From a460e745e8f9c75a0525ff94154a0629f9d3e05d Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 17 Oct 2006 00:10:03 -0700 Subject: [PATCH] genirq: clean up irq-flow-type naming Introduce desc->name and eliminate the handle_irq_name() hack. Add set_irq_chip_and_handler_name() to set the flow type and name at once. Signed-off-by: Ingo Molnar Acked-by: Thomas Gleixner Cc: "Eric W. Biederman" Cc: Matthew Wilcox Cc: Kyle McMartin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/i8259.c | 7 ++++--- arch/x86_64/kernel/io_apic.c | 15 ++++++++------- arch/x86_64/kernel/irq.c | 2 +- 3 files changed, 13 insertions(+), 11 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/i8259.c b/arch/x86_64/kernel/i8259.c index 0612a33bb89..c4ef801b765 100644 --- a/arch/x86_64/kernel/i8259.c +++ b/arch/x86_64/kernel/i8259.c @@ -178,7 +178,8 @@ void make_8259A_irq(unsigned int irq) { disable_irq_nosync(irq); io_apic_irqs &= ~(1<name); - seq_printf(p, "-%s", handle_irq_name(irq_desc[i].handle_irq)); + seq_printf(p, "-%-8s", irq_desc[i].name); seq_printf(p, " %s", action->name); for (action=action->next; action; action = action->next) -- cgit v1.2.3 From f248b6a34fdd726dd12b549fceba907b6df2771c Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:00 +0200 Subject: [PATCH] x86-64: Update defconfig Signed-off-by: Andi Kleen --- arch/x86_64/defconfig | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/defconfig b/arch/x86_64/defconfig index 47bfba6e9dc..0f5d44e86be 100644 --- a/arch/x86_64/defconfig +++ b/arch/x86_64/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.19-rc1 -# Thu Oct 5 13:04:43 2006 +# Linux kernel version: 2.6.19-rc2-git4 +# Sat Oct 21 03:38:52 2006 # CONFIG_X86_64=y CONFIG_64BIT=y @@ -335,8 +335,8 @@ CONFIG_IPV6=y # CONFIG_INET6_XFRM_MODE_TUNNEL is not set # CONFIG_INET6_XFRM_MODE_BEET is not set # CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set +CONFIG_IPV6_SIT=y # CONFIG_IPV6_TUNNEL is not set -# CONFIG_IPV6_SUBTREES is not set # CONFIG_IPV6_MULTIPLE_TABLES is not set # CONFIG_NETWORK_SECMARK is not set # CONFIG_NETFILTER is not set @@ -437,6 +437,13 @@ CONFIG_BLK_DEV_INITRD=y # CONFIG_CDROM_PKTCDVD is not set # CONFIG_ATA_OVER_ETH is not set +# +# Misc devices +# +# CONFIG_IBM_ASM is not set +# CONFIG_SGI_IOC4 is not set +# CONFIG_TIFM_CORE is not set + # # ATA/ATAPI/MFM/RLL support # @@ -1008,6 +1015,7 @@ CONFIG_I2C_ISA=m # # Dallas's 1-wire bus # +# CONFIG_W1 is not set # # Hardware Monitoring support @@ -1058,12 +1066,6 @@ CONFIG_SENSORS_SMSC47B397=m # CONFIG_SENSORS_HDAPS is not set # CONFIG_HWMON_DEBUG_CHIP is not set -# -# Misc devices -# -# CONFIG_IBM_ASM is not set -# CONFIG_TIFM_CORE is not set - # # Multimedia devices # @@ -1196,7 +1198,6 @@ CONFIG_USB_HIDINPUT=y # CONFIG_USB_ATI_REMOTE2 is not set # CONFIG_USB_KEYSPAN_REMOTE is not set # CONFIG_USB_APPLETOUCH is not set -# CONFIG_USB_TRANCEVIBRATOR is not set # # USB Imaging devices @@ -1242,6 +1243,7 @@ CONFIG_USB_MON=y # CONFIG_USB_APPLEDISPLAY is not set # CONFIG_USB_SISUSBVGA is not set # CONFIG_USB_LD is not set +# CONFIG_USB_TRANCEVIBRATOR is not set # CONFIG_USB_TEST is not set # @@ -1318,6 +1320,7 @@ CONFIG_EXT3_FS=y CONFIG_EXT3_FS_XATTR=y CONFIG_EXT3_FS_POSIX_ACL=y # CONFIG_EXT3_FS_SECURITY is not set +# CONFIG_EXT4DEV_FS is not set CONFIG_JBD=y # CONFIG_JBD_DEBUG is not set CONFIG_FS_MBCACHE=y @@ -1341,6 +1344,7 @@ CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set CONFIG_AUTOFS4_FS=y # CONFIG_FUSE_FS is not set +CONFIG_GENERIC_ACL=y # # CD-ROM/DVD Filesystems @@ -1418,7 +1422,6 @@ CONFIG_SUNRPC=y # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set # CONFIG_9P_FS is not set -CONFIG_GENERIC_ACL=y # # Partition Types @@ -1470,10 +1473,6 @@ CONFIG_NLS_ISO8859_15=y # CONFIG_NLS_KOI8_U is not set CONFIG_NLS_UTF8=y -# -# Distributed Lock Manager -# - # # Instrumentation Support # @@ -1512,6 +1511,7 @@ CONFIG_DEBUG_FS=y CONFIG_UNWIND_INFO=y CONFIG_STACK_UNWIND=y # CONFIG_FORCED_INLINING is not set +# CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set # CONFIG_LKDTM is not set # CONFIG_DEBUG_RODATA is not set -- cgit v1.2.3 From 926fafebc48a3218fac675f12974f9a46473bd40 Mon Sep 17 00:00:00 2001 From: keith mannthey Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] x86-64: x86_64 hot-add memory srat.c fix This patch corrects the logic used in srat.c to figure out what parsing what action to take when registering hot-add areas. Hot-add areas should only be added to the node information for the MEMORY_HOTPLUG_RESERVE case. When booting MEMORY_HOTPLUG_SPARSE hot-add areas on everything but the last node are getting include in the node data and during kernel boot the pages are setup then the kernel dies when the pages are used. This patch fixes this issue. Signed-off-by: Keith Mannthey Signed-off-by: Andi Kleen --- arch/x86_64/mm/srat.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c index 3cc0544e25f..1087e150a21 100644 --- a/arch/x86_64/mm/srat.c +++ b/arch/x86_64/mm/srat.c @@ -207,7 +207,7 @@ static inline int save_add_info(void) return hotadd_percent > 0; } #else -int update_end_of_memory(unsigned long end) {return 0;} +int update_end_of_memory(unsigned long end) {return -1;} static int hotadd_enough_memory(struct bootnode *nd) {return 1;} #ifdef CONFIG_MEMORY_HOTPLUG_SPARSE static inline int save_add_info(void) {return 1;} @@ -337,7 +337,7 @@ acpi_numa_memory_affinity_init(struct acpi_table_memory_affinity *ma) push_node_boundaries(node, nd->start >> PAGE_SHIFT, nd->end >> PAGE_SHIFT); - if (ma->flags.hot_pluggable && !reserve_hotadd(node, start, end) < 0) { + if (ma->flags.hot_pluggable && (reserve_hotadd(node, start, end) < 0)) { /* Ignore hotadd region. Undo damage */ printk(KERN_NOTICE "SRAT: Hotplug region ignored\n"); *nd = oldnode; -- cgit v1.2.3 From 45edfd1db02f818b3dc7e4743ee8585af6b78f78 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] x86-64: typo in __assign_irq_vector when updating pos for vector and offset typo with cpu instead of new_cpu Signed-off-by: Yinghai Lu Signed-off-by: Andi Kleen --- arch/x86_64/kernel/io_apic.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 49e94f7994c..b848f480851 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -651,12 +651,12 @@ next: if (vector == IA32_SYSCALL_VECTOR) goto next; for_each_cpu_mask(new_cpu, domain) - if (per_cpu(vector_irq, cpu)[vector] != -1) + if (per_cpu(vector_irq, new_cpu)[vector] != -1) goto next; /* Found one! */ for_each_cpu_mask(new_cpu, domain) { - pos[cpu].vector = vector; - pos[cpu].offset = offset; + pos[new_cpu].vector = vector; + pos[new_cpu].offset = offset; } if (old_vector >= 0) { int old_cpu; -- cgit v1.2.3 From 469b1d8741a5970ad49f2b5a837811579ba0b6f2 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] x86-64: Fix for arch/x86_64/pci/Makefile CFLAGS The arch/x86_64/pci directory was giving problems in a wierd cross-compile environment. The exact cause is unknown, but the Makefile used CFLAGS instead of EXTRA_CFLAGS. From what I can tell from Documentation/kbuild/makefiles.txt, CFLAGS should not be used for this, it should be EXTRA_CFLAGS. And it solves the cross-compile problem. Signed-off-by: Corey Minyard Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: Vojtech Pavlik Cc: Sam Ravnborg Signed-off-by: Andrew Morton --- arch/x86_64/pci/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/pci/Makefile b/arch/x86_64/pci/Makefile index 1eb18f421ed..149aba05a5b 100644 --- a/arch/x86_64/pci/Makefile +++ b/arch/x86_64/pci/Makefile @@ -3,7 +3,7 @@ # # Reuse the i386 PCI subsystem # -CFLAGS += -Iarch/i386/pci +EXTRA_CFLAGS += -Iarch/i386/pci obj-y := i386.o obj-$(CONFIG_PCI_DIRECT)+= direct.o -- cgit v1.2.3 From 73bb8919b33d42cf75a0ed89bc9ca6a7128665be Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] x86-64: fix page align in e820 allocator Currently some code pieces assume that address returned by find_e820_area() are page aligned. But looks like find_e820_area() had no such intention and hence one might end up stomping over some of the data. One such case is bootmem allocator initialization code stomped over bss. This patch modified find_e820_area() to return page aligned address. This might be little wasteful of memory but at the same time probably it is easier to handle page aligned memory. Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/x86_64/kernel/e820.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c index b3f0908668e..a75c829c2b0 100644 --- a/arch/x86_64/kernel/e820.c +++ b/arch/x86_64/kernel/e820.c @@ -54,13 +54,13 @@ static inline int bad_addr(unsigned long *addrp, unsigned long size) /* various gunk below that needed for SMP startup */ if (addr < 0x8000) { - *addrp = 0x8000; + *addrp = PAGE_ALIGN(0x8000); return 1; } /* direct mapping tables of the kernel */ if (last >= table_start<= INITRD_START && addr < INITRD_START+INITRD_SIZE) { - *addrp = INITRD_START + INITRD_SIZE; + *addrp = PAGE_ALIGN(INITRD_START + INITRD_SIZE); return 1; } #endif /* kernel code */ - if (last >= __pa_symbol(&_text) && last < __pa_symbol(&_end)) { - *addrp = __pa_symbol(&_end); + if (last >= __pa_symbol(&_text) && addr < __pa_symbol(&_end)) { + *addrp = PAGE_ALIGN(__pa_symbol(&_end)); return 1; } if (last >= ebda_addr && addr < ebda_addr + ebda_size) { - *addrp = ebda_addr + ebda_size; + *addrp = PAGE_ALIGN(ebda_addr + ebda_size); return 1; } @@ -152,7 +152,7 @@ unsigned long __init find_e820_area(unsigned long start, unsigned long end, unsi continue; while (bad_addr(&addr, size) && addr+size <= ei->addr+ei->size) ; - last = addr + size; + last = PAGE_ALIGN(addr) + size; if (last > ei->addr + ei->size) continue; if (last > end) -- cgit v1.2.3 From cdfce1f5714fec7b24715f569b2fee1607350a6d Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] x86: Use -maccumulate-outgoing-args This avoids some problems with gcc 4.x and earlier generating invalid unwind information. In 4.1 the option is default when unwind information is enabled. And it seems to generate smaller code too, so it's probably a good thing on its own. With gcc 4.0: i386: 4683198 902112 480868 6066178 5c9002 vmlinux (before) 4449895 902112 480868 5832875 5900ab vmlinux (after) x86-64: 4939761 1449584 648216 7037561 6b6279 vmlinux (before) 4854193 1449584 648216 6951993 6a1439 vmlinux (after) On 4.1 it shouldn't make much difference because it is default when unwind is enabled anyways. Suggested by Michael Matz and Jan Beulich Cc: jbeulich@novell.com Signed-off-by: Andi Kleen --- arch/x86_64/Makefile | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile index 1c0f18d4f88..13972148058 100644 --- a/arch/x86_64/Makefile +++ b/arch/x86_64/Makefile @@ -54,6 +54,10 @@ endif cflags-y += $(call cc-option,-funit-at-a-time) # prevent gcc from generating any FP code by mistake cflags-y += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,) +# this works around some issues with generating unwind tables in older gccs +# newer gccs do it by default +cflags-y += -maccumulate-outgoing-args + # do binutils support CFI? cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) -- cgit v1.2.3 From cc7d479fe56133e79840beffe9cb4fd193af93aa Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] x86-64: Fix ENOSYS in system call tracing This patch: - out of range system calls failing to return -ENOSYS under system call tracing [AK: split out from another patch by Jan as separate bugfix] Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen --- arch/x86_64/kernel/entry.S | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index 38a7b2d528e..038dcf733a1 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -315,6 +315,8 @@ tracesys: LOAD_ARGS ARGOFFSET /* reload args from stack in case ptrace changed it */ RESTORE_REST cmpq $__NR_syscall_max,%rax + movq $-ENOSYS,%rcx + cmova %rcx,%rax ja 1f movq %r10,%rcx /* fixup for C */ call *sys_call_table(,%rax,8) -- cgit v1.2.3 From 581910e2eb952e541b8ca9b5f551d6c124903b61 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] x86-64: Revert interrupt backlink changes They break more than they fix Signed-off-by: Andi Kleen --- arch/x86_64/kernel/entry.S | 3 --- 1 file changed, 3 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index 038dcf733a1..e3eddde3d3b 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -537,8 +537,6 @@ END(stub_rt_sigreturn) 1: incl %gs:pda_irqcount cmoveq %gs:pda_irqstackptr,%rsp push %rbp # backlink for old unwinder - CFI_ADJUST_CFA_OFFSET 8 - CFI_REL_OFFSET rbp,0 /* * We entered an interrupt context - irqs are off: */ @@ -1178,7 +1176,6 @@ ENTRY(call_softirq) incl %gs:pda_irqcount cmove %gs:pda_irqstackptr,%rsp push %rbp # backlink for old unwinder - CFI_ADJUST_CFA_OFFSET 8 call __do_softirq leaveq CFI_DEF_CFA_REGISTER rsp -- cgit v1.2.3 From 6bf2dafad18c119beb534cbb3d882fe7a6c3f529 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] x86-64: Use irq_domain in ioapic_retrigger_irq Thanks to YH Lu for spotting this. It appears I missed this function when I refactored allocate_irq_vector and introduced irq_domain, with the result that all retriggered irqs would go to cpu 0 even if we were not prepared to receive them there. While reviewing YH's patch I also noticed that this function was missing locking, and since I am now reading two values from two diffrent arrays that looks like a race we might be able to hit in the real world. Cc: Yinghai Lu Signed-off-by: Eric W. Biederman Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/x86_64/kernel/io_apic.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index b848f480851..8a9a357875b 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -1255,12 +1255,15 @@ static int ioapic_retrigger_irq(unsigned int irq) { cpumask_t mask; unsigned vector; + unsigned long flags; + spin_lock_irqsave(&vector_lock, flags); vector = irq_vector[irq]; cpus_clear(mask); - cpu_set(vector >> 8, mask); + cpu_set(first_cpu(irq_domain[irq]), mask); - send_IPI_mask(mask, vector & 0xff); + send_IPI_mask(mask, vector); + spin_unlock_irqrestore(&vector_lock, flags); return 1; } -- cgit v1.2.3 From 8cf2c51927bbeefafc25193d01b91f9ed3806e96 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] x86: Revert new unwind kernel stack termination Jan convinced me that it was unnecessary because the assembly stubs do this already on the stack. Cc: jbeulich@novell.com Signed-off-by: Andi Kleen --- arch/x86_64/kernel/entry.S | 5 ----- 1 file changed, 5 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index e3eddde3d3b..7d401b00d82 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -980,11 +980,6 @@ ENTRY(kernel_thread) call do_fork movq %rax,RAX(%rsp) xorl %edi,%edi - test %rax,%rax - jnz 1f - /* terminate stack in child */ - movq %rdi,RIP(%rsp) -1: /* * It isn't worth to check for reschedule here, -- cgit v1.2.3 From 84f404f695b16bd142c8dd9910d5a398f54fb044 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] x86-64: Put more than one cpu in TARGET_CPUS TARGET_CPUS is the default irq routing poicy. It specifies which cpus the kernel should aim an irq at. In physflat delivery mode we can route an irq to a single cpu. But that doesn't mean our default policy should only be a single cpu is allowed. By allowing the irq routing code to select from multiple cpus this enables systems with more irqs then we can service on a single processor to actually work. I just audited and tested the code and irqbalance doesn't care, and the io_apic.c doesn't care if we have extra cpus in the mask. Everything will use or assume we are using the lowest numbered cpu in the mask if we can't use them all. So this should result in no behavior changes except on systems that need it. Thanks for YH Lu for spotting this problem in his testing. Cc: Yinghai Lu Signed-off-by: Eric W. Biederman Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/x86_64/kernel/genapic_flat.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/genapic_flat.c b/arch/x86_64/kernel/genapic_flat.c index 0dfc223c183..7c01db8fa9d 100644 --- a/arch/x86_64/kernel/genapic_flat.c +++ b/arch/x86_64/kernel/genapic_flat.c @@ -153,7 +153,7 @@ struct genapic apic_flat = { static cpumask_t physflat_target_cpus(void) { - return cpumask_of_cpu(0); + return cpu_online_map; } static cpumask_t physflat_vector_allocation_domain(int cpu) -- cgit v1.2.3 From dbaab49f92ff6ae6255762a948375e4036cbdbd2 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Sat, 21 Oct 2006 18:37:03 +0200 Subject: [PATCH] x86-64: Overlapping program headers in physical addr space fix o A recent change to vmlinux.ld.S file broke kexec as now resulting vmlinux program headers are overlapping in physical address space. o Now all the vsyscall related sections are placed after data and after that mostly init data sections are placed. To avoid physical overlap among phdrs, there are three possible solutions. - Place vsyscall sections also in data phdrs instead of user - move vsyscal sections after init data in bss. - create another phdrs say data.init and move all the sections after vsyscall into this new phdr. o This patch implements the third solution. Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen Cc: Magnus Damm Cc: Andi Kleen Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton --- arch/x86_64/kernel/vmlinux.lds.S | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index b9df2ab6529..1283614c9b2 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -17,6 +17,7 @@ PHDRS { text PT_LOAD FLAGS(5); /* R_E */ data PT_LOAD FLAGS(7); /* RWE */ user PT_LOAD FLAGS(7); /* RWE */ + data.init PT_LOAD FLAGS(7); /* RWE */ note PT_NOTE FLAGS(4); /* R__ */ } SECTIONS @@ -131,7 +132,7 @@ SECTIONS . = ALIGN(8192); /* init_task */ .data.init_task : AT(ADDR(.data.init_task) - LOAD_OFFSET) { *(.data.init_task) - } :data + }:data.init . = ALIGN(4096); .data.page_aligned : AT(ADDR(.data.page_aligned) - LOAD_OFFSET) { -- cgit v1.2.3 From e70ea8c09db0e25ab58f84ba7f393e5c9125a8ee Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:03 +0200 Subject: [PATCH] x86-64: Revert timer routing behaviour back to 2.6.16 state By default route the 8254 over the 8259 and only disable it on ATI boards where this causes double timer interrupts. This should unbreak some Nvidia boards where the timer doesn't seem to tick of it isn't enabled in the 8259. At least one VIA board also seemed to have a little trouble with the disabled 8259. For 2.6.20 we'll try both dynamically without black listing, but I think for .19 this is the safer approach because it has been already well tested in earlier kernels. This also makes the x86-64 behaviour the same as i386. Command line options can change all this of course. Signed-off-by: Andi Kleen --- arch/x86_64/kernel/early-quirks.c | 9 +++++---- arch/x86_64/kernel/io_apic.c | 2 +- 2 files changed, 6 insertions(+), 5 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/early-quirks.c b/arch/x86_64/kernel/early-quirks.c index 208e38a372c..2b1245d8625 100644 --- a/arch/x86_64/kernel/early-quirks.c +++ b/arch/x86_64/kernel/early-quirks.c @@ -61,10 +61,11 @@ static void nvidia_bugs(void) static void ati_bugs(void) { -#if 1 /* for testing */ - printk("ATI board detected\n"); -#endif - /* No bugs right now */ + if (timer_over_8254 == 1) { + timer_over_8254 = 0; + printk(KERN_INFO + "ATI board detected. Disabling timer routing over 8254.\n"); + } } struct chipset { diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 8a9a357875b..b000017e4b5 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -57,7 +57,7 @@ static int no_timer_check; static int disable_timer_pin_1 __initdata; -int timer_over_8254 __initdata = 0; +int timer_over_8254 __initdata = 1; /* Where if anywhere is the i8259 connect in external int mode */ static struct { int pin, apic; } ioapic_i8259 = { -1, -1 }; -- cgit v1.2.3 From aa026ede513b7d672fa7d9106b2f2a475455dcf2 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sun, 22 Oct 2006 00:41:15 +0200 Subject: [PATCH] x86-64: Fix C3 timer test There was a typo in the C3 latency test to decide of the TSC should be used or not. It used the C2 latency threshold, not the C3 one. Fix that. This should fix the time on various dual core laptops. Signed-off-by: Andi Kleen --- arch/x86_64/kernel/time.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c index 1ba5a442ac3..88722f11ca1 100644 --- a/arch/x86_64/kernel/time.c +++ b/arch/x86_64/kernel/time.c @@ -948,7 +948,7 @@ __cpuinit int unsynchronized_tsc(void) if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) { #ifdef CONFIG_ACPI /* But TSC doesn't tick in C3 so don't use it there */ - if (acpi_fadt.length > 0 && acpi_fadt.plvl3_lat < 100) + if (acpi_fadt.length > 0 && acpi_fadt.plvl3_lat < 1000) return 1; #endif return 0; -- cgit v1.2.3 From cb01fc720c629261b9c616b2d5fcc3d93cd8bb09 Mon Sep 17 00:00:00 2001 From: Muli Ben-Yehuda Date: Sun, 22 Oct 2006 00:41:15 +0200 Subject: [PATCH] x86-64: increase PHB1 split transaction timeout This patch increases the timeout for PCI split transactions on PHB1 on the first Calgary to work around an issue with the aic94xx adapter. Fixes kernel.org bugzilla #7180 (http://bugzilla.kernel.org/show_bug.cgi?id=7180) Based on excellent debugging and a patch by Darrick J. Wong Signed-off-by: Muli Ben-Yehuda Signed-off-by: Jon Mason Signed-off-by: Andi Kleen Acked-by: Darrick J. Wong --- arch/x86_64/kernel/pci-calgary.c | 44 +++++++++++++++++++++++++++++++++++++++- 1 file changed, 43 insertions(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c index b3296cc2f2f..37a770859e7 100644 --- a/arch/x86_64/kernel/pci-calgary.c +++ b/arch/x86_64/kernel/pci-calgary.c @@ -52,7 +52,8 @@ #define ONE_BASED_CHASSIS_NUM 1 /* register offsets inside the host bridge space */ -#define PHB_CSR_OFFSET 0x0110 +#define CALGARY_CONFIG_REG 0x0108 +#define PHB_CSR_OFFSET 0x0110 /* Channel Status */ #define PHB_PLSSR_OFFSET 0x0120 #define PHB_CONFIG_RW_OFFSET 0x0160 #define PHB_IOBASE_BAR_LOW 0x0170 @@ -83,6 +84,8 @@ #define TAR_VALID 0x0000000000000008UL /* CSR (Channel/DMA Status Register) */ #define CSR_AGENT_MASK 0xffe0ffff +/* CCR (Calgary Configuration Register) */ +#define CCR_2SEC_TIMEOUT 0x000000000000000EUL #define MAX_NUM_OF_PHBS 8 /* how many PHBs in total? */ #define MAX_NUM_CHASSIS 8 /* max number of chassis */ @@ -732,6 +735,38 @@ static void calgary_watchdog(unsigned long data) } } +static void __init calgary_increase_split_completion_timeout(void __iomem *bbar, + unsigned char busnum) +{ + u64 val64; + void __iomem *target; + unsigned long phb_shift = -1; + u64 mask; + + switch (busno_to_phbid(busnum)) { + case 0: phb_shift = (63 - 19); + break; + case 1: phb_shift = (63 - 23); + break; + case 2: phb_shift = (63 - 27); + break; + case 3: phb_shift = (63 - 35); + break; + default: + BUG_ON(busno_to_phbid(busnum)); + } + + target = calgary_reg(bbar, CALGARY_CONFIG_REG); + val64 = be64_to_cpu(readq(target)); + + /* zero out this PHB's timer bits */ + mask = ~(0xFUL << phb_shift); + val64 &= mask; + val64 |= (CCR_2SEC_TIMEOUT << phb_shift); + writeq(cpu_to_be64(val64), target); + readq(target); /* flush */ +} + static void __init calgary_enable_translation(struct pci_dev *dev) { u32 val32; @@ -756,6 +791,13 @@ static void __init calgary_enable_translation(struct pci_dev *dev) writel(cpu_to_be32(val32), target); readl(target); /* flush */ + /* + * Give split completion a longer timeout on bus 1 for aic94xx + * http://bugzilla.kernel.org/show_bug.cgi?id=7180 + */ + if (busnum == 1) + calgary_increase_split_completion_timeout(bbar, busnum); + init_timer(&tbl->watchdog_timer); tbl->watchdog_timer.function = &calgary_watchdog; tbl->watchdog_timer.data = (unsigned long)dev; -- cgit v1.2.3 From d1752aa884ec0ac3027c1a3d456bf69bf765c8b8 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 25 Oct 2006 01:00:22 +0200 Subject: [PATCH] x86-64: Simplify the vector allocator. There is no reason to remember a per cpu position of which vector to try. Keeping a global position is simpler and more likely to result in a global vector allocation even if I don't need or require it. For level triggered interrupts this means we are less likely to acknowledge another cpus irq, and cause the level triggered irq to harmlessly refire. This simplification makes it easier to only access data structures of online cpus, by having fewer special cases to deal with. Signed-off-by: Eric W. Biederman Signed-off-by: Andi Kleen --- arch/x86_64/kernel/io_apic.c | 20 +++++++------------- 1 file changed, 7 insertions(+), 13 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index b000017e4b5..0e89ae7e7b2 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -612,10 +612,7 @@ static int __assign_irq_vector(int irq, cpumask_t mask, cpumask_t *result) * Also, we've got to be careful not to trash gate * 0x80, because int 0x80 is hm, kind of importantish. ;) */ - static struct { - int vector; - int offset; - } pos[NR_CPUS] = { [ 0 ... NR_CPUS - 1] = {FIRST_DEVICE_VECTOR, 0} }; + static int current_vector = FIRST_DEVICE_VECTOR, current_offset = 0; int old_vector = -1; int cpu; @@ -631,14 +628,13 @@ static int __assign_irq_vector(int irq, cpumask_t mask, cpumask_t *result) for_each_cpu_mask(cpu, mask) { cpumask_t domain; - int first, new_cpu; + int new_cpu; int vector, offset; domain = vector_allocation_domain(cpu); - first = first_cpu(domain); - vector = pos[first].vector; - offset = pos[first].offset; + vector = current_vector; + offset = current_offset; next: vector += 8; if (vector >= FIRST_SYSTEM_VECTOR) { @@ -646,7 +642,7 @@ next: offset = (offset + 1) % 8; vector = FIRST_DEVICE_VECTOR + offset; } - if (unlikely(pos[first].vector == vector)) + if (unlikely(current_vector == vector)) continue; if (vector == IA32_SYSCALL_VECTOR) goto next; @@ -654,10 +650,8 @@ next: if (per_cpu(vector_irq, new_cpu)[vector] != -1) goto next; /* Found one! */ - for_each_cpu_mask(new_cpu, domain) { - pos[new_cpu].vector = vector; - pos[new_cpu].offset = offset; - } + current_vector = vector; + current_offset = offset; if (old_vector >= 0) { int old_cpu; for_each_cpu_mask(old_cpu, irq_domain[irq]) -- cgit v1.2.3 From 70a0a5357db20c291d46c04011d646d5d84c868c Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 25 Oct 2006 01:00:23 +0200 Subject: [PATCH] x86-64: Only look at per_cpu data for online cpus. When I generalized __assign_irq_vector I failed to pay attention to what happens when you access a per cpu data structure for a cpu that is not online. It is an undefined case making any code that does it have undefined behavior as well. The code still needs to be able to allocate a vector across cpus that are not online to properly handle combinations like lowest priority interrupt delivery and cpu_hotplug. Not that we can do that today but the infrastructure shouldn't prevent it. So this patch updates the places where we touch per cpu data to only touch online cpus, it makes cpu vector allocation an atomic operation with respect to cpu hotplug, and it updates the cpu start code to properly initialize vector_irq so we don't have inconsistencies. Signed-off-by: Eric W. Biederman Signed-off-by: Andi Kleen --- arch/x86_64/kernel/io_apic.c | 42 +++++++++++++++++++++++++++++++++++++----- arch/x86_64/kernel/smpboot.c | 7 ++++++- 2 files changed, 43 insertions(+), 6 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 0e89ae7e7b2..fe429e5d6b2 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -63,7 +63,7 @@ int timer_over_8254 __initdata = 1; static struct { int pin, apic; } ioapic_i8259 = { -1, -1 }; static DEFINE_SPINLOCK(ioapic_lock); -static DEFINE_SPINLOCK(vector_lock); +DEFINE_SPINLOCK(vector_lock); /* * # of IRQ routing registers @@ -618,6 +618,9 @@ static int __assign_irq_vector(int irq, cpumask_t mask, cpumask_t *result) BUG_ON((unsigned)irq >= NR_IRQ_VECTORS); + /* Only try and allocate irqs on cpus that are present */ + cpus_and(mask, mask, cpu_online_map); + if (irq_vector[irq] > 0) old_vector = irq_vector[irq]; if (old_vector > 0) { @@ -627,11 +630,12 @@ static int __assign_irq_vector(int irq, cpumask_t mask, cpumask_t *result) } for_each_cpu_mask(cpu, mask) { - cpumask_t domain; + cpumask_t domain, new_mask; int new_cpu; int vector, offset; domain = vector_allocation_domain(cpu); + cpus_and(new_mask, domain, cpu_online_map); vector = current_vector; offset = current_offset; @@ -646,18 +650,20 @@ next: continue; if (vector == IA32_SYSCALL_VECTOR) goto next; - for_each_cpu_mask(new_cpu, domain) + for_each_cpu_mask(new_cpu, new_mask) if (per_cpu(vector_irq, new_cpu)[vector] != -1) goto next; /* Found one! */ current_vector = vector; current_offset = offset; if (old_vector >= 0) { + cpumask_t old_mask; int old_cpu; - for_each_cpu_mask(old_cpu, irq_domain[irq]) + cpus_and(old_mask, irq_domain[irq], cpu_online_map); + for_each_cpu_mask(old_cpu, old_mask) per_cpu(vector_irq, old_cpu)[old_vector] = -1; } - for_each_cpu_mask(new_cpu, domain) + for_each_cpu_mask(new_cpu, new_mask) per_cpu(vector_irq, new_cpu)[vector] = irq; irq_vector[irq] = vector; irq_domain[irq] = domain; @@ -678,6 +684,32 @@ static int assign_irq_vector(int irq, cpumask_t mask, cpumask_t *result) return vector; } +void __setup_vector_irq(int cpu) +{ + /* Initialize vector_irq on a new cpu */ + /* This function must be called with vector_lock held */ + unsigned long flags; + int irq, vector; + + + /* Mark the inuse vectors */ + for (irq = 0; irq < NR_IRQ_VECTORS; ++irq) { + if (!cpu_isset(cpu, irq_domain[irq])) + continue; + vector = irq_vector[irq]; + per_cpu(vector_irq, cpu)[vector] = irq; + } + /* Mark the free vectors */ + for (vector = 0; vector < NR_VECTORS; ++vector) { + irq = per_cpu(vector_irq, cpu)[vector]; + if (irq < 0) + continue; + if (!cpu_isset(cpu, irq_domain[irq])) + per_cpu(vector_irq, cpu)[vector] = -1; + } +} + + extern void (*interrupt[NR_IRQS])(void); static struct irq_chip ioapic_chip; diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c index 7b7a6870288..62c2e747af5 100644 --- a/arch/x86_64/kernel/smpboot.c +++ b/arch/x86_64/kernel/smpboot.c @@ -581,12 +581,16 @@ void __cpuinit start_secondary(void) * smp_call_function(). */ lock_ipi_call_lock(); + spin_lock(&vector_lock); + /* Setup the per cpu irq handling data structures */ + __setup_vector_irq(smp_processor_id()); /* * Allow the master to continue. */ cpu_set(smp_processor_id(), cpu_online_map); per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE; + spin_unlock(&vector_lock); unlock_ipi_call_lock(); cpu_idle(); @@ -799,7 +803,6 @@ static int __cpuinit do_boot_cpu(int cpu, int apicid) cpu, node); } - alternatives_smp_switch(1); c_idle.idle = get_idle_for_cpu(cpu); @@ -1246,8 +1249,10 @@ int __cpu_disable(void) local_irq_disable(); remove_siblinginfo(cpu); + spin_lock(&vector_lock); /* It's now safe to remove this processor from the online map */ cpu_clear(cpu, cpu_online_map); + spin_unlock(&vector_lock); remove_cpu_from_maps(); fixup_irqs(cpu_online_map); return 0; -- cgit v1.2.3 From 3095fc0c9772b4afb3c81f76664f341ef716d380 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Fri, 20 Oct 2006 14:45:33 -0700 Subject: PCI: x86-64: mmconfig missing printk levels Signed-off-by: Dave Jones Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- arch/x86_64/pci/mmconfig.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c index 7732f4254d2..e61093b34c2 100644 --- a/arch/x86_64/pci/mmconfig.c +++ b/arch/x86_64/pci/mmconfig.c @@ -220,7 +220,7 @@ void __init pci_mmcfg_init(int type) pci_mmcfg_virt = kmalloc(sizeof(*pci_mmcfg_virt) * pci_mmcfg_config_num, GFP_KERNEL); if (pci_mmcfg_virt == NULL) { - printk("PCI: Can not allocate memory for mmconfig structures\n"); + printk(KERN_ERR "PCI: Can not allocate memory for mmconfig structures\n"); return; } for (i = 0; i < pci_mmcfg_config_num; ++i) { @@ -228,7 +228,8 @@ void __init pci_mmcfg_init(int type) pci_mmcfg_virt[i].virt = ioremap_nocache(pci_mmcfg_config[i].base_address, MMCONFIG_APER_MAX); if (!pci_mmcfg_virt[i].virt) { - printk("PCI: Cannot map mmconfig aperture for segment %d\n", + printk(KERN_ERR "PCI: Cannot map mmconfig aperture for " + "segment %d\n", pci_mmcfg_config[i].pci_segment_group_number); return; } -- cgit v1.2.3 From 61ce1efe6e40233663d27ab8ac9ba9710eebcaad Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Fri, 27 Oct 2006 11:41:44 -0700 Subject: [PATCH] vmlinux.lds: consolidate initcall sections Add a vmlinux.lds.h helper macro for defining the eight-level initcall table, teach all the architectures to use it. This is a prerequisite for a patch which performs initcall synchronisation for multithreaded-probing. Cc: Greg KH Signed-off-by: Andrew Morton [ Added AVR32 as well ] Signed-off-by: Haavard Skinnemoen Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/vmlinux.lds.S | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index 1283614c9b2..edb24aa714b 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -175,13 +175,7 @@ SECTIONS __setup_end = .; __initcall_start = .; .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) { - *(.initcall1.init) - *(.initcall2.init) - *(.initcall3.init) - *(.initcall4.init) - *(.initcall5.init) - *(.initcall6.init) - *(.initcall7.init) + INITCALLS } __initcall_end = .; __con_initcall_start = .; -- cgit v1.2.3 From a7aacdf9ea45bf6139cfd750e558a3dcbc6f16c3 Mon Sep 17 00:00:00 2001 From: Albert Cahalan Date: Sun, 29 Oct 2006 22:26:17 -0500 Subject: [PATCH] fix i386 regparm=3 RT signal handlers on x86_64 The recent change to make x86_64 support i386 binaries compiled with -mregparm=3 only covered signal handlers without SA_SIGINFO. (the 3-arg "real-time" ones) This is useful for klibc at least. Signed-off-by: Albert Cahalan Signed-off-by: Linus Torvalds --- arch/x86_64/ia32/ia32_signal.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/ia32/ia32_signal.c b/arch/x86_64/ia32/ia32_signal.c index a6ba9951e86..0e0a266d976 100644 --- a/arch/x86_64/ia32/ia32_signal.c +++ b/arch/x86_64/ia32/ia32_signal.c @@ -579,6 +579,11 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, regs->rsp = (unsigned long) frame; regs->rip = (unsigned long) ka->sa.sa_handler; + /* Make -mregparm=3 work */ + regs->rax = sig; + regs->rdx = (unsigned long) &frame->info; + regs->rcx = (unsigned long) &frame->uc; + asm volatile("movl %0,%%ds" :: "r" (__USER32_DS)); asm volatile("movl %0,%%es" :: "r" (__USER32_DS)); -- cgit v1.2.3 From 6c0ffb9d2fd987c79c6cbb81c3f3011c63749b1a Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 8 Nov 2006 10:23:03 -0800 Subject: x86-64: clean up io-apic accesses This is just commit 130fe05dbc0114609cfef9815c0c5580b42decfa ported to x86-64, for all the same reasons. It cleans up the IO-APIC accesses in order to then fix the ordering issues. We move the accessor functions (that were only used by io_apic.c) out of a header file, and use proper memory-mapped accesses rather than making up our own "volatile" pointers. Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/io_apic.c | 46 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 46 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index fe429e5d6b2..96e02d84571 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -88,6 +88,52 @@ static struct irq_pin_list { short apic, pin, next; } irq_2_pin[PIN_MAP_SIZE]; +struct io_apic { + unsigned int index; + unsigned int unused[3]; + unsigned int data; +}; + +static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) +{ + return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx) + + (mp_ioapics[idx].mpc_apicaddr & ~PAGE_MASK); +} + +static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(reg, &io_apic->index); + return readl(&io_apic->data); +} + +static inline void io_apic_write(unsigned int apic, unsigned int reg, unsigned int value) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(reg, &io_apic->index); + writel(value, &io_apic->data); +} + +/* + * Re-write a value: to be used for read-modify-write + * cycles where the read already set up the index register. + */ +static inline void io_apic_modify(unsigned int apic, unsigned int value) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(value, &io_apic->data); +} + +/* + * Synchronize the IO-APIC and the CPU by doing + * a dummy read from the IO-APIC + */ +static inline void io_apic_sync(unsigned int apic) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + readl(&io_apic->data); +} + #define __DO_ACTION(R, ACTION, FINAL) \ \ { \ -- cgit v1.2.3 From 48797ebd9e8b16fddcd4ef062f792314a6b9219a Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 8 Nov 2006 10:27:54 -0800 Subject: x86-64: write IO APIC irq routing entries in correct order This is the x86-64 version of f9dadfa71bc594df09044da61d1c72701121d802 that did the same thing on i386. Since the "mask" bit is in the low word, when we write a new entry, we need to write the high word first, before we potentially unmask it. The exception is when we actually want to mask the interrupt, in which case we want to write the low word first to make sure that the high word doesn't change while the interrupt routing is still active. Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/io_apic.c | 26 +++++++++++++++++++++++--- 1 file changed, 23 insertions(+), 3 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 96e02d84571..3b8f9c68ad3 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -172,11 +172,33 @@ static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin) return eu.entry; } +/* + * When we write a new IO APIC routing entry, we need to write the high + * word first! If the mask bit in the low word is clear, we will enable + * the interrupt, and we need to make sure the entry is fully populated + * before that happens. + */ static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) { unsigned long flags; union entry_union eu; eu.entry = e; + spin_lock_irqsave(&ioapic_lock, flags); + io_apic_write(apic, 0x11 + 2*pin, eu.w2); + io_apic_write(apic, 0x10 + 2*pin, eu.w1); + spin_unlock_irqrestore(&ioapic_lock, flags); +} + +/* + * When we mask an IO APIC routing entry, we need to write the low + * word first, in order to set the mask bit before we change the + * high bits! + */ +static void ioapic_mask_entry(int apic, int pin) +{ + unsigned long flags; + union entry_union eu = { .entry.mask = 1 }; + spin_lock_irqsave(&ioapic_lock, flags); io_apic_write(apic, 0x10 + 2*pin, eu.w1); io_apic_write(apic, 0x11 + 2*pin, eu.w2); @@ -302,9 +324,7 @@ static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) /* * Disable it in the IO-APIC irq-routing table: */ - memset(&entry, 0, sizeof(entry)); - entry.mask = 1; - ioapic_write_entry(apic, pin, entry); + ioapic_mask_entry(apic, pin); } static void clear_IO_APIC (void) -- cgit v1.2.3 From ec68307cc5a8dc499e48693843bb42f6b6028458 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 8 Nov 2006 17:44:57 -0800 Subject: [PATCH] htirq: refactor so we only have one function that writes to the chip This refactoring actually optimizes the code a little by caching the value that we think the device is programmed with instead of reading it back from the hardware. Which simplifies the code a little and should speed things up a bit. This patch introduces the concept of a ht_irq_msg and modifies the architecture read/write routines to update this code. There is a minor consistency fix here as well as x86_64 forgot to initialize the htirq as masked. Signed-off-by: Eric W. Biederman Cc: Andi Kleen Acked-by: Bryan O'Sullivan Cc: Cc: Roland Dreier Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/io_apic.c | 31 +++++++++++++++---------------- 1 file changed, 15 insertions(+), 16 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 3b8f9c68ad3..41bfc49301a 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -1955,18 +1955,16 @@ void arch_teardown_msi_irq(unsigned int irq) static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector) { - u32 low, high; - low = read_ht_irq_low(irq); - high = read_ht_irq_high(irq); + struct ht_irq_msg msg; + fetch_ht_irq_msg(irq, &msg); - low &= ~(HT_IRQ_LOW_VECTOR_MASK | HT_IRQ_LOW_DEST_ID_MASK); - high &= ~(HT_IRQ_HIGH_DEST_ID_MASK); + msg.address_lo &= ~(HT_IRQ_LOW_VECTOR_MASK | HT_IRQ_LOW_DEST_ID_MASK); + msg.address_hi &= ~(HT_IRQ_HIGH_DEST_ID_MASK); - low |= HT_IRQ_LOW_VECTOR(vector) | HT_IRQ_LOW_DEST_ID(dest); - high |= HT_IRQ_HIGH_DEST_ID(dest); + msg.address_lo |= HT_IRQ_LOW_VECTOR(vector) | HT_IRQ_LOW_DEST_ID(dest); + msg.address_hi |= HT_IRQ_HIGH_DEST_ID(dest); - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); } static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) @@ -1987,7 +1985,7 @@ static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) dest = cpu_mask_to_apicid(tmp); - target_ht_irq(irq, dest, vector & 0xff); + target_ht_irq(irq, dest, vector); set_native_irq_info(irq, mask); } #endif @@ -2010,14 +2008,15 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) vector = assign_irq_vector(irq, TARGET_CPUS, &tmp); if (vector >= 0) { - u32 low, high; + struct ht_irq_msg msg; unsigned dest; dest = cpu_mask_to_apicid(tmp); - high = HT_IRQ_HIGH_DEST_ID(dest); + msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest); - low = HT_IRQ_LOW_BASE | + msg.address_lo = + HT_IRQ_LOW_BASE | HT_IRQ_LOW_DEST_ID(dest) | HT_IRQ_LOW_VECTOR(vector) | ((INT_DEST_MODE == 0) ? @@ -2026,10 +2025,10 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) HT_IRQ_LOW_RQEOI_EDGE | ((INT_DELIVERY_MODE != dest_LowestPrio) ? HT_IRQ_LOW_MT_FIXED : - HT_IRQ_LOW_MT_ARBITRATED); + HT_IRQ_LOW_MT_ARBITRATED) | + HT_IRQ_LOW_IRQ_MASKED; - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); set_irq_chip_and_handler_name(irq, &ht_irq_chip, handle_edge_irq, "edge"); -- cgit v1.2.3 From 64e72e41acae0dab733fb0d5d789b76d115210c0 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 14 Nov 2006 16:56:33 +0100 Subject: Revert "[PATCH] MMCONFIG and new Intel motherboards" This reverts 4c6e052adfe285ede5884e4e8c4d33af33932c13 commit. Following Linus' i386 change: revert resource reservation for mmcfg config now. Will be revisited in .20 hopefully. --- arch/x86_64/pci/mmconfig.c | 32 -------------------------------- 1 file changed, 32 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c index e61093b34c2..f8b6b2800a6 100644 --- a/arch/x86_64/pci/mmconfig.c +++ b/arch/x86_64/pci/mmconfig.c @@ -163,37 +163,6 @@ static __init void unreachable_devices(void) } } -static __init void pci_mmcfg_insert_resources(void) -{ -#define PCI_MMCFG_RESOURCE_NAME_LEN 19 - int i; - struct resource *res; - char *names; - unsigned num_buses; - - res = kcalloc(PCI_MMCFG_RESOURCE_NAME_LEN + sizeof(*res), - pci_mmcfg_config_num, GFP_KERNEL); - - if (!res) { - printk(KERN_ERR "PCI: Unable to allocate MMCONFIG resources\n"); - return; - } - - names = (void *)&res[pci_mmcfg_config_num]; - for (i = 0; i < pci_mmcfg_config_num; i++, res++) { - num_buses = pci_mmcfg_config[i].end_bus_number - - pci_mmcfg_config[i].start_bus_number + 1; - res->name = names; - snprintf(names, PCI_MMCFG_RESOURCE_NAME_LEN, "PCI MMCONFIG %u", - pci_mmcfg_config[i].pci_segment_group_number); - res->start = pci_mmcfg_config[i].base_address; - res->end = res->start + (num_buses << 20) - 1; - res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; - insert_resource(&iomem_resource, res); - names += PCI_MMCFG_RESOURCE_NAME_LEN; - } -} - void __init pci_mmcfg_init(int type) { int i; @@ -237,7 +206,6 @@ void __init pci_mmcfg_init(int type) } unreachable_devices(); - pci_mmcfg_insert_resources(); raw_pci_ops = &pci_mmcfg; pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF; -- cgit v1.2.3 From 14f448e36192d6d2cd7dfd81cb044977b2f9dd9b Mon Sep 17 00:00:00 2001 From: Aaron Durbin Date: Tue, 14 Nov 2006 16:57:45 +0100 Subject: [PATCH] x86-64: Fix partial page check to ensure unusable memory is not being marked usable. Fix partial page check in e820_register_active_regions to ensure partial pages are not being marked as active in the memory pool. Signed-off-by: Aaron Durbin Signed-off-by: Andi Kleen --- arch/x86_64/kernel/e820.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c index a75c829c2b0..855b5611318 100644 --- a/arch/x86_64/kernel/e820.c +++ b/arch/x86_64/kernel/e820.c @@ -278,7 +278,7 @@ e820_register_active_regions(int nid, unsigned long start_pfn, >> PAGE_SHIFT; /* Skip map entries smaller than a page */ - if (ei_startpfn > ei_endpfn) + if (ei_startpfn >= ei_endpfn) continue; /* Check if end_pfn_map should be updated */ -- cgit v1.2.3 From 14679eb3c50897889ba62f9a37e3bcd8a205b5e7 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86-64: Fix PTRACE_[SG]ET_THREAD_AREA regression with ia32 emulation. ptrace(PTRACE_[SG]ET_THREAD_AREA) calls from ia32 code should be passed onto the x86_64 implementation. The default case in sys32_ptrace used to call to sys_ptrace(), but is now EINVAL. This patch fixes a regression caused by that changed. Signed-off-by: Mike McCormack Signed-off-by: Andi Kleen --- arch/x86_64/ia32/ptrace32.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/ia32/ptrace32.c b/arch/x86_64/ia32/ptrace32.c index 3a7561d4703..04566fe5de4 100644 --- a/arch/x86_64/ia32/ptrace32.c +++ b/arch/x86_64/ia32/ptrace32.c @@ -244,6 +244,8 @@ asmlinkage long sys32_ptrace(long request, u32 pid, u32 addr, u32 data) case PTRACE_DETACH: case PTRACE_SYSCALL: case PTRACE_SETOPTIONS: + case PTRACE_SET_THREAD_AREA: + case PTRACE_GET_THREAD_AREA: return sys_ptrace(request, pid, addr, data); default: -- cgit v1.2.3 From 51d67a488b53a5cc8401460480c124eaec71e2d4 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86-64: shorten the x86_64 boot setup GDT to what the comment says Stephen Tweedie, Herbert Xu, and myself have been struggling with a very nasty bug in Xen. But it also pointed out a small bug in the x86_64 kernel boot setup. The GDT limit being setup by the initial bzImage code when entering into protected mode is way too big. The comment by the code states that the size of the GDT is 2048, but the actual size being set up is much bigger (32768). This happens simply because of one extra '0'. Instead of setting up a 0x800 size, 0x8000 is set up. On bare metal this is fine because the CPU wont load any segments unless they are explicitly used. But unfortunately, this breaks Xen on vmx FV, since it (for now) blindly loads all the segments into the VMCS if they are less than the gdt limit. Since the real mode segments are around 0x3000, we are getting junk into the VMCS and that later causes an exception. Stephen Tweedie has written up a patch to fix the Xen side and will be submitting that to those folks. But that doesn't excuse the GDT limit being a magnitude too big. AK: changed to compute true gdt size in assembler, fixed comment Signed-off-by: Steven Rostedt Signed-off-by: Andi Kleen --- arch/x86_64/boot/setup.S | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/boot/setup.S b/arch/x86_64/boot/setup.S index c3bfd223ab4..770940cc010 100644 --- a/arch/x86_64/boot/setup.S +++ b/arch/x86_64/boot/setup.S @@ -836,13 +836,12 @@ gdt: .word 0x9200 # data read/write .word 0x00CF # granularity = 4096, 386 # (+5th nibble of limit) +gdt_end: idt_48: .word 0 # idt limit = 0 .word 0, 0 # idt base = 0L gdt_48: - .word 0x8000 # gdt limit=2048, - # 256 GDT entries - + .word gdt_end-gdt-1 # gdt limit .word 0, 0 # gdt base (filled in later) # Include video setup & detection code -- cgit v1.2.3 From 5e58a02a8f6a7a1c9ae41f39286bcd3aea0d6f24 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86-64: Handle reserve_bootmem_generic beyond end_pfn This can happen on kexec kernels with some configurations, in particularly on Unisys ES7000 systems. Analysis by Amul Shah Cc: Amul Shah Signed-off-by: Andi Kleen --- arch/x86_64/mm/init.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c index 971dc1181e6..f1f977aafae 100644 --- a/arch/x86_64/mm/init.c +++ b/arch/x86_64/mm/init.c @@ -655,9 +655,22 @@ void free_initrd_mem(unsigned long start, unsigned long end) void __init reserve_bootmem_generic(unsigned long phys, unsigned len) { - /* Should check here against the e820 map to avoid double free */ #ifdef CONFIG_NUMA int nid = phys_to_nid(phys); +#endif + unsigned long pfn = phys >> PAGE_SHIFT; + if (pfn >= end_pfn) { + /* This can happen with kdump kernels when accessing firmware + tables. */ + if (pfn < end_pfn_map) + return; + printk(KERN_ERR "reserve_bootmem: illegal reserve %lx %u\n", + phys, len); + return; + } + + /* Should check here against the e820 map to avoid double free */ +#ifdef CONFIG_NUMA reserve_bootmem_node(NODE_DATA(nid), phys, len); #else reserve_bootmem(phys, len); -- cgit v1.2.3 From 15803a43288da434d34d41c4ed650c3c1728d42c Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86-64: setup saved_max_pfn correctly (kdump) x86_64: setup saved_max_pfn correctly 2.6.19-rc4 has broken CONFIG_CRASH_DUMP support on x86_64. It is impossible to read out the kernel contents from /proc/vmcore because saved_max_pfn is set to zero instead of the max_pfn value before the user map is setup. This happens because saved_max_pfn is initialized at parse_early_param() time, and at this time no active regions have been registered. save_max_pfn is setup from e820_end_of_ram(), more exact find_max_pfn_with_active_regions() which returns 0 because no regions exist. This patch fixes this by registering before and removing after the call to e820_end_of_ram(). Signed-off-by: Magnus Damm Signed-off-by: Andi Kleen --- arch/x86_64/kernel/e820.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c index 855b5611318..6fe191c5808 100644 --- a/arch/x86_64/kernel/e820.c +++ b/arch/x86_64/kernel/e820.c @@ -594,7 +594,9 @@ static int __init parse_memmap_opt(char *p) * size before original memory map is * reset. */ + e820_register_active_regions(0, 0, -1UL); saved_max_pfn = e820_end_of_ram(); + remove_all_active_ranges(); #endif end_pfn_map = 0; e820.nr_map = 0; -- cgit v1.2.3 From fa18f477d0987c011cce047a7c3cd1284f547a14 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86: Add acpi_user_timer_override option for Asus boards Timer overrides are normally disabled on Nvidia board because they are commonly wrong, except on new ones with HPET support. Unfortunately there are quite some Asus boards around that don't have HPET, but need a timer override. We don't know yet how to handle this transparently, but at least add a command line option to force the timer override and let them boot. Cc: len.brown@intel.com Signed-off-by: Andi Kleen --- arch/x86_64/kernel/early-quirks.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/early-quirks.c b/arch/x86_64/kernel/early-quirks.c index 2b1245d8625..68273bff58c 100644 --- a/arch/x86_64/kernel/early-quirks.c +++ b/arch/x86_64/kernel/early-quirks.c @@ -45,7 +45,13 @@ static void nvidia_bugs(void) /* * All timer overrides on Nvidia are * wrong unless HPET is enabled. + * Unfortunately that's not true on many Asus boards. + * We don't know yet how to detect this automatically, but + * at least allow a command line override. */ + if (acpi_use_timer_override) + return; + nvidia_hpet_detected = 0; acpi_table_parse(ACPI_HPET, nvidia_hpet_check); if (nvidia_hpet_detected == 0) { @@ -53,6 +59,8 @@ static void nvidia_bugs(void) printk(KERN_INFO "Nvidia board " "detected. Ignoring ACPI " "timer override.\n"); + printk(KERN_INFO "If you got timer trouble " + "try acpi_use_timer_override\n"); } #endif /* RED-PEN skip them on mptables too? */ -- cgit v1.2.3 From 8c131af1db510793f87dc43edbc8950a35370df3 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86-64: Fix vgetcpu when CONFIG_HOTPLUG_CPU is disabled The vgetcpu per CPU initialization previously relied on CPU hotplug events for all CPUs to initialize the per CPU state. That only worked only on kernels with CONFIG_HOTPLUG_CPU enabled. On the others some CPUs didn't get their state initialized properly and vgetcpu wouldn't work. Change the initialization sequence to instead run in a normal initcall (which runs after the normal CPU bootup) and initialize all running CPUs there. Later hotplug CPUs are still handled with an hotplug notifier. This actually simplifies the code somewhat. Signed-off-by: Andi Kleen --- arch/x86_64/kernel/smp.c | 3 +-- arch/x86_64/kernel/time.c | 11 ----------- arch/x86_64/kernel/vsyscall.c | 45 ++++++++++++++++++++++++------------------- 3 files changed, 26 insertions(+), 33 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c index 4f67697f503..9f74c883568 100644 --- a/arch/x86_64/kernel/smp.c +++ b/arch/x86_64/kernel/smp.c @@ -376,9 +376,8 @@ int smp_call_function_single (int cpu, void (*func) (void *info), void *info, /* prevent preemption and reschedule on another processor */ int me = get_cpu(); if (cpu == me) { - WARN_ON(1); put_cpu(); - return -EBUSY; + return 0; } spin_lock_bh(&call_lock); __smp_call_function_single(cpu, func, info, nonatomic, wait); diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c index 88722f11ca1..e3ef544d2cf 100644 --- a/arch/x86_64/kernel/time.c +++ b/arch/x86_64/kernel/time.c @@ -876,15 +876,6 @@ static struct irqaction irq0 = { timer_interrupt, IRQF_DISABLED, CPU_MASK_NONE, "timer", NULL, NULL }; -static int __cpuinit -time_cpu_notifier(struct notifier_block *nb, unsigned long action, void *hcpu) -{ - unsigned cpu = (unsigned long) hcpu; - if (action == CPU_ONLINE) - vsyscall_set_cpu(cpu); - return NOTIFY_DONE; -} - void __init time_init(void) { if (nohpet) @@ -925,8 +916,6 @@ void __init time_init(void) vxtime.last_tsc = get_cycles_sync(); set_cyc2ns_scale(cpu_khz); setup_irq(0, &irq0); - hotcpu_notifier(time_cpu_notifier, 0); - time_cpu_notifier(NULL, CPU_ONLINE, (void *)(long)smp_processor_id()); #ifndef CONFIG_SMP time_init_gtod(); diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c index a98b460af6a..a730bacecb0 100644 --- a/arch/x86_64/kernel/vsyscall.c +++ b/arch/x86_64/kernel/vsyscall.c @@ -27,6 +27,9 @@ #include #include #include +#include +#include +#include #include #include @@ -243,32 +246,17 @@ static ctl_table kernel_root_table2[] = { #endif -static void __cpuinit write_rdtscp_cb(void *info) -{ - write_rdtscp_aux((unsigned long)info); -} - -void __cpuinit vsyscall_set_cpu(int cpu) +/* Assume __initcall executes before all user space. Hopefully kmod + doesn't violate that. We'll find out if it does. */ +static void __cpuinit vsyscall_set_cpu(int cpu) { unsigned long *d; unsigned long node = 0; #ifdef CONFIG_NUMA node = cpu_to_node[cpu]; #endif - if (cpu_has(&cpu_data[cpu], X86_FEATURE_RDTSCP)) { - void *info = (void *)((node << 12) | cpu); - /* Can happen on preemptive kernel */ - if (get_cpu() == cpu) - write_rdtscp_cb(info); -#ifdef CONFIG_SMP - else { - /* the notifier is unfortunately not executed on the - target CPU */ - smp_call_function_single(cpu,write_rdtscp_cb,info,0,1); - } -#endif - put_cpu(); - } + if (cpu_has(&cpu_data[cpu], X86_FEATURE_RDTSCP)) + write_rdtscp_aux((node << 12) | cpu); /* Store cpu number in limit so that it can be loaded quickly in user space in vgetcpu. @@ -280,6 +268,21 @@ void __cpuinit vsyscall_set_cpu(int cpu) *d |= (node >> 4) << 48; } +static void __cpuinit cpu_vsyscall_init(void *arg) +{ + /* preemption should be already off */ + vsyscall_set_cpu(raw_smp_processor_id()); +} + +static int __cpuinit +cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg) +{ + long cpu = (long)arg; + if (action == CPU_ONLINE) + smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 0, 1); + return NOTIFY_DONE; +} + static void __init map_vsyscall(void) { extern char __vsyscall_0; @@ -299,6 +302,8 @@ static int __init vsyscall_init(void) #ifdef CONFIG_SYSCTL register_sysctl_table(kernel_root_table2, 0); #endif + on_each_cpu(cpu_vsyscall_init, NULL, 0, 1); + hotcpu_notifier(cpu_vsyscall_notifier, 0); return 0; } -- cgit v1.2.3 From 9446868b5383eb87f76b2d4389dea4bb968a6657 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 14 Nov 2006 16:57:46 +0100 Subject: [PATCH] x86-64: Fix race in exit_idle When another interrupt happens in exit_idle the exit idle notifier could be called an incorrect number of times. Add a test_and_clear_bit_pda and use it handle the bit atomically against interrupts to avoid this. Pointed out by Stephane Eranian Signed-off-by: Andi Kleen --- arch/x86_64/kernel/process.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 49f7fac6229..f6226055d53 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -88,9 +88,8 @@ void enter_idle(void) static void __exit_idle(void) { - if (read_pda(isidle) == 0) + if (test_and_clear_bit_pda(0, isidle) == 0) return; - write_pda(isidle, 0); atomic_notifier_call_chain(&idle_notifier, IDLE_END, NULL); } -- cgit v1.2.3 From 45c99533252ef2297f37c5fdd672a3e0eb566870 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Tue, 14 Nov 2006 10:52:12 -0700 Subject: [PATCH] Use delayed disable mode of ioapic edge triggered interrupts Komuro reports that ISA interrupts do not work after a disable_irq(), causing some PCMCIA drivers to not work, with messages like eth0: Asix AX88190: io 0x300, irq 3, hw_addr xx:xx:xx:xx:xx:xx eth0: found link beat eth0: autonegotiation complete: 100baseT-FD selected eth0: interrupt(s) dropped! eth0: interrupt(s) dropped! eth0: interrupt(s) dropped! ... Linus Torvalds said: "Now, edge-triggered interrupts are a _lot_ harder to mask, because the Intel APIC is an unbelievable piece of sh*t, and has the edge-detect logic _before_ the mask logic, so if a edge happens _while_ the device is masked, you'll never ever see the edge ever again (unmasking will not cause a new edge, so you simply lost the interrupt). So when you "mask" an edge-triggered IRQ, you can't really mask it at all, because if you did that, you'd lose it forever if the IRQ comes in while you masked it. Instead, we're supposed to leave it active, and set a flag, and IF the IRQ comes in, we just remember it, and mask it at that point instead, and then on unmasking, we have to replay it by sending a self-IPI." This trivial patch solves the problem. Signed-off-by: Eric W. Biederman Cc: Ingo Molnar Acked-by: Komuro Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/io_apic.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 41bfc49301a..14654e68241 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -790,9 +790,11 @@ static void ioapic_register_intr(int irq, int vector, unsigned long trigger) trigger == IOAPIC_LEVEL) set_irq_chip_and_handler_name(irq, &ioapic_chip, handle_fasteoi_irq, "fasteoi"); - else + else { + irq_desc[irq].status |= IRQ_DELAYED_DISABLE; set_irq_chip_and_handler_name(irq, &ioapic_chip, handle_edge_irq, "edge"); + } } static void __init setup_IO_APIC_irqs(void) -- cgit v1.2.3 From 6b3d1a95ba714bfb1cc81362f7f3e01b7654b4f3 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 16 Nov 2006 10:22:03 +0100 Subject: [PATCH] x86-64: Fix vsyscall.c compilation on UP Broken by earlier patch by me. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/vsyscall.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c index a730bacecb0..92546c1526f 100644 --- a/arch/x86_64/kernel/vsyscall.c +++ b/arch/x86_64/kernel/vsyscall.c @@ -274,6 +274,7 @@ static void __cpuinit cpu_vsyscall_init(void *arg) vsyscall_set_cpu(raw_smp_processor_id()); } +#ifdef CONFIG_HOTPLUG_CPU static int __cpuinit cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg) { @@ -282,6 +283,7 @@ cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg) smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 0, 1); return NOTIFY_DONE; } +#endif static void __init map_vsyscall(void) { -- cgit v1.2.3 From ccf9ff524ccb195d648ecb0b168340560b42532c Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 16 Nov 2006 11:49:16 +0100 Subject: [PATCH] x86_64: fix CONFIG_CC_STACKPROTECTOR build bug on x86_64, the CONFIG_CC_STACKPROTECTOR build fails if used in a distcc setup that has "CC" defined to "distcc gcc": gcc: gcc: linker input file unused because linking not done gcc: gcc: linker input file unused because linking not done gcc: gcc: linker input file unused because linking not done this is because the gcc-x86_64-has-stack-protector.sh script has a 2-parameters assumption. Fix this by passing $(CC) as a single parameter. Signed-off-by: Ingo Molnar Please-Use-Me-More: make randconfig Signed-off-by: Linus Torvalds --- arch/x86_64/Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile index 13972148058..6e38d4daeed 100644 --- a/arch/x86_64/Makefile +++ b/arch/x86_64/Makefile @@ -66,8 +66,8 @@ AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) cflags-y += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,) AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,) -cflags-$(CONFIG_CC_STACKPROTECTOR) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh $(CC) -fstack-protector ) -cflags-$(CONFIG_CC_STACKPROTECTOR_ALL) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh $(CC) -fstack-protector-all ) +cflags-$(CONFIG_CC_STACKPROTECTOR) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh "$(CC)" -fstack-protector ) +cflags-$(CONFIG_CC_STACKPROTECTOR_ALL) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh "$(CC)" -fstack-protector-all ) CFLAGS += $(cflags-y) CFLAGS_KERNEL += $(cflags-kernel-y) -- cgit v1.2.3 From 0796bdb7e9e4a48b401f4fba1ee5dc79a45528ef Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 17 Nov 2006 05:57:49 +0100 Subject: [PATCH] x86_64: stack unwinder crash fix the new dwarf2 unwinder crashes while trying to dump the stack: Leftover inexact backtrace: Unable to handle kernel paging request at ffffffff82800000 RIP: [] dump_trace+0x35b/0x3d2 PGD 203027 PUD 205027 PMD 0 Oops: 0000 [2] PREEMPT SMP CPU 0 Modules linked in: Pid: 30, comm: khelper Not tainted 2.6.19-rc6-rt1 #11 RIP: 0010:[] [] dump_trace+0x35b/0x3d2 RSP: 0000:ffff81003fb9d848 EFLAGS: 00010006 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffffffff805b3520 RDI: 0000000000000000 RBP: ffffffff827ffff9 R08: ffffffff80aad000 R09: 0000000000000005 R10: ffffffff80aae000 R11: ffffffff8037961b R12: ffff81003fb9d858 R13: 0000000000000000 R14: ffffffff80598460 R15: ffffffff80ab1fc0 FS: 0000000000000000(0000) GS:ffffffff806c4200(0000) knlGS:0000000000000000 CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b CR2: ffffffff82800000 CR3: 0000000000201000 CR4: 00000000000006e0 this crash happened because it did not sanitize the dwarf2 data it got, and got an unaligned stack pointer - which happily walked past the process stack (and eventually reached the end of kernel memory and pagefaulted there) due to this naive iteration condition: HANDLE_STACK (((long) stack & (THREAD_SIZE-1)) != 0); note that i386 is alot more conservative when it comes to trusting stack pointers: static inline int valid_stack_ptr(struct thread_info *tinfo, void *p) { return p > (void *)tinfo && p < (void *)tinfo + THREAD_SIZE - 3; } but the x86_64 code did not take this bit of i386 code. The fix is to align the stack pointer. Signed-off-by: Ingo Molnar Cc: Andi Kleen Cc: Jan Beulich Cc: Thomas Gleixner Cc: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/traps.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 7819022a8db..a153d0a01b7 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -290,6 +290,12 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s if (tsk && tsk != current) stack = (unsigned long *)tsk->thread.rsp; } + /* + * Align the stack pointer on word boundary, later loops + * rely on that (and corruption / debug info bugs can cause + * unaligned values here): + */ + stack = (unsigned long *)((unsigned long)stack & ~(sizeof(long)-1)); /* * Print function call entries within a stack. 'cond' is the -- cgit v1.2.3 From dc1829a4c378d793fb3b95d56135d89a0d7ff72a Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 17 Nov 2006 14:26:18 +0100 Subject: [PATCH] i386/x86_64: ACPI cpu_idle_wait() fix The scheduler on Andreas Friedrich's hyperthreading system stopped working properly: the scheduler would never move tasks to another CPU! The lask known working kernel was 2.6.8. After a couple of attempts to corner the bug, the following smoking gun was found: BIOS reported wrong ACPI idfor the processor CPU#1: set_cpus_allowed(), swapper:1, 3 -> 2 [] show_trace_log_lvl+0x34/0x4a [] show_trace+0x2c/0x2e [] dump_stack+0x2b/0x2d [] set_cpus_allowed+0x52/0xec [] cpu_idle_wait+0x2e/0x100 [] acpi_processor_power_exit+0x45/0x58 [] acpi_processor_remove+0x46/0xea [] acpi_start_single_object+0x47/0x54 [] acpi_bus_register_driver+0xa4/0xd3 [] acpi_processor_init+0x57/0x77 [] init+0x146/0x2fd [] kernel_thread_helper+0x7/0x10 a quick look at cpu_idle_wait() shows how broken that code is on i386: it changes the init task's affinity map but never restores it ... and because all userspace tasks get forked by init, they all inherited that single-CPU affinity mask. x86_64 cloned this bug too. Signed-off-by: Ingo Molnar Cc: Andreas Friedrich Cc: Wolfgang Erig Cc: Andrew Morton Cc: Adrian Bunk Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/process.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index f6226055d53..7451a4c43c1 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -144,7 +144,7 @@ static void poll_idle (void) void cpu_idle_wait(void) { unsigned int cpu, this_cpu = get_cpu(); - cpumask_t map; + cpumask_t map, tmp = current->cpus_allowed; set_cpus_allowed(current, cpumask_of_cpu(this_cpu)); put_cpu(); @@ -167,6 +167,8 @@ void cpu_idle_wait(void) } cpus_and(map, map, cpu_online_map); } while (!cpus_empty(map)); + + set_cpus_allowed(current, tmp); } EXPORT_SYMBOL_GPL(cpu_idle_wait); -- cgit v1.2.3 From 8243229f0940ab4e9f501879d3ffb7476b02ee6a Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Sat, 18 Nov 2006 22:19:40 -0800 Subject: [PATCH] x86_64: fix memory hotplug build with NUMA=n This is to fix compile error of x86-64 memory hotplug without any NUMA option. CC arch/x86_64/mm/init.o arch/x86_64/mm/init.c:501: error: redefinition of 'memory_add_physaddr_to_nid' include/linux/memory_hotplug.h:71: error: previous definition of 'memory_add_phys addr_to_nid' was here arch/x86_64/mm/init.c:509: error: redefinition of 'memory_add_physaddr_to_nid' arch/x86_64/mm/init.c:501: error: previous definition of 'memory_add_physaddr_to_ nid' was here I confirmed compile completion with !NUMA, (NUMA & !ACPI_NUMA), or (NUMA & ACPI_NUMA). Signed-off-by: Yasunori Goto Acked-by: Andi Kleen Cc: "Randy.Dunlap" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/mm/init.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c index f1f977aafae..4c0c00ef3ca 100644 --- a/arch/x86_64/mm/init.c +++ b/arch/x86_64/mm/init.c @@ -496,7 +496,7 @@ int remove_memory(u64 start, u64 size) } EXPORT_SYMBOL_GPL(remove_memory); -#ifndef CONFIG_ACPI_NUMA +#if !defined(CONFIG_ACPI_NUMA) && defined(CONFIG_NUMA) int memory_add_physaddr_to_nid(u64 start) { return 0; @@ -504,13 +504,6 @@ int memory_add_physaddr_to_nid(u64 start) EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid); #endif -#ifndef CONFIG_ACPI_NUMA -int memory_add_physaddr_to_nid(u64 start) -{ - return 0; -} -#endif - #endif /* CONFIG_MEMORY_HOTPLUG */ #ifdef CONFIG_MEMORY_HOTPLUG_RESERVE -- cgit v1.2.3 From 9a14f2964b459c18198ee59ff7212321def82df7 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Mon, 20 Nov 2006 11:29:09 -0500 Subject: [PATCH] x86_64: Align data segment to PAGE_SIZE boundary o Explicitly align data segment to PAGE_SIZE boundary otherwise depending on config options and tool chain it might be placed on a non PAGE_SIZE aligned boundary and vmlinux loaders like kexec fail when they encounter a PT_LOAD type segment which is not aligned to PAGE_SIZE boundary. Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen --- arch/x86_64/kernel/vmlinux.lds.S | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index edb24aa714b..d9534e750d4 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -60,6 +60,7 @@ SECTIONS } #endif + . = ALIGN(PAGE_SIZE); /* Align data segment to page size boundary */ /* Data */ .data : AT(ADDR(.data) - LOAD_OFFSET) { *(.data) -- cgit v1.2.3 From 3af9815328bba76e8d11d71d6dbbd6f38beafe58 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Mon, 20 Nov 2006 11:29:09 -0500 Subject: [PATCH] x86_64: Align data segment to PAGE_SIZE boundary o Explicitly align data segment to PAGE_SIZE boundary otherwise depending on config options and tool chain it might be placed on a non PAGE_SIZE aligned boundary and vmlinux loaders like kexec fail when they encounter a PT_LOAD type segment which is not aligned to PAGE_SIZE boundary. Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen --- arch/x86_64/kernel/vmlinux.lds.S | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index edb24aa714b..d9534e750d4 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -60,6 +60,7 @@ SECTIONS } #endif + . = ALIGN(PAGE_SIZE); /* Align data segment to page size boundary */ /* Data */ .data : AT(ADDR(.data) - LOAD_OFFSET) { *(.data) -- cgit v1.2.3 From 52bad64d95bd89e08c49ec5a071fa6dcbe5a1a9c Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 22 Nov 2006 14:54:01 +0000 Subject: WorkStruct: Separate delayable and non-delayable events. Separate delayable work items from non-delayable work items be splitting them into a separate structure (delayed_work), which incorporates a work_struct and the timer_list removed from work_struct. The work_struct struct is huge, and this limits it's usefulness. On a 64-bit architecture it's nearly 100 bytes in size. This reduces that by half for the non-delayable type of event. Signed-Off-By: David Howells --- arch/x86_64/kernel/mce.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c index bbea88801d8..5306f263090 100644 --- a/arch/x86_64/kernel/mce.c +++ b/arch/x86_64/kernel/mce.c @@ -307,7 +307,7 @@ void mce_log_therm_throt_event(unsigned int cpu, __u64 status) static int check_interval = 5 * 60; /* 5 minutes */ static void mcheck_timer(void *data); -static DECLARE_WORK(mcheck_work, mcheck_timer, NULL); +static DECLARE_DELAYED_WORK(mcheck_work, mcheck_timer, NULL); static void mcheck_check_cpu(void *info) { -- cgit v1.2.3 From 65f27f38446e1976cc98fd3004b110fedcddd189 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 22 Nov 2006 14:55:48 +0000 Subject: WorkStruct: Pass the work_struct pointer instead of context data Pass the work_struct pointer to the work function rather than context data. The work function can use container_of() to work out the data. For the cases where the container of the work_struct may go away the moment the pending bit is cleared, it is made possible to defer the release of the structure by deferring the clearing of the pending bit. To make this work, an extra flag is introduced into the management side of the work_struct. This governs auto-release of the structure upon execution. Ordinarily, the work queue executor would release the work_struct for further scheduling or deallocation by clearing the pending bit prior to jumping to the work function. This means that, unless the driver makes some guarantee itself that the work_struct won't go away, the work function may not access anything else in the work_struct or its container lest they be deallocated.. This is a problem if the auxiliary data is taken away (as done by the last patch). However, if the pending bit is *not* cleared before jumping to the work function, then the work function *may* access the work_struct and its container with no problems. But then the work function must itself release the work_struct by calling work_release(). In most cases, automatic release is fine, so this is the default. Special initiators exist for the non-auto-release case (ending in _NAR). Signed-Off-By: David Howells --- arch/x86_64/kernel/mce.c | 6 +++--- arch/x86_64/kernel/smpboot.c | 12 +++++++----- arch/x86_64/kernel/time.c | 4 ++-- 3 files changed, 12 insertions(+), 10 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c index 5306f263090..c7587fc3901 100644 --- a/arch/x86_64/kernel/mce.c +++ b/arch/x86_64/kernel/mce.c @@ -306,8 +306,8 @@ void mce_log_therm_throt_event(unsigned int cpu, __u64 status) */ static int check_interval = 5 * 60; /* 5 minutes */ -static void mcheck_timer(void *data); -static DECLARE_DELAYED_WORK(mcheck_work, mcheck_timer, NULL); +static void mcheck_timer(struct work_struct *work); +static DECLARE_DELAYED_WORK(mcheck_work, mcheck_timer); static void mcheck_check_cpu(void *info) { @@ -315,7 +315,7 @@ static void mcheck_check_cpu(void *info) do_machine_check(NULL, 0); } -static void mcheck_timer(void *data) +static void mcheck_timer(struct work_struct *work) { on_each_cpu(mcheck_check_cpu, NULL, 1, 1); schedule_delayed_work(&mcheck_work, check_interval * HZ); diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c index 62c2e747af5..9800147c4c6 100644 --- a/arch/x86_64/kernel/smpboot.c +++ b/arch/x86_64/kernel/smpboot.c @@ -753,14 +753,16 @@ static int __cpuinit wakeup_secondary_via_INIT(int phys_apicid, unsigned int sta } struct create_idle { + struct work_struct work; struct task_struct *idle; struct completion done; int cpu; }; -void do_fork_idle(void *_c_idle) +void do_fork_idle(struct work_struct *work) { - struct create_idle *c_idle = _c_idle; + struct create_idle *c_idle = + container_of(work, struct create_idle, work); c_idle->idle = fork_idle(c_idle->cpu); complete(&c_idle->done); @@ -775,10 +777,10 @@ static int __cpuinit do_boot_cpu(int cpu, int apicid) int timeout; unsigned long start_rip; struct create_idle c_idle = { + .work = __WORK_INITIALIZER(c_idle.work, do_fork_idle), .cpu = cpu, .done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done), }; - DECLARE_WORK(work, do_fork_idle, &c_idle); /* allocate memory for gdts of secondary cpus. Hotplug is considered */ if (!cpu_gdt_descr[cpu].address && @@ -825,9 +827,9 @@ static int __cpuinit do_boot_cpu(int cpu, int apicid) * thread. */ if (!keventd_up() || current_is_keventd()) - work.func(work.data); + c_idle.work.func(&c_idle.work); else { - schedule_work(&work); + schedule_work(&c_idle.work); wait_for_completion(&c_idle.done); } diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c index e3ef544d2cf..9f05bc9b2da 100644 --- a/arch/x86_64/kernel/time.c +++ b/arch/x86_64/kernel/time.c @@ -563,7 +563,7 @@ static unsigned int cpufreq_delayed_issched = 0; static unsigned int cpufreq_init = 0; static struct work_struct cpufreq_delayed_get_work; -static void handle_cpufreq_delayed_get(void *v) +static void handle_cpufreq_delayed_get(struct work_struct *v) { unsigned int cpu; for_each_online_cpu(cpu) { @@ -639,7 +639,7 @@ static struct notifier_block time_cpufreq_notifier_block = { static int __init cpufreq_tsc(void) { - INIT_WORK(&cpufreq_delayed_get_work, handle_cpufreq_delayed_get, NULL); + INIT_WORK(&cpufreq_delayed_get_work, handle_cpufreq_delayed_get); if (!cpufreq_register_notifier(&time_cpufreq_notifier_block, CPUFREQ_TRANSITION_NOTIFIER)) cpufreq_init = 1; -- cgit v1.2.3 From 24d7bb3396c51ceb2285e0e7b0c1bd1865652c43 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 28 Nov 2006 09:14:05 +0100 Subject: [PATCH] x86_64: fix 'earlyprintk=...,keep' regression Commit 2c8c0e6b8d7700a990da8d24eff767f9ca223b96 ("[PATCH] Convert x86-64 to early param") broke the earlyprintk=...,keep feature. This restores that functionality. Tested on x86_64. Must-have for v2.6.19, no risk. Signed-off-by: Ingo Molnar Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/early_printk.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/early_printk.c b/arch/x86_64/kernel/early_printk.c index e22ecd54870..47b6d90349d 100644 --- a/arch/x86_64/kernel/early_printk.c +++ b/arch/x86_64/kernel/early_printk.c @@ -224,7 +224,7 @@ static int __init setup_early_printk(char *buf) return 0; early_console_initialized = 1; - if (!strcmp(buf,"keep")) + if (strstr(buf, "keep")) keep_early = 1; if (!strncmp(buf, "serial", 6)) { -- cgit v1.2.3 From f7a23328a738b45124400d85eaf78a76939da726 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 28 Nov 2006 20:12:59 +0100 Subject: [PATCH] x86-64: Fix warning in io_apic.c --- arch/x86_64/kernel/io_apic.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 14654e68241..c80081a6ba4 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -754,10 +754,8 @@ void __setup_vector_irq(int cpu) { /* Initialize vector_irq on a new cpu */ /* This function must be called with vector_lock held */ - unsigned long flags; int irq, vector; - /* Mark the inuse vectors */ for (irq = 0; irq < NR_IRQ_VECTORS; ++irq) { if (!cpu_isset(cpu, irq_domain[irq])) -- cgit v1.2.3 From c547c77ee4d0408907847f64c403df1bf2f9c7a0 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 28 Nov 2006 20:12:59 +0100 Subject: [PATCH] x86-64: Use stricter in process stack check for unwinder Previously it would check for alignment only, which could break if the stack pointer was unaligned. Now explicitely check if the stack pointer is in the stack page of the current process. Ported from i386. Signed-off-by: Andi Kleen --- arch/x86_64/kernel/traps.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index a153d0a01b7..0d65b22f229 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -242,12 +242,19 @@ static int dump_trace_unwind(struct unwind_frame_info *info, void *context) * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack */ +static inline int valid_stack_ptr(struct thread_info *tinfo, void *p) +{ + void *t = (void *)tinfo; + return p > t && p < t + THREAD_SIZE - 3; +} + void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * stack, struct stacktrace_ops *ops, void *data) { const unsigned cpu = smp_processor_id(); unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr; unsigned used = 0; + struct thread_info *tinfo; if (!tsk) tsk = current; @@ -370,7 +377,8 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s /* * This handles the process stack: */ - HANDLE_STACK (((long) stack & (THREAD_SIZE-1)) != 0); + tinfo = current_thread_info(); + HANDLE_STACK (valid_stack_ptr(tinfo, stack)); #undef HANDLE_STACK } EXPORT_SYMBOL(dump_trace); -- cgit v1.2.3 From a4f89fb7c072b8592b296c2ba216269c0c96db43 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:20:08 -0800 Subject: [NET]: X86_64 checksum annotations and cleanups. * sanitize prototypes, annotate * usual ntohs->shift Signed-off-by: Al Viro Signed-off-by: David S. Miller --- arch/x86_64/lib/csum-partial.c | 7 ++++--- arch/x86_64/lib/csum-wrappers.c | 37 +++++++++++++++++++++---------------- 2 files changed, 25 insertions(+), 19 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/lib/csum-partial.c b/arch/x86_64/lib/csum-partial.c index c493735218d..06ae630de82 100644 --- a/arch/x86_64/lib/csum-partial.c +++ b/arch/x86_64/lib/csum-partial.c @@ -132,9 +132,10 @@ static __force_inline unsigned do_csum(const unsigned char *buff, unsigned len) * * it's best to have buff aligned on a 64-bit boundary */ -unsigned csum_partial(const unsigned char *buff, unsigned len, unsigned sum) +__wsum csum_partial(const void *buff, int len, __wsum sum) { - return add32_with_carry(do_csum(buff, len), sum); + return (__force __wsum)add32_with_carry(do_csum(buff, len), + (__force u32)sum); } EXPORT_SYMBOL(csum_partial); @@ -143,7 +144,7 @@ EXPORT_SYMBOL(csum_partial); * this routine is used for miscellaneous IP-like checksums, mainly * in icmp.c */ -unsigned short ip_compute_csum(unsigned char * buff, int len) +__sum16 ip_compute_csum(const void *buff, int len) { return csum_fold(csum_partial(buff,len,0)); } diff --git a/arch/x86_64/lib/csum-wrappers.c b/arch/x86_64/lib/csum-wrappers.c index b1320ec5842..fd42a4a095f 100644 --- a/arch/x86_64/lib/csum-wrappers.c +++ b/arch/x86_64/lib/csum-wrappers.c @@ -18,9 +18,9 @@ * Returns an 32bit unfolded checksum of the buffer. * src and dst are best aligned to 64bits. */ -unsigned int -csum_partial_copy_from_user(const unsigned char __user *src, unsigned char *dst, - int len, unsigned int isum, int *errp) +__wsum +csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum isum, int *errp) { might_sleep(); *errp = 0; @@ -34,17 +34,19 @@ csum_partial_copy_from_user(const unsigned char __user *src, unsigned char *dst, if (unlikely((unsigned long)src & 6)) { while (((unsigned long)src & 6) && len >= 2) { __u16 val16; - *errp = __get_user(val16, (__u16 __user *)src); + *errp = __get_user(val16, (const __u16 __user *)src); if (*errp) return isum; *(__u16 *)dst = val16; - isum = add32_with_carry(isum, val16); + isum = (__force __wsum)add32_with_carry( + (__force unsigned)isum, val16); src += 2; dst += 2; len -= 2; } } - isum = csum_partial_copy_generic((__force void *)src,dst,len,isum,errp,NULL); + isum = csum_partial_copy_generic((__force const void *)src, + dst, len, isum, errp, NULL); if (likely(*errp == 0)) return isum; } @@ -66,9 +68,9 @@ EXPORT_SYMBOL(csum_partial_copy_from_user); * Returns an 32bit unfolded checksum of the buffer. * src and dst are best aligned to 64bits. */ -unsigned int -csum_partial_copy_to_user(unsigned const char *src, unsigned char __user *dst, - int len, unsigned int isum, int *errp) +__wsum +csum_partial_copy_to_user(const void *src, void __user *dst, + int len, __wsum isum, int *errp) { might_sleep(); if (unlikely(!access_ok(VERIFY_WRITE, dst, len))) { @@ -79,7 +81,8 @@ csum_partial_copy_to_user(unsigned const char *src, unsigned char __user *dst, if (unlikely((unsigned long)dst & 6)) { while (((unsigned long)dst & 6) && len >= 2) { __u16 val16 = *(__u16 *)src; - isum = add32_with_carry(isum, val16); + isum = (__force __wsum)add32_with_carry( + (__force unsigned)isum, val16); *errp = __put_user(val16, (__u16 __user *)dst); if (*errp) return isum; @@ -104,19 +107,21 @@ EXPORT_SYMBOL(csum_partial_copy_to_user); * * Returns an 32bit unfolded checksum of the buffer. */ -unsigned int -csum_partial_copy_nocheck(const unsigned char *src, unsigned char *dst, int len, unsigned int sum) +__wsum +csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum) { return csum_partial_copy_generic(src,dst,len,sum,NULL,NULL); } EXPORT_SYMBOL(csum_partial_copy_nocheck); -unsigned short csum_ipv6_magic(struct in6_addr *saddr, struct in6_addr *daddr, - __u32 len, unsigned short proto, unsigned int sum) +__sum16 csum_ipv6_magic(const struct in6_addr *saddr, + const struct in6_addr *daddr, + __u32 len, unsigned short proto, __wsum sum) { __u64 rest, sum64; - rest = (__u64)htonl(len) + (__u64)htons(proto) + (__u64)sum; + rest = (__force __u64)htonl(len) + (__force __u64)htons(proto) + + (__force __u64)sum; asm(" addq (%[saddr]),%[sum]\n" " adcq 8(%[saddr]),%[sum]\n" " adcq (%[daddr]),%[sum]\n" @@ -124,7 +129,7 @@ unsigned short csum_ipv6_magic(struct in6_addr *saddr, struct in6_addr *daddr, " adcq $0,%[sum]\n" : [sum] "=r" (sum64) : "[sum]" (rest),[saddr] "r" (saddr), [daddr] "r" (daddr)); - return csum_fold(add32_with_carry(sum64 & 0xffffffff, sum64>>32)); + return csum_fold((__force __wsum)add32_with_carry(sum64 & 0xffffffff, sum64>>32)); } EXPORT_SYMBOL(csum_ipv6_magic); -- cgit v1.2.3 From 8f820e976057261e249367514e9920cf20048c76 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:00 +0100 Subject: [PATCH] x86-64: Update defconfig Signed-off-by: Andi Kleen --- arch/x86_64/defconfig | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/defconfig b/arch/x86_64/defconfig index 0f5d44e86be..96f226cfb33 100644 --- a/arch/x86_64/defconfig +++ b/arch/x86_64/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.19-rc2-git4 -# Sat Oct 21 03:38:52 2006 +# Linux kernel version: 2.6.19-git7 +# Wed Dec 6 23:50:47 2006 # CONFIG_X86_64=y CONFIG_64BIT=y @@ -47,13 +47,14 @@ CONFIG_POSIX_MQUEUE=y CONFIG_IKCONFIG=y CONFIG_IKCONFIG_PROC=y # CONFIG_CPUSETS is not set +CONFIG_SYSFS_DEPRECATED=y # CONFIG_RELAY is not set CONFIG_INITRAMFS_SOURCE="" CONFIG_CC_OPTIMIZE_FOR_SIZE=y CONFIG_SYSCTL=y # CONFIG_EMBEDDED is not set CONFIG_UID16=y -# CONFIG_SYSCTL_SYSCALL is not set +CONFIG_SYSCTL_SYSCALL=y CONFIG_KALLSYMS=y CONFIG_KALLSYMS_ALL=y # CONFIG_KALLSYMS_EXTRA_PASS is not set @@ -87,9 +88,7 @@ CONFIG_STOP_MACHINE=y # Block layer # CONFIG_BLOCK=y -CONFIG_LBD=y # CONFIG_BLK_DEV_IO_TRACE is not set -# CONFIG_LSF is not set # # IO Schedulers @@ -111,10 +110,11 @@ CONFIG_X86_PC=y # CONFIG_X86_VSMP is not set # CONFIG_MK8 is not set # CONFIG_MPSC is not set -CONFIG_GENERIC_CPU=y -CONFIG_X86_L1_CACHE_BYTES=128 -CONFIG_X86_L1_CACHE_SHIFT=7 -CONFIG_X86_INTERNODE_CACHE_BYTES=128 +CONFIG_MCORE2=y +# CONFIG_GENERIC_CPU is not set +CONFIG_X86_L1_CACHE_BYTES=64 +CONFIG_X86_L1_CACHE_SHIFT=6 +CONFIG_X86_INTERNODE_CACHE_BYTES=64 CONFIG_X86_TSC=y CONFIG_X86_GOOD_APIC=y # CONFIG_MICROCODE is not set @@ -322,6 +322,7 @@ CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set CONFIG_TCP_CONG_CUBIC=y CONFIG_DEFAULT_TCP_CONG="cubic" +# CONFIG_TCP_MD5SIG is not set CONFIG_IPV6=y # CONFIG_IPV6_PRIVACY is not set # CONFIG_IPV6_ROUTER_PREF is not set @@ -624,6 +625,7 @@ CONFIG_SATA_INTEL_COMBINED=y # CONFIG_PATA_IT821X is not set # CONFIG_PATA_JMICRON is not set # CONFIG_PATA_TRIFLEX is not set +# CONFIG_PATA_MARVELL is not set # CONFIG_PATA_MPIIX is not set # CONFIG_PATA_OLDPIIX is not set # CONFIG_PATA_NETCELL is not set @@ -795,6 +797,7 @@ CONFIG_BNX2=y CONFIG_S2IO=m # CONFIG_S2IO_NAPI is not set # CONFIG_MYRI10GE is not set +# CONFIG_NETXEN_NIC is not set # # Token Ring devices @@ -927,10 +930,6 @@ CONFIG_RTC=y # CONFIG_DTLK is not set # CONFIG_R3964 is not set # CONFIG_APPLICOM is not set - -# -# Ftape, the floppy tape device driver -# CONFIG_AGP=y CONFIG_AGP_AMD64=y CONFIG_AGP_INTEL=y @@ -1135,6 +1134,7 @@ CONFIG_USB_DEVICEFS=y # CONFIG_USB_BANDWIDTH is not set # CONFIG_USB_DYNAMIC_MINORS is not set # CONFIG_USB_SUSPEND is not set +# CONFIG_USB_MULTITHREAD_PROBE is not set # CONFIG_USB_OTG is not set # @@ -1212,6 +1212,7 @@ CONFIG_USB_HIDINPUT=y # CONFIG_USB_KAWETH is not set # CONFIG_USB_PEGASUS is not set # CONFIG_USB_RTL8150 is not set +# CONFIG_USB_USBNET_MII is not set # CONFIG_USB_USBNET is not set CONFIG_USB_MON=y -- cgit v1.2.3 From b615ebdac97c648a2ae7d23c5a0bbb3972adf928 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:00 +0100 Subject: [PATCH] x86: shorten lines in unwinder to be <= 80 characters Andrew complained about > 80 character lines in the new unwinder. Fix that. Signed-off-by: Andi Kleen --- arch/x86_64/kernel/traps.c | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 0d65b22f229..d3f43c958ca 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -235,6 +235,8 @@ static int dump_trace_unwind(struct unwind_frame_info *info, void *context) return n; } +#define MSG(txt) ops->warning(data, txt) + /* * x86-64 can have upto three kernel stacks: * process stack @@ -248,11 +250,12 @@ static inline int valid_stack_ptr(struct thread_info *tinfo, void *p) return p > t && p < t + THREAD_SIZE - 3; } -void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * stack, +void dump_trace(struct task_struct *tsk, struct pt_regs *regs, + unsigned long *stack, struct stacktrace_ops *ops, void *data) { const unsigned cpu = smp_processor_id(); - unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr; + unsigned long *irqstack_end = (unsigned long*)cpu_pda(cpu)->irqstackptr; unsigned used = 0; struct thread_info *tinfo; @@ -268,28 +271,30 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s if (unwind_init_frame_info(&info, tsk, regs) == 0) unw_ret = dump_trace_unwind(&info, &oad); } else if (tsk == current) - unw_ret = unwind_init_running(&info, dump_trace_unwind, &oad); + unw_ret = unwind_init_running(&info, dump_trace_unwind, + &oad); else { if (unwind_init_blocked(&info, tsk) == 0) unw_ret = dump_trace_unwind(&info, &oad); } if (unw_ret > 0) { if (call_trace == 1 && !arch_unw_user_mode(&info)) { - ops->warning_symbol(data, "DWARF2 unwinder stuck at %s\n", + ops->warning_symbol(data, + "DWARF2 unwinder stuck at %s\n", UNW_PC(&info)); if ((long)UNW_SP(&info) < 0) { - ops->warning(data, "Leftover inexact backtrace:\n"); + MSG("Leftover inexact backtrace:"); stack = (unsigned long *)UNW_SP(&info); if (!stack) return; } else - ops->warning(data, "Full inexact backtrace again:\n"); + MSG("Full inexact backtrace again:\n"); } else if (call_trace >= 1) return; else - ops->warning(data, "Full inexact backtrace again:\n"); + MSG("Full inexact backtrace again:\n"); } else - ops->warning(data, "Inexact backtrace:\n"); + MSG("Inexact backtrace:\n"); } if (!stack) { unsigned long dummy; -- cgit v1.2.3 From dd315df1767cf56bd4fb8d730fdff4a3d7e15d84 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:00 +0100 Subject: [PATCH] x86: Compress stack unwinder output The unwinder has some extra newlines, which eat up loads of screen space when it spews. (See https://bugzilla.redhat.com/bugzilla/attachment.cgi?id=137900 for a nasty example). warning_symbol-> and warning-> already printk a newline, so don't add one in the strings passed to them. [AK: redone for new code] Signed-off-by: Dave Jones Signed-off-by: Andi Kleen --- arch/x86_64/kernel/traps.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index d3f43c958ca..eedd4e759c3 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -280,7 +280,7 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, if (unw_ret > 0) { if (call_trace == 1 && !arch_unw_user_mode(&info)) { ops->warning_symbol(data, - "DWARF2 unwinder stuck at %s\n", + "DWARF2 unwinder stuck at %s", UNW_PC(&info)); if ((long)UNW_SP(&info) < 0) { MSG("Leftover inexact backtrace:"); @@ -288,13 +288,13 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, if (!stack) return; } else - MSG("Full inexact backtrace again:\n"); + MSG("Full inexact backtrace again:"); } else if (call_trace >= 1) return; else - MSG("Full inexact backtrace again:\n"); + MSG("Full inexact backtrace again:"); } else - MSG("Inexact backtrace:\n"); + MSG("Inexact backtrace:"); } if (!stack) { unsigned long dummy; -- cgit v1.2.3 From 36b2a8d5aff4cb3ee83d5e40447a8f073bcfe2fb Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Thu, 7 Dec 2006 02:14:01 +0100 Subject: [PATCH] x86-64: add X86_FEATURE_PEBS and detection Here is a patch (used by perfmon2) to detect the presence of the Precise Event Based Sampling (PEBS) feature for Intel 64-bit processors. The patch also adds the cpu_has_pebs macro. changelog: - adds X86_FEATURE_PEBS - adds cpu_has_pebs to test for X86_FEATURE_PEBS Signed-off-by: stephane eranian Signed-off-by: Andi Kleen --- arch/x86_64/kernel/setup.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index fc944b5e8f4..619af2e2fa2 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -835,6 +835,13 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) set_bit(X86_FEATURE_ARCH_PERFMON, &c->x86_capability); } + if (cpu_has_ds) { + unsigned int l1, l2; + rdmsr(MSR_IA32_MISC_ENABLE, l1, l2); + if (!(l1 & (1<<12))) + set_bit(X86_FEATURE_PEBS, c->x86_capability); + } + n = c->extended_cpuid_level; if (n >= 0x80000008) { unsigned eax = cpuid_eax(0x80000008); -- cgit v1.2.3 From e2764a1e306c986053a52b33748c33463cf888de Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 7 Dec 2006 02:14:01 +0100 Subject: [PATCH] x86-64: use BUILD_BUG_ON in FPU code Signed-off-by: Alexey Dobriyan Signed-off-by: Andi Kleen Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/x86_64/kernel/i387.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/i387.c b/arch/x86_64/kernel/i387.c index 3aa1e9bb781..1d58c13bc6b 100644 --- a/arch/x86_64/kernel/i387.c +++ b/arch/x86_64/kernel/i387.c @@ -82,11 +82,8 @@ int save_i387(struct _fpstate __user *buf) struct task_struct *tsk = current; int err = 0; - { - extern void bad_user_i387_struct(void); - if (sizeof(struct user_i387_struct) != sizeof(tsk->thread.i387.fxsave)) - bad_user_i387_struct(); - } + BUILD_BUG_ON(sizeof(struct user_i387_struct) != + sizeof(tsk->thread.i387.fxsave)); if ((unsigned long)buf % 16) printk("save_i387: bad fpstate %p\n",buf); -- cgit v1.2.3 From bb81a09e55eaf7e5f798468ab971469b6f66a259 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 7 Dec 2006 02:14:01 +0100 Subject: [PATCH] x86: all cpu backtrace When a spinlock lockup occurs, arrange for the NMI code to emit an all-cpu backtrace, so we get to see which CPU is holding the lock, and where. Cc: Andi Kleen Cc: Ingo Molnar Cc: Badari Pulavarty Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/x86_64/kernel/nmi.c | 29 ++++++++++++++++++++++++++++- 1 file changed, 28 insertions(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c index 7af9cb3e2d9..27e95e7922c 100644 --- a/arch/x86_64/kernel/nmi.c +++ b/arch/x86_64/kernel/nmi.c @@ -12,14 +12,15 @@ * Mikael Pettersson : PM converted to driver model. Disable/enable API. */ +#include #include #include #include #include #include -#include #include #include +#include #include #include @@ -41,6 +42,8 @@ int panic_on_unrecovered_nmi; static DEFINE_PER_CPU(unsigned, perfctr_nmi_owner); static DEFINE_PER_CPU(unsigned, evntsel_nmi_owner[2]); +static cpumask_t backtrace_mask = CPU_MASK_NONE; + /* this number is calculated from Intel's MSR_P4_CRU_ESCR5 register and it's * offset from MSR_P4_BSU_ESCR0. It will be the max for all platforms (for now) */ @@ -782,6 +785,7 @@ int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason) { int sum; int touched = 0; + int cpu = smp_processor_id(); struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); u64 dummy; int rc=0; @@ -799,6 +803,16 @@ int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason) touched = 1; } + if (cpu_isset(cpu, backtrace_mask)) { + static DEFINE_SPINLOCK(lock); /* Serialise the printks */ + + spin_lock(&lock); + printk("NMI backtrace for cpu %d\n", cpu); + dump_stack(); + spin_unlock(&lock); + cpu_clear(cpu, backtrace_mask); + } + #ifdef CONFIG_X86_MCE /* Could check oops_in_progress here too, but it's safer not too */ @@ -931,6 +945,19 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file, #endif +void __trigger_all_cpu_backtrace(void) +{ + int i; + + backtrace_mask = cpu_online_map; + /* Wait for up to 10 seconds for all CPUs to do the backtrace */ + for (i = 0; i < 10 * 1000; i++) { + if (cpus_empty(backtrace_mask)) + break; + mdelay(1); + } +} + EXPORT_SYMBOL(nmi_active); EXPORT_SYMBOL(nmi_watchdog); EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi); -- cgit v1.2.3 From 399287229c775a8962a852a761d65dc9475dec7c Mon Sep 17 00:00:00 2001 From: Aaron Durbin Date: Thu, 7 Dec 2006 02:14:01 +0100 Subject: [PATCH] x86-64: Insert Local and IO APIC(s) into resource map Insert the Local APIC and IO APIC(s) into the resource tree. It allows the APIC resources to be visible within /proc/iomem. The patch also takes into account IO APIC(s) mapped in the PCI space by deferring the insertion until after PCI has allocated its necessary resources. Signed-off-by: Aaron Durbin Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton --- arch/x86_64/kernel/apic.c | 78 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 78 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c index 4d9d5ed942b..5c468971e64 100644 --- a/arch/x86_64/kernel/apic.c +++ b/arch/x86_64/kernel/apic.c @@ -25,6 +25,7 @@ #include #include #include +#include #include #include @@ -45,6 +46,12 @@ int apic_calibrate_pmtmr __initdata; int disable_apic_timer __initdata; +static struct resource *ioapic_resources; +static struct resource lapic_resource = { + .name = "Local APIC", + .flags = IORESOURCE_MEM | IORESOURCE_BUSY, +}; + /* * cpu_mask that denotes the CPUs that needs timer interrupt coming in as * IPIs in place of local APIC timers @@ -585,6 +592,64 @@ static int __init detect_init_APIC (void) return 0; } +#ifdef CONFIG_X86_IO_APIC +static struct resource * __init ioapic_setup_resources(void) +{ +#define IOAPIC_RESOURCE_NAME_SIZE 11 + unsigned long n; + struct resource *res; + char *mem; + int i; + + if (nr_ioapics <= 0) + return NULL; + + n = IOAPIC_RESOURCE_NAME_SIZE + sizeof(struct resource); + n *= nr_ioapics; + + mem = alloc_bootmem(n); + res = (void *)mem; + + if (mem != NULL) { + memset(mem, 0, n); + mem += sizeof(struct resource) * nr_ioapics; + + for (i = 0; i < nr_ioapics; i++) { + res[i].name = mem; + res[i].flags = IORESOURCE_MEM | IORESOURCE_BUSY; + sprintf(mem, "IOAPIC %u", i); + mem += IOAPIC_RESOURCE_NAME_SIZE; + } + } + + ioapic_resources = res; + + return res; +} + +static int __init ioapic_insert_resources(void) +{ + int i; + struct resource *r = ioapic_resources; + + if (!r) { + printk("IO APIC resources could be not be allocated.\n"); + return -1; + } + + for (i = 0; i < nr_ioapics; i++) { + insert_resource(&iomem_resource, r); + r++; + } + + return 0; +} + +/* Insert the IO APIC resources after PCI initialization has occured to handle + * IO APICS that are mapped in on a BAR in PCI space. */ +late_initcall(ioapic_insert_resources); +#endif + void __init init_apic_mappings(void) { unsigned long apic_phys; @@ -604,6 +669,11 @@ void __init init_apic_mappings(void) apic_mapped = 1; apic_printk(APIC_VERBOSE,"mapped APIC to %16lx (%16lx)\n", APIC_BASE, apic_phys); + /* Put local APIC into the resource map. */ + lapic_resource.start = apic_phys; + lapic_resource.end = lapic_resource.start + PAGE_SIZE - 1; + insert_resource(&iomem_resource, &lapic_resource); + /* * Fetch the APIC ID of the BSP in case we have a * default configuration (or the MP table is broken). @@ -613,7 +683,9 @@ void __init init_apic_mappings(void) { unsigned long ioapic_phys, idx = FIX_IO_APIC_BASE_0; int i; + struct resource *ioapic_res; + ioapic_res = ioapic_setup_resources(); for (i = 0; i < nr_ioapics; i++) { if (smp_found_config) { ioapic_phys = mp_ioapics[i].mpc_apicaddr; @@ -625,6 +697,12 @@ void __init init_apic_mappings(void) apic_printk(APIC_VERBOSE,"mapped IOAPIC to %016lx (%016lx)\n", __fix_to_virt(idx), ioapic_phys); idx++; + + if (ioapic_res != NULL) { + ioapic_res->start = ioapic_phys; + ioapic_res->end = ioapic_phys + (4 * 1024) - 1; + ioapic_res++; + } } } } -- cgit v1.2.3 From da68933e0a999fb13636653c710cca701b457ad2 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 7 Dec 2006 02:14:02 +0100 Subject: [PATCH] x86-64: dump_trace() atomicity fix Fix BUG: using smp_processor_id() in preemptible [00000001] code: in backtracer on preemptible debug kernels. Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/x86_64/kernel/traps.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index eedd4e759c3..e37b4d77d5a 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -254,7 +254,7 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long *stack, struct stacktrace_ops *ops, void *data) { - const unsigned cpu = smp_processor_id(); + const unsigned cpu = get_cpu(); unsigned long *irqstack_end = (unsigned long*)cpu_pda(cpu)->irqstackptr; unsigned used = 0; struct thread_info *tinfo; @@ -286,11 +286,11 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, MSG("Leftover inexact backtrace:"); stack = (unsigned long *)UNW_SP(&info); if (!stack) - return; + goto out; } else MSG("Full inexact backtrace again:"); } else if (call_trace >= 1) - return; + goto out; else MSG("Full inexact backtrace again:"); } else @@ -385,6 +385,8 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, tinfo = current_thread_info(); HANDLE_STACK (valid_stack_ptr(tinfo, stack)); #undef HANDLE_STACK +out: + put_cpu(); } EXPORT_SYMBOL(dump_trace); -- cgit v1.2.3 From bcddc0155f351ab3f06c6ede6d91fd399ef9e18f Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 7 Dec 2006 02:14:02 +0100 Subject: [PATCH] x86-64: miscellaneous entry.S adjustments This patch: - makes ret_from_sys_call no longer global (all external users were previously switched to use int_ret_from_sys_call) - adjusts placement of a CFI_{REMEMBER,RESTORE}_STATE pair to better fit logic flow - eliminates an unnecessary pair of CFI_{REMEMBER,RESTORE}_STATE - glues together function- and unwinder-wise the previously separate system_call and int_ret_from_sys_call function fragments Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen --- arch/x86_64/kernel/entry.S | 36 ++++++------------------------------ 1 file changed, 6 insertions(+), 30 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index 7d401b00d82..601d332c4b7 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -230,7 +230,6 @@ ENTRY(system_call) CFI_REL_OFFSET rip,RIP-ARGOFFSET GET_THREAD_INFO(%rcx) testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%rcx) - CFI_REMEMBER_STATE jnz tracesys cmpq $__NR_syscall_max,%rax ja badsys @@ -241,7 +240,6 @@ ENTRY(system_call) * Syscall return path ending with SYSRET (fast path) * Has incomplete stack frame and undefined top of stack. */ - .globl ret_from_sys_call ret_from_sys_call: movl $_TIF_ALLWORK_MASK,%edi /* edi: flagmask */ @@ -251,8 +249,8 @@ sysret_check: TRACE_IRQS_OFF movl threadinfo_flags(%rcx),%edx andl %edi,%edx - CFI_REMEMBER_STATE jnz sysret_careful + CFI_REMEMBER_STATE /* * sysretq will re-enable interrupts: */ @@ -265,10 +263,10 @@ sysret_check: swapgs sysretq + CFI_RESTORE_STATE /* Handle reschedules */ /* edx: work, edi: workmask */ sysret_careful: - CFI_RESTORE_STATE bt $TIF_NEED_RESCHED,%edx jnc sysret_signal TRACE_IRQS_ON @@ -306,7 +304,6 @@ badsys: /* Do syscall tracing */ tracesys: - CFI_RESTORE_STATE SAVE_REST movq $-ENOSYS,RAX(%rsp) FIXUP_TOP_OF_STACK %rdi @@ -322,32 +319,13 @@ tracesys: call *sys_call_table(,%rax,8) 1: movq %rax,RAX-ARGOFFSET(%rsp) /* Use IRET because user could have changed frame */ - jmp int_ret_from_sys_call - CFI_ENDPROC -END(system_call) /* * Syscall return path ending with IRET. * Has correct top of stack, but partial stack frame. - */ -ENTRY(int_ret_from_sys_call) - CFI_STARTPROC simple - CFI_SIGNAL_FRAME - CFI_DEF_CFA rsp,SS+8-ARGOFFSET - /*CFI_REL_OFFSET ss,SS-ARGOFFSET*/ - CFI_REL_OFFSET rsp,RSP-ARGOFFSET - /*CFI_REL_OFFSET rflags,EFLAGS-ARGOFFSET*/ - /*CFI_REL_OFFSET cs,CS-ARGOFFSET*/ - CFI_REL_OFFSET rip,RIP-ARGOFFSET - CFI_REL_OFFSET rdx,RDX-ARGOFFSET - CFI_REL_OFFSET rcx,RCX-ARGOFFSET - CFI_REL_OFFSET rax,RAX-ARGOFFSET - CFI_REL_OFFSET rdi,RDI-ARGOFFSET - CFI_REL_OFFSET rsi,RSI-ARGOFFSET - CFI_REL_OFFSET r8,R8-ARGOFFSET - CFI_REL_OFFSET r9,R9-ARGOFFSET - CFI_REL_OFFSET r10,R10-ARGOFFSET - CFI_REL_OFFSET r11,R11-ARGOFFSET + */ + .globl int_ret_from_sys_call +int_ret_from_sys_call: cli TRACE_IRQS_OFF testl $3,CS-ARGOFFSET(%rsp) @@ -394,8 +372,6 @@ int_very_careful: popq %rdi CFI_ADJUST_CFA_OFFSET -8 andl $~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP),%edi - cli - TRACE_IRQS_OFF jmp int_restore_rest int_signal: @@ -411,7 +387,7 @@ int_restore_rest: TRACE_IRQS_OFF jmp int_with_check CFI_ENDPROC -END(int_ret_from_sys_call) +END(system_call) /* * Certain special system calls that need to save a complete full stack frame. -- cgit v1.2.3 From 72690a21188586022a9e65cb6f1cc8845167555a Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:03 +0100 Subject: [PATCH] x86: Don't use nested idle loops Currently the idle loop has two nested loops -- one high level in cpu_idle and in some low level idle functions another one. Looping in the low level idle functions breaks the idle notifiers because interrupts waking up sleep states need to execute exit_idle() which is only in cpu_idle(). So don't do that, only loop in cpu_idle(). This only removes code. In some cases e.g. poll_idle the idle loop is a little longer now because cpu_idle checks more things. I hope that isn't a problem ACPI idle doesn't change behaviour because it never looped anyways. Cc: len.brown@intel.com Cc: eranian@hpl.hp.com Signed-off-by: Andi Kleen --- arch/x86_64/kernel/process.c | 30 +++++++++--------------------- 1 file changed, 9 insertions(+), 21 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 7451a4c43c1..0b7b4caa4f7 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -108,17 +108,15 @@ void exit_idle(void) */ static void default_idle(void) { - local_irq_enable(); - current_thread_info()->status &= ~TS_POLLING; smp_mb__after_clear_bit(); - while (!need_resched()) { - local_irq_disable(); - if (!need_resched()) - safe_halt(); - else - local_irq_enable(); - } + local_irq_disable(); + if (!need_resched()) { + /* Enables interrupts one instruction before HLT. + x86 special cases this so there is no race. */ + safe_halt(); + } else + local_irq_enable(); current_thread_info()->status |= TS_POLLING; } @@ -129,16 +127,7 @@ static void default_idle(void) */ static void poll_idle (void) { - local_irq_enable(); - - asm volatile( - "2:" - "testl %0,%1;" - "rep; nop;" - "je 2b;" - : : - "i" (_TIF_NEED_RESCHED), - "m" (current_thread_info()->flags)); + cpu_relax(); } void cpu_idle_wait(void) @@ -257,8 +246,7 @@ void mwait_idle_with_hints(unsigned long eax, unsigned long ecx) static void mwait_idle(void) { local_irq_enable(); - while (!need_resched()) - mwait_idle_with_hints(0,0); + mwait_idle_with_hints(0,0); } void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c) -- cgit v1.2.3 From 7cd8b6861eb586aabe4c725cc0c259ce2e653695 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Thu, 7 Dec 2006 02:14:03 +0100 Subject: [PATCH] x86: remove last two pci_find offenders in the core code Resending as I believe the discussion about them established they were correct. Signed-off-by: Alan Cox Signed-off-by: Andi Kleen Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/x86_64/kernel/pci-calgary.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c index 37a770859e7..d2ea87a9526 100644 --- a/arch/x86_64/kernel/pci-calgary.c +++ b/arch/x86_64/kernel/pci-calgary.c @@ -921,7 +921,7 @@ static int __init calgary_init(void) error: do { - dev = pci_find_device_reverse(PCI_VENDOR_ID_IBM, + dev = pci_get_device_reverse(PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CALGARY, dev); if (!dev) -- cgit v1.2.3 From 9c5f8be4625e73f17e28fea89399ed871a30e064 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:03 +0100 Subject: [PATCH] x86: Mention PCI instead of RAM in NMI parity error message On modern systems RAM errors don't cause NMIs, but it's usually caused by PCI SERR. Mention PCI instead of RAM in the printk. Reported by r_hayashi@ctc-g.co.jp (Ryutaro Hayashi) Cc: r_hayashi@ctc-g.co.jp Signed-off-by: Andi Kleen --- arch/x86_64/kernel/traps.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index e37b4d77d5a..70bfaab9822 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -793,8 +793,7 @@ mem_parity_error(unsigned char reason, struct pt_regs * regs) { printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n", reason); - printk(KERN_EMERG "You probably have a hardware problem with your " - "RAM chips\n"); + printk(KERN_EMERG "You have some hardware problem, likely on the PCI bus.\n"); if (panic_on_unrecovered_nmi) panic("NMI: Not continuing"); -- cgit v1.2.3 From ea7322decb974a4a3e804f96a0201e893ff88ce3 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:05 +0100 Subject: [PATCH] x86-64: Speed and clean up cache flushing in change_page_attr CLFLUSH is a lot faster than WBINVD so avoid the later if at all possible. Always pass the complete list of pages to other CPUs to cut down the number of IPIs. Minor other cleanup and sync with i386 version. Signed-off-by: Andi Kleen --- arch/x86_64/mm/pageattr.c | 58 ++++++++++++++++++++++++++--------------------- 1 file changed, 32 insertions(+), 26 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/mm/pageattr.c b/arch/x86_64/mm/pageattr.c index 3e231d762aa..ccb91dd996a 100644 --- a/arch/x86_64/mm/pageattr.c +++ b/arch/x86_64/mm/pageattr.c @@ -61,34 +61,40 @@ static struct page *split_large_page(unsigned long address, pgprot_t prot, return base; } - -static void flush_kernel_map(void *address) +static void cache_flush_page(void *adr) { - if (0 && address && cpu_has_clflush) { - /* is this worth it? */ - int i; - for (i = 0; i < PAGE_SIZE; i += boot_cpu_data.x86_clflush_size) - asm volatile("clflush (%0)" :: "r" (address + i)); - } else - asm volatile("wbinvd":::"memory"); - if (address) - __flush_tlb_one(address); - else - __flush_tlb_all(); + int i; + for (i = 0; i < PAGE_SIZE; i += boot_cpu_data.x86_clflush_size) + asm volatile("clflush (%0)" :: "r" (adr + i)); } +static void flush_kernel_map(void *arg) +{ + struct list_head *l = (struct list_head *)arg; + struct page *pg; + + /* When clflush is available always use it because it is + much cheaper than WBINVD */ + if (!cpu_has_clflush) + asm volatile("wbinvd" ::: "memory"); + list_for_each_entry(pg, l, lru) { + void *adr = page_address(pg); + if (cpu_has_clflush) + cache_flush_page(adr); + __flush_tlb_one(adr); + } +} -static inline void flush_map(unsigned long address) +static inline void flush_map(struct list_head *l) { - on_each_cpu(flush_kernel_map, (void *)address, 1, 1); + on_each_cpu(flush_kernel_map, l, 1, 1); } -static struct page *deferred_pages; /* protected by init_mm.mmap_sem */ +static LIST_HEAD(deferred_pages); /* protected by init_mm.mmap_sem */ static inline void save_page(struct page *fpage) { - fpage->lru.next = (struct list_head *)deferred_pages; - deferred_pages = fpage; + list_add(&fpage->lru, &deferred_pages); } /* @@ -207,18 +213,18 @@ int change_page_attr(struct page *page, int numpages, pgprot_t prot) void global_flush_tlb(void) { - struct page *dpage; + struct page *pg, *next; + struct list_head l; down_read(&init_mm.mmap_sem); - dpage = xchg(&deferred_pages, NULL); + list_replace_init(&deferred_pages, &l); up_read(&init_mm.mmap_sem); - flush_map((dpage && !dpage->lru.next) ? (unsigned long)page_address(dpage) : 0); - while (dpage) { - struct page *tmp = dpage; - dpage = (struct page *)dpage->lru.next; - ClearPagePrivate(tmp); - __free_page(tmp); + flush_map(&l); + + list_for_each_entry_safe(pg, next, &l, lru) { + ClearPagePrivate(pg); + __free_page(pg); } } -- cgit v1.2.3 From 5df0287ecc4f53e68bbd188fa8258b555e6b734f Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Thu, 7 Dec 2006 02:14:05 +0100 Subject: [PATCH] x86-64: Extend clear_irq_vector Clear the irq releated entries in irq_vector, irq_domain and vector_irq instead of clearing irq_vector only. So when new irq is created, it could reuse that vector. (actually is the second loop scanning from FIRST_DEVICE_VECTOR+8). This could avoid the vectors are used up with enough module inserting and removing Cc: Eric W. Biedierman Cc: Muli Ben-Yehuda Signed-off-By: Yinghai Lu Signed-off-by: Andi Kleen --- arch/x86_64/kernel/io_apic.c | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index c80081a6ba4..f71461b1f03 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -750,6 +750,22 @@ static int assign_irq_vector(int irq, cpumask_t mask, cpumask_t *result) return vector; } +static void __clear_irq_vector(int irq) +{ + cpumask_t mask; + int cpu, vector; + + BUG_ON(!irq_vector[irq]); + + vector = irq_vector[irq]; + cpus_and(mask, irq_domain[irq], cpu_online_map); + for_each_cpu_mask(cpu, mask) + per_cpu(vector_irq, cpu)[vector] = -1; + + irq_vector[irq] = 0; + irq_domain[irq] = CPU_MASK_NONE; +} + void __setup_vector_irq(int cpu) { /* Initialize vector_irq on a new cpu */ @@ -1837,7 +1853,7 @@ void destroy_irq(unsigned int irq) dynamic_irq_cleanup(irq); spin_lock_irqsave(&vector_lock, flags); - irq_vector[irq] = 0; + __clear_irq_vector(irq); spin_unlock_irqrestore(&vector_lock, flags); } -- cgit v1.2.3 From ab2bf0c1c689905b628dca94d0acd9c50e152468 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:06 +0100 Subject: [PATCH] x86-64: Use probe_kernel_address in arch/x86_64/* Instead of open coded __get_user Signed-off-by: Andi Kleen --- arch/x86_64/kernel/traps.c | 2 +- arch/x86_64/mm/fault.c | 10 +++++----- 2 files changed, 6 insertions(+), 6 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 70bfaab9822..264db33476a 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -30,9 +30,9 @@ #include #include #include +#include #include -#include #include #include #include diff --git a/arch/x86_64/mm/fault.c b/arch/x86_64/mm/fault.c index 3751b4788e2..a65fc6f1dca 100644 --- a/arch/x86_64/mm/fault.c +++ b/arch/x86_64/mm/fault.c @@ -23,9 +23,9 @@ #include #include #include +#include #include -#include #include #include #include @@ -96,7 +96,7 @@ void bust_spinlocks(int yes) static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr, unsigned long error_code) { - unsigned char __user *instr; + unsigned char *instr; int scan_more = 1; int prefetch = 0; unsigned char *max_instr; @@ -116,7 +116,7 @@ static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr, unsigned char instr_hi; unsigned char instr_lo; - if (__get_user(opcode, (char __user *)instr)) + if (probe_kernel_address(instr, opcode)) break; instr_hi = opcode & 0xf0; @@ -154,7 +154,7 @@ static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr, case 0x00: /* Prefetch instruction is 0x0F0D or 0x0F18 */ scan_more = 0; - if (__get_user(opcode, (char __user *)instr)) + if (probe_kernel_address(instr, opcode)) break; prefetch = (instr_lo == 0xF) && (opcode == 0x0D || opcode == 0x18); @@ -170,7 +170,7 @@ static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr, static int bad_address(void *p) { unsigned long dummy; - return __get_user(dummy, (unsigned long __user *)p); + return probe_kernel_address((unsigned long *)p, dummy); } void dump_pagetable(unsigned long address) -- cgit v1.2.3 From b026872601976f666bae77b609dc490d1834bf77 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:06 +0100 Subject: [PATCH] x86-64: Try multiple timer variants in check_timer Instead of adding all kinds of more quirks try various timer routing variants in check_timer. In particular this tries to handle quirks from: - Nvidia NF2-4 reference BIOS: wrong timer override - Asus: Wrong timer override but no HPET table - ATI: require timer disabled in 8259 - Some boards: require timer enabled in 8259 We just try many of the the known variants in the hopefully right order in check_timer. Trying pin 0/2 on Nvidia suggested by Tim Hockin. TBD Experimental. Needs a lot of testing Signed-off-by: Andi Kleen --- arch/x86_64/kernel/early-quirks.c | 5 -- arch/x86_64/kernel/io_apic.c | 124 ++++++++++++++++---------------------- 2 files changed, 51 insertions(+), 78 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/early-quirks.c b/arch/x86_64/kernel/early-quirks.c index 68273bff58c..fb0c6da41b7 100644 --- a/arch/x86_64/kernel/early-quirks.c +++ b/arch/x86_64/kernel/early-quirks.c @@ -69,11 +69,6 @@ static void nvidia_bugs(void) static void ati_bugs(void) { - if (timer_over_8254 == 1) { - timer_over_8254 = 0; - printk(KERN_INFO - "ATI board detected. Disabling timer routing over 8254.\n"); - } } struct chipset { diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index f71461b1f03..88fcc4ebbf6 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -55,10 +55,6 @@ int sis_apic_bug; /* not actually supported, dummy for compile */ static int no_timer_check; -static int disable_timer_pin_1 __initdata; - -int timer_over_8254 __initdata = 1; - /* Where if anywhere is the i8259 connect in external int mode */ static struct { int pin, apic; } ioapic_i8259 = { -1, -1 }; @@ -348,29 +344,6 @@ static int __init disable_ioapic_setup(char *str) } early_param("noapic", disable_ioapic_setup); -/* Actually the next is obsolete, but keep it for paranoid reasons -AK */ -static int __init disable_timer_pin_setup(char *arg) -{ - disable_timer_pin_1 = 1; - return 1; -} -__setup("disable_timer_pin_1", disable_timer_pin_setup); - -static int __init setup_disable_8254_timer(char *s) -{ - timer_over_8254 = -1; - return 1; -} -static int __init setup_enable_8254_timer(char *s) -{ - timer_over_8254 = 2; - return 1; -} - -__setup("disable_8254_timer", setup_disable_8254_timer); -__setup("enable_8254_timer", setup_enable_8254_timer); - - /* * Find the IRQ entry number of a certain pin. */ @@ -1579,10 +1552,33 @@ static inline void unlock_ExtINT_logic(void) * a wide range of boards and BIOS bugs. Fortunately only the timer IRQ * is so screwy. Thanks to Brian Perkins for testing/hacking this beast * fanatically on his truly buggy board. - * - * FIXME: really need to revamp this for modern platforms only. */ -static inline void check_timer(void) + +static int try_apic_pin(int apic, int pin, char *msg) +{ + apic_printk(APIC_VERBOSE, KERN_INFO + "..TIMER: trying IO-APIC=%d PIN=%d %s", + apic, pin, msg); + + /* + * Ok, does IRQ0 through the IOAPIC work? + */ + if (!no_timer_check && timer_irq_works()) { + nmi_watchdog_default(); + if (nmi_watchdog == NMI_IO_APIC) { + disable_8259A_irq(0); + setup_nmi(); + enable_8259A_irq(0); + } + return 1; + } + clear_IO_APIC_pin(apic, pin); + apic_printk(APIC_QUIET, KERN_ERR " .. failed\n"); + return 0; +} + +/* The function from hell */ +static void check_timer(void) { int apic1, pin1, apic2, pin2; int vector; @@ -1603,61 +1599,43 @@ static inline void check_timer(void) */ apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT); init_8259A(1); - if (timer_over_8254 > 0) - enable_8259A_irq(0); pin1 = find_isa_irq_pin(0, mp_INT); apic1 = find_isa_irq_apic(0, mp_INT); pin2 = ioapic_i8259.pin; apic2 = ioapic_i8259.apic; - apic_printk(APIC_VERBOSE,KERN_INFO "..TIMER: vector=0x%02X apic1=%d pin1=%d apic2=%d pin2=%d\n", - vector, apic1, pin1, apic2, pin2); + /* Do this first, otherwise we get double interrupts on ATI boards */ + if ((pin1 != -1) && try_apic_pin(apic1, pin1,"with 8259 IRQ0 disabled")) + return; - if (pin1 != -1) { - /* - * Ok, does IRQ0 through the IOAPIC work? - */ - unmask_IO_APIC_irq(0); - if (!no_timer_check && timer_irq_works()) { - nmi_watchdog_default(); - if (nmi_watchdog == NMI_IO_APIC) { - disable_8259A_irq(0); - setup_nmi(); - enable_8259A_irq(0); - } - if (disable_timer_pin_1 > 0) - clear_IO_APIC_pin(0, pin1); - return; - } - clear_IO_APIC_pin(apic1, pin1); - apic_printk(APIC_QUIET,KERN_ERR "..MP-BIOS bug: 8254 timer not " - "connected to IO-APIC\n"); - } + /* Now try again with IRQ0 8259A enabled. + Assumes timer is on IO-APIC 0 ?!? */ + enable_8259A_irq(0); + unmask_IO_APIC_irq(0); + if (try_apic_pin(apic1, pin1, "with 8259 IRQ0 enabled")) + return; + disable_8259A_irq(0); - apic_printk(APIC_VERBOSE,KERN_INFO "...trying to set up timer (IRQ0) " - "through the 8259A ... "); + /* Always try pin0 and pin2 on APIC 0 to handle buggy timer overrides + on Nvidia boards */ + if (!(apic1 == 0 && pin1 == 0) && + try_apic_pin(0, 0, "fallback with 8259 IRQ0 disabled")) + return; + if (!(apic1 == 0 && pin1 == 2) && + try_apic_pin(0, 2, "fallback with 8259 IRQ0 disabled")) + return; + + /* Then try pure 8259A routing on the 8259 as reported by BIOS*/ + enable_8259A_irq(0); if (pin2 != -1) { - apic_printk(APIC_VERBOSE,"\n..... (found apic %d pin %d) ...", - apic2, pin2); - /* - * legacy devices should be connected to IO APIC #0 - */ setup_ExtINT_IRQ0_pin(apic2, pin2, vector); - if (timer_irq_works()) { - apic_printk(APIC_VERBOSE," works.\n"); - nmi_watchdog_default(); - if (nmi_watchdog == NMI_IO_APIC) { - setup_nmi(); - } + if (try_apic_pin(apic2,pin2,"8259A broadcast ExtINT from BIOS")) return; - } - /* - * Cleanup, just in case ... - */ - clear_IO_APIC_pin(apic2, pin2); } - apic_printk(APIC_VERBOSE," failed.\n"); + + /* Tried all possibilities to go through the IO-APIC. Now come the + really cheesy fallbacks. */ if (nmi_watchdog == NMI_IO_APIC) { printk(KERN_WARNING "timer doesn't work through the IO-APIC - disabling NMI Watchdog!\n"); -- cgit v1.2.3 From 8e3de538eec95b57a5b86038988451c38ba83f7e Mon Sep 17 00:00:00 2001 From: Albert Cahalan Date: Thu, 7 Dec 2006 02:14:06 +0100 Subject: [PATCH] x86-64: Support -mregparm arguments for signals with SA_SIGINFO in compat mode The recent change to make x86_64 support i386 binaries compiled with -mregparm=3 only covered signal handlers without SA_SIGINFO. (the 3-arg "real-time" ones) To be compatible with i386, both types should be supported. Signed-off-by: Albert Cahalan Signed-off-by: Andi Kleen --- arch/x86_64/ia32/ia32_signal.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/ia32/ia32_signal.c b/arch/x86_64/ia32/ia32_signal.c index 0e0a266d976..ff499ef2a1b 100644 --- a/arch/x86_64/ia32/ia32_signal.c +++ b/arch/x86_64/ia32/ia32_signal.c @@ -584,6 +584,11 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, regs->rdx = (unsigned long) &frame->info; regs->rcx = (unsigned long) &frame->uc; + /* Make -mregparm=3 work */ + regs->rax = sig; + regs->rdx = (unsigned long) &frame->info; + regs->rcx = (unsigned long) &frame->uc; + asm volatile("movl %0,%%ds" :: "r" (__USER32_DS)); asm volatile("movl %0,%%es" :: "r" (__USER32_DS)); -- cgit v1.2.3 From 58db85482743f5e3495d168c641c60ce1d3dfb06 Mon Sep 17 00:00:00 2001 From: Muli Ben-Yehuda Date: Thu, 7 Dec 2006 02:14:06 +0100 Subject: [PATCH] calgary: phb_shift can be int Signed-off-by: Muli Ben-Yehuda Signed-off-by: Jon Mason Signed-off-by: Andi Kleen --- arch/x86_64/kernel/pci-calgary.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c index d2ea87a9526..f53b581dfd0 100644 --- a/arch/x86_64/kernel/pci-calgary.c +++ b/arch/x86_64/kernel/pci-calgary.c @@ -740,7 +740,7 @@ static void __init calgary_increase_split_completion_timeout(void __iomem *bbar, { u64 val64; void __iomem *target; - unsigned long phb_shift = -1; + unsigned int phb_shift = ~0; /* silence gcc */ u64 mask; switch (busno_to_phbid(busnum)) { -- cgit v1.2.3 From b34e90b8f0f30151349134f87b5dc6ef75a5218c Mon Sep 17 00:00:00 2001 From: Laurent Vivier Date: Thu, 7 Dec 2006 02:14:06 +0100 Subject: [PATCH] Calgary: use BIOS supplied BBARs and topology information Find the BBAR register address of each Calgary using the "Extended BIOS Data Area" rather than calculating it ourselves. Also get the bus topology (what PHB each bus is on) from Calgary rather than calculating it ourselves. This patch fixes http://bugzilla.kernel.org/show_bug.cgi?id=7407. Signed-off-by: Laurent Vivier Signed-off-by: Muli Ben-Yehuda Signed-off-by: Jon Mason Signed-off-by: Andi Kleen --- arch/x86_64/kernel/pci-calgary.c | 161 ++++++++++++++++++++++++++++++--------- 1 file changed, 125 insertions(+), 36 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c index f53b581dfd0..afc0a53505f 100644 --- a/arch/x86_64/kernel/pci-calgary.c +++ b/arch/x86_64/kernel/pci-calgary.c @@ -41,6 +41,7 @@ #include #include #include +#include #define PCI_DEVICE_ID_IBM_CALGARY 0x02a1 #define PCI_VENDOR_DEVICE_ID_CALGARY \ @@ -115,14 +116,35 @@ static const unsigned long phb_offsets[] = { 0xB000 /* PHB3 */ }; +/* PHB debug registers */ + +static const unsigned long phb_debug_offsets[] = { + 0x4000 /* PHB 0 DEBUG */, + 0x5000 /* PHB 1 DEBUG */, + 0x6000 /* PHB 2 DEBUG */, + 0x7000 /* PHB 3 DEBUG */ +}; + +/* + * STUFF register for each debug PHB, + * byte 1 = start bus number, byte 2 = end bus number + */ + +#define PHB_DEBUG_STUFF_OFFSET 0x0020 + unsigned int specified_table_size = TCE_TABLE_SIZE_UNSPECIFIED; static int translate_empty_slots __read_mostly = 0; static int calgary_detected __read_mostly = 0; +static struct rio_table_hdr *rio_table_hdr __initdata; +static struct scal_detail *scal_devs[MAX_NUMNODES] __initdata; +static struct rio_detail *rio_devs[MAX_NUMNODES*4] __initdata; + struct calgary_bus_info { void *tce_space; unsigned char translation_disabled; signed char phbid; + void __iomem *bbar; }; static struct calgary_bus_info bus_info[MAX_PHB_BUS_NUM] = { { NULL, 0, 0 }, }; @@ -475,6 +497,11 @@ static struct dma_mapping_ops calgary_dma_ops = { .unmap_sg = calgary_unmap_sg, }; +static inline void __iomem * busno_to_bbar(unsigned char num) +{ + return bus_info[num].bbar; +} + static inline int busno_to_phbid(unsigned char num) { return bus_info[num].phbid; @@ -828,31 +855,9 @@ static void __init calgary_disable_translation(struct pci_dev *dev) del_timer_sync(&tbl->watchdog_timer); } -static inline unsigned int __init locate_register_space(struct pci_dev *dev) +static inline void __iomem * __init locate_register_space(struct pci_dev *dev) { - int rionodeid; - u32 address; - - /* - * Each Calgary has four busses. The first four busses (first Calgary) - * have RIO node ID 2, then the next four (second Calgary) have RIO - * node ID 3, the next four (third Calgary) have node ID 2 again, etc. - * We use a gross hack - relying on the dev->bus->number ordering, - * modulo 14 - to decide which Calgary a given bus is on. Busses 0, 1, - * 2 and 4 are on the first Calgary (id 2), 6, 8, a and c are on the - * second (id 3), and then it repeats modulo 14. - */ - rionodeid = (dev->bus->number % 14 > 4) ? 3 : 2; - /* - * register space address calculation as follows: - * FE0MB-8MB*OneBasedChassisNumber+1MB*(RioNodeId-ChassisBase) - * ChassisBase is always zero for x366/x260/x460 - * RioNodeId is 2 for first Calgary, 3 for second Calgary - */ - address = START_ADDRESS - - (0x800000 * (ONE_BASED_CHASSIS_NUM + dev->bus->number / 14)) + - (0x100000) * (rionodeid - CHASSIS_BASE); - return address; + return busno_to_bbar(dev->bus->number); } static void __init calgary_init_one_nontraslated(struct pci_dev *dev) @@ -864,15 +869,12 @@ static void __init calgary_init_one_nontraslated(struct pci_dev *dev) static int __init calgary_init_one(struct pci_dev *dev) { - u32 address; void __iomem *bbar; int ret; BUG_ON(dev->bus->number >= MAX_PHB_BUS_NUM); - address = locate_register_space(dev); - /* map entire 1MB of Calgary config space */ - bbar = ioremap_nocache(address, 1024 * 1024); + bbar = locate_register_space(dev); if (!bbar) { ret = -ENODATA; goto done; @@ -898,6 +900,35 @@ static int __init calgary_init(void) { int ret = -ENODEV; struct pci_dev *dev = NULL; + int rio, phb, bus; + void __iomem *bbar; + void __iomem *target; + u8 start_bus, end_bus; + u32 val; + + for (rio = 0; rio < rio_table_hdr->num_rio_dev; rio++) { + + if ( (rio_devs[rio]->type != COMPAT_CALGARY) && + (rio_devs[rio]->type != ALT_CALGARY) ) + continue; + + /* map entire 1MB of Calgary config space */ + bbar = ioremap_nocache(rio_devs[rio]->BBAR, 1024 * 1024); + + for (phb = 0; phb < PHBS_PER_CALGARY; phb++) { + + target = calgary_reg(bbar, phb_debug_offsets[phb] | + PHB_DEBUG_STUFF_OFFSET); + val = be32_to_cpu(readl(target)); + start_bus = (u8)((val & 0x00FF0000) >> 16); + end_bus = (u8)((val & 0x0000FF00) >> 8); + for (bus = start_bus; bus <= end_bus; bus++) { + bus_info[bus].bbar = bbar; + bus_info[bus].phbid = phb; + } + } + } + do { dev = pci_get_device(PCI_VENDOR_ID_IBM, @@ -962,13 +993,55 @@ static inline int __init determine_tce_table_size(u64 ram) return ret; } +static int __init build_detail_arrays(void) +{ + unsigned long ptr; + int i, scal_detail_size, rio_detail_size; + + if (rio_table_hdr->num_scal_dev > MAX_NUMNODES){ + printk(KERN_WARNING + "Calgary: MAX_NUMNODES too low! Defined as %d, " + "but system has %d nodes.\n", + MAX_NUMNODES, rio_table_hdr->num_scal_dev); + return -ENODEV; + } + + switch (rio_table_hdr->version){ + default: + printk(KERN_WARNING + "Calgary: Invalid Rio Grande Table Version: %d\n", + rio_table_hdr->version); + return -ENODEV; + case 2: + scal_detail_size = 11; + rio_detail_size = 13; + break; + case 3: + scal_detail_size = 12; + rio_detail_size = 15; + break; + } + + ptr = ((unsigned long)rio_table_hdr) + 3; + for (i = 0; i < rio_table_hdr->num_scal_dev; + i++, ptr += scal_detail_size) + scal_devs[i] = (struct scal_detail *)ptr; + + for (i = 0; i < rio_table_hdr->num_rio_dev; + i++, ptr += rio_detail_size) + rio_devs[i] = (struct rio_detail *)ptr; + + return 0; +} + void __init detect_calgary(void) { u32 val; int bus; void *tbl; int calgary_found = 0; - int phb = -1; + unsigned long ptr; + int offset; /* * if the user specified iommu=off or iommu=soft or we found @@ -980,6 +1053,29 @@ void __init detect_calgary(void) if (!early_pci_allowed()) return; + ptr = (unsigned long)phys_to_virt(get_bios_ebda()); + + rio_table_hdr = NULL; + offset = 0x180; + while (offset) { + /* The block id is stored in the 2nd word */ + if (*((unsigned short *)(ptr + offset + 2)) == 0x4752){ + /* set the pointer past the offset & block id */ + rio_table_hdr = (struct rio_table_hdr *)(ptr+offset+4); + break; + } + /* The next offset is stored in the 1st word. 0 means no more */ + offset = *((unsigned short *)(ptr + offset)); + } + if (!rio_table_hdr){ + printk(KERN_ERR "Calgary: Unable to locate " + "Rio Grande Table in EBDA - bailing!\n"); + return; + } + + if (build_detail_arrays()) + return; + specified_table_size = determine_tce_table_size(end_pfn * PAGE_SIZE); for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) { @@ -990,12 +1086,6 @@ void __init detect_calgary(void) if (read_pci_config(bus, 0, 0, 0) != PCI_VENDOR_DEVICE_ID_CALGARY) continue; - /* - * There are 4 PHBs per Calgary chip. Set phb to which phb (0-3) - * it is connected to releative to the clagary chip. - */ - phb = (phb + 1) % PHBS_PER_CALGARY; - if (info->translation_disabled) continue; @@ -1010,7 +1100,6 @@ void __init detect_calgary(void) if (!tbl) goto cleanup; info->tce_space = tbl; - info->phbid = phb; calgary_found = 1; break; } -- cgit v1.2.3 From eae93755540bae18aff46b8a0e621b5d65bd5380 Mon Sep 17 00:00:00 2001 From: Muli Ben-Yehuda Date: Thu, 7 Dec 2006 02:14:06 +0100 Subject: [PATCH] Calgary: check BBAR ioremap success when ioremapping This patch cleans up the previous "Use BIOS supplied BBAR information" patch. Mostly stylistic clenaups, but also check for ioremap failure when we ioremap the BBAR rather than when trying to use it. Signed-off-by: Muli Ben-Yehuda Signed-off-by: Jon Mason Signed-off-by: Andi Kleen Acked-by: Laurent Vivier --- arch/x86_64/kernel/pci-calgary.c | 85 +++++++++++++++++++++++----------------- 1 file changed, 49 insertions(+), 36 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c index afc0a53505f..8a1e4f35bc3 100644 --- a/arch/x86_64/kernel/pci-calgary.c +++ b/arch/x86_64/kernel/pci-calgary.c @@ -138,7 +138,7 @@ static int calgary_detected __read_mostly = 0; static struct rio_table_hdr *rio_table_hdr __initdata; static struct scal_detail *scal_devs[MAX_NUMNODES] __initdata; -static struct rio_detail *rio_devs[MAX_NUMNODES*4] __initdata; +static struct rio_detail *rio_devs[MAX_NUMNODES * 4] __initdata; struct calgary_bus_info { void *tce_space; @@ -855,11 +855,6 @@ static void __init calgary_disable_translation(struct pci_dev *dev) del_timer_sync(&tbl->watchdog_timer); } -static inline void __iomem * __init locate_register_space(struct pci_dev *dev) -{ - return busno_to_bbar(dev->bus->number); -} - static void __init calgary_init_one_nontraslated(struct pci_dev *dev) { pci_dev_get(dev); @@ -874,15 +869,10 @@ static int __init calgary_init_one(struct pci_dev *dev) BUG_ON(dev->bus->number >= MAX_PHB_BUS_NUM); - bbar = locate_register_space(dev); - if (!bbar) { - ret = -ENODATA; - goto done; - } - + bbar = busno_to_bbar(dev->bus->number); ret = calgary_setup_tar(dev, bbar); if (ret) - goto iounmap; + goto done; pci_dev_get(dev); dev->bus->self = dev; @@ -890,38 +880,39 @@ static int __init calgary_init_one(struct pci_dev *dev) return 0; -iounmap: - iounmap(bbar); done: return ret; } -static int __init calgary_init(void) +static int __init calgary_locate_bbars(void) { - int ret = -ENODEV; - struct pci_dev *dev = NULL; - int rio, phb, bus; + int ret; + int rioidx, phb, bus; void __iomem *bbar; void __iomem *target; + unsigned long offset; u8 start_bus, end_bus; u32 val; - for (rio = 0; rio < rio_table_hdr->num_rio_dev; rio++) { + ret = -ENODATA; + for (rioidx = 0; rioidx < rio_table_hdr->num_rio_dev; rioidx++) { + struct rio_detail *rio = rio_devs[rioidx]; - if ( (rio_devs[rio]->type != COMPAT_CALGARY) && - (rio_devs[rio]->type != ALT_CALGARY) ) + if ((rio->type != COMPAT_CALGARY) && (rio->type != ALT_CALGARY)) continue; /* map entire 1MB of Calgary config space */ - bbar = ioremap_nocache(rio_devs[rio]->BBAR, 1024 * 1024); + bbar = ioremap_nocache(rio->BBAR, 1024 * 1024); + if (!bbar) + goto error; for (phb = 0; phb < PHBS_PER_CALGARY; phb++) { + offset = phb_debug_offsets[phb] | PHB_DEBUG_STUFF_OFFSET; + target = calgary_reg(bbar, offset); - target = calgary_reg(bbar, phb_debug_offsets[phb] | - PHB_DEBUG_STUFF_OFFSET); val = be32_to_cpu(readl(target)); start_bus = (u8)((val & 0x00FF0000) >> 16); - end_bus = (u8)((val & 0x0000FF00) >> 8); + end_bus = (u8)((val & 0x0000FF00) >> 8); for (bus = start_bus; bus <= end_bus; bus++) { bus_info[bus].bbar = bbar; bus_info[bus].phbid = phb; @@ -929,6 +920,25 @@ static int __init calgary_init(void) } } + return 0; + +error: + /* scan bus_info and iounmap any bbars we previously ioremap'd */ + for (bus = 0; bus < ARRAY_SIZE(bus_info); bus++) + if (bus_info[bus].bbar) + iounmap(bus_info[bus].bbar); + + return ret; +} + +static int __init calgary_init(void) +{ + int ret; + struct pci_dev *dev = NULL; + + ret = calgary_locate_bbars(); + if (ret) + return ret; do { dev = pci_get_device(PCI_VENDOR_ID_IBM, @@ -1000,18 +1010,13 @@ static int __init build_detail_arrays(void) if (rio_table_hdr->num_scal_dev > MAX_NUMNODES){ printk(KERN_WARNING - "Calgary: MAX_NUMNODES too low! Defined as %d, " + "Calgary: MAX_NUMNODES too low! Defined as %d, " "but system has %d nodes.\n", MAX_NUMNODES, rio_table_hdr->num_scal_dev); return -ENODEV; } switch (rio_table_hdr->version){ - default: - printk(KERN_WARNING - "Calgary: Invalid Rio Grande Table Version: %d\n", - rio_table_hdr->version); - return -ENODEV; case 2: scal_detail_size = 11; rio_detail_size = 13; @@ -1020,6 +1025,11 @@ static int __init build_detail_arrays(void) scal_detail_size = 12; rio_detail_size = 15; break; + default: + printk(KERN_WARNING + "Calgary: Invalid Rio Grande Table Version: %d\n", + rio_table_hdr->version); + return -EPROTO; } ptr = ((unsigned long)rio_table_hdr) + 3; @@ -1042,6 +1052,7 @@ void __init detect_calgary(void) int calgary_found = 0; unsigned long ptr; int offset; + int ret; /* * if the user specified iommu=off or iommu=soft or we found @@ -1061,27 +1072,29 @@ void __init detect_calgary(void) /* The block id is stored in the 2nd word */ if (*((unsigned short *)(ptr + offset + 2)) == 0x4752){ /* set the pointer past the offset & block id */ - rio_table_hdr = (struct rio_table_hdr *)(ptr+offset+4); + rio_table_hdr = (struct rio_table_hdr *)(ptr + offset + 4); break; } /* The next offset is stored in the 1st word. 0 means no more */ offset = *((unsigned short *)(ptr + offset)); } - if (!rio_table_hdr){ + if (!rio_table_hdr) { printk(KERN_ERR "Calgary: Unable to locate " "Rio Grande Table in EBDA - bailing!\n"); return; } - if (build_detail_arrays()) + ret = build_detail_arrays(); + if (ret) { + printk(KERN_ERR "Calgary: build_detail_arrays ret %d\n", ret); return; + } specified_table_size = determine_tce_table_size(end_pfn * PAGE_SIZE); for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) { int dev; struct calgary_bus_info *info = &bus_info[bus]; - info->phbid = -1; if (read_pci_config(bus, 0, 0, 0) != PCI_VENDOR_DEVICE_ID_CALGARY) continue; -- cgit v1.2.3 From bff6547bb6a4e82c399d74e7fba78b12d2f162ed Mon Sep 17 00:00:00 2001 From: Muli Ben-Yehuda Date: Thu, 7 Dec 2006 02:14:07 +0100 Subject: [PATCH] Calgary: allow compiling Calgary in but not using it by default This patch makes it possible to compile Calgary in but not use it by default. In this mode, use 'iommu=calgary' to activate it. Signed-off-by: Muli Ben-Yehuda Signed-off-by: Jon Mason Signed-off-by: Andi Kleen --- arch/x86_64/Kconfig | 11 +++++++++++ arch/x86_64/kernel/pci-calgary.c | 9 +++++++++ arch/x86_64/kernel/pci-dma.c | 5 +++++ 3 files changed, 25 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index 010d2265f1c..5cb509dbffe 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -455,6 +455,17 @@ config CALGARY_IOMMU Normally the kernel will make the right choice by itself. If unsure, say Y. +config CALGARY_IOMMU_ENABLED_BY_DEFAULT + bool "Should Calgary be enabled by default?" + default y + depends on CALGARY_IOMMU + help + Should Calgary be enabled by default? if you choose 'y', Calgary + will be used (if it exists). If you choose 'n', Calgary will not be + used even if it exists. If you choose 'n' and would like to use + Calgary anyway, pass 'iommu=calgary' on the kernel command line. + If unsure, say Y. + # need this always selected by IOMMU for the VIA workaround config SWIOTLB bool diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c index 8a1e4f35bc3..0ddf29dae7e 100644 --- a/arch/x86_64/kernel/pci-calgary.c +++ b/arch/x86_64/kernel/pci-calgary.c @@ -43,6 +43,12 @@ #include #include +#ifdef CONFIG_CALGARY_IOMMU_ENABLED_BY_DEFAULT +int use_calgary __read_mostly = 1; +#else +int use_calgary __read_mostly = 0; +#endif /* CONFIG_CALGARY_DEFAULT_ENABLED */ + #define PCI_DEVICE_ID_IBM_CALGARY 0x02a1 #define PCI_VENDOR_DEVICE_ID_CALGARY \ (PCI_VENDOR_ID_IBM | PCI_DEVICE_ID_IBM_CALGARY << 16) @@ -1061,6 +1067,9 @@ void __init detect_calgary(void) if (swiotlb || no_iommu || iommu_detected) return; + if (!use_calgary) + return; + if (!early_pci_allowed()) return; diff --git a/arch/x86_64/kernel/pci-dma.c b/arch/x86_64/kernel/pci-dma.c index f8d857453f8..683b7a5c1ab 100644 --- a/arch/x86_64/kernel/pci-dma.c +++ b/arch/x86_64/kernel/pci-dma.c @@ -296,6 +296,11 @@ __init int iommu_setup(char *p) gart_parse_options(p); #endif +#ifdef CONFIG_CALGARY_IOMMU + if (!strncmp(p, "calgary", 7)) + use_calgary = 1; +#endif /* CONFIG_CALGARY_IOMMU */ + p += strcspn(p, ","); if (*p == ',') ++p; -- cgit v1.2.3 From b9a8d94a47f8a41766f6f7944adfb1d641349903 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Thu, 7 Dec 2006 02:14:07 +0100 Subject: [PATCH] x86-64: Make x86_64 udelay() round up instead of down. Port two patches from i386 to x86_64 delay.c to make sure all rounding is done upward instead of downward. There is no sign in commit messages that the mismatch was done on purpose, and "delay() guarantees sleeping at least for the specified time" is still a valid rule IMHO. The original x86 patches are both from pre-GIT era, i.e.: "[PATCH] round up in __udelay()" in commit 54c7e1f5cc6771ff644d7bc21a2b829308bd126f "[PATCH] add 1 in __const_udelay()" in commit 42c77a9801b8877d8b90f65f75db758822a0bccc (both commits are from converted BK repository to x86_64). AK: fixed gcc warning linux/arch/x86_64/lib/delay.c:43: warning: suggest parentheses around + or - inside shift (did this actually work?) Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andi Kleen --- arch/x86_64/lib/delay.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/lib/delay.c b/arch/x86_64/lib/delay.c index 50be90975d0..2dbebd30834 100644 --- a/arch/x86_64/lib/delay.c +++ b/arch/x86_64/lib/delay.c @@ -40,13 +40,13 @@ EXPORT_SYMBOL(__delay); inline void __const_udelay(unsigned long xloops) { - __delay((xloops * HZ * cpu_data[raw_smp_processor_id()].loops_per_jiffy) >> 32); + __delay(((xloops * HZ * cpu_data[raw_smp_processor_id()].loops_per_jiffy) >> 32) + 1); } EXPORT_SYMBOL(__const_udelay); void __udelay(unsigned long usecs) { - __const_udelay(usecs * 0x000010c6); /* 2**32 / 1000000 */ + __const_udelay(usecs * 0x000010c7); /* 2**32 / 1000000 (rounded up) */ } EXPORT_SYMBOL(__udelay); -- cgit v1.2.3 From 516d2836434d279fef164bb1b964e83425d312d2 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:07 +0100 Subject: [PATCH] x86-64: Fix race in IO-APIC routing entry setup. Interrupt could happen between setting the IO-APIC entry and setting its interrupt data. Pointed out by Linus. Signed-off-by: Andi Kleen --- arch/x86_64/kernel/io_apic.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 88fcc4ebbf6..eaf0b708e67 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -174,14 +174,20 @@ static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin) * the interrupt, and we need to make sure the entry is fully populated * before that happens. */ -static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) +static void +__ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) { - unsigned long flags; union entry_union eu; eu.entry = e; - spin_lock_irqsave(&ioapic_lock, flags); io_apic_write(apic, 0x11 + 2*pin, eu.w2); io_apic_write(apic, 0x10 + 2*pin, eu.w1); +} + +static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) +{ + unsigned long flags; + spin_lock_irqsave(&ioapic_lock, flags); + __ioapic_write_entry(apic, pin, e); spin_unlock_irqrestore(&ioapic_lock, flags); } -- cgit v1.2.3 From b6bcc4bb1cdfbc3c8612aad63a8703ac3d59f61a Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:07 +0100 Subject: [PATCH] x86-64: Don't force inlining of do_csum It's two big and used by two callers. Calls should be cheap enough anyways. Signed-off-by: Andi Kleen --- arch/x86_64/lib/csum-partial.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/lib/csum-partial.c b/arch/x86_64/lib/csum-partial.c index 06ae630de82..bc503f50690 100644 --- a/arch/x86_64/lib/csum-partial.c +++ b/arch/x86_64/lib/csum-partial.c @@ -9,8 +9,6 @@ #include #include -#define __force_inline inline __attribute__((always_inline)) - static inline unsigned short from32to16(unsigned a) { unsigned short b = a >> 16; @@ -33,7 +31,7 @@ static inline unsigned short from32to16(unsigned a) * Unrolling to an 128 bytes inner loop. * Using interleaving with more registers to break the carry chains. */ -static __force_inline unsigned do_csum(const unsigned char *buff, unsigned len) +static unsigned do_csum(const unsigned char *buff, unsigned len) { unsigned odd, count; unsigned long result = 0; -- cgit v1.2.3 From 713819989aa4dd141a37074dbc369e7c620bc619 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:09 +0100 Subject: [PATCH] x86-64: Add option to compile for Core2 Add an option to compile for Intel's Core 2 The Kconfig help is a mouthful due to the inventiveness of Intel's product naming department. Mainly for the 64bit cache line sizes because gcc doesn't support optimizing for core2 yet. However it will and then the kernel should be ready by passing the right option Also fix the old MPSC help text to confirm better to reality. Signed-off-by: Andi Kleen --- arch/x86_64/Kconfig | 27 +++++++++++++++++++++------ arch/x86_64/Makefile | 4 ++++ 2 files changed, 25 insertions(+), 6 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index 5cb509dbffe..6eece27d579 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -122,7 +122,7 @@ endchoice choice prompt "Processor family" - default MK8 + default GENERIC_CPU config MK8 bool "AMD-Opteron/Athlon64" @@ -130,16 +130,31 @@ config MK8 Optimize for AMD Opteron/Athlon64/Hammer/K8 CPUs. config MPSC - bool "Intel EM64T" + bool "Intel P4 / older Netburst based Xeon" help - Optimize for Intel Pentium 4 and Xeon CPUs with Intel - Extended Memory 64 Technology(EM64T). For details see + Optimize for Intel Pentium 4 and older Nocona/Dempsey Xeon CPUs + with Intel Extended Memory 64 Technology(EM64T). For details see . + Note the the latest Xeons (Xeon 51xx and 53xx) are not based on the + Netburst core and shouldn't use this option. You can distingush them + using the cpu family field + in /proc/cpuinfo. Family 15 is a older Xeon, Family 6 a newer one + (this rule only applies to system that support EM64T) + +config MCORE2 + bool "Intel Core2 / newer Xeon" + help + Optimize for Intel Core2 and newer Xeons (51xx) + You can distingush the newer Xeons from the older ones using + the cpu family field in /proc/cpuinfo. 15 is a older Xeon + (use CONFIG_MPSC then), 6 is a newer one. This rule only + applies to CPUs that support EM64T. config GENERIC_CPU bool "Generic-x86-64" help Generic x86-64 CPU. + Run equally well on all x86-64 CPUs. endchoice @@ -149,12 +164,12 @@ endchoice config X86_L1_CACHE_BYTES int default "128" if GENERIC_CPU || MPSC - default "64" if MK8 + default "64" if MK8 || MCORE2 config X86_L1_CACHE_SHIFT int default "7" if GENERIC_CPU || MPSC - default "6" if MK8 + default "6" if MK8 || MCORE2 config X86_INTERNODE_CACHE_BYTES int diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile index 6e38d4daeed..b471b8550d0 100644 --- a/arch/x86_64/Makefile +++ b/arch/x86_64/Makefile @@ -30,6 +30,10 @@ cflags-y := cflags-kernel-y := cflags-$(CONFIG_MK8) += $(call cc-option,-march=k8) cflags-$(CONFIG_MPSC) += $(call cc-option,-march=nocona) +# gcc doesn't support -march=core2 yet as of gcc 4.3, but I hope it +# will eventually. Use -mtune=generic as fallback +cflags-$(CONFIG_MCORE2) += \ + $(call cc-option,-march=core2,$(call cc-option,-mtune=generic)) cflags-$(CONFIG_GENERIC_CPU) += $(call cc-option,-mtune=generic) cflags-y += -m64 -- cgit v1.2.3 From 103efcd9aac1de4da6a1477f2f3b9fcf35273a18 Mon Sep 17 00:00:00 2001 From: Ernie Petrides Date: Thu, 7 Dec 2006 02:14:09 +0100 Subject: [PATCH] x86-64: fix perms/range of vsyscall vma in /proc/*/maps The final line of /proc//maps on x86_64 for native 64-bit tasks shows an incorrect ending address and incorrect permissions. There is only a single page mapped in this vsyscall region, and it is accessible for both read and execute. The patch below fixes this. (Since 32-bit-compat tasks have a real vma with correct perms/range, no change is necessary for that scenario.) Before the patch, a "cat /proc/self/maps | tail -1" shows this: ffffffffff600000-ffffffffffe00000 ---p 00000000 [...] After the patch, this is the output: ffffffffff600000-ffffffffff601000 r-xp 00000000 [...] Signed-off-by: Ernie Petrides Signed-off-by: Andi Kleen --- arch/x86_64/kernel/vsyscall.c | 1 + arch/x86_64/mm/init.c | 7 ++++--- 2 files changed, 5 insertions(+), 3 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c index 92546c1526f..c3de9a09cd9 100644 --- a/arch/x86_64/kernel/vsyscall.c +++ b/arch/x86_64/kernel/vsyscall.c @@ -290,6 +290,7 @@ static void __init map_vsyscall(void) extern char __vsyscall_0; unsigned long physaddr_page0 = __pa_symbol(&__vsyscall_0); + /* Note that VSYSCALL_MAPPED_PAGES must agree with the code below. */ __set_fixmap(VSYSCALL_FIRST_PAGE, physaddr_page0, PAGE_KERNEL_VSYSCALL); } diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c index 4c0c00ef3ca..2968b90ef8a 100644 --- a/arch/x86_64/mm/init.c +++ b/arch/x86_64/mm/init.c @@ -730,14 +730,15 @@ static __init int x8664_sysctl_init(void) __initcall(x8664_sysctl_init); #endif -/* A pseudo VMAs to allow ptrace access for the vsyscall page. This only +/* A pseudo VMA to allow ptrace access for the vsyscall page. This only covers the 64bit vsyscall page now. 32bit has a real VMA now and does not need special handling anymore. */ static struct vm_area_struct gate_vma = { .vm_start = VSYSCALL_START, - .vm_end = VSYSCALL_END, - .vm_page_prot = PAGE_READONLY + .vm_end = VSYSCALL_START + (VSYSCALL_MAPPED_PAGES << PAGE_SHIFT), + .vm_page_prot = PAGE_READONLY_EXEC, + .vm_flags = VM_READ | VM_EXEC }; struct vm_area_struct *get_gate_vma(struct task_struct *tsk) -- cgit v1.2.3 From 9a457324229db34d3bcb0b67360130c287289401 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:09 +0100 Subject: [PATCH] x86-64: Rate limit no irq handler messages Signed-off-by: Andi Kleen --- arch/x86_64/kernel/irq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/irq.c b/arch/x86_64/kernel/irq.c index e46c55856d4..0c06af6c13b 100644 --- a/arch/x86_64/kernel/irq.c +++ b/arch/x86_64/kernel/irq.c @@ -120,7 +120,7 @@ asmlinkage unsigned int do_IRQ(struct pt_regs *regs) if (likely(irq < NR_IRQS)) generic_handle_irq(irq); - else + else if (printk_ratelimit()) printk(KERN_EMERG "%s: %d.%d No irq handler for vector\n", __func__, smp_processor_id(), vector); -- cgit v1.2.3 From 9899f826fc90beba4f78083f6230e06cbe1050c9 Mon Sep 17 00:00:00 2001 From: "Siddha, Suresh B" Date: Thu, 7 Dec 2006 02:14:10 +0100 Subject: [PATCH] x86-64: add genapic_force Add genapic_force. Used by the next Intel quirks patch. Signed-off-by: Suresh Siddha Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: "Li, Shaohua" Signed-off-by: Andrew Morton --- arch/x86_64/kernel/genapic.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/genapic.c b/arch/x86_64/kernel/genapic.c index 8e78a75d186..b007433f96b 100644 --- a/arch/x86_64/kernel/genapic.c +++ b/arch/x86_64/kernel/genapic.c @@ -33,7 +33,7 @@ extern struct genapic apic_flat; extern struct genapic apic_physflat; struct genapic *genapic = &apic_flat; - +struct genapic *genapic_force; /* * Check the APIC IDs in bios_cpu_apicid and choose the APIC mode. @@ -46,6 +46,13 @@ void __init clustered_apic_check(void) u8 cluster_cnt[NUM_APIC_CLUSTERS]; int max_apic = 0; + /* genapic selection can be forced because of certain quirks. + */ + if (genapic_force) { + genapic = genapic_force; + goto print; + } + #if defined(CONFIG_ACPI) /* * Some x86_64 machines use physical APIC mode regardless of how many -- cgit v1.2.3 From b0d0a4ba45760b10ecee9035ed45b442c1a6cc84 Mon Sep 17 00:00:00 2001 From: "Siddha, Suresh B" Date: Thu, 7 Dec 2006 02:14:10 +0100 Subject: [PATCH] x86: fix the irqbalance quirk for E7320/E7520/E7525 Move the irqbalance quirks for E7320/E7520/E7525(Errata 23 in http://download.intel.com/design/chipsets/specupdt/30304203.pdf) to early quirks. And add a PCI quirk for these platforms to check(which happens very late during the boot) if the APIC routing is indeed set to default flat mode. This fixes the breakage(in x86_64) of this quirk due to cpu hotplug which selects physical mode instead of the logical flat(as needed for this errata workaround). Signed-off-by: Suresh Siddha Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: "Li, Shaohua" Signed-off-by: Andrew Morton --- arch/x86_64/kernel/early-quirks.c | 13 +++++++++++++ arch/x86_64/kernel/smpboot.c | 8 ++++++++ 2 files changed, 21 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/early-quirks.c b/arch/x86_64/kernel/early-quirks.c index fb0c6da41b7..829698f6d04 100644 --- a/arch/x86_64/kernel/early-quirks.c +++ b/arch/x86_64/kernel/early-quirks.c @@ -71,6 +71,18 @@ static void ati_bugs(void) { } +static void intel_bugs(void) +{ + u16 device = read_pci_config_16(0, 0, 0, PCI_DEVICE_ID); + +#ifdef CONFIG_SMP + if (device == PCI_DEVICE_ID_INTEL_E7320_MCH || + device == PCI_DEVICE_ID_INTEL_E7520_MCH || + device == PCI_DEVICE_ID_INTEL_E7525_MCH) + quirk_intel_irqbalance(); +#endif +} + struct chipset { u16 vendor; void (*f)(void); @@ -80,6 +92,7 @@ static struct chipset early_qrk[] = { { PCI_VENDOR_ID_NVIDIA, nvidia_bugs }, { PCI_VENDOR_ID_VIA, via_bugs }, { PCI_VENDOR_ID_ATI, ati_bugs }, + { PCI_VENDOR_ID_INTEL, intel_bugs}, {} }; diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c index 62c2e747af5..4c161c208d5 100644 --- a/arch/x86_64/kernel/smpboot.c +++ b/arch/x86_64/kernel/smpboot.c @@ -60,6 +60,7 @@ #include #include #include +#include /* Number of siblings per CPU package */ int smp_num_siblings = 1; @@ -1167,6 +1168,13 @@ int __cpuinit __cpu_up(unsigned int cpu) while (!cpu_isset(cpu, cpu_online_map)) cpu_relax(); + + if (num_online_cpus() > 8 && genapic == &apic_flat) { + printk(KERN_WARNING + "flat APIC routing can't be used with > 8 cpus\n"); + BUG(); + } + err = 0; return err; -- cgit v1.2.3 From ee58fad51a2a767cb2567706ace967705233d881 Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] x86-64: x86-64 add Intel BTS cpufeature bit and detection (take 2) Here is a small patch for x86-64 which adds a cpufeature flag and detection code for Intel's Branch Trace Store (BTS) feature. This feature can be found on Intel P4 and Core 2 processors among others. It can also be used by perfmon. changelog: - add CPU_FEATURE_BTS - add Branch Trace Store detection signed-off-by: stephane eranian Signed-off-by: Andi Kleen --- arch/x86_64/kernel/setup.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index 619af2e2fa2..a570c81c831 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -838,6 +838,8 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) if (cpu_has_ds) { unsigned int l1, l2; rdmsr(MSR_IA32_MISC_ENABLE, l1, l2); + if (!(l1 & (1<<11))) + set_bit(X86_FEATURE_BTS, c->x86_capability); if (!(l1 & (1<<12))) set_bit(X86_FEATURE_PEBS, c->x86_capability); } -- cgit v1.2.3 From f990fff427d68af3e4e1d16fe799c106abc0bf53 Mon Sep 17 00:00:00 2001 From: Karsten Wiese Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] x86: Regard MSRs in lapic_suspend()/lapic_resume() Read/Write APIC_LVTPC and APIC_LVTTHMR only, if get_maxlvt() returns certain values. This is done like everywhere else in i386/kernel/apic.c, so I guess its correct. Suspends/Resumes to disk fine and eleminates an smp_error_interrupt() here on a K8. AK: ported to x86-64 too Signed-off-by: Karsten Wiese Signed-off-by: Andi Kleen --- arch/x86_64/kernel/apic.c | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c index 5c468971e64..f0b00d8731c 100644 --- a/arch/x86_64/kernel/apic.c +++ b/arch/x86_64/kernel/apic.c @@ -459,23 +459,30 @@ static struct { static int lapic_suspend(struct sys_device *dev, pm_message_t state) { unsigned long flags; + int maxlvt; if (!apic_pm_state.active) return 0; + maxlvt = get_maxlvt(); + apic_pm_state.apic_id = apic_read(APIC_ID); apic_pm_state.apic_taskpri = apic_read(APIC_TASKPRI); apic_pm_state.apic_ldr = apic_read(APIC_LDR); apic_pm_state.apic_dfr = apic_read(APIC_DFR); apic_pm_state.apic_spiv = apic_read(APIC_SPIV); apic_pm_state.apic_lvtt = apic_read(APIC_LVTT); - apic_pm_state.apic_lvtpc = apic_read(APIC_LVTPC); + if (maxlvt >= 4) + apic_pm_state.apic_lvtpc = apic_read(APIC_LVTPC); apic_pm_state.apic_lvt0 = apic_read(APIC_LVT0); apic_pm_state.apic_lvt1 = apic_read(APIC_LVT1); apic_pm_state.apic_lvterr = apic_read(APIC_LVTERR); apic_pm_state.apic_tmict = apic_read(APIC_TMICT); apic_pm_state.apic_tdcr = apic_read(APIC_TDCR); - apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR); +#ifdef CONFIG_X86_MCE_INTEL + if (maxlvt >= 5) + apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR); +#endif local_irq_save(flags); disable_local_APIC(); local_irq_restore(flags); @@ -486,10 +493,13 @@ static int lapic_resume(struct sys_device *dev) { unsigned int l, h; unsigned long flags; + int maxlvt; if (!apic_pm_state.active) return 0; + maxlvt = get_maxlvt(); + local_irq_save(flags); rdmsr(MSR_IA32_APICBASE, l, h); l &= ~MSR_IA32_APICBASE_BASE; @@ -503,8 +513,12 @@ static int lapic_resume(struct sys_device *dev) apic_write(APIC_SPIV, apic_pm_state.apic_spiv); apic_write(APIC_LVT0, apic_pm_state.apic_lvt0); apic_write(APIC_LVT1, apic_pm_state.apic_lvt1); - apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr); - apic_write(APIC_LVTPC, apic_pm_state.apic_lvtpc); +#ifdef CONFIG_X86_MCE_INTEL + if (maxlvt >= 5) + apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr); +#endif + if (maxlvt >= 4) + apic_write(APIC_LVTPC, apic_pm_state.apic_lvtpc); apic_write(APIC_LVTT, apic_pm_state.apic_lvtt); apic_write(APIC_TDCR, apic_pm_state.apic_tdcr); apic_write(APIC_TMICT, apic_pm_state.apic_tmict); -- cgit v1.2.3 From 86bd58bf4c383fde4e99b83ce917b091a072040d Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] x86-64: Remove unused GET_APIC_VERSION call from clear_local_APIC Remove unused GET_APIC_VERSION call from clear_local_APIC() and __setup_APIC_LVTT(). Reported by D Binderman . Cc: Andi Kleen Cc: Ingo Molnar Signed-off-by: David Rientjes Signed-off-by: Andi Kleen --- arch/x86_64/kernel/apic.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c index f0b00d8731c..124b2d27b4a 100644 --- a/arch/x86_64/kernel/apic.c +++ b/arch/x86_64/kernel/apic.c @@ -140,7 +140,6 @@ void clear_local_APIC(void) apic_write(APIC_LVTERR, APIC_LVT_MASKED); if (maxlvt >= 4) apic_write(APIC_LVTPC, APIC_LVT_MASKED); - v = GET_APIC_VERSION(apic_read(APIC_LVR)); apic_write(APIC_ESR, 0); apic_read(APIC_ESR); } @@ -736,10 +735,9 @@ void __init init_apic_mappings(void) static void __setup_APIC_LVTT(unsigned int clocks) { - unsigned int lvtt_value, tmp_value, ver; + unsigned int lvtt_value, tmp_value; int cpu = smp_processor_id(); - ver = GET_APIC_VERSION(apic_read(APIC_LVR)); lvtt_value = APIC_LVT_TIMER_PERIODIC | LOCAL_TIMER_VECTOR; if (cpu_isset(cpu, timer_interrupt_broadcast_ipi_mask)) -- cgit v1.2.3 From 0741f4d207a644482d7a040f05cd264c98cf7ee8 Mon Sep 17 00:00:00 2001 From: Chuck Ebbert <76306.1226@compuserve.com> Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] x86: add sysctl for kstack_depth_to_print Add sysctl for kstack_depth_to_print. This lets users change the amount of raw stack data printed in dump_stack() without having to reboot. Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com> Signed-off-by: Andi Kleen --- arch/x86_64/kernel/traps.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 264db33476a..75ceccee178 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -108,7 +108,7 @@ static inline void preempt_conditional_cli(struct pt_regs *regs) preempt_enable_no_resched(); } -static int kstack_depth_to_print = 12; +int kstack_depth_to_print = 12; #ifdef CONFIG_STACK_UNWIND static int call_trace = 1; #else -- cgit v1.2.3 From 3df0af0eb064a16bbdbe81b46bc72a4089f88d54 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Thu, 7 Dec 2006 02:14:12 +0100 Subject: [PATCH] x86_64: clear_bss before set_intr_gate with early_idt_handler idt_table is in the .bss section, so clear_bss need to called at first Signed-off-by: Yinghai Lu Signed-off-by: Andi Kleen --- arch/x86_64/kernel/head64.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/head64.c b/arch/x86_64/kernel/head64.c index 9561eb3c5b5..cc230b93cd1 100644 --- a/arch/x86_64/kernel/head64.c +++ b/arch/x86_64/kernel/head64.c @@ -57,10 +57,12 @@ void __init x86_64_start_kernel(char * real_mode_data) { int i; - for (i = 0; i < 256; i++) + /* clear bss before set_intr_gate with early_idt_handler */ + clear_bss(); + + for (i = 0; i < IDT_ENTRIES; i++) set_intr_gate(i, early_idt_handler); asm volatile("lidt %0" :: "m" (idt_descr)); - clear_bss(); early_printk("Kernel alive\n"); -- cgit v1.2.3 From 8fb6e5f5db860113e71ce7b854382ed40559395b Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Thu, 7 Dec 2006 02:14:12 +0100 Subject: [PATCH] x86_64: interrupt array size should be aligned to NR_VECTORS interrupt array is referred for idt vectors instead of NR_IRQS, so change size to NR_VECTORS - FIRST_EXTERNAL_VECTOR. Also change to static. Signed-off-by: Yinghai Lu Signed-off-by: Andi Kleen --- arch/x86_64/kernel/i8259.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/i8259.c b/arch/x86_64/kernel/i8259.c index c4ef801b765..d73c79e821f 100644 --- a/arch/x86_64/kernel/i8259.c +++ b/arch/x86_64/kernel/i8259.c @@ -76,7 +76,8 @@ BUILD_16_IRQS(0xc) BUILD_16_IRQS(0xd) BUILD_16_IRQS(0xe) BUILD_16_IRQS(0xf) IRQ(x,8), IRQ(x,9), IRQ(x,a), IRQ(x,b), \ IRQ(x,c), IRQ(x,d), IRQ(x,e), IRQ(x,f) -void (*interrupt[NR_IRQS])(void) = { +/* for the irq vectors */ +static void (*interrupt[NR_VECTORS - FIRST_EXTERNAL_VECTOR])(void) = { IRQLIST_16(0x2), IRQLIST_16(0x3), IRQLIST_16(0x4), IRQLIST_16(0x5), IRQLIST_16(0x6), IRQLIST_16(0x7), IRQLIST_16(0x8), IRQLIST_16(0x9), IRQLIST_16(0xa), IRQLIST_16(0xb), -- cgit v1.2.3 From 026c66bdda5f07959da7d74d29b18a7c480242f7 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 7 Dec 2006 02:14:12 +0100 Subject: [PATCH] x86-64: remove duplicate ARCH_DISCONTIGMEM_ENABLE option One ARCH_DISCONTIGMEM_ENABLE option is enough. Signed-off-by: Adrian Bunk Signed-off-by: Andi Kleen --- arch/x86_64/Kconfig | 5 ----- 1 file changed, 5 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index 6eece27d579..bfbb9bcae12 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -359,11 +359,6 @@ config ARCH_DISCONTIGMEM_ENABLE depends on NUMA default y - -config ARCH_DISCONTIGMEM_ENABLE - def_bool y - depends on NUMA - config ARCH_DISCONTIGMEM_DEFAULT def_bool y depends on NUMA -- cgit v1.2.3 From d4c45718b3c0d3045eab803cd15fabbed1ea5bcd Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 7 Dec 2006 02:14:12 +0100 Subject: [PATCH] x86-64: Fix kobject_init() WARN_ON on resume from disk Make mce_remove_device() clean up the kobject in per_cpu(device_mce, cpu) after it has been unregistered. Signed-off-by: Rafael J. Wysocki Signed-off-by: Andi Kleen --- arch/x86_64/kernel/mce.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c index bbea88801d8..a7440cb9c14 100644 --- a/arch/x86_64/kernel/mce.c +++ b/arch/x86_64/kernel/mce.c @@ -652,6 +652,7 @@ static void mce_remove_device(unsigned int cpu) sysdev_remove_file(&per_cpu(device_mce,cpu), &attr_tolerant); sysdev_remove_file(&per_cpu(device_mce,cpu), &attr_check_interval); sysdev_unregister(&per_cpu(device_mce,cpu)); + memset(&per_cpu(device_mce, cpu).kobj, 0, sizeof(struct kobject)); } /* Get notified when a cpu comes on/off. Be hotplug friendly. */ -- cgit v1.2.3 From 73ad8355d7db6a3cdcf313d4a4586a8f81b19c2f Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Thu, 7 Dec 2006 02:14:12 +0100 Subject: [PATCH] x86-64: remove unused acpi_found_madt in mparse. remove unused acpi_found_madt in mparse.c Signed-off-by: Yinghai Lu Signed-off-by: Andi Kleen --- arch/x86_64/kernel/mpparse.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c index b147ab19fbd..08072568847 100644 --- a/arch/x86_64/kernel/mpparse.c +++ b/arch/x86_64/kernel/mpparse.c @@ -35,8 +35,6 @@ int smp_found_config; unsigned int __initdata maxcpus = NR_CPUS; -int acpi_found_madt; - /* * Various Linux-internal data structures created from the * MP-table. -- cgit v1.2.3 From 616779656989cb8c59177e35cb13e87028b1edc8 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:12 +0100 Subject: [PATCH] x86-64: Synchronize RDTSC on single core AMD There is no guarantee that two RDTSCs in a row are monotonic, so don't assume it on single core AMD systems. This will make gettimeofday slower again Signed-off-by: Andi Kleen --- arch/x86_64/kernel/setup.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index a570c81c831..05eaca41802 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -732,11 +732,8 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) /* Fix cpuid4 emulation for more */ num_cache_leaves = 3; - /* When there is only one core no need to synchronize RDTSC */ - if (num_possible_cpus() == 1) - set_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability); - else - clear_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability); + /* RDTSC can be speculated around */ + clear_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability); } static void __cpuinit detect_ht(struct cpuinfo_x86 *c) -- cgit v1.2.3 From e496a0da7f8110054d0ad4039245b0f49f9540f5 Mon Sep 17 00:00:00 2001 From: Muli Ben-Yehuda Date: Thu, 7 Dec 2006 02:14:12 +0100 Subject: [PATCH] Calgary: remove unused variables Spotted by d binderman . Signed-off-by: Muli Ben-Yehuda Signed-off-by: Andi Kleen --- arch/x86_64/kernel/pci-calgary.c | 5 ----- 1 file changed, 5 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c index 0ddf29dae7e..3215675ab12 100644 --- a/arch/x86_64/kernel/pci-calgary.c +++ b/arch/x86_64/kernel/pci-calgary.c @@ -653,14 +653,9 @@ static void __init calgary_reserve_peripheral_mem_2(struct pci_dev *dev) static void __init calgary_reserve_regions(struct pci_dev *dev) { unsigned int npages; - void __iomem *bbar; - unsigned char busnum; u64 start; struct iommu_table *tbl = dev->sysdata; - bbar = tbl->bbar; - busnum = dev->bus->number; - /* reserve bad_dma_address in case it's a legal address */ iommu_range_reserve(tbl, bad_dma_address, 1); -- cgit v1.2.3 From f3d73707a1e84f0687a05144b70b660441e999c7 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Thu, 7 Dec 2006 02:14:12 +0100 Subject: [PATCH] x86-64: Mark rdtsc as sync only for netburst, not for core2 On the Core2 cpus, the rdtsc instruction is not serializing (as defined in the architecture reference since rdtsc exists) and due to the deep speculation of these cores, it's possible that you can observe time go backwards between cores due to this speculation. Since the kernel already deals with this with the SYNC_RDTSC flag, the solution is simple, only assume that the instruction is serializing on family 15... The price one pays for this is a slightly slower gettimeofday (by a dozen or two cycles), but that increase is quite small to pay for a really-going-forward tsc counter. Signed-off-by: Arjan van de Ven Signed-off-by: Andi Kleen --- arch/x86_64/kernel/setup.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index 05eaca41802..6595a4ebe7f 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -860,7 +860,10 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) set_bit(X86_FEATURE_CONSTANT_TSC, &c->x86_capability); if (c->x86 == 6) set_bit(X86_FEATURE_REP_GOOD, &c->x86_capability); - set_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability); + if (c->x86 == 15) + set_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability); + else + clear_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability); c->x86_max_cores = intel_num_cpu_cores(c); srat_detect_node(); -- cgit v1.2.3 From 446f713ba1afd68568139ae4691335ba273fa7f4 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:12 +0100 Subject: [PATCH] unwinder: always use unlocked module list access in unwinder fallback We're already well protected against module unloads because module unload uses stop_machine(). The only exception is NMIs, but other users already risk lockless accesses here. This avoids some hackery in lockdep and also a potential deadlock This matches what i386 does. Signed-off-by: Andi Kleen --- arch/x86_64/kernel/traps.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 75ceccee178..9864d195c40 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -317,9 +317,9 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, #define HANDLE_STACK(cond) \ do while (cond) { \ unsigned long addr = *stack++; \ - if (oops_in_progress ? \ - __kernel_text_address(addr) : \ - kernel_text_address(addr)) { \ + /* Use unlocked access here because except for NMIs \ + we should be already protected against module unloads */ \ + if (__kernel_text_address(addr)) { \ /* \ * If the address is either in the text segment of the \ * kernel, or in the region which contains vmalloc'ed \ -- cgit v1.2.3 From 359ad0d4015a9ab39243f2ebc4eb07915bd618b2 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 7 Dec 2006 02:14:13 +0100 Subject: [PATCH] unwinder: more sanity checks in Dwarf2 unwinder Tighten the requirements on both input to and output from the Dwarf2 unwinder. Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen --- arch/x86_64/kernel/traps.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 9864d195c40..4fdd162f0be 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -225,12 +225,19 @@ static int dump_trace_unwind(struct unwind_frame_info *info, void *context) { struct ops_and_data *oad = (struct ops_and_data *)context; int n = 0; + unsigned long sp = UNW_SP(info); + if (arch_unw_user_mode(info)) + return -1; while (unwind(info) == 0 && UNW_PC(info)) { n++; oad->ops->address(oad->data, UNW_PC(info)); if (arch_unw_user_mode(info)) break; + if ((sp & ~(PAGE_SIZE - 1)) == (UNW_SP(info) & ~(PAGE_SIZE - 1)) + && sp > UNW_SP(info)) + break; + sp = UNW_SP(info); } return n; } -- cgit v1.2.3 From a0429d0d7a6116dedcb71d9128da904bf135f189 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:13 +0100 Subject: [PATCH] x86-64: Remove unwind stack pointer alignment forcing again This was added as a workaround for the fallback unwinder not supporting unaligned stack pointers properly. But now it was fixed to do that, so it's not needed anymore Cc: mingo@elte.hu Signed-off-by: Andi Kleen --- arch/x86_64/kernel/traps.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 4fdd162f0be..a1641ffdffc 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -309,12 +309,6 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, if (tsk && tsk != current) stack = (unsigned long *)tsk->thread.rsp; } - /* - * Align the stack pointer on word boundary, later loops - * rely on that (and corruption / debug info bugs can cause - * unaligned values here): - */ - stack = (unsigned long *)((unsigned long)stack & ~(sizeof(long)-1)); /* * Print function call entries within a stack. 'cond' is the -- cgit v1.2.3 From d331e739f5ad2aaa9d8553891ba6ca823bdbce37 Mon Sep 17 00:00:00 2001 From: Venkatesh Pallipadi Date: Thu, 7 Dec 2006 02:14:13 +0100 Subject: [PATCH] x86-64: Fix interrupt race in idle callback (3rd try) Idle callbacks has some races when enter_idle() sets isidle and subsequent interrupts that can happen on that CPU, before CPU goes to idle. Due to this, an IDLE_END can get called before IDLE_START. To avoid these races, disable interrupts before enter_idle and make sure that all idle routines do not enable interrupts before entering idle. Note that poll_idle() still has a this race as it has to enable interrupts before going to idle. But, all other idle routines have the race fixed. Signed-off-by: Venkatesh Pallipadi Signed-off-by: Andi Kleen --- arch/x86_64/kernel/process.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 0b7b4caa4f7..a418ee4c8c6 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -127,6 +127,7 @@ static void default_idle(void) */ static void poll_idle (void) { + local_irq_enable(); cpu_relax(); } @@ -208,6 +209,12 @@ void cpu_idle (void) idle = default_idle; if (cpu_is_offline(smp_processor_id())) play_dead(); + /* + * Idle routines should keep interrupts disabled + * from here on, until they go to idle. + * Otherwise, idle callbacks can misfire. + */ + local_irq_disable(); enter_idle(); idle(); /* In many cases the interrupt that ended idle @@ -245,8 +252,16 @@ void mwait_idle_with_hints(unsigned long eax, unsigned long ecx) /* Default MONITOR/MWAIT with no hints, used for default C1 state */ static void mwait_idle(void) { - local_irq_enable(); - mwait_idle_with_hints(0,0); + if (!need_resched()) { + __monitor((void *)¤t_thread_info()->flags, 0, 0); + smp_mb(); + if (!need_resched()) + __sti_mwait(0, 0); + else + local_irq_enable(); + } else { + local_irq_enable(); + } } void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c) -- cgit v1.2.3 From 3807fd46e94ab9f09e5ee3bff5e6515a94e9b3c7 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:13 +0100 Subject: [PATCH] x86-64: Clarify error message in GART code - Remove "Disabling IOMMU" message because it confuses people - Clarify that the GART IOMMU is refered to in other message Signed-off-by: Andi Kleen --- arch/x86_64/kernel/pci-gart.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/pci-gart.c b/arch/x86_64/kernel/pci-gart.c index 16261a8a330..fc1960f1f24 100644 --- a/arch/x86_64/kernel/pci-gart.c +++ b/arch/x86_64/kernel/pci-gart.c @@ -601,10 +601,9 @@ void __init gart_iommu_init(void) (!force_iommu && end_pfn <= MAX_DMA32_PFN) || !iommu_aperture || (no_agp && init_k8_gatt(&info) < 0)) { - printk(KERN_INFO "PCI-DMA: Disabling IOMMU.\n"); if (end_pfn > MAX_DMA32_PFN) { printk(KERN_ERR "WARNING more than 4GB of memory " - "but IOMMU not available.\n" + "but GART IOMMU not available.\n" KERN_ERR "WARNING 32bit PCI may malfunction.\n"); } return; -- cgit v1.2.3 From ad892f5e0d01f3c3b475a688d1ddc211cf3ea56d Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Thu, 7 Dec 2006 02:14:19 +0100 Subject: [PATCH] x86-64: check vector in setup_ioapic_dest to verify if need setup_IO_APIC_irq setup_IO_APIC_irqs could fail to get vector for some device when you have too many devices, because at that time only boot cpu is online. So check vector for irq in setup_ioapic_dest and call setup_IO_APIC_irq to make sure IO-APIC irq-routing table is initialized. Also seperate setup_IO_APIC_irq from setup_IO_APIC_irqs. Signed-off-by: Yinghai Lu Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: "Eric W. Biederman" Cc: Ingo Molnar Signed-off-by: Andrew Morton --- arch/x86_64/kernel/io_apic.c | 104 +++++++++++++++++++++++++------------------ 1 file changed, 61 insertions(+), 43 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index eaf0b708e67..2a1dcd5f69c 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -789,27 +789,65 @@ static void ioapic_register_intr(int irq, int vector, unsigned long trigger) handle_edge_irq, "edge"); } } - -static void __init setup_IO_APIC_irqs(void) +static void __init setup_IO_APIC_irq(int apic, int pin, int idx, int irq) { struct IO_APIC_route_entry entry; - int apic, pin, idx, irq, first_notcon = 1, vector; + int vector; unsigned long flags; - apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n"); - for (apic = 0; apic < nr_ioapics; apic++) { - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + /* + * add it to the IO-APIC irq-routing table: + */ + memset(&entry,0,sizeof(entry)); - /* - * add it to the IO-APIC irq-routing table: - */ - memset(&entry,0,sizeof(entry)); + entry.delivery_mode = INT_DELIVERY_MODE; + entry.dest_mode = INT_DEST_MODE; + entry.mask = 0; /* enable IRQ */ + entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS); + + entry.trigger = irq_trigger(idx); + entry.polarity = irq_polarity(idx); - entry.delivery_mode = INT_DELIVERY_MODE; - entry.dest_mode = INT_DEST_MODE; - entry.mask = 0; /* enable IRQ */ + if (irq_trigger(idx)) { + entry.trigger = 1; + entry.mask = 1; entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS); + } + + if (!apic && !IO_APIC_IRQ(irq)) + return; + + if (IO_APIC_IRQ(irq)) { + cpumask_t mask; + vector = assign_irq_vector(irq, TARGET_CPUS, &mask); + if (vector < 0) + return; + + entry.dest.logical.logical_dest = cpu_mask_to_apicid(mask); + entry.vector = vector; + + ioapic_register_intr(irq, vector, IOAPIC_AUTO); + if (!apic && (irq < 16)) + disable_8259A_irq(irq); + } + + ioapic_write_entry(apic, pin, entry); + + spin_lock_irqsave(&ioapic_lock, flags); + set_native_irq_info(irq, TARGET_CPUS); + spin_unlock_irqrestore(&ioapic_lock, flags); + +} + +static void __init setup_IO_APIC_irqs(void) +{ + int apic, pin, idx, irq, first_notcon = 1; + + apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n"); + + for (apic = 0; apic < nr_ioapics; apic++) { + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { idx = find_irq_entry(apic,pin,mp_INT); if (idx == -1) { @@ -821,39 +859,11 @@ static void __init setup_IO_APIC_irqs(void) continue; } - entry.trigger = irq_trigger(idx); - entry.polarity = irq_polarity(idx); - - if (irq_trigger(idx)) { - entry.trigger = 1; - entry.mask = 1; - entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS); - } - irq = pin_2_irq(idx, apic, pin); add_pin_to_irq(irq, apic, pin); - if (!apic && !IO_APIC_IRQ(irq)) - continue; + setup_IO_APIC_irq(apic, pin, idx, irq); - if (IO_APIC_IRQ(irq)) { - cpumask_t mask; - vector = assign_irq_vector(irq, TARGET_CPUS, &mask); - if (vector < 0) - continue; - - entry.dest.logical.logical_dest = cpu_mask_to_apicid(mask); - entry.vector = vector; - - ioapic_register_intr(irq, vector, IOAPIC_AUTO); - if (!apic && (irq < 16)) - disable_8259A_irq(irq); - } - ioapic_write_entry(apic, pin, entry); - - spin_lock_irqsave(&ioapic_lock, flags); - set_native_irq_info(irq, TARGET_CPUS); - spin_unlock_irqrestore(&ioapic_lock, flags); } } @@ -2139,7 +2149,15 @@ void __init setup_ioapic_dest(void) if (irq_entry == -1) continue; irq = pin_2_irq(irq_entry, ioapic, pin); - set_ioapic_affinity_irq(irq, TARGET_CPUS); + + /* setup_IO_APIC_irqs could fail to get vector for some device + * when you have too many devices, because at that time only boot + * cpu is online. + */ + if(!irq_vector[irq]) + setup_IO_APIC_irq(ioapic, pin, irq_entry, irq); + else + set_ioapic_affinity_irq(irq, TARGET_CPUS); } } -- cgit v1.2.3 From b65780e123ba9b762276482bbfb52836e4d41fd9 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 7 Dec 2006 02:14:19 +0100 Subject: [PATCH] unwinder: move .eh_frame to RODATA The .eh_frame section contents is never written to, so it can as well benefit from CONFIG_DEBUG_RODATA. Diff-ed against firstfloor tree. Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen --- arch/x86_64/kernel/vmlinux.lds.S | 9 --------- 1 file changed, 9 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index d9534e750d4..6a1f8f491e5 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -51,15 +51,6 @@ SECTIONS RODATA -#ifdef CONFIG_STACK_UNWIND - . = ALIGN(8); - .eh_frame : AT(ADDR(.eh_frame) - LOAD_OFFSET) { - __start_unwind = .; - *(.eh_frame) - __end_unwind = .; - } -#endif - . = ALIGN(PAGE_SIZE); /* Align data segment to page size boundary */ /* Data */ .data : AT(ADDR(.data) - LOAD_OFFSET) { -- cgit v1.2.3 From 64a26a731235b59c9d73bbe82c1f896d57400d37 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:19 +0100 Subject: [PATCH] x86-64: Export smp_call_function_single smp_call_function() is exported, makes sense to export this one too. Signed-off-by: Andi Kleen --- arch/x86_64/kernel/smp.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c index 9f74c883568..1bf0e094f43 100644 --- a/arch/x86_64/kernel/smp.c +++ b/arch/x86_64/kernel/smp.c @@ -385,6 +385,7 @@ int smp_call_function_single (int cpu, void (*func) (void *info), void *info, put_cpu(); return 0; } +EXPORT_SYMBOL(smp_call_function_single); /* * this function sends a 'generic call function' IPI to all other CPUs -- cgit v1.2.3 From e94b1766097d53e6f3ccfb36c8baa562ffeda3fc Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:33:17 -0800 Subject: [PATCH] slab: remove SLAB_KERNEL SLAB_KERNEL is an alias of GFP_KERNEL. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/ia32/ia32_binfmt.c | 2 +- arch/x86_64/ia32/syscall32.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/ia32/ia32_binfmt.c b/arch/x86_64/ia32/ia32_binfmt.c index 82ef182de6a..932a62ad6c8 100644 --- a/arch/x86_64/ia32/ia32_binfmt.c +++ b/arch/x86_64/ia32/ia32_binfmt.c @@ -351,7 +351,7 @@ int ia32_setup_arg_pages(struct linux_binprm *bprm, unsigned long stack_top, bprm->loader += stack_base; bprm->exec += stack_base; - mpnt = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + mpnt = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!mpnt) return -ENOMEM; diff --git a/arch/x86_64/ia32/syscall32.c b/arch/x86_64/ia32/syscall32.c index 3a01329473a..3e5ed20cba4 100644 --- a/arch/x86_64/ia32/syscall32.c +++ b/arch/x86_64/ia32/syscall32.c @@ -49,7 +49,7 @@ int syscall32_setup_pages(struct linux_binprm *bprm, int exstack) struct mm_struct *mm = current->mm; int ret; - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!vma) return -ENOMEM; -- cgit v1.2.3 From f5738ceed46782aea7663d62cb6398eb05fc4ce0 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 6 Dec 2006 20:37:29 -0800 Subject: [PATCH] remove kernel syscalls The last thing we agreed on was to remove the macros entirely for 2.6.19, on all architectures. Unfortunately, I think nobody actually _did_ that, so they are still there. [akpm@osdl.org: x86_64 fix] Cc: David Woodhouse Cc: Greg Schafer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/vsyscall.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c index 92546c1526f..630036c06c7 100644 --- a/arch/x86_64/kernel/vsyscall.c +++ b/arch/x86_64/kernel/vsyscall.c @@ -42,6 +42,7 @@ #include #define __vsyscall(nr) __attribute__ ((unused,__section__(".vsyscall_" #nr))) +#define __syscall_clobber "r11","rcx","memory" int __sysctl_vsyscall __section_sysctl_vsyscall = 1; seqlock_t __xtime_lock __section_xtime_lock = SEQLOCK_UNLOCKED; -- cgit v1.2.3 From 19e5d9c0d2194b4b47189cbec2921cbf72b0bd1c Mon Sep 17 00:00:00 2001 From: Henry Nestler Date: Wed, 6 Dec 2006 20:37:45 -0800 Subject: [PATCH] initrd: remove unused false condition for initrd_start After LOADER_TYPE && INITRD_START are true, the short if-condition for INITRD_START can never be false. Remove unused code from the else condition. Signed-off-by: Henry Nestler Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/setup.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index fc944b5e8f4..f12f266f3e9 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -471,8 +471,7 @@ void __init setup_arch(char **cmdline_p) if (LOADER_TYPE && INITRD_START) { if (INITRD_START + INITRD_SIZE <= (end_pfn << PAGE_SHIFT)) { reserve_bootmem_generic(INITRD_START, INITRD_SIZE); - initrd_start = - INITRD_START ? INITRD_START + PAGE_OFFSET : 0; + initrd_start = INITRD_START + PAGE_OFFSET; initrd_end = initrd_start+INITRD_SIZE; } else { -- cgit v1.2.3 From 386d9a7edd9f3492c99124b0a659e9ed7abb30f9 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Wed, 6 Dec 2006 20:37:53 -0800 Subject: [PATCH] elf: Always define elf_addr_t in linux/elf.h Define elf_addr_t in linux/elf.h. The size of the type is determined using ELF_CLASS. This allows us to remove the defines that today are spread all over .c and .h files. Signed-off-by: Magnus Damm Cc: Daniel Jacobowitz Cc: Roland McGrath Cc: Jakub Jelinek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/ia32/ia32_binfmt.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/ia32/ia32_binfmt.c b/arch/x86_64/ia32/ia32_binfmt.c index 932a62ad6c8..543ef4f405e 100644 --- a/arch/x86_64/ia32/ia32_binfmt.c +++ b/arch/x86_64/ia32/ia32_binfmt.c @@ -305,8 +305,6 @@ MODULE_AUTHOR("Eric Youngdale, Andi Kleen"); #undef MODULE_DESCRIPTION #undef MODULE_AUTHOR -#define elf_addr_t __u32 - static void elf32_init(struct pt_regs *); #define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1 -- cgit v1.2.3 From b4c6c34a530b4d1c626f4ac0a884e0a9b849378c Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Wed, 6 Dec 2006 20:38:11 -0800 Subject: [PATCH] kprobes: enable booster on the preemptible kernel When we are unregistering a kprobe-booster, we can't release its instruction buffer immediately on the preemptive kernel, because some processes might be preempted on the buffer. The freeze_processes() and thaw_processes() functions can clean most of processes up from the buffer. There are still some non-frozen threads who have the PF_NOFREEZE flag. If those threads are sleeping (not preempted) at the known place outside the buffer, we can ensure safety of freeing. However, the processing of this check routine takes a long time. So, this patch introduces the garbage collection mechanism of insn_slot. It also introduces the "dirty" flag to free_insn_slot because of efficiency. The "clean" instruction slots (dirty flag is cleared) are released immediately. But the "dirty" slots which are used by boosted kprobes, are marked as garbages. collect_garbage_slots() will be invoked to release "dirty" slots if there are more than INSNS_PER_PAGE garbage slots or if there are no unused slots. Cc: "Keshavamurthy, Anil S" Cc: Ananth N Mavinakayanahalli Cc: "bibo,mao" Cc: Prasanna S Panchamukhi Cc: Yumiko Sugita Cc: Satoshi Oshima Cc: Hideo Aoki Signed-off-by: Masami Hiramatsu Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/kprobes.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c index ac241567e68..209c8c0bec7 100644 --- a/arch/x86_64/kernel/kprobes.c +++ b/arch/x86_64/kernel/kprobes.c @@ -224,7 +224,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) void __kprobes arch_remove_kprobe(struct kprobe *p) { mutex_lock(&kprobe_mutex); - free_insn_slot(p->ainsn.insn); + free_insn_slot(p->ainsn.insn, 0); mutex_unlock(&kprobe_mutex); } -- cgit v1.2.3 From a38a44c1a93078fc5fadc4ac2df8dea4697069e2 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 6 Dec 2006 20:38:16 -0800 Subject: [PATCH] smp_call_function_single() check that local interrupts are enabled smp_call_function_single() can deadlock if the caller disabled local interrupts (the target CPU could be spinning on call_lock). Check for that. Why on earth do these functions use spin_lock_bh()?? Cc: "Randy.Dunlap" Cc: Andi Kleen Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/smp.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c index 9f74c883568..32f4d7e2a06 100644 --- a/arch/x86_64/kernel/smp.c +++ b/arch/x86_64/kernel/smp.c @@ -379,6 +379,10 @@ int smp_call_function_single (int cpu, void (*func) (void *info), void *info, put_cpu(); return 0; } + + /* Can deadlock when called with interrupts disabled */ + WARN_ON(irqs_disabled()); + spin_lock_bh(&call_lock); __smp_call_function_single(cpu, func, info, nonatomic, wait); spin_unlock_bh(&call_lock); -- cgit v1.2.3 From 02316067852187b8bec781bec07410e91af79627 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 6 Dec 2006 20:38:17 -0800 Subject: [PATCH] hotplug CPU: clean up hotcpu_notifier() use There was lots of #ifdef noise in the kernel due to hotcpu_notifier(fn, prio) not correctly marking 'fn' as used in the !HOTPLUG_CPU case, and thus generating compiler warnings of unused symbols, hence forcing people to add #ifdefs. the compiler can skip truly unused functions just fine: text data bss dec hex filename 1624412 728710 3674856 6027978 5bfaca vmlinux.before 1624412 728710 3674856 6027978 5bfaca vmlinux.after [akpm@osdl.org: topology.c fix] Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/mce.c | 2 -- arch/x86_64/kernel/mce_amd.c | 4 ++-- arch/x86_64/kernel/vsyscall.c | 2 -- 3 files changed, 2 insertions(+), 6 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c index c7587fc3901..bc863c464a1 100644 --- a/arch/x86_64/kernel/mce.c +++ b/arch/x86_64/kernel/mce.c @@ -641,7 +641,6 @@ static __cpuinit int mce_create_device(unsigned int cpu) return err; } -#ifdef CONFIG_HOTPLUG_CPU static void mce_remove_device(unsigned int cpu) { int i; @@ -674,7 +673,6 @@ mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) static struct notifier_block mce_cpu_notifier = { .notifier_call = mce_cpu_callback, }; -#endif static __init int mce_init_device(void) { diff --git a/arch/x86_64/kernel/mce_amd.c b/arch/x86_64/kernel/mce_amd.c index 883fe747f64..fa09debad4b 100644 --- a/arch/x86_64/kernel/mce_amd.c +++ b/arch/x86_64/kernel/mce_amd.c @@ -551,7 +551,6 @@ out: return err; } -#ifdef CONFIG_HOTPLUG_CPU /* * let's be hotplug friendly. * in case of multiple core processors, the first core always takes ownership @@ -594,12 +593,14 @@ static void threshold_remove_bank(unsigned int cpu, int bank) sprintf(name, "threshold_bank%i", bank); +#ifdef CONFIG_SMP /* sibling symlink */ if (shared_bank[bank] && b->blocks->cpu != cpu) { sysfs_remove_link(&per_cpu(device_mce, cpu).kobj, name); per_cpu(threshold_banks, cpu)[bank] = NULL; return; } +#endif /* remove all sibling symlinks before unregistering */ for_each_cpu_mask(i, b->cpus) { @@ -656,7 +657,6 @@ static int threshold_cpu_callback(struct notifier_block *nfb, static struct notifier_block threshold_cpu_notifier = { .notifier_call = threshold_cpu_callback, }; -#endif /* CONFIG_HOTPLUG_CPU */ static __init int threshold_init_device(void) { diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c index 630036c06c7..3785e495473 100644 --- a/arch/x86_64/kernel/vsyscall.c +++ b/arch/x86_64/kernel/vsyscall.c @@ -275,7 +275,6 @@ static void __cpuinit cpu_vsyscall_init(void *arg) vsyscall_set_cpu(raw_smp_processor_id()); } -#ifdef CONFIG_HOTPLUG_CPU static int __cpuinit cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg) { @@ -284,7 +283,6 @@ cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg) smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 0, 1); return NOTIFY_DONE; } -#endif static void __init map_vsyscall(void) { -- cgit v1.2.3 From 85916f8166b59eeac63d2b4f7f1df8de849334b4 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Wed, 6 Dec 2006 20:40:41 -0800 Subject: [PATCH] Kexec / Kdump: Unify elf note code The elf note saving code is currently duplicated over several architectures. This cleanup patch simply adds code to a common file and then replaces the arch-specific code with calls to the newly added code. The only drawback with this approach is that s390 doesn't fully support kexec-on-panic which for that arch leads to introduction of unused code. Signed-off-by: Magnus Damm Cc: Vivek Goyal Cc: Andi Kleen Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/crash.c | 69 ++-------------------------------------------- 1 file changed, 2 insertions(+), 67 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c index 3525f884af8..95a7a2c1313 100644 --- a/arch/x86_64/kernel/crash.c +++ b/arch/x86_64/kernel/crash.c @@ -28,71 +28,6 @@ /* This keeps a track of which one is crashing cpu. */ static int crashing_cpu; -static u32 *append_elf_note(u32 *buf, char *name, unsigned type, - void *data, size_t data_len) -{ - struct elf_note note; - - note.n_namesz = strlen(name) + 1; - note.n_descsz = data_len; - note.n_type = type; - memcpy(buf, ¬e, sizeof(note)); - buf += (sizeof(note) +3)/4; - memcpy(buf, name, note.n_namesz); - buf += (note.n_namesz + 3)/4; - memcpy(buf, data, note.n_descsz); - buf += (note.n_descsz + 3)/4; - - return buf; -} - -static void final_note(u32 *buf) -{ - struct elf_note note; - - note.n_namesz = 0; - note.n_descsz = 0; - note.n_type = 0; - memcpy(buf, ¬e, sizeof(note)); -} - -static void crash_save_this_cpu(struct pt_regs *regs, int cpu) -{ - struct elf_prstatus prstatus; - u32 *buf; - - if ((cpu < 0) || (cpu >= NR_CPUS)) - return; - - /* Using ELF notes here is opportunistic. - * I need a well defined structure format - * for the data I pass, and I need tags - * on the data to indicate what information I have - * squirrelled away. ELF notes happen to provide - * all of that, no need to invent something new. - */ - - buf = (u32*)per_cpu_ptr(crash_notes, cpu); - - if (!buf) - return; - - memset(&prstatus, 0, sizeof(prstatus)); - prstatus.pr_pid = current->pid; - elf_core_copy_regs(&prstatus.pr_reg, regs); - buf = append_elf_note(buf, "CORE", NT_PRSTATUS, &prstatus, - sizeof(prstatus)); - final_note(buf); -} - -static void crash_save_self(struct pt_regs *regs) -{ - int cpu; - - cpu = smp_processor_id(); - crash_save_this_cpu(regs, cpu); -} - #ifdef CONFIG_SMP static atomic_t waiting_for_crash_ipi; @@ -117,7 +52,7 @@ static int crash_nmi_callback(struct notifier_block *self, return NOTIFY_STOP; local_irq_disable(); - crash_save_this_cpu(regs, cpu); + crash_save_cpu(regs, cpu); disable_local_APIC(); atomic_dec(&waiting_for_crash_ipi); /* Assume hlt works */ @@ -196,5 +131,5 @@ void machine_crash_shutdown(struct pt_regs *regs) disable_IO_APIC(); - crash_save_self(regs); + crash_save_cpu(regs, smp_processor_id()); } -- cgit v1.2.3 From c31a0bf3e1bc581676618db7492f18798fd0a73f Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Fri, 8 Dec 2006 02:36:22 -0800 Subject: [PATCH] Generic BUG for x86-64 This makes x86-64 use the generic BUG machinery. The main advantage in using the generic BUG machinery for x86-64 is that the inlined overhead of BUG is just the ud2a instruction; the file+line information are no longer inlined into the instruction stream. This reduces cache pollution. Signed-off-by: Jeremy Fitzhardinge Cc: Andi Kleen Cc: Hugh Dickens Cc: Michael Ellerman Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: Rusty Russell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/Kconfig | 5 +++++ arch/x86_64/kernel/module.c | 5 ++++- arch/x86_64/kernel/traps.c | 36 ++++++++++++------------------------ arch/x86_64/kernel/vmlinux.lds.S | 2 ++ 4 files changed, 23 insertions(+), 25 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index bfbb9bcae12..3254a616c69 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -96,6 +96,11 @@ config AUDIT_ARCH bool default y +config GENERIC_BUG + bool + default y + depends on BUG + source "init/Kconfig" diff --git a/arch/x86_64/kernel/module.c b/arch/x86_64/kernel/module.c index 9d0958ff547..a888e67f587 100644 --- a/arch/x86_64/kernel/module.c +++ b/arch/x86_64/kernel/module.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include @@ -173,10 +174,12 @@ int module_finalize(const Elf_Ehdr *hdr, lseg, lseg + locks->sh_size, tseg, tseg + text->sh_size); } - return 0; + + return module_bug_finalize(hdr, sechdrs, me); } void module_arch_cleanup(struct module *mod) { alternatives_smp_module_del(mod); + module_bug_cleanup(mod); } diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index a1641ffdffc..b54ccc07f37 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -31,6 +31,7 @@ #include #include #include +#include #include #include @@ -524,30 +525,15 @@ bad: printk("\n"); } -void handle_BUG(struct pt_regs *regs) -{ - struct bug_frame f; - long len; - const char *prefix = ""; +int is_valid_bugaddr(unsigned long rip) +{ + unsigned short ud2; - if (user_mode(regs)) - return; - if (__copy_from_user(&f, (const void __user *) regs->rip, - sizeof(struct bug_frame))) - return; - if (f.filename >= 0 || - f.ud2[0] != 0x0f || f.ud2[1] != 0x0b) - return; - len = __strnlen_user((char *)(long)f.filename, PATH_MAX) - 1; - if (len < 0 || len >= PATH_MAX) - f.filename = (int)(long)"unmapped filename"; - else if (len > 50) { - f.filename += len - 50; - prefix = "..."; - } - printk("----------- [cut here ] --------- [please bite here ] ---------\n"); - printk(KERN_ALERT "Kernel BUG at %s%.50s:%d\n", prefix, (char *)(long)f.filename, f.line); -} + if (__copy_from_user(&ud2, (const void __user *) rip, sizeof(ud2))) + return 0; + + return ud2 == 0x0b0f; +} #ifdef CONFIG_BUG void out_of_line_bug(void) @@ -627,7 +613,9 @@ void die(const char * str, struct pt_regs * regs, long err) { unsigned long flags = oops_begin(); - handle_BUG(regs); + if (!user_mode(regs)) + report_bug(regs->rip); + __die(str, regs, err); oops_end(flags); do_exit(SIGSEGV); diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index 6a1f8f491e5..6c417788c54 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -51,6 +51,8 @@ SECTIONS RODATA + BUG_TABLE + . = ALIGN(PAGE_SIZE); /* Align data segment to page size boundary */ /* Data */ .data : AT(ADDR(.data) - LOAD_OFFSET) { -- cgit v1.2.3 From c941192aafe053ae445fb22f5d89bd34794c5442 Mon Sep 17 00:00:00 2001 From: "Josef \"Jeff\" Sipek" Date: Fri, 8 Dec 2006 02:36:43 -0800 Subject: [PATCH] x86_64: change uses of f_{dentry, vfsmnt} to use f_path Change all the uses of f_{dentry,vfsmnt} to f_path.{dentry,mnt} in the x86_64 arch code. Signed-off-by: Josef "Jeff" Sipek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/ia32/ia32_aout.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/ia32/ia32_aout.c b/arch/x86_64/ia32/ia32_aout.c index 396d3c10001..be87df506f3 100644 --- a/arch/x86_64/ia32/ia32_aout.c +++ b/arch/x86_64/ia32/ia32_aout.c @@ -272,7 +272,7 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs) if ((N_MAGIC(ex) != ZMAGIC && N_MAGIC(ex) != OMAGIC && N_MAGIC(ex) != QMAGIC && N_MAGIC(ex) != NMAGIC) || N_TRSIZE(ex) || N_DRSIZE(ex) || - i_size_read(bprm->file->f_dentry->d_inode) < ex.a_text+ex.a_data+N_SYMSIZE(ex)+N_TXTOFF(ex)) { + i_size_read(bprm->file->f_path.dentry->d_inode) < ex.a_text+ex.a_data+N_SYMSIZE(ex)+N_TXTOFF(ex)) { return -ENOEXEC; } @@ -357,7 +357,7 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs) { printk(KERN_WARNING "fd_offset is not page aligned. Please convert program: %s\n", - bprm->file->f_dentry->d_name.name); + bprm->file->f_path.dentry->d_name.name); error_time = jiffies; } #endif @@ -440,7 +440,7 @@ static int load_aout_library(struct file *file) int retval; struct exec ex; - inode = file->f_dentry->d_inode; + inode = file->f_path.dentry->d_inode; retval = -ENOEXEC; error = kernel_read(file, 0, (char *) &ex, sizeof(ex)); @@ -471,7 +471,7 @@ static int load_aout_library(struct file *file) { printk(KERN_WARNING "N_TXTOFF is not page aligned. Please convert library: %s\n", - file->f_dentry->d_name.name); + file->f_path.dentry->d_name.name); error_time = jiffies; } #endif -- cgit v1.2.3 From f0d1b0b30d250a07627ad8b9fbbb5c7cc08422e8 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 8 Dec 2006 02:37:49 -0800 Subject: [PATCH] LOG2: Implement a general integer log2 facility in the kernel This facility provides three entry points: ilog2() Log base 2 of unsigned long ilog2_u32() Log base 2 of u32 ilog2_u64() Log base 2 of u64 These facilities can either be used inside functions on dynamic data: int do_something(long q) { ...; y = ilog2(x) ...; } Or can be used to statically initialise global variables with constant values: unsigned n = ilog2(27); When performing static initialisation, the compiler will report "error: initializer element is not constant" if asked to take a log of zero or of something not reducible to a constant. They treat negative numbers as unsigned. When not dealing with a constant, they fall back to using fls() which permits them to use arch-specific log calculation instructions - such as BSR on x86/x86_64 or SCAN on FRV - if available. [akpm@osdl.org: MMC fix] Signed-off-by: David Howells Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Herbert Xu Cc: David Howells Cc: Wojtek Kaniewski Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/Kconfig | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index 3254a616c69..3ac581d1720 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -101,6 +101,14 @@ config GENERIC_BUG default y depends on BUG +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + source "init/Kconfig" -- cgit v1.2.3 From 9f25441fe6a4d8fc080a1f814078ac5d67adeea4 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 9 Dec 2006 21:33:35 +0100 Subject: [PATCH] x86-64: Update defconfig Signed-off-by: Andi Kleen --- arch/x86_64/defconfig | 29 ++++++++++++++++++++++------- 1 file changed, 22 insertions(+), 7 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/defconfig b/arch/x86_64/defconfig index 96f226cfb33..1a1c6a1a299 100644 --- a/arch/x86_64/defconfig +++ b/arch/x86_64/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.19-git7 -# Wed Dec 6 23:50:47 2006 +# Linux kernel version: 2.6.19-git14 +# Sat Dec 9 21:23:09 2006 # CONFIG_X86_64=y CONFIG_64BIT=y @@ -22,6 +22,9 @@ CONFIG_ARCH_MAY_HAVE_PC_FDC=y CONFIG_ARCH_POPULATES_NODE_MAP=y CONFIG_DMI=y CONFIG_AUDIT_ARCH=y +CONFIG_GENERIC_BUG=y +# CONFIG_ARCH_HAS_ILOG2_U32 is not set +# CONFIG_ARCH_HAS_ILOG2_U64 is not set CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" # @@ -170,6 +173,7 @@ CONFIG_SECCOMP=y # CONFIG_CC_STACKPROTECTOR is not set # CONFIG_HZ_100 is not set CONFIG_HZ_250=y +# CONFIG_HZ_300 is not set # CONFIG_HZ_1000 is not set CONFIG_HZ=250 # CONFIG_REORDER is not set @@ -514,6 +518,7 @@ CONFIG_IDEDMA_AUTO=y # # CONFIG_RAID_ATTRS is not set CONFIG_SCSI=y +# CONFIG_SCSI_TGT is not set CONFIG_SCSI_NETLINK=y # CONFIG_SCSI_PROC_FS is not set @@ -534,6 +539,7 @@ CONFIG_CHR_DEV_SG=y # CONFIG_SCSI_MULTI_LUN is not set CONFIG_SCSI_CONSTANTS=y # CONFIG_SCSI_LOGGING is not set +# CONFIG_SCSI_SCAN_ASYNC is not set # # SCSI Transports @@ -587,6 +593,7 @@ CONFIG_MEGARAID_SAS=y # CONFIG_SCSI_DC395x is not set # CONFIG_SCSI_DC390T is not set # CONFIG_SCSI_DEBUG is not set +# CONFIG_SCSI_SRP is not set # # Serial ATA (prod) and Parallel ATA (experimental) drivers @@ -1106,10 +1113,7 @@ CONFIG_SOUND=y # Open Sound System # CONFIG_SOUND_PRIME=y -CONFIG_OSS_OBSOLETE_DRIVER=y # CONFIG_SOUND_BT878 is not set -# CONFIG_SOUND_EMU10K1 is not set -# CONFIG_SOUND_FUSION is not set # CONFIG_SOUND_ES1371 is not set CONFIG_SOUND_ICH=y # CONFIG_SOUND_TRIDENT is not set @@ -1118,6 +1122,11 @@ CONFIG_SOUND_ICH=y # CONFIG_SOUND_VIA82CXXX is not set # CONFIG_SOUND_OSS is not set +# +# HID Devices +# +CONFIG_HID=y + # # USB support # @@ -1182,8 +1191,7 @@ CONFIG_USB_STORAGE=y # USB Input Devices # CONFIG_USB_HID=y -CONFIG_USB_HIDINPUT=y -# CONFIG_USB_HIDINPUT_POWERBOOK is not set +# CONFIG_USB_HID_POWERBOOK is not set # CONFIG_HID_FF is not set # CONFIG_USB_HIDDEV is not set # CONFIG_USB_AIPTEK is not set @@ -1474,6 +1482,11 @@ CONFIG_NLS_ISO8859_15=y # CONFIG_NLS_KOI8_U is not set CONFIG_NLS_UTF8=y +# +# Distributed Lock Manager +# +# CONFIG_DLM is not set + # # Instrumentation Support # @@ -1504,6 +1517,7 @@ CONFIG_DETECT_SOFTLOCKUP=y # CONFIG_DEBUG_SPINLOCK_SLEEP is not set # CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set # CONFIG_DEBUG_KOBJECT is not set +CONFIG_DEBUG_BUGVERBOSE=y # CONFIG_DEBUG_INFO is not set CONFIG_DEBUG_FS=y # CONFIG_DEBUG_VM is not set @@ -1534,6 +1548,7 @@ CONFIG_DEBUG_STACKOVERFLOW=y # # Library routines # +CONFIG_BITREVERSE=y # CONFIG_CRC_CCITT is not set # CONFIG_CRC16 is not set CONFIG_CRC32=y -- cgit v1.2.3 From 92715e282be7c7488f892703c8d39b08976a833b Mon Sep 17 00:00:00 2001 From: Ravikiran G Thirumalai Date: Sat, 9 Dec 2006 21:33:35 +0100 Subject: [PATCH] x86: Fix boot hang due to nmi watchdog init code 2.6.19 stopped booting (or booted based on build/config) on our x86_64 systems due to a bug introduced in 2.6.19. check_nmi_watchdog schedules an IPI on all cpus to busy wait on a flag, but fails to set the busywait flag if NMI functionality is disabled. This causes the secondary cpus to spin in an endless loop, causing the kernel bootup to hang. Depending upon the build, the busywait flag got overwritten (stack variable) and caused the kernel to bootup on certain builds. Following patch fixes the bug by setting the busywait flag before returning from check_nmi_watchdog. I guess using a stack variable is not good here as the calling function could potentially return while the busy wait loop is still spinning on the flag. AK: I redid the patch significantly to be cleaner Signed-off-by: Ravikiran Thirumalai Signed-off-by: Shai Fultheim Signed-off-by: Andi Kleen --- arch/x86_64/kernel/nmi.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c index 27e95e7922c..186aebbae32 100644 --- a/arch/x86_64/kernel/nmi.c +++ b/arch/x86_64/kernel/nmi.c @@ -193,6 +193,8 @@ void nmi_watchdog_default(void) nmi_watchdog = NMI_IO_APIC; } +static int endflag __initdata = 0; + #ifdef CONFIG_SMP /* The performance counters used by NMI_LOCAL_APIC don't trigger when * the CPU is idle. To make sure the NMI watchdog really ticks on all @@ -200,7 +202,6 @@ void nmi_watchdog_default(void) */ static __init void nmi_cpu_busy(void *data) { - volatile int *endflag = data; local_irq_enable_in_hardirq(); /* Intentionally don't use cpu_relax here. This is to make sure that the performance counter really ticks, @@ -208,14 +209,13 @@ static __init void nmi_cpu_busy(void *data) pause instruction. On a real HT machine this is fine because all other CPUs are busy with "useless" delay loops and don't care if they get somewhat less cycles. */ - while (*endflag == 0) - barrier(); + while (endflag == 0) + mb(); } #endif int __init check_nmi_watchdog (void) { - volatile int endflag = 0; int *counts; int cpu; @@ -256,6 +256,7 @@ int __init check_nmi_watchdog (void) if (!atomic_read(&nmi_active)) { kfree(counts); atomic_set(&nmi_active, -1); + endflag = 1; return -1; } endflag = 1; -- cgit v1.2.3 From 1bac3b383a93a4a920ffc57441eb133c78567fbd Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 9 Dec 2006 21:33:36 +0100 Subject: [PATCH] x86: Work around gcc 4.2 over aggressive optimizer The new PDA code uses a dummy _proxy_pda variable to describe memory references to the PDA. It is never referenced in inline assembly, but exists as input/output arguments. gcc 4.2 in some cases can CSE references to this which causes unresolved symbols. Define it to zero to avoid this. Signed-off-by: Andi Kleen --- arch/x86_64/kernel/vmlinux.lds.S | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S index 6c417788c54..514be5dd230 100644 --- a/arch/x86_64/kernel/vmlinux.lds.S +++ b/arch/x86_64/kernel/vmlinux.lds.S @@ -13,6 +13,7 @@ OUTPUT_FORMAT("elf64-x86-64", "elf64-x86-64", "elf64-x86-64") OUTPUT_ARCH(i386:x86-64) ENTRY(phys_startup_64) jiffies_64 = jiffies; +_proxy_pda = 0; PHDRS { text PT_LOAD FLAGS(5); /* R_E */ data PT_LOAD FLAGS(7); /* RWE */ -- cgit v1.2.3 From 1f29bcd739972f71f2fd5d5d265daf3e1208fa5e Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sun, 10 Dec 2006 02:19:10 -0800 Subject: [PATCH] sysctl: remove unused "context" param Signed-off-by: Alexey Dobriyan Cc: Andi Kleen Cc: "David S. Miller" Cc: David Howells Cc: Ralf Baechle Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/kernel/vsyscall.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c index 4a673f5397a..2433d6fc68b 100644 --- a/arch/x86_64/kernel/vsyscall.c +++ b/arch/x86_64/kernel/vsyscall.c @@ -225,8 +225,7 @@ out: static int vsyscall_sysctl_nostrat(ctl_table *t, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context) + void __user *newval, size_t newlen) { return -ENOSYS; } -- cgit v1.2.3 From e45116b8d71ece9dbe41b114368ff7aebe3ae41a Mon Sep 17 00:00:00 2001 From: Brice Goglin Date: Mon, 11 Dec 2006 20:14:15 +0100 Subject: [PATCH] Fix typo in 'EXPERIMENTAL' in CC_STACKPROTECTOR on x86_64 Fix typo in 'EXPERIMENTAL' in config CC_STACKPROTECTOR in arch/x86_64/Kconfig. Signed-off-by: Brice Goglin Signed-off-by: Linus Torvalds --- arch/x86_64/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index 3ac581d1720..d4275537b25 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -584,7 +584,7 @@ config SECCOMP If unsure, say Y. Only embedded should say N here. config CC_STACKPROTECTOR - bool "Enable -fstack-protector buffer overflow detection (EXPRIMENTAL)" + bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)" depends on EXPERIMENTAL help This option turns on the -fstack-protector GCC feature. This -- cgit v1.2.3