struct max6875_data *data; + int err; if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_WRITE_BYTE_DATA | I2C_FUNC_SMBUS_READ_BYTE)) return -ENODEV; - /* Only check even addresses */ + /* Only bind to even addresses */ if (client->addr & 1) return -ENODEV; - strlcpy(info->type, "max6875", I2C_NAME_SIZE); - - return 0; -} - -static int max6875_probe(struct i2c_client *client, - const struct i2c_device_id *id) -{ - struct max6875_data *data; - int err; - if (!(data = kzalloc(sizeof(struct max6875_data), GFP_KERNEL))) return -ENOMEM; @@ -222,9 +206,6 @@ static struct i2c_driver max6875_driver = { .probe = max6875_probe, .remove = max6875_remove, .id_table = max6875_id, - - .detect = max6875_detect, - .address_data = &addr_data, }; static int __init max6875_init(void) diff --git a/drivers/mtd/mtd_blkdevs.c b/drivers/mtd/mtd_blkdevs.c index 0acbf4f5be5..8ca17a3e96e 100644 --- a/drivers/mtd/mtd_blkdevs.c +++ b/drivers/mtd/mtd_blkdevs.c @@ -32,14 +32,6 @@ struct mtd_blkcore_priv { spinlock_t queue_lock; }; -static int blktrans_discard_request(struct request_queue *q, - struct request *req) -{ - req->cmd_type = REQ_TYPE_LINUX_BLOCK; - req->cmd[0] = REQ_LB_OP_DISCARD; - return 0; -} - static int do_blktrans_request(struct mtd_blktrans_ops *tr, struct mtd_blktrans_dev *dev, struct request *req) @@ -52,10 +44,6 @@ static int do_blktrans_request(struct mtd_blktrans_ops *tr, buf = req->buffer; - if (req->cmd_type == REQ_TYPE_LINUX_BLOCK && - req->cmd[0] == REQ_LB_OP_DISCARD) - return tr->discard(dev, block, nsect); - if (!blk_fs_request(req)) return -EIO; @@ -63,6 +51,9 @@ static int do_blktrans_request(struct mtd_blktrans_ops *tr, get_capacity(req->rq_disk)) return -EIO; + if (blk_discard_rq(req)) + return tr->discard(dev, block, nsect); + switch(rq_data_dir(req)) { case READ: for (; nsect > 0; nsect--, block++, buf += tr->blksize) @@ -380,8 +371,8 @@ int register_mtd_blktrans(struct mtd_blktrans_ops *tr) tr->blkcore_priv->rq->queuedata = tr; blk_queue_logical_block_size(tr->blkcore_priv->rq, tr->blksize); if (tr->discard) - blk_queue_set_discard(tr->blkcore_priv->rq, - blktrans_discard_request); + queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, + tr->blkcore_priv->rq); tr->blkshift = ffs(tr->blksize) - 1; diff --git a/drivers/platform/x86/sony-laptop.c b/drivers/platform/x86/sony-laptop.c index afdbdaaf80c..a2a742c8ff7 100644 --- a/drivers/platform/x86/sony-laptop.c +++ b/drivers/platform/x86/sony-laptop.c @@ -1211,15 +1211,6 @@ static int sony_nc_add(struct acpi_device *device) } } - /* try to _INI the device if such method exists (ACPI spec 3.0-6.5.1 - * should be respected as we already checked for the device presence above */ - if (ACPI_SUCCESS(acpi_get_handle(sony_nc_acpi_handle, METHOD_NAME__INI, &handle))) { - dprintk("Invoking _INI\n"); - if (ACPI_FAILURE(acpi_evaluate_object(sony_nc_acpi_handle, METHOD_NAME__INI, - NULL, NULL))) - dprintk("_INI Method failed\n"); - } - if (ACPI_SUCCESS(acpi_get_handle(sony_nc_acpi_handle, "ECON", &handle))) { if (acpi_callsetfunc(sony_nc_acpi_handle, "ECON", 1, NULL)) @@ -1399,27 +1390,20 @@ struct sonypi_eventtypes { struct sonypi_event *events; }; -struct device_ctrl { +struct sony_pic_dev { + struct acpi_device *acpi_dev; + struct sony_pic_irq *cur_irq; + struct sony_pic_ioport *cur_ioport; + struct list_head interrupts; + struct list_head ioports; + struct mutex lock; + struct sonypi_eventtypes *event_types; + int (*handle_irq)(const u8, const u8); int model; - int (*handle_irq)(const u8, const u8); u16 evport_offset; - u8 has_camera; - u8 has_bluetooth; - u8 has_wwan; - struct sonypi_eventtypes *event_types; -}; - -struct sony_pic_dev { - struct device_ctrl *control; - struct acpi_device *acpi_dev; - struct sony_pic_irq *cur_irq; - struct sony_pic_ioport *cur_ioport; - struct list_head interrupts; - struct list_head ioports; - struct mutex lock; - u8 camera_power; - u8 bluetooth_power; - u8 wwan_power; + u8 camera_power; + u8 bluetooth_power; + u8 wwan_power; }; static struct sony_pic_dev spic_dev = { @@ -1427,6 +1411,8 @@ static struct sony_pic_dev spic_dev = { .ioports = LIST_HEAD_INIT(spic_dev.ioports), }; +static int spic_drv_registered; + /* Event masks */ #define SONYPI_JOGGER_MASK 0x00000001 #define SONYPI_CAPTURE_MASK 0x00000002 @@ -1724,27 +1710,6 @@ static int type3_handle_irq(const u8 data_mask, const u8 ev) return 1; } -static struct device_ctrl spic_types[] = { - { - .model = SONYPI_DEVICE_TYPE1, - .handle_irq = NULL, - .evport_offset = SONYPI_TYPE1_OFFSET, - .event_types = type1_events, - }, - { - .model = SONYPI_DEVICE_TYPE2, - .handle_irq = NULL, - .evport_offset = SONYPI_TYPE2_OFFSET, - .event_types = type2_events, - }, - { - .model = SONYPI_DEVICE_TYPE3, - .handle_irq = type3_handle_irq, - .evport_offset = SONYPI_TYPE3_OFFSET, - .event_types = type3_events, - }, -}; - static void sony_pic_detect_device_type(struct sony_pic_dev *dev) { struct pci_dev *pcidev; @@ -1752,48 +1717,63 @@ static void sony_pic_detect_device_type(struct sony_pic_dev *dev) pcidev = pci_get_device(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82371AB_3, NULL); if (pcidev) { - dev->control = &spic_types[0]; + dev->model = SONYPI_DEVICE_TYPE1; + dev->evport_offset = SONYPI_TYPE1_OFFSET; + dev->event_types = type1_events; goto out; } pcidev = pci_get_device(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_1, NULL); if (pcidev) { - dev->control = &spic_types[2]; + dev->model = SONYPI_DEVICE_TYPE2; + dev->evport_offset = SONYPI_TYPE2_OFFSET; + dev->event_types = type2_events; goto out; } pcidev = pci_get_device(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_1, NULL); if (pcidev) { - dev->control = &spic_types[2]; + dev->model = SONYPI_DEVICE_TYPE3; + dev->handle_irq = type3_handle_irq; + dev->evport_offset = SONYPI_TYPE3_OFFSET; + dev->event_types = type3_events; goto out; } pcidev = pci_get_device(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_4, NULL); if (pcidev) { - dev->control = &spic_types[2]; + dev->model = SONYPI_DEVICE_TYPE3; + dev->handle_irq = type3_handle_irq; + dev->evport_offset = SONYPI_TYPE3_OFFSET; + dev->event_types = type3_events; goto out; } pcidev = pci_get_device(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH9_1, NULL); if (pcidev) { - dev->control = &spic_types[2]; + dev->model = SONYPI_DEVICE_TYPE3; + dev->handle_irq = type3_handle_irq; + dev->evport_offset = SONYPI_TYPE3_OFFSET; + dev->event_types = type3_events; goto out; } /* default */ - dev->control = &spic_types[1]; + dev->model = SONYPI_DEVICE_TYPE2; + dev->evport_offset = SONYPI_TYPE2_OFFSET; + dev->event_types = type2_events; out: if (pcidev) pci_dev_put(pcidev); printk(KERN_INFO DRV_PFX "detected Type%d model\n", - dev->control->model == SONYPI_DEVICE_TYPE1 ? 1 : - dev->control->model == SONYPI_DEVICE_TYPE2 ? 2 : 3); + dev->model == SONYPI_DEVICE_TYPE1 ? 1 : + dev->model == SONYPI_DEVICE_TYPE2 ? 2 : 3); } /* camera tests and poweron/poweroff */ @@ -2566,7 +2546,7 @@ static int sony_pic_enable(struct acpi_device *device, buffer.pointer = resource; /* setup Type 1 resources */ - if (spic_dev.control->model == SONYPI_DEVICE_TYPE1) { + if (spic_dev.model == SONYPI_DEVICE_TYPE1) { /* setup io resources */ resource->res1.type = ACPI_RESOURCE_TYPE_IO; @@ -2649,29 +2629,28 @@ static irqreturn_t sony_pic_irq(int irq, void *dev_id) data_mask = inb_p(dev->cur_ioport->io2.minimum); else data_mask = inb_p(dev->cur_ioport->io1.minimum + - dev->control->evport_offset); + dev->evport_offset); dprintk("event ([%.2x] [%.2x]) at port 0x%.4x(+0x%.2x)\n", ev, data_mask, dev->cur_ioport->io1.minimum, - dev->control->evport_offset); + dev->evport_offset); if (ev == 0x00 || ev == 0xff) return IRQ_HANDLED; - for (i = 0; dev->control->event_types[i].mask; i++) { + for (i = 0; dev->event_types[i].mask; i++) { - if ((data_mask & dev->control->event_types[i].data) != - dev->control->event_types[i].data) + if ((data_mask & dev->event_types[i].data) != + dev->event_types[i].data) continue; - if (!(mask & dev->control->event_types[i].mask)) + if (!(mask & dev->event_types[i].mask)) continue; - for (j = 0; dev->control->event_types[i].events[j].event; j++) { - if (ev == dev->control->event_types[i].events[j].data) { + for (j = 0; dev->event_types[i].events[j].event; j++) { + if (ev == dev->event_types[i].events[j].data) { device_event = - dev->control-> - event_types[i].events[j].event; + dev->event_types[i].events[j].event; goto found; } } @@ -2679,13 +2658,12 @@ static irqreturn_t sony_pic_irq(int irq, void *dev_id) /* Still not able to decode the event try to pass * it over to the minidriver */ - if (dev->control->handle_irq && - dev->control->handle_irq(data_mask, ev) == 0) + if (dev->handle_irq && dev->handle_irq(data_mask, ev) == 0) return IRQ_HANDLED; dprintk("unknown event ([%.2x] [%.2x]) at port 0x%.4x(+0x%.2x)\n", ev, data_mask, dev->cur_ioport->io1.minimum, - dev->control->evport_offset); + dev->evport_offset); return IRQ_HANDLED; found: @@ -2816,7 +2794,7 @@ static int sony_pic_add(struct acpi_device *device) /* request IRQ */ list_for_each_entry_reverse(irq, &spic_dev.interrupts, list) { if (!request_irq(irq->irq.interrupts[0], sony_pic_irq, - IRQF_SHARED, "sony-laptop", &spic_dev)) { + IRQF_DISABLED, "sony-laptop", &spic_dev)) { dprintk("IRQ: %d - triggering: %d - " "polarity: %d - shr: %d\n", irq->irq.interrupts[0], @@ -2949,6 +2927,7 @@ static int __init sony_laptop_init(void) "Unable to register SPIC driver."); goto out; } + spic_drv_registered = 1; } result = acpi_bus_register_driver(&sony_nc_driver); @@ -2960,7 +2939,7 @@ static int __init sony_laptop_init(void) return 0; out_unregister_pic: - if (!no_spic) + if (spic_drv_registered) acpi_bus_unregister_driver(&sony_pic_driver); out: return result; @@ -2969,7 +2948,7 @@ out: static void __exit sony_laptop_exit(void) { acpi_bus_unregister_driver(&sony_nc_driver); - if (!no_spic) + if (spic_drv_registered) acpi_bus_unregister_driver(&sony_pic_driver); } diff --git a/drivers/sfi/sfi_core.c b/drivers/sfi/sfi_core.c index d3b49680047..b204a092913 100644 --- a/drivers/sfi/sfi_core.c +++ b/drivers/sfi/sfi_core.c @@ -90,7 +90,11 @@ static struct sfi_table_simple *syst_va __read_mostly; */ static u32 sfi_use_ioremap __read_mostly; -static void __iomem *sfi_map_memory(u64 phys, u32 size) +/* + * sfi_un/map_memory calls early_ioremap/iounmap which is a __init function + * and introduces section mismatch. So use __ref to make it calm. + */ +static void __iomem * __ref sfi_map_memory(u64 phys, u32 size) { if (!phys || !size) return NULL; @@ -101,7 +105,7 @@ static void __iomem *sfi_map_memory(u64 phys, u32 size) return early_ioremap(phys, size); } -static void sfi_unmap_memory(void __iomem *virt, u32 size) +static void __ref sfi_unmap_memory(void __iomem *virt, u32 size) { if (!virt || !size) return; @@ -125,7 +129,7 @@ static void sfi_print_table_header(unsigned long long pa, * sfi_verify_table() * Sanity check table lengh, calculate checksum */ -static __init int sfi_verify_table(struct sfi_table_header *table) +static int sfi_verify_table(struct sfi_table_header *table) { u8 checksum = 0; @@ -213,12 +217,17 @@ static int sfi_table_check_key(struct sfi_table_header *th, * the mapped virt address will be returned, and the virt space * will be released by call sfi_put_table() later * + * This two cases are from two different functions with two different + * sections and causes section mismatch warning. So use __ref to tell + * modpost not to make any noise. + * * Return value: * NULL: when can't find a table matching the key * ERR_PTR(error): error value * virt table address: when a matched table is found */ -struct sfi_table_header *sfi_check_table(u64 pa, struct sfi_table_key *key) +struct sfi_table_header * + __ref sfi_check_table(u64 pa, struct sfi_table_key *key) { struct sfi_table_header *th; void *ret = NULL; diff --git a/drivers/staging/dst/dcore.c b/drivers/staging/dst/dcore.c index ee1601026fb..c24e4e0367a 100644 --- a/drivers/staging/dst/dcore.c +++ b/drivers/staging/dst/dcore.c @@ -102,7 +102,7 @@ static int dst_request(struct request_queue *q, struct bio *bio) struct dst_node *n = q->queuedata; int err = -EIO; - if (bio_empty_barrier(bio) && !q->prepare_discard_fn) { + if (bio_empty_barrier(bio) && !blk_queue_discard(q)) { /* * This is a dirty^Wnice hack, but if we complete this * operation with -EOPNOTSUPP like intended, XFS diff --git a/drivers/staging/iio/light/tsl2561.c b/drivers/staging/iio/light/tsl2561.c index ea8a5efc19b..fc2107f4c04 100644 --- a/drivers/staging/iio/light/tsl2561.c +++ b/drivers/staging/iio/light/tsl2561.c @@ -239,10 +239,6 @@ static int __devexit tsl2561_remove(struct i2c_client *client) return tsl2561_powerdown(client); } -static unsigned short normal_i2c[] = { 0x29, 0x39, 0x49, I2C_CLIENT_END }; - -I2C_CLIENT_INSMOD; - static const struct i2c_device_id tsl2561_id[] = { { "tsl2561", 0 }, { } diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c index c44367fea18..bf0f6520c6d 100644 --- a/drivers/usb/gadget/inode.c +++ b/drivers/usb/gadget/inode.c @@ -30,6 +30,7 @@ #include <linux/wait.h> #include <linux/compiler.h> #include <asm/uaccess.h> +#include <linux/sched.h> #include <linux/slab.h> #include <linux/poll.h> #include <linux/smp_lock.h> diff --git a/drivers/w1/masters/ds2482.c b/drivers/w1/masters/ds2482.c index df52cb355f7..406caa6a71c 100644 --- a/drivers/w1/masters/ds2482.c +++ b/drivers/w1/masters/ds2482.c @@ -24,19 +24,6 @@ #include "../w1_int.h" /** - * Address is selected using 2 pins, resulting in 4 possible addresses. - * 0x18, 0x19, 0x1a, 0x1b - * However, the chip cannot be detected without doing an i2c write, - * so use the force module parameter. - */ -static const unsigned short normal_i2c[] = { I2C_CLIENT_END }; - -/** - * Insmod parameters - */ -I2C_CLIENT_INSMOD_1(ds2482); - -/** * The DS2482 registers - there are 3 registers that are addressed by a read * pointer. The read pointer is set by the last command executed. * @@ -96,8 +83,6 @@ static const u8 ds2482_chan_rd[8] = static int ds2482_probe(struct i2c_client *client, const struct i2c_device_id *id); -static int ds2482_detect(struct i2c_client *client, int kind, - struct i2c_board_info *info); static int ds2482_remove(struct i2c_client *client); @@ -117,8 +102,6 @@ static struct i2c_driver ds2482_driver = { .probe = ds2482_probe, .remove = ds2482_remove, .id_table = ds2482_id, - .detect = ds2482_detect, - .address_data = &addr_data, }; /* @@ -425,19 +408,6 @@ static u8 ds2482_w1_reset_bus(void *data) } -static int ds2482_detect(struct i2c_client *client, int kind, - struct i2c_board_info *info) -{ - if (!i2c_check_functionality(client->adapter, - I2C_FUNC_SMBUS_WRITE_BYTE_DATA | - I2C_FUNC_SMBUS_BYTE)) - return -ENODEV; - - strlcpy(info->type, "ds2482", I2C_NAME_SIZE); - - return 0; -} - static int ds2482_probe(struct i2c_client *client, const struct i2c_device_id *id) { @@ -446,6 +416,11 @@ static int ds2482_probe(struct i2c_client *client, int temp1; int idx; + if (!i2c_check_functionality(client->adapter, + I2C_FUNC_SMBUS_WRITE_BYTE_DATA | + I2C_FUNC_SMBUS_BYTE)) + return -ENODEV; + if (!(data = kzalloc(sizeof(struct ds2482_data), GFP_KERNEL))) { err = -ENOMEM; goto exit; diff --git a/drivers/xen/xenfs/xenbus.c b/drivers/xen/xenfs/xenbus.c index a9592d981b1..6c4269b836b 100644 --- a/drivers/xen/xenfs/xenbus.c +++ b/drivers/xen/xenfs/xenbus.c @@ -43,6 +43,7 @@ #include <linux/fs.h> #include <linux/poll.h> #include <linux/mutex.h> +#include <linux/sched.h> #include <linux/spinlock.h> #include <linux/mount.h> #include <linux/pagemap.h> diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c index d11c51fc2a3..2ca7a7cafdb 100644 --- a/fs/anon_inodes.c +++ b/fs/anon_inodes.c @@ -8,8 +8,10 @@ * */ +#include <linux/cred.h> #include <linux/file.h> #include <linux/poll.h> +#include <linux/sched.h> #include <linux/slab.h> #include <linux/init.h> #include <linux/fs.h> @@ -249,6 +249,7 @@ void bio_free(struct bio *bio, struct bio_set *bs) mempool_free(p, bs->bio_pool); } +EXPORT_SYMBOL(bio_free); void bio_init(struct bio *bio) { @@ -257,6 +258,7 @@ void bio_init(struct bio *bio) bio->bi_comp_cpu = -1; atomic_set(&bio->bi_cnt, 1); } +EXPORT_SYMBOL(bio_init); /** * bio_alloc_bioset - allocate a bio for I/O @@ -311,6 +313,7 @@ err_free: mempool_free(p, bs->bio_pool); return NULL; } +EXPORT_SYMBOL(bio_alloc_bioset); static void bio_fs_destructor(struct bio *bio) { @@ -337,6 +340,7 @@ struct bio *bio_alloc(gfp_t gfp_mask, int nr_iovecs) return bio; } +EXPORT_SYMBOL(bio_alloc); static void bio_kmalloc_destructor(struct bio *bio) { @@ -380,6 +384,7 @@ struct bio *bio_kmalloc(gfp_t gfp_mask, int nr_iovecs) return bio; } +EXPORT_SYMBOL(bio_kmalloc); void zero_fill_bio(struct bio *bio) { @@ -416,6 +421,7 @@ void bio_put(struct bio *bio) bio->bi_destructor(bio); } } +EXPORT_SYMBOL(bio_put); inline int bio_phys_segments(struct request_queue *q, struct bio *bio) { @@ -424,6 +430,7 @@ inline int bio_phys_segments(struct request_queue *q, struct bio *bio) return bio->bi_phys_segments; } +EXPORT_SYMBOL(bio_phys_segments); /** * __bio_clone - clone a bio @@ -451,6 +458,7 @@ void __bio_clone(struct bio *bio, struct bio *bio_src) bio->bi_size = bio_src->bi_size; bio->bi_idx = bio_src->bi_idx; } +EXPORT_SYMBOL(__bio_clone); /** * bio_clone - clone a bio @@ -482,6 +490,7 @@ struct bio *bio_clone(struct bio *bio, gfp_t gfp_mask) return b; } +EXPORT_SYMBOL(bio_clone); /** * bio_get_nr_vecs - return approx number of vecs @@ -505,6 +514,7 @@ int bio_get_nr_vecs(struct block_device *bdev) return nr_pages; } +EXPORT_SYMBOL(bio_get_nr_vecs); static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page *page, unsigned int len, unsigned int offset, @@ -635,6 +645,7 @@ int bio_add_pc_page(struct request_queue *q, struct bio *bio, struct page *page, return __bio_add_page(q, bio, page, len, offset, queue_max_hw_sectors(q)); } +EXPORT_SYMBOL(bio_add_pc_page); /** * bio_add_page - attempt to add page to bio @@ -655,6 +666,7 @@ int bio_add_page(struct bio *bio, struct page *page, unsigned int len, struct request_queue *q = bdev_get_queue(bio->bi_bdev); return __bio_add_page(q, bio, page, len, offset, queue_max_sectors(q)); } +EXPORT_SYMBOL(bio_add_page); struct bio_map_data { struct bio_vec *iovecs; @@ -776,6 +788,7 @@ int bio_uncopy_user(struct bio *bio) bio_put(bio); return ret; } +EXPORT_SYMBOL(bio_uncopy_user); /** * bio_copy_user_iov - copy user data to bio @@ -920,6 +933,7 @@ struct bio *bio_copy_user(struct request_queue *q, struct rq_map_data *map_data, return bio_copy_user_iov(q, map_data, &iov, 1, write_to_vm, gfp_mask); } +EXPORT_SYMBOL(bio_copy_user); static struct bio *__bio_map_user_iov(struct request_queue *q, struct block_device *bdev, @@ -1050,6 +1064,7 @@ struct bio *bio_map_user(struct request_queue *q, struct block_device *bdev, return bio_map_user_iov(q, bdev, &iov, 1, write_to_vm, gfp_mask); } +EXPORT_SYMBOL(bio_map_user); /** * bio_map_user_iov - map user sg_iovec table into bio @@ -1117,13 +1132,13 @@ void bio_unmap_user(struct bio *bio) __bio_unmap_user(bio); bio_put(bio); } +EXPORT_SYMBOL(bio_unmap_user); static void bio_map_kern_endio(struct bio *bio, int err) { bio_put(bio); } - static struct bio *__bio_map_kern(struct request_queue *q, void *data, unsigned int len, gfp_t gfp_mask) { @@ -1189,6 +1204,7 @@ struct bio *bio_map_kern(struct request_queue *q, void *data, unsigned int len, bio_put(bio); return ERR_PTR(-EINVAL); } +EXPORT_SYMBOL(bio_map_kern); static void bio_copy_kern_endio(struct bio *bio, int err) { @@ -1250,6 +1266,7 @@ struct bio *bio_copy_kern(struct request_queue *q, void *data, unsigned int len, return bio; } +EXPORT_SYMBOL(bio_copy_kern); /* * bio_set_pages_dirty() and bio_check_pages_dirty() are support functions @@ -1400,6 +1417,7 @@ void bio_endio(struct bio *bio, int error) if (bio->bi_end_io) bio->bi_end_io(bio, error); } +EXPORT_SYMBOL(bio_endio); void bio_pair_release(struct bio_pair *bp) { @@ -1410,6 +1428,7 @@ void bio_pair_release(struct bio_pair *bp) mempool_free(bp, bp->bio2.bi_private); } } +EXPORT_SYMBOL(bio_pair_release); static void bio_pair_end_1(struct bio *bi, int err) { @@ -1477,6 +1496,7 @@ struct bio_pair *bio_split(struct bio *bi, int first_sectors) return bp; } +EXPORT_SYMBOL(bio_split); /** * bio_sector_offset - Find hardware sector offset in bio @@ -1547,6 +1567,7 @@ void bioset_free(struct bio_set *bs) kfree(bs); } +EXPORT_SYMBOL(bioset_free); /** * bioset_create - Create a bio_set @@ -1592,6 +1613,7 @@ bad: bioset_free(bs); return NULL; } +EXPORT_SYMBOL(bioset_create); static void __init biovec_init_slabs(void) { @@ -1636,29 +1658,4 @@ static int __init init_bio(void) return 0; } - subsys_initcall(init_bio); - -EXPORT_SYMBOL(bio_alloc); -EXPORT_SYMBOL(bio_kmalloc); -EXPORT_SYMBOL(bio_put); -EXPORT_SYMBOL(bio_free); -EXPORT_SYMBOL(bio_endio); -EXPORT_SYMBOL(bio_init); -EXPORT_SYMBOL(__bio_clone); -EXPORT_SYMBOL(bio_clone); -EXPORT_SYMBOL(bio_phys_segments); -EXPORT_SYMBOL(bio_add_page); -EXPORT_SYMBOL(bio_add_pc_page); -EXPORT_SYMBOL(bio_get_nr_vecs); -EXPORT_SYMBOL(bio_map_user); -EXPORT_SYMBOL(bio_unmap_user); -EXPORT_SYMBOL(bio_map_kern); -EXPORT_SYMBOL(bio_copy_kern); -EXPORT_SYMBOL(bio_pair_release); -EXPORT_SYMBOL(bio_split); -EXPORT_SYMBOL(bio_copy_user); -EXPORT_SYMBOL(bio_uncopy_user); -EXPORT_SYMBOL(bioset_create); -EXPORT_SYMBOL(bioset_free); -EXPORT_SYMBOL(bio_alloc_bioset); diff --git a/fs/coda/psdev.c b/fs/coda/psdev.c index 0376ac66c44..be4392ca209 100644 --- a/fs/coda/psdev.c +++ b/fs/coda/psdev.c @@ -22,6 +22,7 @@ #include <linux/kernel.h> #include <linux/major.h> #include <linux/time.h> +#include <linux/sched.h> #include <linux/slab.h> #include <linux/ioport.h> #include <linux/fcntl.h> diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 7b685e10cba..f38fee0311a 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -248,19 +248,11 @@ ssize_t part_stat_show(struct device *dev, part_stat_read(p, merges[WRITE]), (unsigned long long)part_stat_read(p, sectors[WRITE]), jiffies_to_msecs(part_stat_read(p, ticks[WRITE])), - part_in_flight(p), + p->in_flight, jiffies_to_msecs(part_stat_read(p, io_ticks)), jiffies_to_msecs(part_stat_read(p, time_in_queue))); } -ssize_t part_inflight_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hd_struct *p = dev_to_part(dev); - - return sprintf(buf, "%8u %8u\n", p->in_flight[0], p->in_flight[1]); -} - #ifdef CONFIG_FAIL_MAKE_REQUEST ssize_t part_fail_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -289,7 +281,6 @@ static DEVICE_ATTR(start, S_IRUGO, part_start_show, NULL); static DEVICE_ATTR(size, S_IRUGO, part_size_show, NULL); static DEVICE_ATTR(alignment_offset, S_IRUGO, part_alignment_offset_show, NULL); static DEVICE_ATTR(stat, S_IRUGO, part_stat_show, NULL); -static DEVICE_ATTR(inflight, S_IRUGO, part_inflight_show, NULL); #ifdef CONFIG_FAIL_MAKE_REQUEST static struct device_attribute dev_attr_fail = __ATTR(make-it-fail, S_IRUGO|S_IWUSR, part_fail_show, part_fail_store); @@ -301,7 +292,6 @@ static struct attribute *part_attrs[] = { &dev_attr_size.attr, &dev_attr_alignment_offset.attr, &dev_attr_stat.attr, - &dev_attr_inflight.attr, #ifdef CONFIG_FAIL_MAKE_REQUEST &dev_attr_fail.attr, #endif diff --git a/fs/select.c b/fs/select.c index a201fc37022..fd38ce2e32e 100644 --- a/fs/select.c +++ b/fs/select.c @@ -15,6 +15,7 @@ */ #include <linux/kernel.h> +#include <linux/sched.h> #include <linux/syscalls.h> #include <linux/module.h> #include <linux/slab.h> diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index e23a86cae5a..25119041e03 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -82,7 +82,6 @@ enum rq_cmd_type_bits { enum { REQ_LB_OP_EJECT = 0x40, /* eject request */ REQ_LB_OP_FLUSH = 0x41, /* flush request */ - REQ_LB_OP_DISCARD = 0x42, /* discard sectors */ }; /* @@ -261,7 +260,6 @@ typedef void (request_fn_proc) (struct request_queue *q); typedef int (make_request_fn) (struct request_queue *q, struct bio *bio); typedef int (prep_rq_fn) (struct request_queue *, struct request *); typedef void (unplug_fn) (struct request_queue *); -typedef int (prepare_discard_fn) (struct request_queue *, struct request *); struct bio_vec; struct bvec_merge_data { @@ -313,6 +311,7 @@ struct queue_limits { unsigned int alignment_offset; unsigned int io_min; unsigned int io_opt; + unsigned int max_discard_sectors; unsigned short logical_block_size; unsigned short max_hw_segments; @@ -340,7 +339,6 @@ struct request_queue make_request_fn *make_request_fn; prep_rq_fn *prep_rq_fn; unplug_fn *unplug_fn; - prepare_discard_fn *prepare_discard_fn; merge_bvec_fn *merge_bvec_fn; prepare_flush_fn *prepare_flush_fn; softirq_done_fn *softirq_done_fn; @@ -460,6 +458,7 @@ struct request_queue #define QUEUE_FLAG_VIRT QUEUE_FLAG_NONROT /* paravirt device */ #define QUEUE_FLAG_IO_STAT 15 /* do IO stats */ #define QUEUE_FLAG_CQ 16 /* hardware does queuing */ +#define QUEUE_FLAG_DISCARD 17 /* supports DISCARD */ #define QUEUE_FLAG_DEFAULT ((1 << QUEUE_FLAG_IO_STAT) | \ (1 << QUEUE_FLAG_CLUSTER) | \ @@ -591,6 +590,7 @@ enum { #define blk_queue_flushing(q) ((q)->ordseq) #define blk_queue_stackable(q) \ test_bit(QUEUE_FLAG_STACKABLE, &(q)->queue_flags) +#define blk_queue_discard(q) test_bit(QUEUE_FLAG_DISCARD, &(q)->queue_flags) #define blk_fs_request(rq) ((rq)->cmd_type == REQ_TYPE_FS) #define blk_pc_request(rq) ((rq)->cmd_type == REQ_TYPE_BLOCK_PC) @@ -929,6 +929,8 @@ extern void blk_queue_max_hw_sectors(struct request_queue *, unsigned int); extern void blk_queue_max_phys_segments(struct request_queue *, unsigned short); extern void blk_queue_max_hw_segments(struct request_queue *, unsigned short); extern void blk_queue_max_segment_size(struct request_queue *, unsigned int); +extern void blk_queue_max_discard_sectors(struct request_queue *q, + unsigned int max_discard_sectors); extern void blk_queue_logical_block_size(struct request_queue *, unsigned short); extern void blk_queue_physical_block_size(struct request_queue *, unsigned short); extern void blk_queue_alignment_offset(struct request_queue *q, @@ -955,7 +957,6 @@ extern void blk_queue_merge_bvec(struct request_queue *, merge_bvec_fn *); extern void blk_queue_dma_alignment(struct request_queue *, int); extern void blk_queue_update_dma_alignment(struct request_queue *, int); extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *); -extern void blk_queue_set_discard(struct request_queue *, prepare_discard_fn *); extern void blk_queue_rq_timed_out(struct request_queue *, rq_timed_out_fn *); extern void blk_queue_rq_timeout(struct request_queue *, unsigned int); extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev); @@ -1080,25 +1081,37 @@ static inline unsigned int queue_physical_block_size(struct request_queue *q) return q->limits.physical_block_size; } +static inline int bdev_physical_block_size(struct block_device *bdev) +{ + return queue_physical_block_size(bdev_get_queue(bdev)); +} + static inline unsigned int queue_io_min(struct request_queue *q) { return q->limits.io_min; } +static inline int bdev_io_min(struct block_device *bdev) +{ + return queue_io_min(bdev_get_queue(bdev)); +} + static inline unsigned int queue_io_opt(struct request_queue *q) { return q->limits.io_opt; } +static inline int bdev_io_opt(struct block_device *bdev) +{ + return queue_io_opt(bdev_get_queue(bdev)); +} + static inline int queue_alignment_offset(struct request_queue *q) { - if (q && q->limits.misaligned) + if (q->limits.misaligned) return -1; - if (q && q->limits.alignment_offset) - return q->limits.alignment_offset; - - return 0; + return q->limits.alignment_offset; } static inline int queue_sector_alignment_offset(struct request_queue *q, @@ -1108,6 +1121,19 @@ static inline int queue_sector_alignment_offset(struct request_queue *q, & (q->limits.io_min - 1); } +static inline int bdev_alignment_offset(struct block_device *bdev) +{ + struct request_queue *q = bdev_get_queue(bdev); + + if (q->limits.misaligned) + return -1; + + if (bdev != bdev->bd_contains) + return bdev->bd_part->alignment_offset; + + return q->limits.alignment_offset; +} + static inline int queue_dma_alignment(struct request_queue *q) { return q ? q->dma_alignment : 511; @@ -1146,7 +1172,11 @@ static inline void put_dev_sector(Sector p) } struct work_struct; +struct delayed_work; int kblockd_schedule_work(struct request_queue *q, struct work_struct *work); +int kblockd_schedule_delayed_work(struct request_queue *q, + struct delayed_work *work, + unsigned long delay); #define MODULE_ALIAS_BLOCKDEV(major,minor) \ MODULE_ALIAS("block-major-" __stringify(major) "-" __stringify(minor)) diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h index 7e4350ece0f..3b73b9992b2 100644 --- a/include/linux/blktrace_api.h +++ b/include/linux/blktrace_api.h @@ -198,6 +198,7 @@ extern int blk_trace_setup(struct request_queue *q, char *name, dev_t dev, char __user *arg); extern int blk_trace_startstop(struct request_queue *q, int start); extern int blk_trace_remove(struct request_queue *q); +extern void blk_trace_remove_sysfs(struct device *dev); extern int blk_trace_init_sysfs(struct device *dev); extern struct attribute_group blk_trace_attr_group; @@ -211,6 +212,7 @@ extern struct attribute_group blk_trace_attr_group; # define blk_trace_startstop(q, start) (-ENOTTY) # define blk_trace_remove(q) (-ENOTTY) # define blk_add_trace_msg(q, fmt, ...) do { } while (0) +# define blk_trace_remove_sysfs(dev) do { } while (0) static inline int blk_trace_init_sysfs(struct device *dev) { return 0; diff --git a/include/linux/fs.h b/include/linux/fs.h index a1e6899d4b6..2620a8c6357 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -300,6 +300,10 @@ struct inodes_stat_t { #define BLKTRACESTOP _IO(0x12,117) #define BLKTRACETEARDOWN _IO(0x12,118) #define BLKDISCARD _IO(0x12,119) +#define BLKIOMIN _IO(0x12,120) +#define BLKIOOPT _IO(0x12,121) +#define BLKALIGNOFF _IO(0x12,122) +#define BLKPBSZGET _IO(0x12,123) #define BMAP_IOCTL 1 /* obsolete - kept for compatibility */ #define FIBMAP _IO(0x00,1) /* bmap access */ diff --git a/include/linux/genhd.h b/include/linux/genhd.h index 297df45ffd0..7beaa21b388 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -98,7 +98,7 @@ struct hd_struct { int make_it_fail; #endif unsigned long stamp; - int in_flight[2]; + int in_flight; #ifdef CONFIG_SMP struct disk_stats *dkstats; #else @@ -322,23 +322,18 @@ static inline void free_part_stats(struct hd_struct *part) #define part_stat_sub(cpu, gendiskp, field, subnd) \ part_stat_add(cpu, gendiskp, field, -subnd) -static inline void part_inc_in_flight(struct hd_struct *part, int rw) +static inline void part_inc_in_flight(struct hd_struct *part) { - part->in_flight[rw]++; + part->in_flight++; if (part->partno) - part_to_disk(part)->part0.in_flight[rw]++; + part_to_disk(part)->part0.in_flight++; } -static inline void part_dec_in_flight(struct hd_struct *part, int rw) +static inline void part_dec_in_flight(struct hd_struct *part) { - part->in_flight[rw]--; + part->in_flight--; if (part->partno) - part_to_disk(part)->part0.in_flight[rw]--; -} - -static inline int part_in_flight(struct hd_struct *part) -{ - return part->in_flight[0] + part->in_flight[1]; + part_to_disk(part)->part0.in_flight--; } /* block/blk-core.c */ @@ -551,8 +546,6 @@ extern ssize_t part_size_show(struct device *dev, struct device_attribute *attr, char *buf); extern ssize_t part_stat_show(struct device *dev, struct device_attribute *attr, char *buf); -extern ssize_t part_inflight_show(struct device *dev, - struct device_attribute *attr, char *buf); #ifdef CONFIG_FAIL_MAKE_REQUEST extern ssize_t part_fail_show(struct device *dev, struct device_attribute *attr, char *buf); diff --git a/include/linux/poll.h b/include/linux/poll.h index fa287f25138..6673743946f 100644 --- a/include/linux/poll.h +++ b/include/linux/poll.h @@ -6,10 +6,10 @@ #ifdef __KERNEL__ #include <linux/compiler.h> +#include <linux/ktime.h> #include <linux/wait.h> #include <linux/string.h> #include <linux/fs.h> -#include <linux/sched.h> #include <asm/uaccess.h> /* ~832 bytes of stack space used max in sys_select/sys_poll before allocating diff --git a/include/trace/events/block.h b/include/trace/events/block.h index d86af94691c..00405b5f624 100644 --- a/include/trace/events/block.h +++ b/include/trace/events/block.h @@ -488,6 +488,39 @@ TRACE_EVENT(block_remap, (unsigned long long)__entry->old_sector) ); +TRACE_EVENT(block_rq_remap, + + TP_PROTO(struct request_queue *q, struct request *rq, dev_t dev, + sector_t from), + + TP_ARGS(q, rq, dev, from), + + TP_STRUCT__entry( + __field( dev_t, dev ) + __field( sector_t, sector ) + __field( unsigned int, nr_sector ) + __field( dev_t, old_dev ) + __field( sector_t, old_sector ) + __array( char, rwbs, 6 ) + ), + + TP_fast_assign( + __entry->dev = disk_devt(rq->rq_disk); + __entry->sector = blk_rq_pos(rq); + __entry->nr_sector = blk_rq_sectors(rq); + __entry->old_dev = dev; + __entry->old_sector = from; + blk_fill_rwbs_rq(__entry->rwbs, rq); + ), + + TP_printk("%d,%d %s %llu + %u <- (%d,%d) %llu", + MAJOR(__entry->dev), MINOR(__entry->dev), __entry->rwbs, + (unsigned long long)__entry->sector, + __entry->nr_sector, + MAJOR(__entry->old_dev), MINOR(__entry->old_dev), + (unsigned long long)__entry->old_sector) +); + #endif /* _TRACE_BLOCK_H */ /* This part must be outside protection */ diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index 6d7020490f9..3e1c36e7998 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -726,8 +726,6 @@ static int hrtimer_switch_to_hres(void) /* "Retrigger" the interrupt to get things going */ retrigger_next_event(NULL); local_irq_restore(flags); - printk(KERN_DEBUG "Switched to high resolution mode on CPU %d\n", - smp_processor_id()); return 1; } diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 0f86feb6db0..e491fb08793 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -1030,14 +1030,10 @@ void __perf_event_sched_out(struct perf_event_context *ctx, update_context_time(ctx); perf_disable(); - if (ctx->nr_active) { - list_for_each_entry(event, &ctx->group_list, group_entry) { - if (event != event->group_leader) - event_sched_out(event, cpuctx, ctx); - else - group_sched_out(event, cpuctx, ctx); - } - } + if (ctx->nr_active) + list_for_each_entry(event, &ctx->group_list, group_entry) + group_sched_out(event, cpuctx, ctx); + perf_enable(); out: spin_unlock(&ctx->lock); @@ -1258,12 +1254,8 @@ __perf_event_sched_in(struct perf_event_context *ctx, if (event->cpu != -1 && event->cpu != cpu) continue; - if (event != event->group_leader) - event_sched_in(event, cpuctx, ctx, cpu); - else { - if (group_can_go_on(event, cpuctx, 1)) - group_sched_in(event, cpuctx, ctx, cpu); - } + if (group_can_go_on(event, cpuctx, 1)) + group_sched_in(event, cpuctx, ctx, cpu); /* * If this pinned group hasn't been scheduled, @@ -1291,15 +1283,9 @@ __perf_event_sched_in(struct perf_event_context *ctx, if (event->cpu != -1 && event->cpu != cpu) continue; - if (event != event->group_leader) { - if (event_sched_in(event, cpuctx, ctx, cpu)) + if (group_can_go_on(event, cpuctx, can_add_hw)) + if (group_sched_in(event, cpuctx, ctx, cpu)) can_add_hw = 0; - } else { - if (group_can_go_on(event, cpuctx, can_add_hw)) { - if (group_sched_in(event, cpuctx, ctx, cpu)) - can_add_hw = 0; - } - } } perf_enable(); out: @@ -4781,9 +4767,7 @@ int perf_event_init_task(struct task_struct *child) * We dont have to disable NMIs - we are only looking at * the list, not manipulating it: */ - list_for_each_entry_rcu(event, &parent_ctx->event_list, event_entry) { - if (event != event->group_leader) - continue; + list_for_each_entry(event, &parent_ctx->group_list, group_entry) { if (!event->attr.inherit) { inherited_all = 0; diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 3eb159c277c..d9d6206e0b1 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -856,6 +856,37 @@ static void blk_add_trace_remap(struct request_queue *q, struct bio *bio, } /** + * blk_add_trace_rq_remap - Add a trace for a request-remap operation + * @q: queue the io is for + * @rq: the source request + * @dev: target device + * @from: source sector + * + * Description: + * Device mapper remaps request to other devices. + * Add a trace for that action. + * + **/ +static void blk_add_trace_rq_remap(struct request_queue *q, + struct request *rq, dev_t dev, + sector_t from) +{ + struct blk_trace *bt = q->blk_trace; + struct blk_io_trace_remap r; + + if (likely(!bt)) + return; + + r.device_from = cpu_to_be32(dev); + r.device_to = cpu_to_be32(disk_devt(rq->rq_disk)); + r.sector_from = cpu_to_be64(from); + + __blk_add_trace(bt, blk_rq_pos(rq), blk_rq_bytes(rq), + rq_data_dir(rq), BLK_TA_REMAP, !!rq->errors, + sizeof(r), &r); +} + +/** * blk_add_driver_data - Add binary message with driver-specific data * @q: queue the io is for * @rq: io request @@ -922,10 +953,13 @@ static void blk_register_tracepoints(void) WARN_ON(ret); ret = register_trace_block_remap(blk_add_trace_remap); WARN_ON(ret); + ret = register_trace_block_rq_remap(blk_add_trace_rq_remap); + WARN_ON(ret); } static void blk_unregister_tracepoints(void) { + unregister_trace_block_rq_remap(blk_add_trace_rq_remap); unregister_trace_block_remap(blk_add_trace_remap); unregister_trace_block_split(blk_add_trace_split); unregister_trace_block_unplug_io(blk_add_trace_unplug_io); @@ -1657,6 +1691,11 @@ int blk_trace_init_sysfs(struct device *dev) return sysfs_create_group(&dev->kobj, &blk_trace_attr_group); } +void blk_trace_remove_sysfs(struct device *dev) +{ + sysfs_remove_group(&dev->kobj, &blk_trace_attr_group); +} + #endif /* CONFIG_BLK_DEV_IO_TRACE */ #ifdef CONFIG_EVENT_TRACING diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 46592feab5a..3724756e41c 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -225,7 +225,11 @@ static void ftrace_update_pid_func(void) if (ftrace_trace_function == ftrace_stub) return; +#ifdef CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST func = ftrace_trace_function; +#else + func = __ftrace_trace_function; +#endif if (ftrace_pid_trace) { set_ftrace_pid_function(func); diff --git a/kernel/trace/kmemtrace.c b/kernel/trace/kmemtrace.c index 81b1645c854..a91da69f153 100644 --- a/kernel/trace/kmemtrace.c +++ b/kernel/trace/kmemtrace.c @@ -501,7 +501,7 @@ static int __init init_kmem_tracer(void) return 1; } - if (!register_tracer(&kmem_tracer)) { + if (register_tracer(&kmem_tracer) != 0) { pr_warning("Warning: could not register the kmem tracer\n"); return 1; } diff --git a/mm/swapfile.c b/mm/swapfile.c index 4de7f02f820..a1bc6b9af9a 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -1974,12 +1974,14 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags) goto bad_swap; } - if (blk_queue_nonrot(bdev_get_queue(p->bdev))) { - p->flags |= SWP_SOLIDSTATE; - p->cluster_next = 1 + (random32() % p->highest_bit); + if (p->bdev) { + if (blk_queue_nonrot(bdev_get_queue(p->bdev))) { + p->flags |= SWP_SOLIDSTATE; + p->cluster_next = 1 + (random32() % p->highest_bit); + } + if (discard_swap(p) == 0) + p->flags |= SWP_DISCARDABLE; } - if (discard_swap(p) == 0) - p->flags |= SWP_DISCARDABLE; mutex_lock(&swapon_mutex); spin_lock(&swap_lock); diff --git a/net/rfkill/core.c b/net/rfkill/core.c index dbeaf298382..ba2efb960c6 100644 --- a/net/rfkill/core.c +++ b/net/rfkill/core.c @@ -27,6 +27,7 @@ #include <linux/list.h> #include <linux/mutex.h> #include <linux/rfkill.h> +#include <linux/sched.h> #include <linux/spinlock.h> #include <linux/miscdevice.h> #include <linux/wait.h> diff --git a/tools/perf/Documentation/perf-timechart.txt b/tools/perf/Documentation/perf-timechart.txt index 1c2ed3090cc..a7910099d6f 100644 --- a/tools/perf/Documentation/perf-timechart.txt +++ b/tools/perf/Documentation/perf-timechart.txt @@ -31,6 +31,9 @@ OPTIONS -w:: --width=:: Select the width of the SVG file (default: 1000) +-p:: +--power-only:: + Only output the CPU power section of the diagram SEE ALSO diff --git a/tools/perf/Makefile b/tools/perf/Makefile index b5f1953b614..5881943f0c3 100644 --- a/tools/perf/Makefile +++ b/tools/perf/Makefile @@ -728,7 +728,7 @@ $(BUILT_INS): perf$X common-cmds.h: util/generate-cmdlist.sh command-list.txt common-cmds.h: $(wildcard Documentation/perf-*.txt) - $(QUIET_GEN)util/generate-cmdlist.sh > $@+ && mv $@+ $@ + $(QUIET_GEN). util/generate-cmdlist.sh > $@+ && mv $@+ $@ $(patsubst %.sh,%,$(SCRIPT_SH)) : % : %.sh $(QUIET_GEN)$(RM) $@ $@+ && \ diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c index 4405681b313..702d8fe58fb 100644 --- a/tools/perf/builtin-timechart.c +++ b/tools/perf/builtin-timechart.c @@ -46,6 +46,8 @@ static u64 turbo_frequency; static u64 first_time, last_time; +static int power_only; + static struct perf_header *header; @@ -547,7 +549,7 @@ static void end_sample_processing(void) u64 cpu; struct power_event *pwr; - for (cpu = 0; cpu < numcpus; cpu++) { + for (cpu = 0; cpu <= numcpus; cpu++) { pwr = malloc(sizeof(struct power_event)); if (!pwr) return; @@ -871,7 +873,7 @@ static int determine_display_tasks(u64 threshold) /* no exit marker, task kept running to the end */ if (p->end_time == 0) p->end_time = last_time; - if (p->total_time >= threshold) + if (p->total_time >= threshold && !power_only) p->display = 1; c = p->all; @@ -882,7 +884,7 @@ static int determine_display_tasks(u64 threshold) if (c->start_time == 1) c->start_time = first_time; - if (c->total_time >= threshold) { + if (c->total_time >= threshold && !power_only) { c->display = 1; count++; } @@ -1134,6 +1136,8 @@ static const struct option options[] = { "output file name"), OPT_INTEGER('w', "width", &svg_page_width, "page width"), + OPT_BOOLEAN('p', "power-only", &power_only, + "output power data only"), OPT_END() }; diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index 1ca88896eee..37512e93623 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -782,6 +782,7 @@ static const char *skip_symbols[] = { "exit_idle", "mwait_idle", "mwait_idle_with_hints", + "poll_idle", "ppc64_runlatch_off", "pseries_dedicated_idle_sleep", NULL diff --git a/tools/perf/util/svghelper.c b/tools/perf/util/svghelper.c index a778fd0f4ae..856655d8b0b 100644 --- a/tools/perf/util/svghelper.c +++ b/tools/perf/util/svghelper.c @@ -28,7 +28,7 @@ static u64 turbo_frequency, max_freq; int svg_page_width = 1000; -#define MIN_TEXT_SIZE 0.001 +#define MIN_TEXT_SIZE 0.01 static u64 total_height; static FILE *svgfile; @@ -217,6 +217,18 @@ static char *cpu_model(void) } fclose(file); } + + /* CPU type */ + file = fopen("/sys/devices/system/cpu/cpu0/cpufreq/scaling_available_frequencies", "r"); + if (file) { + while (fgets(buf, 255, file)) { + unsigned int freq; + freq = strtoull(buf, NULL, 10); + if (freq > max_freq) + max_freq = freq; + } + fclose(file); + } return cpu_m; } |