From 79f3b3cb7a2586b319a43a7f29924c6c555e4357 Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Thu, 8 Jan 2009 10:04:30 -0500 Subject: x86, mtrr: fix types used in userspace exported header Commit 932d27a7913fc6b3c64c6e6082628b0a1561dec9 exported some mtrr structures without using the exportable __uX types, causing userspace build failures. Signed-off-by: Kyle McMartin Signed-off-by: Ingo Molnar --- arch/x86/include/asm/mtrr.h | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/mtrr.h b/arch/x86/include/asm/mtrr.h index cb988aab716..14080d22edb 100644 --- a/arch/x86/include/asm/mtrr.h +++ b/arch/x86/include/asm/mtrr.h @@ -58,15 +58,15 @@ struct mtrr_gentry { #endif /* !__i386__ */ struct mtrr_var_range { - u32 base_lo; - u32 base_hi; - u32 mask_lo; - u32 mask_hi; + __u32 base_lo; + __u32 base_hi; + __u32 mask_lo; + __u32 mask_hi; }; /* In the Intel processor's MTRR interface, the MTRR type is always held in an 8 bit field: */ -typedef u8 mtrr_type; +typedef __u8 mtrr_type; #define MTRR_NUM_FIXED_RANGES 88 #define MTRR_MAX_VAR_RANGES 256 -- cgit v1.2.3 From 6d612b0f943289856c6e8186c564cda922cd040e Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 12 Jan 2009 12:52:23 +0100 Subject: locking, hpet: annotate false positive warning Alexander Beregalov reported that this warning is caused by the HPET code: > hpet0: at MMIO 0xfed00000, IRQs 2, 8, 0 > hpet0: 3 comparators, 64-bit 14.318180 MHz counter > ODEBUG: object is on stack, but not annotated > ------------[ cut here ]------------ > WARNING: at lib/debugobjects.c:251 __debug_object_init+0x2a4/0x352() > Bisected down to 26afe5f2fbf06ea0765aaa316640c4dd472310c0 > (x86: HPET_MSI Initialise per-cpu HPET timers) The commit is fine - but the on-stack workqueue entry needs annotation. Reported-and-bisected-by: Alexander Beregalov Signed-off-by: Peter Zijlstra Tested-by: Alexander Beregalov Signed-off-by: Ingo Molnar --- arch/x86/kernel/hpet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index cd759ad9069..bb2e0f0975a 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -628,7 +628,7 @@ static int hpet_cpuhp_notify(struct notifier_block *n, switch (action & 0xf) { case CPU_ONLINE: - INIT_DELAYED_WORK(&work.work, hpet_work); + INIT_DELAYED_WORK_ON_STACK(&work.work, hpet_work); init_completion(&work.complete); /* FIXME: add schedule_work_on() */ schedule_delayed_work_on(cpu, &work.work, 0); -- cgit v1.2.3 From f313e12308f7c5ea645f18e759d104d088b18615 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 9 Jan 2009 12:17:43 -0800 Subject: x86: avoid theoretical vmalloc fault loop Ajith Kumar noticed: I was going through the vmalloc fault handling for x86_64 and am unclear about the following lines in the vmalloc_fault() function. pgd = pgd_offset(current->mm ?: &init_mm, address); pgd_ref = pgd_offset_k(address); Here the intention is to get the pgd corresponding to the current process and sync it up with the pgd in init_mm(obtained from pgd_offset_k). However, for kernel threads current->mm is NULL and hence pgd = pgd_offset(init_mm, address) = pgd_ref which means the fault handler returns without setting the pgd entry in the MM structure in the context of which the kernel thread has faulted. This could lead to never-ending faults and busy looping of kernel threads like pdflush. So, shouldn't the pgd = pgd_offset(current->mm ?: &init_mm, address); be pgd = pgd_offset(current->active_mm ?: &init_mm, address); We can use active_mm unconditionally because it should be always set. Signed-off-by: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Ingo Molnar --- arch/x86/mm/fault.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 9e268b6b204..90dfae511a4 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -534,7 +534,7 @@ static int vmalloc_fault(unsigned long address) happen within a race in page table update. In the later case just flush. */ - pgd = pgd_offset(current->mm ?: &init_mm, address); + pgd = pgd_offset(current->active_mm, address); pgd_ref = pgd_offset_k(address); if (pgd_none(*pgd_ref)) return -1; -- cgit v1.2.3 From 4884d8e6a05026ec906355436cea9dc1acb1d09e Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Rajput Date: Sun, 11 Jan 2009 18:38:55 +0530 Subject: x86: fix mpparse.c build error on latest git Fix this by reintroducing asm/smp.h include in mpparse.c - later on I will fix this by removing non-smp data from smp.h. Reported-by: Petr Titera Signed-off-by: Jaswinder Singh Rajput Signed-off-by: Ingo Molnar --- arch/x86/kernel/mpparse.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index c0601c2848a..a649a4ccad4 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #ifdef CONFIG_X86_32 -- cgit v1.2.3 From 2bc1379712e74c5b99adaa6db433c14d8841ab4f Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Rajput Date: Sun, 11 Jan 2009 20:34:47 +0530 Subject: x86: fix apic.c build error on latest git Fix this by reintroducing asm/smp.h include in apic.c - later on I will fix this by removing non-smp data from smp.h Also fix the __inquire_remote_apic() prototype/inline. Signed-off-by: Jaswinder Singh Rajput Signed-off-by: Ingo Molnar --- arch/x86/include/asm/mach-default/mach_wakecpu.h | 6 ++++++ arch/x86/kernel/apic.c | 1 + 2 files changed, 7 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/mach-default/mach_wakecpu.h b/arch/x86/include/asm/mach-default/mach_wakecpu.h index ceb01366014..89897a6a65b 100644 --- a/arch/x86/include/asm/mach-default/mach_wakecpu.h +++ b/arch/x86/include/asm/mach-default/mach_wakecpu.h @@ -24,7 +24,13 @@ static inline void restore_NMI_vector(unsigned short *high, unsigned short *low) { } +#ifdef CONFIG_SMP extern void __inquire_remote_apic(int apicid); +#else /* CONFIG_SMP */ +static inline void __inquire_remote_apic(int apicid) +{ +} +#endif /* CONFIG_SMP */ static inline void inquire_remote_apic(int apicid) { diff --git a/arch/x86/kernel/apic.c b/arch/x86/kernel/apic.c index 566a08466b1..0f830e4f567 100644 --- a/arch/x86/kernel/apic.c +++ b/arch/x86/kernel/apic.c @@ -47,6 +47,7 @@ #include #include #include +#include #include #include -- cgit v1.2.3 From 50c668d678fd01284799a6e4f1b91829d83cb9ed Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Mon, 12 Jan 2009 10:49:53 +0100 Subject: Revert "cpumask: use work_on_cpu in acpi-cpufreq.c for drv_read and drv_write" This reverts commit 7503bfbae89eba07b46441a5d1594647f6b8ab7d. Dieter Ries reported bootup soft-hangs and bisected it back to this commit, and reverting this commit gave him a working system. The commit introduces work_on_cpu() use into the cpufreq code, but that is subtly problematic from a lock hierarchy POV: the hotplug-cpu lock is an highlevel lock that is taken before lowlevel locks, and in this codepath we are called with the policy lock taken. Dieter did not have lockdep enabled so we dont have a nice stack trace proof for this, but using work_on_cpu() in such a lowlevel place certainly looks wrong, so we revert the patch. work_on_cpu() needs to be reworked to be more generally usable. Reported-by: Dieter Ries Tested-by: Dieter Ries Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c | 25 ++++++++++++------------- 1 file changed, 12 insertions(+), 13 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c index 06fcd8f9323..6f11e029e8c 100644 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -150,9 +150,8 @@ struct drv_cmd { u32 val; }; -static long do_drv_read(void *_cmd) +static void do_drv_read(struct drv_cmd *cmd) { - struct drv_cmd *cmd = _cmd; u32 h; switch (cmd->type) { @@ -167,12 +166,10 @@ static long do_drv_read(void *_cmd) default: break; } - return 0; } -static long do_drv_write(void *_cmd) +static void do_drv_write(struct drv_cmd *cmd) { - struct drv_cmd *cmd = _cmd; u32 lo, hi; switch (cmd->type) { @@ -189,23 +186,30 @@ static long do_drv_write(void *_cmd) default: break; } - return 0; } static void drv_read(struct drv_cmd *cmd) { + cpumask_t saved_mask = current->cpus_allowed; cmd->val = 0; - work_on_cpu(cpumask_any(cmd->mask), do_drv_read, cmd); + set_cpus_allowed_ptr(current, cmd->mask); + do_drv_read(cmd); + set_cpus_allowed_ptr(current, &saved_mask); } static void drv_write(struct drv_cmd *cmd) { + cpumask_t saved_mask = current->cpus_allowed; unsigned int i; for_each_cpu(i, cmd->mask) { - work_on_cpu(i, do_drv_write, cmd); + set_cpus_allowed_ptr(current, cpumask_of(i)); + do_drv_write(cmd); } + + set_cpus_allowed_ptr(current, &saved_mask); + return; } static u32 get_cur_val(const struct cpumask *mask) @@ -231,15 +235,10 @@ static u32 get_cur_val(const struct cpumask *mask) return 0; } - if (unlikely(!alloc_cpumask_var(&cmd.mask, GFP_KERNEL))) - return 0; - cpumask_copy(cmd.mask, mask); drv_read(&cmd); - free_cpumask_var(cmd.mask); - dprintk("get_cur_val = %u\n", cmd.val); return cmd.val; -- cgit v1.2.3 From e8cea892dff8e3ebed42954c46730309b617196f Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Mon, 12 Jan 2009 19:36:59 +0100 Subject: Revert "i386: add TRACE_IRQS_OFF for the nmi" This reverts commit e0c7317557c8fc8eacf611e30c2a80f4e24e47a3. This patch was wrong, as lockdep (and thus the irq state tracer) aren't nmi safe. People are already seeing lockdep warnings due to this. Signed-off-by: Ingo Molnar --- arch/x86/kernel/entry_32.S | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index d6f0490a739..46469029e9d 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -1203,7 +1203,6 @@ nmi_stack_correct: pushl %eax CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL - TRACE_IRQS_OFF xorl %edx,%edx # zero error code movl %esp,%eax # pt_regs pointer call do_nmi @@ -1244,7 +1243,6 @@ nmi_espfix_stack: pushl %eax CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL - TRACE_IRQS_OFF FIXUP_ESPFIX_STACK # %eax == %esp xorl %edx,%edx # zero error code call do_nmi -- cgit v1.2.3 From 4a922a969cb0190ce4580d4b064e2ac35f3ac9bf Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 13 Jan 2009 16:11:00 +0100 Subject: x86, cpufreq: remove leftover copymask_copy() Impact: fix potential boot crash on MAXSMP Remove code left over by: 50c668d: Revert "cpumask: use work_on_cpu in acpi-cpufreq.c for drv_read That cmd.cpumask is not allocated anymore. No impact on default !MAXSMP kernels. Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c index 6f11e029e8c..8f3c95c7e61 100644 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -235,8 +235,6 @@ static u32 get_cur_val(const struct cpumask *mask) return 0; } - cpumask_copy(cmd.mask, mask); - drv_read(&cmd); dprintk("get_cur_val = %u\n", cmd.val); -- cgit v1.2.3 From c8399943bdb70fef78798b97f975506ecc99e039 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Mon, 12 Jan 2009 23:01:15 +0100 Subject: x86, generic: mark complex bitops.h inlines as __always_inline Impact: reduce kernel image size Hugh Dickins noticed that older gcc versions when the kernel is built for code size didn't inline some of the bitops. Mark all complex x86 bitops that have more than a single asm statement or two as always inline to avoid this problem. Probably should be done for other architectures too. Ingo then found a better fix that only requires a single line change, but it unfortunately only works on gcc 4.3. On older gccs the original patch still makes a ~0.3% defconfig difference with CONFIG_OPTIMIZE_INLINING=y. With gcc 4.1 and a defconfig like build: 6116998 1138540 883788 8139326 7c323e vmlinux-oi-with-patch 6137043 1138540 883788 8159371 7c808b vmlinux-optimize-inlining ~20k / 0.3% difference. Signed-off-by: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Ingo Molnar --- arch/x86/include/asm/bitops.h | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/bitops.h b/arch/x86/include/asm/bitops.h index e02a359d2aa..02b47a603fc 100644 --- a/arch/x86/include/asm/bitops.h +++ b/arch/x86/include/asm/bitops.h @@ -3,6 +3,9 @@ /* * Copyright 1992, Linus Torvalds. + * + * Note: inlines with more than a single statement should be marked + * __always_inline to avoid problems with older gcc's inlining heuristics. */ #ifndef _LINUX_BITOPS_H @@ -53,7 +56,8 @@ * Note that @nr may be almost arbitrarily large; this function is not * restricted to acting on a single-word quantity. */ -static inline void set_bit(unsigned int nr, volatile unsigned long *addr) +static __always_inline void +set_bit(unsigned int nr, volatile unsigned long *addr) { if (IS_IMMEDIATE(nr)) { asm volatile(LOCK_PREFIX "orb %1,%0" @@ -90,7 +94,8 @@ static inline void __set_bit(int nr, volatile unsigned long *addr) * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit() * in order to ensure changes are visible on other processors. */ -static inline void clear_bit(int nr, volatile unsigned long *addr) +static __always_inline void +clear_bit(int nr, volatile unsigned long *addr) { if (IS_IMMEDIATE(nr)) { asm volatile(LOCK_PREFIX "andb %1,%0" @@ -204,7 +209,8 @@ static inline int test_and_set_bit(int nr, volatile unsigned long *addr) * * This is the same as test_and_set_bit on x86. */ -static inline int test_and_set_bit_lock(int nr, volatile unsigned long *addr) +static __always_inline int +test_and_set_bit_lock(int nr, volatile unsigned long *addr) { return test_and_set_bit(nr, addr); } @@ -300,7 +306,7 @@ static inline int test_and_change_bit(int nr, volatile unsigned long *addr) return oldbit; } -static inline int constant_test_bit(unsigned int nr, const volatile unsigned long *addr) +static __always_inline int constant_test_bit(unsigned int nr, const volatile unsigned long *addr) { return ((1UL << (nr % BITS_PER_LONG)) & (((unsigned long *)addr)[nr / BITS_PER_LONG])) != 0; -- cgit v1.2.3 From afc7d20c8429f32f19d47367fdc36eeed2334ec3 Mon Sep 17 00:00:00 2001 From: "venkatesh.pallipadi@intel.com" Date: Fri, 9 Jan 2009 16:13:10 -0800 Subject: x86 PAT: consolidate old memtype new memtype check into a function Impact: cleanup Move the new memtype old memtype allowed check to header so that is can be shared by other users. Subsequent patch uses this in pat.c in remap_pfn_range() code path. No functionality change in this patch. Signed-off-by: Venkatesh Pallipadi Signed-off-by: Suresh Siddha Signed-off-by: Ingo Molnar --- arch/x86/include/asm/pgtable.h | 19 +++++++++++++++++++ arch/x86/pci/i386.c | 12 +----------- 2 files changed, 20 insertions(+), 11 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index 83e69f4a37f..06bbcbd66e9 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -341,6 +341,25 @@ static inline pgprot_t pgprot_modify(pgprot_t oldprot, pgprot_t newprot) #define canon_pgprot(p) __pgprot(pgprot_val(p) & __supported_pte_mask) +static inline int is_new_memtype_allowed(unsigned long flags, + unsigned long new_flags) +{ + /* + * Certain new memtypes are not allowed with certain + * requested memtype: + * - request is uncached, return cannot be write-back + * - request is write-combine, return cannot be write-back + */ + if ((flags == _PAGE_CACHE_UC_MINUS && + new_flags == _PAGE_CACHE_WB) || + (flags == _PAGE_CACHE_WC && + new_flags == _PAGE_CACHE_WB)) { + return 0; + } + + return 1; +} + #ifndef __ASSEMBLY__ /* Indicate that x86 has its own track and untrack pfn vma functions */ #define __HAVE_PFNMAP_TRACKING diff --git a/arch/x86/pci/i386.c b/arch/x86/pci/i386.c index f884740da31..5ead808dd70 100644 --- a/arch/x86/pci/i386.c +++ b/arch/x86/pci/i386.c @@ -314,17 +314,7 @@ int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma, return retval; if (flags != new_flags) { - /* - * Do not fallback to certain memory types with certain - * requested type: - * - request is uncached, return cannot be write-back - * - request is uncached, return cannot be write-combine - * - request is write-combine, return cannot be write-back - */ - if ((flags == _PAGE_CACHE_UC_MINUS && - (new_flags == _PAGE_CACHE_WB)) || - (flags == _PAGE_CACHE_WC && - new_flags == _PAGE_CACHE_WB)) { + if (!is_new_memtype_allowed(flags, new_flags)) { free_memtype(addr, addr+len); return -EINVAL; } -- cgit v1.2.3 From e4b866ed197cef9989348e0479fed8d864ea465b Mon Sep 17 00:00:00 2001 From: "venkatesh.pallipadi@intel.com" Date: Fri, 9 Jan 2009 16:13:11 -0800 Subject: x86 PAT: change track_pfn_vma_new to take pgprot_t pointer param Impact: cleanup Change the protection parameter for track_pfn_vma_new() into a pgprot_t pointer. Subsequent patch changes the x86 PAT handling to return a compatible memtype in pgprot_t, if what was requested cannot be allowed due to conflicts. No fuctionality change in this patch. Signed-off-by: Venkatesh Pallipadi Signed-off-by: Suresh Siddha Signed-off-by: Ingo Molnar --- arch/x86/mm/pat.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index 85cbd3cd372..f88ac80530c 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -741,7 +741,7 @@ cleanup_ret: * Note that this function can be called with caller trying to map only a * subrange/page inside the vma. */ -int track_pfn_vma_new(struct vm_area_struct *vma, pgprot_t prot, +int track_pfn_vma_new(struct vm_area_struct *vma, pgprot_t *prot, unsigned long pfn, unsigned long size) { int retval = 0; @@ -758,14 +758,14 @@ int track_pfn_vma_new(struct vm_area_struct *vma, pgprot_t prot, if (is_linear_pfn_mapping(vma)) { /* reserve the whole chunk starting from vm_pgoff */ paddr = (resource_size_t)vma->vm_pgoff << PAGE_SHIFT; - return reserve_pfn_range(paddr, vma_size, prot); + return reserve_pfn_range(paddr, vma_size, *prot); } /* reserve page by page using pfn and size */ base_paddr = (resource_size_t)pfn << PAGE_SHIFT; for (i = 0; i < size; i += PAGE_SIZE) { paddr = base_paddr + i; - retval = reserve_pfn_range(paddr, PAGE_SIZE, prot); + retval = reserve_pfn_range(paddr, PAGE_SIZE, *prot); if (retval) goto cleanup_ret; } -- cgit v1.2.3 From cdecff6864a1cd352a41d44a65e7451b8ef5cee2 Mon Sep 17 00:00:00 2001 From: "venkatesh.pallipadi@intel.com" Date: Fri, 9 Jan 2009 16:13:12 -0800 Subject: x86 PAT: return compatible mapping to remap_pfn_range callers Impact: avoid warning message, potentially solve 3D performance regression Change x86 PAT code to return compatible memtype if the exact memtype that was requested in remap_pfn_rage and friends is not available due to some conflict. This is done by returning the compatible type in pgprot parameter of track_pfn_vma_new(), and the caller uses that memtype for page table. Note that track_pfn_vma_copy() which is basically called during fork gets the prot from existing page table and should not have any conflict. Hence we use strict memtype check there and do not allow compatible memtypes. This patch fixes the bug reported here: http://marc.info/?l=linux-kernel&m=123108883716357&w=2 Specifically the error message: X:5010 map pfn expected mapping type write-back for d0000000-d0101000, got write-combining Should go away. Reported-and-bisected-by: Kevin Winchester Signed-off-by: Venkatesh Pallipadi Signed-off-by: Suresh Siddha Signed-off-by: Ingo Molnar --- arch/x86/mm/pat.c | 43 ++++++++++++++++++++++++++++--------------- 1 file changed, 28 insertions(+), 15 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index f88ac80530c..8b08fb95527 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -601,12 +601,13 @@ void unmap_devmem(unsigned long pfn, unsigned long size, pgprot_t vma_prot) * Reserved non RAM regions only and after successful reserve_memtype, * this func also keeps identity mapping (if any) in sync with this new prot. */ -static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t vma_prot) +static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t *vma_prot, + int strict_prot) { int is_ram = 0; int id_sz, ret; unsigned long flags; - unsigned long want_flags = (pgprot_val(vma_prot) & _PAGE_CACHE_MASK); + unsigned long want_flags = (pgprot_val(*vma_prot) & _PAGE_CACHE_MASK); is_ram = pagerange_is_ram(paddr, paddr + size); @@ -625,15 +626,24 @@ static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t vma_prot) return ret; if (flags != want_flags) { - free_memtype(paddr, paddr + size); - printk(KERN_ERR - "%s:%d map pfn expected mapping type %s for %Lx-%Lx, got %s\n", - current->comm, current->pid, - cattr_name(want_flags), - (unsigned long long)paddr, - (unsigned long long)(paddr + size), - cattr_name(flags)); - return -EINVAL; + if (strict_prot || !is_new_memtype_allowed(want_flags, flags)) { + free_memtype(paddr, paddr + size); + printk(KERN_ERR "%s:%d map pfn expected mapping type %s" + " for %Lx-%Lx, got %s\n", + current->comm, current->pid, + cattr_name(want_flags), + (unsigned long long)paddr, + (unsigned long long)(paddr + size), + cattr_name(flags)); + return -EINVAL; + } + /* + * We allow returning different type than the one requested in + * non strict case. + */ + *vma_prot = __pgprot((pgprot_val(*vma_prot) & + (~_PAGE_CACHE_MASK)) | + flags); } /* Need to keep identity mapping in sync */ @@ -689,6 +699,7 @@ int track_pfn_vma_copy(struct vm_area_struct *vma) unsigned long vma_start = vma->vm_start; unsigned long vma_end = vma->vm_end; unsigned long vma_size = vma_end - vma_start; + pgprot_t pgprot; if (!pat_enabled) return 0; @@ -702,7 +713,8 @@ int track_pfn_vma_copy(struct vm_area_struct *vma) WARN_ON_ONCE(1); return -EINVAL; } - return reserve_pfn_range(paddr, vma_size, __pgprot(prot)); + pgprot = __pgprot(prot); + return reserve_pfn_range(paddr, vma_size, &pgprot, 1); } /* reserve entire vma page by page, using pfn and prot from pte */ @@ -710,7 +722,8 @@ int track_pfn_vma_copy(struct vm_area_struct *vma) if (follow_phys(vma, vma_start + i, 0, &prot, &paddr)) continue; - retval = reserve_pfn_range(paddr, PAGE_SIZE, __pgprot(prot)); + pgprot = __pgprot(prot); + retval = reserve_pfn_range(paddr, PAGE_SIZE, &pgprot, 1); if (retval) goto cleanup_ret; } @@ -758,14 +771,14 @@ int track_pfn_vma_new(struct vm_area_struct *vma, pgprot_t *prot, if (is_linear_pfn_mapping(vma)) { /* reserve the whole chunk starting from vm_pgoff */ paddr = (resource_size_t)vma->vm_pgoff << PAGE_SHIFT; - return reserve_pfn_range(paddr, vma_size, *prot); + return reserve_pfn_range(paddr, vma_size, prot, 0); } /* reserve page by page using pfn and size */ base_paddr = (resource_size_t)pfn << PAGE_SHIFT; for (i = 0; i < size; i += PAGE_SIZE) { paddr = base_paddr + i; - retval = reserve_pfn_range(paddr, PAGE_SIZE, *prot); + retval = reserve_pfn_range(paddr, PAGE_SIZE, prot, 0); if (retval) goto cleanup_ret; } -- cgit v1.2.3 From 58dab916dfb57328d50deb0aa9b3fc92efa248ff Mon Sep 17 00:00:00 2001 From: "venkatesh.pallipadi@intel.com" Date: Fri, 9 Jan 2009 16:13:14 -0800 Subject: x86 PAT: remove CPA WARN_ON for zero pte Impact: reduce scope of debug check - avoid warnings The logic to find whether identity map exists or not using high_memory or max_low_pfn_mapped/max_pfn_mapped are not complete as the memory withing the range may not be mapped if there is a unusable hole in e820. Specifically, on my test system I started seeing these warnings with tools like hwinfo, acpidump trying to map ACPI region. [ 27.400018] ------------[ cut here ]------------ [ 27.400344] WARNING: at /home/venkip/src/linus/linux-2.6/arch/x86/mm/pageattr.c:560 __change_page_attr_set_clr+0xf3/0x8b8() [ 27.400821] Hardware name: X7DB8 [ 27.401070] CPA: called for zero pte. vaddr = ffff8800cff6a000 cpa->vaddr = ffff8800cff6a000 [ 27.401569] Modules linked in: [ 27.401882] Pid: 4913, comm: dmidecode Not tainted 2.6.28-05716-gfe0bdec #586 [ 27.402141] Call Trace: [ 27.402488] [] warn_slowpath+0xd3/0x10f [ 27.402749] [] ? find_get_page+0xb3/0xc9 [ 27.403028] [] ? find_get_page+0x0/0xc9 [ 27.403333] [] __change_page_attr_set_clr+0xf3/0x8b8 [ 27.403628] [] ? __purge_vmap_area_lazy+0x192/0x1a1 [ 27.403883] [] ? __purge_vmap_area_lazy+0x4b/0x1a1 [ 27.404172] [] ? vm_unmap_aliases+0x1ab/0x1bb [ 27.404512] [] ? vm_unmap_aliases+0x48/0x1bb [ 27.404766] [] change_page_attr_set_clr+0x13e/0x2e6 [ 27.405026] [] ? _spin_unlock+0x26/0x2a [ 27.405292] [] ? reserve_memtype+0x19b/0x4e3 [ 27.405590] [] _set_memory_wb+0x22/0x24 [ 27.405844] [] ioremap_change_attr+0x26/0x28 [ 27.406097] [] reserve_pfn_range+0x1a3/0x235 [ 27.406427] [] track_pfn_vma_new+0x49/0xb3 [ 27.406686] [] remap_pfn_range+0x94/0x32c [ 27.406940] [] ? phys_mem_access_prot_allowed+0xb5/0x1a8 [ 27.407209] [] mmap_mem+0x75/0x9d [ 27.407523] [] mmap_region+0x2cf/0x53e [ 27.407776] [] do_mmap_pgoff+0x2a9/0x30d [ 27.408034] [] sys_mmap+0x92/0xce [ 27.408339] [] system_call_fastpath+0x16/0x1b [ 27.408614] ---[ end trace 4b16ad70c09a602d ]--- [ 27.408871] dmidecode:4913 reserve_pfn_range ioremap_change_attr failed write-back for cff6a000-cff6b000 This is wih track_pfn_vma_new trying to keep identity map in sync. The address cff6a000 is the ACPI region according to e820. [ 0.000000] BIOS-provided physical RAM map: [ 0.000000] BIOS-e820: 0000000000000000 - 000000000009c000 (usable) [ 0.000000] BIOS-e820: 000000000009c000 - 00000000000a0000 (reserved) [ 0.000000] BIOS-e820: 00000000000cc000 - 00000000000d0000 (reserved) [ 0.000000] BIOS-e820: 00000000000e4000 - 0000000000100000 (reserved) [ 0.000000] BIOS-e820: 0000000000100000 - 00000000cff60000 (usable) [ 0.000000] BIOS-e820: 00000000cff60000 - 00000000cff69000 (ACPI data) [ 0.000000] BIOS-e820: 00000000cff69000 - 00000000cff80000 (ACPI NVS) [ 0.000000] BIOS-e820: 00000000cff80000 - 00000000d0000000 (reserved) [ 0.000000] BIOS-e820: 00000000e0000000 - 00000000f0000000 (reserved) [ 0.000000] BIOS-e820: 00000000fec00000 - 00000000fec10000 (reserved) [ 0.000000] BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved) [ 0.000000] BIOS-e820: 00000000ff000000 - 0000000100000000 (reserved) [ 0.000000] BIOS-e820: 0000000100000000 - 0000000230000000 (usable) And is not mapped as per init_memory_mapping. [ 0.000000] init_memory_mapping: 0000000000000000-00000000cff60000 [ 0.000000] init_memory_mapping: 0000000100000000-0000000230000000 We can add logic to check for this. But, there can also be other holes in identity map when we have 1GB of aligned reserved space in e820. This patch handles it by removing the WARN_ON and returning a specific error value (EFAULT) to indicate that the address does not have any identity mapping. The code that tries to keep identity map in sync can ignore this error, with other callers of cpa still getting error here. Signed-off-by: Venkatesh Pallipadi Signed-off-by: Suresh Siddha Signed-off-by: Ingo Molnar --- arch/x86/mm/pageattr.c | 10 ++++++---- arch/x86/mm/pat.c | 45 ++++++++++++++++++++++++++++++++------------- 2 files changed, 38 insertions(+), 17 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index e89d24815f2..4cf30dee816 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -555,10 +555,12 @@ repeat: if (!pte_val(old_pte)) { if (!primary) return 0; - WARN(1, KERN_WARNING "CPA: called for zero pte. " - "vaddr = %lx cpa->vaddr = %lx\n", address, - *cpa->vaddr); - return -EINVAL; + + /* + * Special error value returned, indicating that the mapping + * did not exist at this address. + */ + return -EFAULT; } if (level == PG_LEVEL_4K) { diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index 8b08fb95527..160c42d3eb8 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -505,6 +505,35 @@ static inline int range_is_allowed(unsigned long pfn, unsigned long size) } #endif /* CONFIG_STRICT_DEVMEM */ +/* + * Change the memory type for the physial address range in kernel identity + * mapping space if that range is a part of identity map. + */ +static int kernel_map_sync_memtype(u64 base, unsigned long size, + unsigned long flags) +{ + unsigned long id_sz; + int ret; + + if (!pat_enabled || base >= __pa(high_memory)) + return 0; + + id_sz = (__pa(high_memory) < base + size) ? + __pa(high_memory) - base : + size; + + ret = ioremap_change_attr((unsigned long)__va(base), id_sz, flags); + /* + * -EFAULT return means that the addr was not valid and did not have + * any identity mapping. That case is a success for + * kernel_map_sync_memtype. + */ + if (ret == -EFAULT) + ret = 0; + + return ret; +} + int phys_mem_access_prot_allowed(struct file *file, unsigned long pfn, unsigned long size, pgprot_t *vma_prot) { @@ -555,9 +584,7 @@ int phys_mem_access_prot_allowed(struct file *file, unsigned long pfn, if (retval < 0) return 0; - if (((pfn < max_low_pfn_mapped) || - (pfn >= (1UL<<(32 - PAGE_SHIFT)) && pfn < max_pfn_mapped)) && - ioremap_change_attr((unsigned long)__va(offset), size, flags) < 0) { + if (kernel_map_sync_memtype(offset, size, flags)) { free_memtype(offset, offset + size); printk(KERN_INFO "%s:%d /dev/mem ioremap_change_attr failed %s for %Lx-%Lx\n", @@ -605,7 +632,7 @@ static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t *vma_prot, int strict_prot) { int is_ram = 0; - int id_sz, ret; + int ret; unsigned long flags; unsigned long want_flags = (pgprot_val(*vma_prot) & _PAGE_CACHE_MASK); @@ -646,15 +673,7 @@ static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t *vma_prot, flags); } - /* Need to keep identity mapping in sync */ - if (paddr >= __pa(high_memory)) - return 0; - - id_sz = (__pa(high_memory) < paddr + size) ? - __pa(high_memory) - paddr : - size; - - if (ioremap_change_attr((unsigned long)__va(paddr), id_sz, flags) < 0) { + if (kernel_map_sync_memtype(paddr, size, flags)) { free_memtype(paddr, paddr + size); printk(KERN_ERR "%s:%d reserve_pfn_range ioremap_change_attr failed %s " -- cgit v1.2.3 From e55380edf68796d75bf41391a781c68ee678587d Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 14 Jan 2009 14:13:55 +0100 Subject: [CVE-2009-0029] Rename old_readdir to sys_old_readdir This way it matches the generic system call name convention. Signed-off-by: Heiko Carstens --- arch/x86/kernel/syscall_table_32.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index d44395ff34c..e2e86a08f31 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -88,7 +88,7 @@ ENTRY(sys_call_table) .long sys_uselib .long sys_swapon .long sys_reboot - .long old_readdir + .long sys_old_readdir .long old_mmap /* 90 */ .long sys_munmap .long sys_truncate -- cgit v1.2.3 From 5cca0cf15a94417f49625ce52e23589eed0a1675 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Fri, 9 Jan 2009 14:35:20 -0800 Subject: x86, pat: fix reserve_memtype() for legacy 1MB range Thierry Vignaud reported: > http://bugzilla.kernel.org/show_bug.cgi?id=12372 > > On P4 with an SiS motherboard (video card is a SiS 651) > X server fails to start with error: > xf86MapVidMem: Could not mmap framebuffer (0x00000000,0x2000) (Invalid > argument) Here X is trying to map first 8KB of memory using /dev/mem. Existing code treats first 0-4KB of memory as non-RAM and 4KB-8KB as RAM. Recent code changes don't allow to map memory with different attributes at the same time. Fix this by treating the first 1MB legacy region as special and always track the attribute requests with in this region using linear linked list (and don't bother if the range is RAM or non-RAM or mixed) Reported-and-tested-by: Thierry Vignaud Signed-off-by: Suresh Siddha Signed-off-by: Venkatesh Pallipadi Signed-off-by: Ingo Molnar --- arch/x86/mm/pat.c | 37 +++++++++++++++++++++++++++---------- 1 file changed, 27 insertions(+), 10 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index 85cbd3cd372..070ee4a3b22 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -333,11 +333,20 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, req_type & _PAGE_CACHE_MASK); } - is_range_ram = pagerange_is_ram(start, end); - if (is_range_ram == 1) - return reserve_ram_pages_type(start, end, req_type, new_type); - else if (is_range_ram < 0) - return -EINVAL; + /* + * For legacy reasons, some parts of the physical address range in the + * legacy 1MB region is treated as non-RAM (even when listed as RAM in + * the e820 tables). So we will track the memory attributes of this + * legacy 1MB region using the linear memtype_list always. + */ + if (end >= ISA_END_ADDRESS) { + is_range_ram = pagerange_is_ram(start, end); + if (is_range_ram == 1) + return reserve_ram_pages_type(start, end, req_type, + new_type); + else if (is_range_ram < 0) + return -EINVAL; + } new = kmalloc(sizeof(struct memtype), GFP_KERNEL); if (!new) @@ -437,11 +446,19 @@ int free_memtype(u64 start, u64 end) if (is_ISA_range(start, end - 1)) return 0; - is_range_ram = pagerange_is_ram(start, end); - if (is_range_ram == 1) - return free_ram_pages_type(start, end); - else if (is_range_ram < 0) - return -EINVAL; + /* + * For legacy reasons, some parts of the physical address range in the + * legacy 1MB region is treated as non-RAM (even when listed as RAM in + * the e820 tables). So we will track the memory attributes of this + * legacy 1MB region using the linear memtype_list always. + */ + if (end >= ISA_END_ADDRESS) { + is_range_ram = pagerange_is_ram(start, end); + if (is_range_ram == 1) + return free_ram_pages_type(start, end); + else if (is_range_ram < 0) + return -EINVAL; + } spin_lock(&memtype_lock); list_for_each_entry(entry, &memtype_list, nd) { -- cgit v1.2.3 From 2ea038917bbdd51a7ae4a898c6a04641324dd033 Mon Sep 17 00:00:00 2001 From: Sam Ravnborg Date: Wed, 14 Jan 2009 21:38:20 +0100 Subject: Revert "kbuild: strip generated symbols from *.ko" This reverts commit ad7a953c522ceb496611d127e51e278bfe0ff483. And commit: ("allow stripping of generated symbols under CONFIG_KALLSYMS_ALL") 9bb482476c6c9d1ae033306440c51ceac93ea80c These stripping patches has caused a set of issues: 1) People have reported compatibility issues with binutils due to lack of support for `--strip-unneeded-symbols' with objcopy 2.15.92.0.2 Reported by: Wenji 2) ccache and distcc no longer works as expeced Reported by: Ted, Roland, + others 3) The installed modules increased a lot in size Reported by: Ted, Davej + others Reported-by: Wenji Huang Reported-by: "Theodore Ts'o" Reported-by: Dave Jones Reported-by: Roland McGrath Signed-off-by: Sam Ravnborg --- arch/x86/scripts/strip-symbols | 1 - 1 file changed, 1 deletion(-) delete mode 100644 arch/x86/scripts/strip-symbols (limited to 'arch/x86') diff --git a/arch/x86/scripts/strip-symbols b/arch/x86/scripts/strip-symbols deleted file mode 100644 index a2f1ccb827c..00000000000 --- a/arch/x86/scripts/strip-symbols +++ /dev/null @@ -1 +0,0 @@ -__cpu_vendor_dev_X86_VENDOR_* -- cgit v1.2.3 From 74d96f018673759d04d032c137d132f6447bfb1e Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Tue, 13 Jan 2009 19:27:09 -0800 Subject: byteorder: make swab.h include asm/swab.h like a regular header Add swab.h to kbuild.asm and remove the individual entries from each arch, mark as unifdef as some arches have some kernel-only bits inside. Signed-off-by: Harvey Harrison Signed-off-by: Linus Torvalds --- arch/x86/include/asm/Kbuild | 1 - arch/x86/include/asm/byteorder.h | 1 - 2 files changed, 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/Kbuild b/arch/x86/include/asm/Kbuild index a9f8a814a1f..4a8e80cdcfa 100644 --- a/arch/x86/include/asm/Kbuild +++ b/arch/x86/include/asm/Kbuild @@ -22,4 +22,3 @@ unifdef-y += unistd_32.h unifdef-y += unistd_64.h unifdef-y += vm86.h unifdef-y += vsyscall.h -unifdef-y += swab.h diff --git a/arch/x86/include/asm/byteorder.h b/arch/x86/include/asm/byteorder.h index 7c49917e3d9..b13a7a88f3e 100644 --- a/arch/x86/include/asm/byteorder.h +++ b/arch/x86/include/asm/byteorder.h @@ -1,7 +1,6 @@ #ifndef _ASM_X86_BYTEORDER_H #define _ASM_X86_BYTEORDER_H -#include #include #endif /* _ASM_X86_BYTEORDER_H */ -- cgit v1.2.3 From 4a13ad0bd8869bb491c67918662f9b1852595af5 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Wed, 14 Jan 2009 12:28:51 +0000 Subject: x86: avoid early crash in disable_local_APIC() E.g. when called due to an early panic. Signed-off-by: Jan Beulich Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic.c b/arch/x86/kernel/apic.c index 0f830e4f567..d19aa3aab62 100644 --- a/arch/x86/kernel/apic.c +++ b/arch/x86/kernel/apic.c @@ -895,6 +895,10 @@ void disable_local_APIC(void) { unsigned int value; + /* APIC hasn't been mapped yet */ + if (!apic_phys) + return; + clear_local_APIC(); /* -- cgit v1.2.3 From 18c07cf530cf4aa8b7551801f68ed40db5ee4e45 Mon Sep 17 00:00:00 2001 From: Cliff Wickman Date: Thu, 15 Jan 2009 09:51:20 -0600 Subject: x86, UV: cpu_relax in uv_wait_completion The function uv_wait_completion() spins on reads of a memory-mapped register, waiting for completion of BAU hardware replies. It should call "cpu_relax()" between those reads to improve performance on hyperthreaded configurations. Signed-off-by: Cliff Wickman Acked-by: Jack Steiner Signed-off-by: Ingo Molnar --- arch/x86/kernel/tlb_uv.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index f885023167e..6812b829ed8 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -200,6 +200,7 @@ static int uv_wait_completion(struct bau_desc *bau_desc, destination_timeouts = 0; } } + cpu_relax(); } return FLUSH_COMPLETE; } -- cgit v1.2.3 From b5db0e38653bfada34a92f360b4111566ede3842 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 15 Jan 2009 15:32:12 -0800 Subject: Revert "x86 PAT: remove CPA WARN_ON for zero pte" This reverts commit 58dab916dfb57328d50deb0aa9b3fc92efa248ff, which makes my Nehalem come to a nasty crawling almost-halt. It looks like it turns off caching of regular kernel RAM, with the understandable slowdown of a few orders of magnitude as a result. Acked-by: Ingo Molnar Cc: Yinghai Lu Cc: Peter Anvin Cc: Venkatesh Pallipadi Cc: Suresh Siddha Signed-off-by: Linus Torvalds --- arch/x86/mm/pageattr.c | 10 ++++------ arch/x86/mm/pat.c | 45 +++++++++++++-------------------------------- 2 files changed, 17 insertions(+), 38 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 4cf30dee816..e89d24815f2 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -555,12 +555,10 @@ repeat: if (!pte_val(old_pte)) { if (!primary) return 0; - - /* - * Special error value returned, indicating that the mapping - * did not exist at this address. - */ - return -EFAULT; + WARN(1, KERN_WARNING "CPA: called for zero pte. " + "vaddr = %lx cpa->vaddr = %lx\n", address, + *cpa->vaddr); + return -EINVAL; } if (level == PG_LEVEL_4K) { diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index 160c42d3eb8..8b08fb95527 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -505,35 +505,6 @@ static inline int range_is_allowed(unsigned long pfn, unsigned long size) } #endif /* CONFIG_STRICT_DEVMEM */ -/* - * Change the memory type for the physial address range in kernel identity - * mapping space if that range is a part of identity map. - */ -static int kernel_map_sync_memtype(u64 base, unsigned long size, - unsigned long flags) -{ - unsigned long id_sz; - int ret; - - if (!pat_enabled || base >= __pa(high_memory)) - return 0; - - id_sz = (__pa(high_memory) < base + size) ? - __pa(high_memory) - base : - size; - - ret = ioremap_change_attr((unsigned long)__va(base), id_sz, flags); - /* - * -EFAULT return means that the addr was not valid and did not have - * any identity mapping. That case is a success for - * kernel_map_sync_memtype. - */ - if (ret == -EFAULT) - ret = 0; - - return ret; -} - int phys_mem_access_prot_allowed(struct file *file, unsigned long pfn, unsigned long size, pgprot_t *vma_prot) { @@ -584,7 +555,9 @@ int phys_mem_access_prot_allowed(struct file *file, unsigned long pfn, if (retval < 0) return 0; - if (kernel_map_sync_memtype(offset, size, flags)) { + if (((pfn < max_low_pfn_mapped) || + (pfn >= (1UL<<(32 - PAGE_SHIFT)) && pfn < max_pfn_mapped)) && + ioremap_change_attr((unsigned long)__va(offset), size, flags) < 0) { free_memtype(offset, offset + size); printk(KERN_INFO "%s:%d /dev/mem ioremap_change_attr failed %s for %Lx-%Lx\n", @@ -632,7 +605,7 @@ static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t *vma_prot, int strict_prot) { int is_ram = 0; - int ret; + int id_sz, ret; unsigned long flags; unsigned long want_flags = (pgprot_val(*vma_prot) & _PAGE_CACHE_MASK); @@ -673,7 +646,15 @@ static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t *vma_prot, flags); } - if (kernel_map_sync_memtype(paddr, size, flags)) { + /* Need to keep identity mapping in sync */ + if (paddr >= __pa(high_memory)) + return 0; + + id_sz = (__pa(high_memory) < paddr + size) ? + __pa(high_memory) - paddr : + size; + + if (ioremap_change_attr((unsigned long)__va(paddr), id_sz, flags) < 0) { free_memtype(paddr, paddr + size); printk(KERN_ERR "%s:%d reserve_pfn_range ioremap_change_attr failed %s " -- cgit v1.2.3 From a3c6018e565dc07cf3738ace6bbe412f97b1bba8 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Fri, 16 Jan 2009 11:59:33 +0000 Subject: x86: fix assumed to be contiguous leaf page tables for kmap_atomic region (take 2) Debugging and original patch from Nick Piggin The early fixmap pmd entry inserted at the very top of the KVA is causing the subsequent fixmap mapping code to not provide physically linear pte pages over the kmap atomic portion of the fixmap (which relies on said property to calculate pte addresses). This has caused weird boot failures in kmap_atomic much later in the boot process (initial userspace faults) on a 32-bit PAE system with a larger number of CPUs (smaller CPU counts tend not to run over into the next page so don't show up the problem). Solve this by attempting to clear out the page table, and copy any of its entries to the new one. Also, add a bug if a nonlinear condition is encountered and can't be resolved, which might save some hours of debugging if this fragile scheme ever breaks again... Once we have such logic, we can also use it to eliminate the early ioremap trickery around the page table setup for the fixmap area. This also fixes potential issues with FIX_* entries sharing the leaf page table with the early ioremap ones getting discarded by early_ioremap_clear() and not restored by early_ioremap_reset(). It at once eliminates the temporary (and configuration, namely NR_CPUS, dependent) unavailability of early fixed mappings during the time the fixmap area page tables get constructed. Finally, also replace the hard coded calculation of the initial table space needed for the fixmap area with a proper one, allowing kernels configured for large CPU counts to actually boot. Based-on: Nick Piggin Signed-off-by: Jan Beulich Signed-off-by: Ingo Molnar --- arch/x86/include/asm/io.h | 1 - arch/x86/mm/init_32.c | 48 ++++++++++++++++++++++++++++++++++++++++++++--- arch/x86/mm/ioremap.c | 25 ------------------------ 3 files changed, 45 insertions(+), 29 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h index 05cfed4485f..1dbbdf4be9b 100644 --- a/arch/x86/include/asm/io.h +++ b/arch/x86/include/asm/io.h @@ -99,7 +99,6 @@ extern void __iomem *ioremap_wc(unsigned long offset, unsigned long size); * A boot-time mapping is currently limited to at most 16 pages. */ extern void early_ioremap_init(void); -extern void early_ioremap_clear(void); extern void early_ioremap_reset(void); extern void __iomem *early_ioremap(unsigned long offset, unsigned long size); extern void __iomem *early_memremap(unsigned long offset, unsigned long size); diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 88f1b10de3b..2cef0507441 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -138,6 +138,47 @@ static pte_t * __init one_page_table_init(pmd_t *pmd) return pte_offset_kernel(pmd, 0); } +static pte_t *__init page_table_kmap_check(pte_t *pte, pmd_t *pmd, + unsigned long vaddr, pte_t *lastpte) +{ +#ifdef CONFIG_HIGHMEM + /* + * Something (early fixmap) may already have put a pte + * page here, which causes the page table allocation + * to become nonlinear. Attempt to fix it, and if it + * is still nonlinear then we have to bug. + */ + int pmd_idx_kmap_begin = fix_to_virt(FIX_KMAP_END) >> PMD_SHIFT; + int pmd_idx_kmap_end = fix_to_virt(FIX_KMAP_BEGIN) >> PMD_SHIFT; + + if (pmd_idx_kmap_begin != pmd_idx_kmap_end + && (vaddr >> PMD_SHIFT) >= pmd_idx_kmap_begin + && (vaddr >> PMD_SHIFT) <= pmd_idx_kmap_end + && ((__pa(pte) >> PAGE_SHIFT) < table_start + || (__pa(pte) >> PAGE_SHIFT) >= table_end)) { + pte_t *newpte; + int i; + + BUG_ON(after_init_bootmem); + newpte = alloc_low_page(); + for (i = 0; i < PTRS_PER_PTE; i++) + set_pte(newpte + i, pte[i]); + + paravirt_alloc_pte(&init_mm, __pa(newpte) >> PAGE_SHIFT); + set_pmd(pmd, __pmd(__pa(newpte)|_PAGE_TABLE)); + BUG_ON(newpte != pte_offset_kernel(pmd, 0)); + __flush_tlb_all(); + + paravirt_release_pte(__pa(pte) >> PAGE_SHIFT); + pte = newpte; + } + BUG_ON(vaddr < fix_to_virt(FIX_KMAP_BEGIN - 1) + && vaddr > fix_to_virt(FIX_KMAP_END) + && lastpte && lastpte + PTRS_PER_PTE != pte); +#endif + return pte; +} + /* * This function initializes a certain range of kernel virtual memory * with new bootmem page tables, everywhere page tables are missing in @@ -154,6 +195,7 @@ page_table_range_init(unsigned long start, unsigned long end, pgd_t *pgd_base) unsigned long vaddr; pgd_t *pgd; pmd_t *pmd; + pte_t *pte = NULL; vaddr = start; pgd_idx = pgd_index(vaddr); @@ -165,7 +207,8 @@ page_table_range_init(unsigned long start, unsigned long end, pgd_t *pgd_base) pmd = pmd + pmd_index(vaddr); for (; (pmd_idx < PTRS_PER_PMD) && (vaddr != end); pmd++, pmd_idx++) { - one_page_table_init(pmd); + pte = page_table_kmap_check(one_page_table_init(pmd), + pmd, vaddr, pte); vaddr += PMD_SIZE; } @@ -508,7 +551,6 @@ static void __init early_ioremap_page_table_range_init(pgd_t *pgd_base) * Fixed mappings, only the page table structure has to be * created - mappings will be set by set_fixmap(): */ - early_ioremap_clear(); vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK; end = (FIXADDR_TOP + PMD_SIZE - 1) & PMD_MASK; page_table_range_init(vaddr, end, pgd_base); @@ -801,7 +843,7 @@ static void __init find_early_table_space(unsigned long end, int use_pse) tables += PAGE_ALIGN(ptes * sizeof(pte_t)); /* for fixmap */ - tables += PAGE_SIZE * 2; + tables += PAGE_ALIGN(__end_of_fixed_addresses * sizeof(pte_t)); /* * RED-PEN putting page tables only on node 0 could diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index bd85d42819e..af750ab973b 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -557,34 +557,9 @@ void __init early_ioremap_init(void) } } -void __init early_ioremap_clear(void) -{ - pmd_t *pmd; - - if (early_ioremap_debug) - printk(KERN_INFO "early_ioremap_clear()\n"); - - pmd = early_ioremap_pmd(fix_to_virt(FIX_BTMAP_BEGIN)); - pmd_clear(pmd); - paravirt_release_pte(__pa(bm_pte) >> PAGE_SHIFT); - __flush_tlb_all(); -} - void __init early_ioremap_reset(void) { - enum fixed_addresses idx; - unsigned long addr, phys; - pte_t *pte; - after_paging_init = 1; - for (idx = FIX_BTMAP_BEGIN; idx >= FIX_BTMAP_END; idx--) { - addr = fix_to_virt(idx); - pte = early_ioremap_pte(addr); - if (pte_present(*pte)) { - phys = pte_val(*pte) & PAGE_MASK; - set_fixmap(idx, phys); - } - } } static void __init __early_set_fixmap(enum fixed_addresses idx, -- cgit v1.2.3 From 5a4ccaf37ffece09ef33f1cfec67efa8ee56f967 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Tue, 6 Jan 2009 21:15:32 +0100 Subject: kprobes: check CONFIG_FREEZER instead of CONFIG_PM Check CONFIG_FREEZER instead of CONFIG_PM because kprobe booster depends on freeze_processes() and thaw_processes() when CONFIG_PREEMPT=y. This fixes a linkage error which occurs when CONFIG_PREEMPT=y, CONFIG_PM=y and CONFIG_FREEZER=n. Reported-by: Cheng Renquan Signed-off-by: Masami Hiramatsu Signed-off-by: Rafael J. Wysocki Acked-by: Ingo Molnar Signed-off-by: Len Brown --- arch/x86/kernel/kprobes.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index 884d985b8b8..e948b28a5a9 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -446,7 +446,7 @@ void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri, static void __kprobes setup_singlestep(struct kprobe *p, struct pt_regs *regs, struct kprobe_ctlblk *kcb) { -#if !defined(CONFIG_PREEMPT) || defined(CONFIG_PM) +#if !defined(CONFIG_PREEMPT) || defined(CONFIG_FREEZER) if (p->ainsn.boostable == 1 && !p->post_handler) { /* Boost up -- we can execute copied instructions directly */ reset_current_kprobe(); -- cgit v1.2.3 From 5d8b532af9e52ea89208f5ef31889f646e67ba28 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 16 Jan 2009 23:09:14 +0100 Subject: ACPI suspend: Fix compilation warnings in drivers/acpi/sleep.c Fix two compilation warnings in drivers/acpi/sleep.c, one triggered by unsetting CONFIG_SUSPEND and the other triggered by unsetting CONFIG_HIBERNATION, by moving some code under the appropriate #ifdefs . Signed-off-by: Rafael J. Wysocki Signed-off-by: Len Brown --- arch/x86/kernel/acpi/sleep.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index 707c1f6f95f..a60c1f3bcb8 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -156,11 +156,11 @@ static int __init acpi_sleep_setup(char *str) #ifdef CONFIG_HIBERNATION if (strncmp(str, "s4_nohwsig", 10) == 0) acpi_no_s4_hw_signature(); + if (strncmp(str, "s4_nonvs", 8) == 0) + acpi_s4_no_nvs(); #endif if (strncmp(str, "old_ordering", 12) == 0) acpi_old_suspend_ordering(); - if (strncmp(str, "s4_nonvs", 8) == 0) - acpi_s4_no_nvs(); str = strchr(str, ','); if (str != NULL) str += strspn(str, ", \t"); -- cgit v1.2.3 From b2b815d80a5c4e5b50be0a98aba8c445ce8f3e1f Mon Sep 17 00:00:00 2001 From: Mike Travis Date: Fri, 16 Jan 2009 15:22:16 -0800 Subject: x86: put trigger in to detect mismatched apic versions Impact: add debug warning Fire off one message if two apic's discovered with different apic versions. (this code is only called during CPU init) The goal of this is to pave the way of the removal of the apic_version[] array. We dont expect any apic version incompatibilities in the x86 landscape of systems [if so we dont handle them very well and probably never will handle deep apic version assymetries well], but it's prudent to have a debug check for one kernel cycle nevertheless. Signed-off-by: Mike Travis Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic.c b/arch/x86/kernel/apic.c index d19aa3aab62..4b6df2469fe 100644 --- a/arch/x86/kernel/apic.c +++ b/arch/x86/kernel/apic.c @@ -1837,6 +1837,11 @@ void __cpuinit generic_processor_info(int apicid, int version) num_processors++; cpu = cpumask_next_zero(-1, cpu_present_mask); + if (version != apic_version[boot_cpu_physical_apicid]) + WARN_ONCE(1, + "ACPI: apic version mismatch, bootcpu: %x cpu %d: %x\n", + apic_version[boot_cpu_physical_apicid], cpu, version); + physid_set(apicid, phys_cpu_present_map); if (apicid == boot_cpu_physical_apicid) { /* -- cgit v1.2.3 From c7f8562a51c2e5dcc1a00a2bdd232b9965ff960d Mon Sep 17 00:00:00 2001 From: Leonardo Potenza Date: Sun, 18 Jan 2009 23:03:56 +0100 Subject: x86: fix section mismatch warnings in kernel/setup_percpu.c The function setup_cpu_local_masks() has been marked __init, in order to remove the following section mismatch messages: WARNING: vmlinux.o(.text+0x3c2c7): Section mismatch in reference from the function setup_cpu_local_masks() to the function .init.text:alloc_bootmem_cpumask_var() The function setup_cpu_local_masks() references the function __init alloc_bootmem_cpumask_var(). This is often because setup_cpu_local_masks lacks a __init annotation or the annotation of alloc_bootmem_cpumask_var is wrong. WARNING: vmlinux.o(.text+0x3c2d3): Section mismatch in reference from the function setup_cpu_local_masks() to the function .init.text:alloc_bootmem_cpumask_var() The function setup_cpu_local_masks() references the function __init alloc_bootmem_cpumask_var(). This is often because setup_cpu_local_masks lacks a __init annotation or the annotation of alloc_bootmem_cpumask_var is wrong. WARNING: vmlinux.o(.text+0x3c2df): Section mismatch in reference from the function setup_cpu_local_masks() to the function .init.text:alloc_bootmem_cpumask_var() The function setup_cpu_local_masks() references the function __init alloc_bootmem_cpumask_var(). This is often because setup_cpu_local_masks lacks a __init annotation or the annotation of alloc_bootmem_cpumask_var is wrong. WARNING: vmlinux.o(.text+0x3c2eb): Section mismatch in reference from the function setup_cpu_local_masks() to the function .init.text:alloc_bootmem_cpumask_var() The function setup_cpu_local_masks() references the function __init alloc_bootmem_cpumask_var(). This is often because setup_cpu_local_masks lacks a __init annotation or the annotation of alloc_bootmem_cpumask_var is wrong. Signed-off-by: Leonardo Potenza Signed-off-by: Ingo Molnar --- arch/x86/kernel/setup_percpu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index 55c46074eba..01161077a49 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -136,7 +136,7 @@ static void __init setup_cpu_pda_map(void) #ifdef CONFIG_X86_64 /* correctly size the local cpu masks */ -static void setup_cpu_local_masks(void) +static void __init setup_cpu_local_masks(void) { alloc_bootmem_cpumask_var(&cpu_initialized_mask); alloc_bootmem_cpumask_var(&cpu_callin_mask); -- cgit v1.2.3 From 72859081851af2bda04117ca3d64206ffa199e5e Mon Sep 17 00:00:00 2001 From: Mike Travis Date: Fri, 16 Jan 2009 15:31:15 -0800 Subject: cpufreq: use work_on_cpu in acpi-cpufreq.c for drv_read and drv_write Impact: use new work_on_cpu function to reduce stack usage Replace the saving of current->cpus_allowed and set_cpus_allowed_ptr() with a work_on_cpu function for drv_read() and drv_write(). Basically converts do_drv_{read,write} into "work_on_cpu" functions that are now called by drv_read and drv_write. Note: This patch basically reverts 50c668d6 which reverted 7503bfba, now that the work_on_cpu() function is more stable. Signed-off-by: Mike Travis Acked-by: Rusty Russell Tested-by: Dieter Ries Tested-by: Maciej Rutecki Cc: Dave Jones Cc: Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c | 22 +++++++++------------- 1 file changed, 9 insertions(+), 13 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c index 8f3c95c7e61..5bf2c834a23 100644 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -150,8 +150,9 @@ struct drv_cmd { u32 val; }; -static void do_drv_read(struct drv_cmd *cmd) +static long do_drv_read(void *_cmd) { + struct drv_cmd *cmd = _cmd; u32 h; switch (cmd->type) { @@ -166,10 +167,12 @@ static void do_drv_read(struct drv_cmd *cmd) default: break; } + return 0; } -static void do_drv_write(struct drv_cmd *cmd) +static long do_drv_write(void *_cmd) { + struct drv_cmd *cmd = _cmd; u32 lo, hi; switch (cmd->type) { @@ -186,30 +189,23 @@ static void do_drv_write(struct drv_cmd *cmd) default: break; } + return 0; } static void drv_read(struct drv_cmd *cmd) { - cpumask_t saved_mask = current->cpus_allowed; cmd->val = 0; - set_cpus_allowed_ptr(current, cmd->mask); - do_drv_read(cmd); - set_cpus_allowed_ptr(current, &saved_mask); + work_on_cpu(cpumask_any(cmd->mask), do_drv_read, cmd); } static void drv_write(struct drv_cmd *cmd) { - cpumask_t saved_mask = current->cpus_allowed; unsigned int i; for_each_cpu(i, cmd->mask) { - set_cpus_allowed_ptr(current, cpumask_of(i)); - do_drv_write(cmd); + work_on_cpu(i, do_drv_write, cmd); } - - set_cpus_allowed_ptr(current, &saved_mask); - return; } static u32 get_cur_val(const struct cpumask *mask) @@ -366,7 +362,7 @@ static unsigned int get_cur_freq_on_cpu(unsigned int cpu) return freq; } -static unsigned int check_freqs(const cpumask_t *mask, unsigned int freq, +static unsigned int check_freqs(const struct cpumask *mask, unsigned int freq, struct acpi_cpufreq_data *data) { unsigned int cur_freq; -- cgit v1.2.3 From bfa318ad52a23f1e303d176b44366cdd2bb71ad2 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 15 Jan 2009 15:46:08 +0100 Subject: fix: crash: IP: __bitmap_intersects+0x48/0x73 -tip testing found this crash: > [ 35.258515] calling acpi_cpufreq_init+0x0/0x127 @ 1 > [ 35.264127] BUG: unable to handle kernel NULL pointer dereference at (null) > [ 35.267554] IP: [] __bitmap_intersects+0x48/0x73 > [ 35.267554] PGD 0 > [ 35.267554] Oops: 0000 [#1] SMP DEBUG_PAGEALLOC arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c is still broken: there's no allocation of the variable mask, so we pass in an uninitialized cmd.mask field to drv_read(), which then passes it to the scheduler which then crashes ... Switch it over to the much simpler constant-cpumask-pointers approach. Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c index 5bf2c834a23..4b1c319d30c 100644 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -145,7 +145,7 @@ typedef union { struct drv_cmd { unsigned int type; - cpumask_var_t mask; + const struct cpumask *mask; drv_addr_union addr; u32 val; }; @@ -231,6 +231,7 @@ static u32 get_cur_val(const struct cpumask *mask) return 0; } + cmd.mask = mask; drv_read(&cmd); dprintk("get_cur_val = %u\n", cmd.val); @@ -397,9 +398,6 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy, return -ENODEV; } - if (unlikely(!alloc_cpumask_var(&cmd.mask, GFP_KERNEL))) - return -ENOMEM; - perf = data->acpi_data; result = cpufreq_frequency_table_target(policy, data->freq_table, @@ -444,9 +442,9 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy, /* cpufreq holds the hotplug lock, so we are safe from here on */ if (policy->shared_type != CPUFREQ_SHARED_TYPE_ANY) - cpumask_and(cmd.mask, cpu_online_mask, policy->cpus); + cmd.mask = policy->cpus; else - cpumask_copy(cmd.mask, cpumask_of(policy->cpu)); + cmd.mask = cpumask_of(policy->cpu); freqs.old = perf->states[perf->state].core_frequency * 1000; freqs.new = data->freq_table[next_state].frequency; @@ -473,7 +471,6 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy, perf->state = next_perf_state; out: - free_cpumask_var(cmd.mask); return result; } -- cgit v1.2.3 From f5495506c3c1300d249d403c36f92de71920dbeb Mon Sep 17 00:00:00 2001 From: Gary Hade Date: Mon, 19 Jan 2009 13:46:41 -0800 Subject: x86: remove kernel_physical_mapping_init() from init section Impact: fix crash with memory hotplug enabled kernel_physical_mapping_init() is called during memory hotplug so it does not belong in the init section. If the kernel is built with CONFIG_DEBUG_SECTION_MISMATCH=y on the make command line, arch/x86/mm/init_64.c is compiled with the -fno-inline-functions-called-once gcc option defeating inlining of kernel_physical_mapping_init() within init_memory_mapping(). When kernel_physical_mapping_init() is not inlined it is placed in the .init.text section according to the __init in it's current declaration. A later call to kernel_physical_mapping_init() during a memory hotplug operation encounters an int3 trap because the .init.text section memory has been freed. This patch eliminates the crash caused by the int3 trap by moving the non-inlined kernel_physical_mapping_init() from .init.text to .meminit.text. Signed-off-by: Gary Hade Signed-off-by: Ingo Molnar --- arch/x86/mm/init_64.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 23f68e77ad1..e6d36b49025 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -596,7 +596,7 @@ static void __init init_gbpages(void) direct_gbpages = 0; } -static unsigned long __init kernel_physical_mapping_init(unsigned long start, +static unsigned long __meminit kernel_physical_mapping_init(unsigned long start, unsigned long end, unsigned long page_size_mask) { -- cgit v1.2.3 From e0a96129db574d6365e3439d16d88517c437ab33 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 16 Jan 2009 15:22:11 +0100 Subject: x86: use early clobbers in usercopy*.c Impact: fix rare (but currently harmless) miscompile with certain configs and gcc versions Hugh Dickins noticed that strncpy_from_user() was miscompiled in some circumstances with gcc 4.3. Thanks to Hugh's excellent analysis it was easy to track down. Hugh writes: > Try building an x86_64 defconfig 2.6.29-rc1 kernel tree, > except not quite defconfig, switch CONFIG_PREEMPT_NONE=y > and CONFIG_PREEMPT_VOLUNTARY off (because it expands a > might_fault() there, which hides the issue): using a > gcc 4.3.2 (I've checked both openSUSE 11.1 and Fedora 10). > > It generates the following: > > 0000000000000000 <__strncpy_from_user>: > 0: 48 89 d1 mov %rdx,%rcx > 3: 48 85 c9 test %rcx,%rcx > 6: 74 0e je 16 <__strncpy_from_user+0x16> > 8: ac lods %ds:(%rsi),%al > 9: aa stos %al,%es:(%rdi) > a: 84 c0 test %al,%al > c: 74 05 je 13 <__strncpy_from_user+0x13> > e: 48 ff c9 dec %rcx > 11: 75 f5 jne 8 <__strncpy_from_user+0x8> > 13: 48 29 c9 sub %rcx,%rcx > 16: 48 89 c8 mov %rcx,%rax > 19: c3 retq > > Observe that "sub %rcx,%rcx; mov %rcx,%rax", whereas gcc 4.2.1 > (and many other configs) say "sub %rcx,%rdx; mov %rdx,%rax". > Isn't it returning 0 when it ought to be returning strlen? The asm constraints for the strncpy_from_user() result were missing an early clobber, which tells gcc that the last output arguments are written before all input arguments are read. Also add more early clobbers in the rest of the file and fix 32-bit usercopy.c in the same way. Signed-off-by: Andi Kleen Signed-off-by: H. Peter Anvin [ since this API is rarely used and no in-kernel user relies on a 'len' return value (they only rely on negative return values) this miscompile was never noticed in the field. But it's worth fixing it nevertheless. ] Signed-off-by: Ingo Molnar --- arch/x86/lib/usercopy_32.c | 4 ++-- arch/x86/lib/usercopy_64.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/lib/usercopy_32.c b/arch/x86/lib/usercopy_32.c index 4a20b2f9a38..7c8ca91bb9e 100644 --- a/arch/x86/lib/usercopy_32.c +++ b/arch/x86/lib/usercopy_32.c @@ -56,7 +56,7 @@ do { \ " jmp 2b\n" \ ".previous\n" \ _ASM_EXTABLE(0b,3b) \ - : "=d"(res), "=c"(count), "=&a" (__d0), "=&S" (__d1), \ + : "=&d"(res), "=&c"(count), "=&a" (__d0), "=&S" (__d1), \ "=&D" (__d2) \ : "i"(-EFAULT), "0"(count), "1"(count), "3"(src), "4"(dst) \ : "memory"); \ @@ -218,7 +218,7 @@ long strnlen_user(const char __user *s, long n) " .align 4\n" " .long 0b,2b\n" ".previous" - :"=r" (n), "=D" (s), "=a" (res), "=c" (tmp) + :"=&r" (n), "=&D" (s), "=&a" (res), "=&c" (tmp) :"0" (n), "1" (s), "2" (0), "3" (mask) :"cc"); return res & mask; diff --git a/arch/x86/lib/usercopy_64.c b/arch/x86/lib/usercopy_64.c index 64d6c84e635..ec13cb5f17e 100644 --- a/arch/x86/lib/usercopy_64.c +++ b/arch/x86/lib/usercopy_64.c @@ -32,7 +32,7 @@ do { \ " jmp 2b\n" \ ".previous\n" \ _ASM_EXTABLE(0b,3b) \ - : "=r"(res), "=c"(count), "=&a" (__d0), "=&S" (__d1), \ + : "=&r"(res), "=&c"(count), "=&a" (__d0), "=&S" (__d1), \ "=&D" (__d2) \ : "i"(-EFAULT), "0"(count), "1"(count), "3"(src), "4"(dst) \ : "memory"); \ @@ -86,7 +86,7 @@ unsigned long __clear_user(void __user *addr, unsigned long size) ".previous\n" _ASM_EXTABLE(0b,3b) _ASM_EXTABLE(1b,2b) - : [size8] "=c"(size), [dst] "=&D" (__d0) + : [size8] "=&c"(size), [dst] "=&D" (__d0) : [size1] "r"(size & 7), "[size8]" (size / 8), "[dst]"(addr), [zero] "r" (0UL), [eight] "r" (8UL)); return size; -- cgit v1.2.3 From 552b8aa4d1edcc1c764ff6f61a7686347a2d1827 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 20 Jan 2009 09:31:49 +0100 Subject: Revert "x86: signal: change type of paramter for sys_rt_sigreturn()" This reverts commit 4217458dafaa57d8e26a46f5d05ab8c53cf64191. Justin Madru bisected this commit, it was causing weird Firefox crashes. The reason is that GCC mis-optimizes (re-uses) the on-stack parameters of the calling frame, which corrupts the syscall return pt_regs state and thus corrupts user-space register state. So we go back to the slightly less clean but more optimization-safe method of getting to pt_regs. Also add a comment to explain this. Resolves: http://bugzilla.kernel.org/show_bug.cgi?id=12505 Reported-and-bisected-by: Justin Madru Tested-by: Justin Madru Signed-off-by: Ingo Molnar --- arch/x86/include/asm/syscalls.h | 2 +- arch/x86/kernel/signal.c | 11 +++++++++-- 2 files changed, 10 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/syscalls.h b/arch/x86/include/asm/syscalls.h index 9c6797c3e56..c0b0bda754e 100644 --- a/arch/x86/include/asm/syscalls.h +++ b/arch/x86/include/asm/syscalls.h @@ -40,7 +40,7 @@ asmlinkage int sys_sigaction(int, const struct old_sigaction __user *, struct old_sigaction __user *); asmlinkage int sys_sigaltstack(unsigned long); asmlinkage unsigned long sys_sigreturn(unsigned long); -asmlinkage int sys_rt_sigreturn(struct pt_regs); +asmlinkage int sys_rt_sigreturn(unsigned long); /* kernel/ioport.c */ asmlinkage long sys_iopl(unsigned long); diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c index 89bb7668041..df0587f24c5 100644 --- a/arch/x86/kernel/signal.c +++ b/arch/x86/kernel/signal.c @@ -632,9 +632,16 @@ badframe: } #ifdef CONFIG_X86_32 -asmlinkage int sys_rt_sigreturn(struct pt_regs regs) +/* + * Note: do not pass in pt_regs directly as with tail-call optimization + * GCC will incorrectly stomp on the caller's frame and corrupt user-space + * register state: + */ +asmlinkage int sys_rt_sigreturn(unsigned long __unused) { - return do_rt_sigreturn(®s); + struct pt_regs *regs = (struct pt_regs *)&__unused; + + return do_rt_sigreturn(regs); } #else /* !CONFIG_X86_32 */ asmlinkage long sys_rt_sigreturn(struct pt_regs *regs) -- cgit v1.2.3 From a1e46212a410793d575718818e81ddc442a65283 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 20 Jan 2009 14:20:21 -0800 Subject: x86: fix page attribute corruption with cpa() Impact: fix sporadic slowdowns and warning messages This patch fixes a performance issue reported by Linus on his Nehalem system. While Linus reverted the PAT patch (commit 58dab916dfb57328d50deb0aa9b3fc92efa248ff) which exposed the issue, existing cpa() code can potentially still cause wrong(page attribute corruption) behavior. This patch also fixes the "WARNING: at arch/x86/mm/pageattr.c:560" that various people reported. In 64bit kernel, kernel identity mapping might have holes depending on the available memory and how e820 reports the address range covering the RAM, ACPI, PCI reserved regions. If there is a 2MB/1GB hole in the address range that is not listed by e820 entries, kernel identity mapping will have a corresponding hole in its 1-1 identity mapping. If cpa() happens on the kernel identity mapping which falls into these holes, existing code fails like this: __change_page_attr_set_clr() __change_page_attr() returns 0 because of if (!kpte). But doesn't set cpa->numpages and cpa->pfn. cpa_process_alias() uses uninitialized cpa->pfn (random value) which can potentially lead to changing the page attribute of kernel text/data, kernel identity mapping of RAM pages etc. oops! This bug was easily exposed by another PAT patch which was doing cpa() more often on kernel identity mapping holes (physical range between max_low_pfn_mapped and 4GB), where in here it was setting the cache disable attribute(PCD) for kernel identity mappings aswell. Fix cpa() to handle the kernel identity mapping holes. Retain the WARN() for cpa() calls to other not present address ranges (kernel-text/data, ioremap() addresses) Signed-off-by: Suresh Siddha Signed-off-by: Venkatesh Pallipadi Cc: Signed-off-by: Ingo Molnar --- arch/x86/mm/pageattr.c | 49 ++++++++++++++++++++++++++++++++++--------------- 1 file changed, 34 insertions(+), 15 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index e89d24815f2..84ba74820ad 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -534,6 +534,36 @@ out_unlock: return 0; } +static int __cpa_process_fault(struct cpa_data *cpa, unsigned long vaddr, + int primary) +{ + /* + * Ignore all non primary paths. + */ + if (!primary) + return 0; + + /* + * Ignore the NULL PTE for kernel identity mapping, as it is expected + * to have holes. + * Also set numpages to '1' indicating that we processed cpa req for + * one virtual address page and its pfn. TBD: numpages can be set based + * on the initial value and the level returned by lookup_address(). + */ + if (within(vaddr, PAGE_OFFSET, + PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT))) { + cpa->numpages = 1; + cpa->pfn = __pa(vaddr) >> PAGE_SHIFT; + return 0; + } else { + WARN(1, KERN_WARNING "CPA: called for zero pte. " + "vaddr = %lx cpa->vaddr = %lx\n", vaddr, + *cpa->vaddr); + + return -EFAULT; + } +} + static int __change_page_attr(struct cpa_data *cpa, int primary) { unsigned long address; @@ -549,17 +579,11 @@ static int __change_page_attr(struct cpa_data *cpa, int primary) repeat: kpte = lookup_address(address, &level); if (!kpte) - return 0; + return __cpa_process_fault(cpa, address, primary); old_pte = *kpte; - if (!pte_val(old_pte)) { - if (!primary) - return 0; - WARN(1, KERN_WARNING "CPA: called for zero pte. " - "vaddr = %lx cpa->vaddr = %lx\n", address, - *cpa->vaddr); - return -EINVAL; - } + if (!pte_val(old_pte)) + return __cpa_process_fault(cpa, address, primary); if (level == PG_LEVEL_4K) { pte_t new_pte; @@ -657,12 +681,7 @@ static int cpa_process_alias(struct cpa_data *cpa) vaddr = *cpa->vaddr; if (!(within(vaddr, PAGE_OFFSET, - PAGE_OFFSET + (max_low_pfn_mapped << PAGE_SHIFT)) -#ifdef CONFIG_X86_64 - || within(vaddr, PAGE_OFFSET + (1UL<<32), - PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT)) -#endif - )) { + PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT)))) { alias_cpa = *cpa; temp_cpa_vaddr = (unsigned long) __va(cpa->pfn << PAGE_SHIFT); -- cgit v1.2.3 From 731f1872f4e8a0f1eabd49c3548207e79a421202 Mon Sep 17 00:00:00 2001 From: Thomas Renninger Date: Tue, 20 Jan 2009 10:37:39 +0100 Subject: x86: mtrr fix debug boot parameter while looking at: http://bugzilla.kernel.org/show_bug.cgi?id=11541 I realized that the mtrr.show param cannot work, because the code is processed much too early. This patch: - Declares mtrr.show as early_param - Stays consistent with the previous param (which I doubt that it ever worked), so mtrr.show=1 would still work - Declares mtrr_show as initdata Signed-off-by: Thomas Renninger Acked-by: Jan Beulich Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/mtrr/generic.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index b59ddcc88cd..0c0a455fe95 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -33,11 +33,13 @@ u64 mtrr_tom2; struct mtrr_state_type mtrr_state = {}; EXPORT_SYMBOL_GPL(mtrr_state); -#undef MODULE_PARAM_PREFIX -#define MODULE_PARAM_PREFIX "mtrr." - -static int mtrr_show; -module_param_named(show, mtrr_show, bool, 0); +static int __initdata mtrr_show; +static int __init mtrr_debug(char *opt) +{ + mtrr_show = 1; + return 0; +} +early_param("mtrr.show", mtrr_debug); /* * Returns the effective MTRR type for the region -- cgit v1.2.3 From 9597134218300c045cf219be3664615e97cb239c Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 13 Jan 2009 10:21:30 -0800 Subject: x86: fix PTE corruption issue while mapping RAM using /dev/mem Beschorner Daniel reported: > hwinfo problem since 2.6.28, showing this in the oops: > Corrupted page table at address 7fd04de3ec00 Also, PaX Team reported a regression with this commit: > commit 9542ada803198e6eba29d3289abb39ea82047b92 > Author: Suresh Siddha > Date: Wed Sep 24 08:53:33 2008 -0700 > > x86: track memtype for RAM in page struct This commit breaks mapping any RAM page through /dev/mem, as the reserve_memtype() was not initializing the return attribute type and as such corrupting the PTE entry that was setup with the return attribute type. Because of this bug, application mapping this RAM page through /dev/mem will die with "Corrupted page table at address xxxx" message in the kernel log and also the kernel identity mapping which maps the underlying RAM page gets converted to UC. Fix this by initializing the return attribute type before calling reserve_ram_pages_type() Reported-by: PaX Team Reported-and-tested-by: Beschorner Daniel Tested-and-Acked-by: PaX Team Signed-off-by: Suresh Siddha Signed-off-by: Venkatesh Pallipadi Signed-off-by: Ingo Molnar --- arch/x86/mm/pat.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index 070ee4a3b22..ffc88cc00fd 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -333,6 +333,9 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, req_type & _PAGE_CACHE_MASK); } + if (new_type) + *new_type = actual_type; + /* * For legacy reasons, some parts of the physical address range in the * legacy 1MB region is treated as non-RAM (even when listed as RAM in @@ -356,9 +359,6 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, new->end = end; new->type = actual_type; - if (new_type) - *new_type = actual_type; - spin_lock(&memtype_lock); if (cached_entry && start >= cached_start) -- cgit v1.2.3 From bdf21a49bab28f0d9613e8d8724ef9c9168b61b9 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Wed, 21 Jan 2009 15:01:56 -0800 Subject: x86: add MSR_IA32_MISC_ENABLE bits to Impact: None (new bit definitions currently unused) Add bit definitions for the MSR_IA32_MISC_ENABLE MSRs to . Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/msr-index.h | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index cb58643947b..358acc59ae0 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -202,6 +202,35 @@ #define MSR_IA32_THERM_STATUS 0x0000019c #define MSR_IA32_MISC_ENABLE 0x000001a0 +/* MISC_ENABLE bits: architectural */ +#define MSR_IA32_MISC_ENABLE_FAST_STRING (1ULL << 0) +#define MSR_IA32_MISC_ENABLE_TCC (1ULL << 1) +#define MSR_IA32_MISC_ENABLE_EMON (1ULL << 7) +#define MSR_IA32_MISC_ENABLE_BTS_UNAVAIL (1ULL << 11) +#define MSR_IA32_MISC_ENABLE_PEBS_UNAVAIL (1ULL << 12) +#define MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP (1ULL << 16) +#define MSR_IA32_MISC_ENABLE_MWAIT (1ULL << 18) +#define MSR_IA32_MISC_ENABLE_LIMIT_CPUID (1ULL << 22) +#define MSR_IA32_MISC_ENABLE_XTPR_DISABLE (1ULL << 23) +#define MSR_IA32_MISC_ENABLE_XD_DISABLE (1ULL << 34) + +/* MISC_ENABLE bits: model-specific, meaning may vary from core to core */ +#define MSR_IA32_MISC_ENABLE_X87_COMPAT (1ULL << 2) +#define MSR_IA32_MISC_ENABLE_TM1 (1ULL << 3) +#define MSR_IA32_MISC_ENABLE_SPLIT_LOCK_DISABLE (1ULL << 4) +#define MSR_IA32_MISC_ENABLE_L3CACHE_DISABLE (1ULL << 6) +#define MSR_IA32_MISC_ENABLE_SUPPRESS_LOCK (1ULL << 8) +#define MSR_IA32_MISC_ENABLE_PREFETCH_DISABLE (1ULL << 9) +#define MSR_IA32_MISC_ENABLE_FERR (1ULL << 10) +#define MSR_IA32_MISC_ENABLE_FERR_MULTIPLEX (1ULL << 10) +#define MSR_IA32_MISC_ENABLE_TM2 (1ULL << 13) +#define MSR_IA32_MISC_ENABLE_ADJ_PREF_DISABLE (1ULL << 19) +#define MSR_IA32_MISC_ENABLE_SPEEDSTEP_LOCK (1ULL << 20) +#define MSR_IA32_MISC_ENABLE_L1D_CONTEXT (1ULL << 24) +#define MSR_IA32_MISC_ENABLE_DCU_PREF_DISABLE (1ULL << 37) +#define MSR_IA32_MISC_ENABLE_TURBO_DISABLE (1ULL << 38) +#define MSR_IA32_MISC_ENABLE_IP_PREF_DISABLE (1ULL << 39) + /* Intel Model 6 */ #define MSR_P6_EVNTSEL0 0x00000186 #define MSR_P6_EVNTSEL1 0x00000187 -- cgit v1.2.3 From 066941bd4eeb159307a5d7d795100d0887c00442 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Wed, 21 Jan 2009 15:04:32 -0800 Subject: x86: unmask CPUID levels on Intel CPUs Impact: Fixes crashes with misconfigured BIOSes on XSAVE hardware Avuton Olrich reported early boot crashes with v2.6.28 and bisected it down to dc1e35c6e95e8923cf1d3510438b63c600fee1e2 ("x86, xsave: enable xsave/xrstor on cpus with xsave support"). If the CPUID limit bit in MSR_IA32_MISC_ENABLE is set, clear it to make all CPUID information available. This is required for some features to work, in particular XSAVE. Reported-and-bisected-by: Avuton Olrich Tested-by: Avuton Olrich Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/intel.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 8ea6929e974..43c1dcf0bec 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -29,6 +29,16 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) { + u64 misc_enable; + + /* Unmask CPUID levels if masked */ + if (!rdmsrl_safe(MSR_IA32_MISC_ENABLE, &misc_enable) && + (misc_enable & MSR_IA32_MISC_ENABLE_LIMIT_CPUID)) { + misc_enable &= ~MSR_IA32_MISC_ENABLE_LIMIT_CPUID; + wrmsrl(MSR_IA32_MISC_ENABLE, misc_enable); + c->cpuid_level = cpuid_eax(0); + } + if ((c->x86 == 0xf && c->x86_model >= 0x03) || (c->x86 == 0x6 && c->x86_model >= 0x0e)) set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); -- cgit v1.2.3 From 336f6c322d87806ef93afad6308ac65083a865e5 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 22 Jan 2009 09:50:44 +0100 Subject: debugobjects: add and use INIT_WORK_ON_STACK Impact: Fix debugobjects warning debugobject enabled kernels spit out a warning in hpet code due to a workqueue which is initialized on stack. Add INIT_WORK_ON_STACK() which calls init_timer_on_stack() and use it in hpet. Signed-off-by: Thomas Gleixner --- arch/x86/kernel/hpet.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index cd759ad9069..64d5ad0b8ad 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -628,11 +628,12 @@ static int hpet_cpuhp_notify(struct notifier_block *n, switch (action & 0xf) { case CPU_ONLINE: - INIT_DELAYED_WORK(&work.work, hpet_work); + INIT_DELAYED_WORK_ON_STACK(&work.work, hpet_work); init_completion(&work.complete); /* FIXME: add schedule_work_on() */ schedule_delayed_work_on(cpu, &work.work, 0); wait_for_completion(&work.complete); + destroy_timer_on_stack(&work.work.timer); break; case CPU_DEAD: if (hdev) { -- cgit v1.2.3 From ba2607fe9c1f2d4ad5a3d4c4ae9117c5bfdca826 Mon Sep 17 00:00:00 2001 From: Markus Metzger Date: Mon, 19 Jan 2009 10:38:35 +0100 Subject: x86, ds, bts: cleanup/fix DS configuration Cleanup the cpuid check for DS configuration. This also fixes a Corei7 CPUID enumeration bug. Signed-off-by: Markus Metzger Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- arch/x86/kernel/ds.c | 31 +++++++++++++++++-------------- 1 file changed, 17 insertions(+), 14 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index da91701a234..169a120587b 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -15,8 +15,8 @@ * - buffer allocation (memory accounting) * * - * Copyright (C) 2007-2008 Intel Corporation. - * Markus Metzger , 2007-2008 + * Copyright (C) 2007-2009 Intel Corporation. + * Markus Metzger , 2007-2009 */ @@ -890,7 +890,7 @@ int ds_set_pebs_reset(struct pebs_tracer *tracer, u64 value) } static const struct ds_configuration ds_cfg_netburst = { - .name = "netburst", + .name = "Netburst", .ctl[dsf_bts] = (1 << 2) | (1 << 3), .ctl[dsf_bts_kernel] = (1 << 5), .ctl[dsf_bts_user] = (1 << 6), @@ -904,7 +904,7 @@ static const struct ds_configuration ds_cfg_netburst = { #endif }; static const struct ds_configuration ds_cfg_pentium_m = { - .name = "pentium m", + .name = "Pentium M", .ctl[dsf_bts] = (1 << 6) | (1 << 7), .sizeof_field = sizeof(long), @@ -915,8 +915,8 @@ static const struct ds_configuration ds_cfg_pentium_m = { .sizeof_rec[ds_pebs] = sizeof(long) * 18, #endif }; -static const struct ds_configuration ds_cfg_core2 = { - .name = "core 2", +static const struct ds_configuration ds_cfg_core2_atom = { + .name = "Core 2/Atom", .ctl[dsf_bts] = (1 << 6) | (1 << 7), .ctl[dsf_bts_kernel] = (1 << 9), .ctl[dsf_bts_user] = (1 << 10), @@ -949,19 +949,22 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c) switch (c->x86) { case 0x6: switch (c->x86_model) { - case 0 ... 0xC: - /* sorry, don't know about them */ - break; - case 0xD: - case 0xE: /* Pentium M */ + case 0x9: + case 0xd: /* Pentium M */ ds_configure(&ds_cfg_pentium_m); break; - default: /* Core2, Atom, ... */ - ds_configure(&ds_cfg_core2); + case 0xf: + case 0x17: /* Core2 */ + case 0x1c: /* Atom */ + ds_configure(&ds_cfg_core2_atom); + break; + case 0x1a: /* i7 */ + default: + /* sorry, don't know about them */ break; } break; - case 0xF: + case 0xf: switch (c->x86_model) { case 0x0: case 0x1: -- cgit v1.2.3 From 42ef73fe134732b2e91c0326df5fd568da17c4b2 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 23 Jan 2009 17:37:49 +0100 Subject: x86, mm: fix pte_free() On -rt we were seeing spurious bad page states like: Bad page state in process 'firefox' page:c1bc2380 flags:0x40000000 mapping:c1bc2390 mapcount:0 count:0 Trying to fix it up, but a reboot is needed Backtrace: Pid: 503, comm: firefox Not tainted 2.6.26.8-rt13 #3 [] ? printk+0x14/0x19 [] bad_page+0x4e/0x79 [] free_hot_cold_page+0x5b/0x1d3 [] free_hot_page+0xf/0x11 [] __free_pages+0x20/0x2b [] __pte_alloc+0x87/0x91 [] handle_mm_fault+0xe4/0x733 [] ? rt_mutex_down_read_trylock+0x57/0x63 [] ? rt_mutex_down_read_trylock+0x57/0x63 [] do_page_fault+0x36f/0x88a This is the case where a concurrent fault already installed the PTE and we get to free the newly allocated one. This is due to pgtable_page_ctor() doing the spin_lock_init(&page->ptl) which is overlaid with the {private, mapping} struct. union { struct { unsigned long private; struct address_space *mapping; }; spinlock_t ptl; struct kmem_cache *slab; struct page *first_page; }; Normally the spinlock is small enough to not stomp on page->mapping, but PREEMPT_RT=y has huge 'spin'locks. But lockdep kernels should also be able to trigger this splat, as the lock tracking code grows the spinlock to cover page->mapping. The obvious fix is calling pgtable_page_dtor() like the regular pte free path __pte_free_tlb() does. It seems all architectures except x86 and nm10300 already do this, and nm10300 doesn't seem to use pgtable_page_ctor(), which suggests it doesn't do SMP or simply doesnt do MMU at all or something. Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar Cc: --- arch/x86/include/asm/pgalloc.h | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/pgalloc.h b/arch/x86/include/asm/pgalloc.h index cb7c151a8bf..dd14c54ac71 100644 --- a/arch/x86/include/asm/pgalloc.h +++ b/arch/x86/include/asm/pgalloc.h @@ -42,6 +42,7 @@ static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte) static inline void pte_free(struct mm_struct *mm, struct page *pte) { + pgtable_page_dtor(pte); __free_page(pte); } -- cgit v1.2.3 From e1b4d1143651fb3838be1117785b6e0386fa151f Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sun, 25 Jan 2009 16:57:00 +0100 Subject: x86: use standard PIT frequency the RDC and ELAN platforms use slighly different PIT clocks, resulting in a timex.h hack that changes PIT_TICK_RATE during build time. But if a tester enables any of these platform support .config options, the PIT will be miscalibrated on standard PC platforms. So use one frequency - in a subsequent patch we'll add a quirk to allow x86 platforms to define different PIT frequencies. Signed-off-by: Ingo Molnar --- arch/x86/include/asm/timex.h | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/timex.h b/arch/x86/include/asm/timex.h index 1287dc1347d..b5c9d45c981 100644 --- a/arch/x86/include/asm/timex.h +++ b/arch/x86/include/asm/timex.h @@ -1,18 +1,13 @@ -/* x86 architecture timex specifications */ #ifndef _ASM_X86_TIMEX_H #define _ASM_X86_TIMEX_H #include #include -#ifdef CONFIG_X86_ELAN -# define PIT_TICK_RATE 1189200 /* AMD Elan has different frequency! */ -#elif defined(CONFIG_X86_RDC321X) -# define PIT_TICK_RATE 1041667 /* Underlying HZ for R8610 */ -#else -# define PIT_TICK_RATE 1193182 /* Underlying HZ */ -#endif -#define CLOCK_TICK_RATE PIT_TICK_RATE +/* The PIT ticks at this frequency (in HZ): */ +#define PIT_TICK_RATE 1193182 + +#define CLOCK_TICK_RATE PIT_TICK_RATE #define ARCH_HAS_READ_CURRENT_TIMER -- cgit v1.2.3 From ef5fa0ab24b87646c7bc98645acbb4b51fc2acd4 Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Fri, 23 Jan 2009 14:14:21 -0800 Subject: x86: work around PAGE_KERNEL_WC not getting WC in iomap_atomic_prot_pfn. In the absence of PAT, PAGE_KERNEL_WC ends up mapping to a memory type that gets UC behavior even in the presence of a WC MTRR covering the area in question. By swapping to PAGE_KERNEL_UC_MINUS, we can get the actual behavior the caller wanted (WC if you can manage it, UC otherwise). This recovers the 40% performance improvement of using WC in the DRM to upload vertex data. Signed-off-by: Eric Anholt Signed-off-by: H. Peter Anvin --- arch/x86/mm/iomap_32.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/mm/iomap_32.c b/arch/x86/mm/iomap_32.c index d0151d8ce45..ca53224fc56 100644 --- a/arch/x86/mm/iomap_32.c +++ b/arch/x86/mm/iomap_32.c @@ -17,6 +17,7 @@ */ #include +#include #include /* Map 'pfn' using fixed map 'type' and protections 'prot' @@ -29,6 +30,15 @@ iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot) pagefault_disable(); + /* + * For non-PAT systems, promote PAGE_KERNEL_WC to PAGE_KERNEL_UC_MINUS. + * PAGE_KERNEL_WC maps to PWT, which translates to uncached if the + * MTRR is UC or WC. UC_MINUS gets the real intention, of the + * user, which is "WC if the MTRR is WC, UC if you can't do that." + */ + if (!pat_enabled && pgprot_val(prot) == pgprot_val(PAGE_KERNEL_WC)) + prot = PAGE_KERNEL_UC_MINUS; + idx = type + KM_TYPE_NR*smp_processor_id(); vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); set_pte(kmap_pte-idx, pfn_pte(pfn, prot)); -- cgit v1.2.3 From 99fb4d349db7e7dacb2099c5cc320a9e2d31c1ef Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Mon, 26 Jan 2009 04:30:41 +0100 Subject: x86: unmask CPUID levels on Intel CPUs, fix Impact: fix boot hang on pre-model-15 Intel CPUs rdmsrl_safe() does not work in very early bootup code yet, because we dont have the pagefault handler installed yet so exception section does not get parsed. rdmsr_safe() will just crash and hang the bootup. So limit the MSR_IA32_MISC_ENABLE MSR read to those CPU types that support it. Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/intel.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 43c1dcf0bec..549f2ada55f 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -29,14 +29,17 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) { - u64 misc_enable; - - /* Unmask CPUID levels if masked */ - if (!rdmsrl_safe(MSR_IA32_MISC_ENABLE, &misc_enable) && - (misc_enable & MSR_IA32_MISC_ENABLE_LIMIT_CPUID)) { - misc_enable &= ~MSR_IA32_MISC_ENABLE_LIMIT_CPUID; - wrmsrl(MSR_IA32_MISC_ENABLE, misc_enable); - c->cpuid_level = cpuid_eax(0); + /* Unmask CPUID levels if masked: */ + if (c->x86 == 6 && c->x86_model >= 15) { + u64 misc_enable; + + rdmsrl(MSR_IA32_MISC_ENABLE, misc_enable); + + if (misc_enable & MSR_IA32_MISC_ENABLE_LIMIT_CPUID) { + misc_enable &= ~MSR_IA32_MISC_ENABLE_LIMIT_CPUID; + wrmsrl(MSR_IA32_MISC_ENABLE, misc_enable); + c->cpuid_level = cpuid_eax(0); + } } if ((c->x86 == 0xf && c->x86_model >= 0x03) || -- cgit v1.2.3 From 659d2618b38f8a4d91bdb19cfc5c7fb330a4c55a Mon Sep 17 00:00:00 2001 From: Rakib Mullick Date: Sat, 24 Jan 2009 01:46:03 +0600 Subject: x86: fix section mismatch warning Here function vmi_activate calls a init function activate_vmi , which causes the following section mismatch warnings: LD arch/x86/kernel/built-in.o WARNING: arch/x86/kernel/built-in.o(.text+0x13ba9): Section mismatch in reference from the function vmi_activate() to the function .init.text:vmi_time_init() The function vmi_activate() references the function __init vmi_time_init(). This is often because vmi_activate lacks a __init annotation or the annotation of vmi_time_init is wrong. WARNING: arch/x86/kernel/built-in.o(.text+0x13bd1): Section mismatch in reference from the function vmi_activate() to the function .devinit.text:vmi_time_bsp_init() The function vmi_activate() references the function __devinit vmi_time_bsp_init(). This is often because vmi_activate lacks a __devinit annotation or the annotation of vmi_time_bsp_init is wrong. WARNING: arch/x86/kernel/built-in.o(.text+0x13bdb): Section mismatch in reference from the function vmi_activate() to the function .devinit.text:vmi_time_ap_init() The function vmi_activate() references the function __devinit vmi_time_ap_init(). This is often because vmi_activate lacks a __devinit annotation or the annotation of vmi_time_ap_init is wrong. Fix it by marking vmi_activate() as __init too. Signed-off-by: Rakib Mullick Signed-off-by: Ingo Molnar --- arch/x86/kernel/vmi_32.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c index 23206ba1687..1d3302cc2dd 100644 --- a/arch/x86/kernel/vmi_32.c +++ b/arch/x86/kernel/vmi_32.c @@ -858,7 +858,7 @@ void __init vmi_init(void) #endif } -void vmi_activate(void) +void __init vmi_activate(void) { unsigned long flags; -- cgit v1.2.3 From 30a0fb947a68ad3ab8a7184e3b3d79dce10e3688 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Mon, 26 Jan 2009 09:40:58 -0800 Subject: x86: correct the CPUID pattern for MSR_IA32_MISC_ENABLE availability Impact: re-enable CPUID unmasking on affected processors As far as I am capable of discerning from the documentation, MSR_IA32_MISC_ENABLE should be available for all family 0xf CPUs, as well as family 6 for model >= 0xd (newer Pentium M). The documentation on this isn't ideal, so we need to be on the lookout for errors, still. Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/intel.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 549f2ada55f..430e5c38a54 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -30,7 +30,7 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) { /* Unmask CPUID levels if masked: */ - if (c->x86 == 6 && c->x86_model >= 15) { + if (c->x86 > 6 || (c->x86 == 6 && c->x86_model >= 0xd)) { u64 misc_enable; rdmsrl(MSR_IA32_MISC_ENABLE, misc_enable); -- cgit v1.2.3 From dd7f8dbe2b3c0611ba969cd867c10cb63d163e25 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Mon, 26 Jan 2009 21:19:57 +0100 Subject: eeprom: More consistent symbol names Now that all EEPROM drivers live in the same place, let's harmonize their symbol names. Also fix eeprom's dependencies, it definitely needs sysfs, and is no longer experimental after many years in the kernel tree. Signed-off-by: Jean Delvare Acked-by: Wolfram Sang Cc: David Brownell --- arch/x86/configs/i386_defconfig | 4 ++-- arch/x86/configs/x86_64_defconfig | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/configs/i386_defconfig b/arch/x86/configs/i386_defconfig index b30a08ed8eb..edba00d98ac 100644 --- a/arch/x86/configs/i386_defconfig +++ b/arch/x86/configs/i386_defconfig @@ -1331,8 +1331,8 @@ CONFIG_I2C_I801=y # Miscellaneous I2C Chip support # # CONFIG_DS1682 is not set -# CONFIG_AT24 is not set -# CONFIG_SENSORS_EEPROM is not set +# CONFIG_EEPROM_AT24 is not set +# CONFIG_EEPROM_LEGACY is not set # CONFIG_SENSORS_PCF8574 is not set # CONFIG_PCF8575 is not set # CONFIG_SENSORS_PCA9539 is not set diff --git a/arch/x86/configs/x86_64_defconfig b/arch/x86/configs/x86_64_defconfig index 0e7dbc0a3e4..322dd2748fc 100644 --- a/arch/x86/configs/x86_64_defconfig +++ b/arch/x86/configs/x86_64_defconfig @@ -1311,8 +1311,8 @@ CONFIG_I2C_I801=y # Miscellaneous I2C Chip support # # CONFIG_DS1682 is not set -# CONFIG_AT24 is not set -# CONFIG_SENSORS_EEPROM is not set +# CONFIG_EEPROM_AT24 is not set +# CONFIG_EEPROM_LEGACY is not set # CONFIG_SENSORS_PCF8574 is not set # CONFIG_PCF8575 is not set # CONFIG_SENSORS_PCA9539 is not set -- cgit v1.2.3 From 57064d213d2e44654d4f13c66df135b5e7389a26 Mon Sep 17 00:00:00 2001 From: Seth Heasley Date: Fri, 23 Jan 2009 12:43:38 -0800 Subject: PCI: irq and pci_ids patch for Intel Tigerpoint DeviceIDs This patch adds the Intel Tigerpoint LPC Controller DeviceIDs. Signed-off-by: Seth Heasley Signed-off-by: Jesse Barnes --- arch/x86/pci/irq.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86') diff --git a/arch/x86/pci/irq.c b/arch/x86/pci/irq.c index 4064345cf14..fecbce6e7d7 100644 --- a/arch/x86/pci/irq.c +++ b/arch/x86/pci/irq.c @@ -572,6 +572,7 @@ static __init int intel_router_probe(struct irq_router *r, struct pci_dev *route case PCI_DEVICE_ID_INTEL_ICH7_1: case PCI_DEVICE_ID_INTEL_ICH7_30: case PCI_DEVICE_ID_INTEL_ICH7_31: + case PCI_DEVICE_ID_INTEL_TGP_LPC: case PCI_DEVICE_ID_INTEL_ESB2_0: case PCI_DEVICE_ID_INTEL_ICH8_0: case PCI_DEVICE_ID_INTEL_ICH8_1: -- cgit v1.2.3 From bf3647c44bc76c43c4b2ebb4c37a559e899ac70e Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 29 Jan 2009 11:45:35 +0100 Subject: x86: tone down mtrr_trim_uncached_memory() warning kerneloops.org is reporting a lot of these warnings that come due to vmware not setting up any MTRRs for emulated CPUs: | Reported 709 times (14696 total reports) | BIOS bug (often in VMWare) where the MTRR's are set up incorrectly | or not at all | | This warning was last seen in version 2.6.29-rc2-git1, and first | seen in 2.6.24. | | More info: | http://www.kerneloops.org/searchweek.php?search=mtrr_trim_uncached_memory Keep a one-liner KERN_INFO about it - so that we have so notice if empty MTRRs are caused by native hardware/BIOS weirdness. Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/mtrr/main.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c index d259e5d2e05..236a401b825 100644 --- a/arch/x86/kernel/cpu/mtrr/main.c +++ b/arch/x86/kernel/cpu/mtrr/main.c @@ -1594,8 +1594,7 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) /* kvm/qemu doesn't have mtrr set right, don't trim them all */ if (!highest_pfn) { - WARN(!kvm_para_available(), KERN_WARNING - "WARNING: strange, CPU MTRRs all blank?\n"); + printk(KERN_INFO "CPU MTRRs all blank - virtualized system.\n"); return 0; } -- cgit v1.2.3 From 72410af921cbc9018da388ca1ddf75880a033ac1 Mon Sep 17 00:00:00 2001 From: Atsushi SAKAI Date: Fri, 16 Jan 2009 20:39:14 +0900 Subject: lguest: typos fix 3 points lguest_asm.S => i386_head.S LHCALL_BREAK => LHREQ_BREAK perferred => preferred Signed-off-by: Atsushi SAKAI Signed-off-by: Rusty Russell --- arch/x86/lguest/boot.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index a7ed208f81e..92f1c6f3e19 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -931,7 +931,7 @@ static void lguest_restart(char *reason) * that we can fit comfortably. * * First we need assembly templates of each of the patchable Guest operations, - * and these are in lguest_asm.S. */ + * and these are in i386_head.S. */ /*G:060 We construct a table from the assembler templates: */ static const struct lguest_insns @@ -1093,7 +1093,7 @@ __init void lguest_init(void) acpi_ht = 0; #endif - /* We set the perferred console to "hvc". This is the "hypervisor + /* We set the preferred console to "hvc". This is the "hypervisor * virtual console" driver written by the PowerPC people, which we also * adapted for lguest's use. */ add_preferred_console("hvc", 0, NULL); -- cgit v1.2.3 From 5872fb94f85d2e4fdef94657bd14e1a492df9825 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 29 Jan 2009 16:28:02 -0800 Subject: Documentation: move DMA-mapping.txt to Doc/PCI/ Move DMA-mapping.txt to Documentation/PCI/. DMA-mapping.txt was supposed to be moved from Documentation/ to Documentation/PCI/. The 00-INDEX files in those two directories were updated, along with a few other text files, but the file itself somehow escaped being moved, so move it and update more text files and source files with its new location. Signed-off-by: Randy Dunlap Acked-by: Greg Kroah-Hartman cc: Jesse Barnes Signed-off-by: Linus Torvalds --- arch/x86/include/asm/dma-mapping.h | 4 ++-- arch/x86/kernel/pci-gart_64.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/dma-mapping.h b/arch/x86/include/asm/dma-mapping.h index 4035357f5b9..132a134d12f 100644 --- a/arch/x86/include/asm/dma-mapping.h +++ b/arch/x86/include/asm/dma-mapping.h @@ -2,8 +2,8 @@ #define _ASM_X86_DMA_MAPPING_H /* - * IOMMU interface. See Documentation/DMA-mapping.txt and DMA-API.txt for - * documentation. + * IOMMU interface. See Documentation/PCI/PCI-DMA-mapping.txt and + * Documentation/DMA-API.txt for documentation. */ #include diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 00c2bcd4146..d5768b1af08 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -5,7 +5,7 @@ * This allows to use PCI devices that only support 32bit addresses on systems * with more than 4GB. * - * See Documentation/DMA-mapping.txt for the interface specification. + * See Documentation/PCI/PCI-DMA-mapping.txt for the interface specification. * * Copyright 2002 Andi Kleen, SuSE Labs. * Subject to the GNU General Public License v2 only. -- cgit v1.2.3 From 999721ca6d0c2540341acb73ac9048cbd6b05d3a Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Rajput Date: Fri, 30 Jan 2009 22:44:58 +0530 Subject: headers_check fix: x86, e820.h fix the following 'make headers_check' warning: usr/include/asm/e820.h:44: found __[us]{8,16,32,64} type without #include Signed-off-by: Jaswinder Singh Rajput --- arch/x86/include/asm/e820.h | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/e820.h b/arch/x86/include/asm/e820.h index 3d8ceddbd40..00d41ce4c84 100644 --- a/arch/x86/include/asm/e820.h +++ b/arch/x86/include/asm/e820.h @@ -49,6 +49,7 @@ #define E820_RESERVED_KERN 128 #ifndef __ASSEMBLY__ +#include struct e820entry { __u64 addr; /* start of memory segment */ __u64 size; /* size of memory segment */ -- cgit v1.2.3 From cef3767852a9b1a7ff4a8dfe0969e2d32eb728df Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Rajput Date: Fri, 30 Jan 2009 22:46:08 +0530 Subject: headers_check fix: x86, kvm.h fix the following 'make headers_check' warnings: usr/include/asm/kvm.h:9: include of is preferred over usr/include/asm/kvm.h:16: found __[us]{8,16,32,64} type without #include Signed-off-by: Jaswinder Singh Rajput --- arch/x86/include/asm/kvm.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/kvm.h b/arch/x86/include/asm/kvm.h index b95162af0bf..d2e3bf3608a 100644 --- a/arch/x86/include/asm/kvm.h +++ b/arch/x86/include/asm/kvm.h @@ -6,7 +6,7 @@ * */ -#include +#include #include /* Architectural interrupt line count. */ -- cgit v1.2.3 From 999b697b9d8b15756e65da72c816ef4363a945a5 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Rajput Date: Fri, 30 Jan 2009 22:47:27 +0530 Subject: headers_check fix: x86, mce.h fix the following 'make headers_check' warnings: usr/include/asm/mce.h:7: include of is preferred over usr/include/asm/mce.h:29: found __[us]{8,16,32,64} type without #include Signed-off-by: Jaswinder Singh Rajput --- arch/x86/include/asm/mce.h | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index 1d6e17c2f23..32c6e17b960 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -3,8 +3,8 @@ #ifdef __x86_64__ +#include #include -#include /* * Machine Check support for x86 @@ -115,8 +115,6 @@ extern int mce_notify_user(void); #endif /* !CONFIG_X86_32 */ - - #ifdef CONFIG_X86_MCE extern void mcheck_init(struct cpuinfo_x86 *c); #else @@ -126,5 +124,4 @@ extern void stop_mce(void); extern void restart_mce(void); #endif /* __KERNEL__ */ - #endif /* _ASM_X86_MCE_H */ -- cgit v1.2.3 From d122072cc079d299e5191c9cbb6162ba8791624c Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Rajput Date: Fri, 30 Jan 2009 22:48:17 +0530 Subject: headers_check fix: x86, mtrr.h fix the following 'make headers_check' warning: usr/include/asm/mtrr.h:61: found __[us]{8,16,32,64} type without #include Signed-off-by: Jaswinder Singh Rajput --- arch/x86/include/asm/mtrr.h | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/mtrr.h b/arch/x86/include/asm/mtrr.h index 14080d22edb..a51ada8467d 100644 --- a/arch/x86/include/asm/mtrr.h +++ b/arch/x86/include/asm/mtrr.h @@ -23,6 +23,7 @@ #ifndef _ASM_X86_MTRR_H #define _ASM_X86_MTRR_H +#include #include #include -- cgit v1.2.3 From 420ab35eef206d147973d26db14b5618868726be Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Rajput Date: Fri, 30 Jan 2009 22:52:16 +0530 Subject: headers_check fix: x86, ptrace-abi.h fix the following 'make headers_check' warnings: usr/include/asm/ptrace-abi.h:86: include of is preferred over usr/include/asm/ptrace-abi.h:93: found __[us]{8,16,32,64} type without #include Signed-off-by: Jaswinder Singh Rajput --- arch/x86/include/asm/ptrace-abi.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/ptrace-abi.h b/arch/x86/include/asm/ptrace-abi.h index 25f1bb8fc62..8e0f8d199e0 100644 --- a/arch/x86/include/asm/ptrace-abi.h +++ b/arch/x86/include/asm/ptrace-abi.h @@ -83,7 +83,7 @@ #ifdef CONFIG_X86_PTRACE_BTS #ifndef __ASSEMBLY__ -#include +#include /* configuration/status structure used in PTRACE_BTS_CONFIG and PTRACE_BTS_STATUS commands. -- cgit v1.2.3 From e59afe6a21dce7bb3c63ba4f894a3195ae3d5529 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Rajput Date: Fri, 30 Jan 2009 22:53:49 +0530 Subject: headers_check fix: x86, sigcontext.h fix the following 'make headers_check' warnings: usr/include/asm/sigcontext.h:5: include of is preferred over usr/include/asm/sigcontext.h:24: found __[us]{8,16,32,64} type without #include Signed-off-by: Jaswinder Singh Rajput --- arch/x86/include/asm/sigcontext.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/sigcontext.h b/arch/x86/include/asm/sigcontext.h index 0afcb5e58ac..ec666491aaa 100644 --- a/arch/x86/include/asm/sigcontext.h +++ b/arch/x86/include/asm/sigcontext.h @@ -2,7 +2,7 @@ #define _ASM_X86_SIGCONTEXT_H #include -#include +#include #define FP_XSTATE_MAGIC1 0x46505853U #define FP_XSTATE_MAGIC2 0x46505845U -- cgit v1.2.3 From 2de548faa78c650bb20c4680ee3a225cca33a45d Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Rajput Date: Fri, 30 Jan 2009 22:55:20 +0530 Subject: headers_check fix: x86, sigcontext32.h fix the following 'make headers_check' warning: usr/include/asm/sigcontext32.h:20: found __[us]{8,16,32,64} type without #include Signed-off-by: Jaswinder Singh Rajput --- arch/x86/include/asm/sigcontext32.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/sigcontext32.h b/arch/x86/include/asm/sigcontext32.h index 6126188cf3a..ad1478c4ae1 100644 --- a/arch/x86/include/asm/sigcontext32.h +++ b/arch/x86/include/asm/sigcontext32.h @@ -1,6 +1,8 @@ #ifndef _ASM_X86_SIGCONTEXT32_H #define _ASM_X86_SIGCONTEXT32_H +#include + /* signal context for 32bit programs. */ #define X86_FXSR_MAGIC 0x0000 -- cgit v1.2.3 From 7cff3608d2553a045b676fa81b0cf54e4f2cc5ce Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Rajput Date: Fri, 30 Jan 2009 22:57:38 +0530 Subject: headers_check fix: x86, swab.h fix the following 'make headers_check' warnings: usr/include/asm/swab.h:4: include of is preferred over usr/include/asm/swab.h:7: found __[us]{8,16,32,64} type without #include Signed-off-by: Jaswinder Singh Rajput --- arch/x86/include/asm/swab.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/swab.h b/arch/x86/include/asm/swab.h index 306d4178ffc..557cd9f0066 100644 --- a/arch/x86/include/asm/swab.h +++ b/arch/x86/include/asm/swab.h @@ -1,7 +1,7 @@ #ifndef _ASM_X86_SWAB_H #define _ASM_X86_SWAB_H -#include +#include #include static inline __attribute_const__ __u32 __arch_swab32(__u32 val) -- cgit v1.2.3 From 7fc49f19813030f2e15ad2ccec5cb701f7f4a3ec Mon Sep 17 00:00:00 2001 From: Andreas Schwab Date: Tue, 27 Jan 2009 21:45:57 +0100 Subject: x86 setup: fix asm constraints in vesa_store_edid Impact: fix potential miscompile (currently believed non-manifest) As the comment explains, the VBE DDC call can clobber any register. Tell the compiler about that fact. Signed-off-by: Andreas Schwab Signed-off-by: H. Peter Anvin Signed-off-by: Ingo Molnar --- arch/x86/boot/video-vesa.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/boot/video-vesa.c b/arch/x86/boot/video-vesa.c index 75115849af3..4a58c8ce3f6 100644 --- a/arch/x86/boot/video-vesa.c +++ b/arch/x86/boot/video-vesa.c @@ -269,9 +269,8 @@ void vesa_store_edid(void) we genuinely have to assume all registers are destroyed here. */ asm("pushw %%es; movw %2,%%es; "INT10"; popw %%es" - : "+a" (ax), "+b" (bx) - : "c" (cx), "D" (di) - : "esi"); + : "+a" (ax), "+b" (bx), "+c" (cx), "+D" (di) + : : "esi", "edx"); if (ax != 0x004f) return; /* No EDID */ @@ -285,9 +284,9 @@ void vesa_store_edid(void) dx = 0; /* EDID block number */ di =(size_t) &boot_params.edid_info; /* (ES:)Pointer to block */ asm(INT10 - : "+a" (ax), "+b" (bx), "+d" (dx), "=m" (boot_params.edid_info) - : "c" (cx), "D" (di) - : "esi"); + : "+a" (ax), "+b" (bx), "+d" (dx), "=m" (boot_params.edid_info), + "+c" (cx), "+D" (di) + : : "esi"); #endif /* CONFIG_FIRMWARE_EDID */ } -- cgit v1.2.3 From 92ab78315c638515d0e81b0c70b2082f713582d9 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Sat, 31 Jan 2009 17:24:43 +0100 Subject: x86/Voyager: make it build and boot [ mingo@elte.hu: these fixes are a subset of changes cherry-picked from: git://git.kernel.org:/pub/scm/linux/kernel/git/jejb/voyager-2.6.git They fix various problems that recent x86 changes caused in the Voyager subarchitecture: both APIC changes and cpumask changes and certain cleanups caused subarch assumptions to break. Most of these changes are obsolete as the subarch code has been removed from the x86 development tree - but we merge them upstream to make Voyager build and boot. ] Signed-off-by: James Bottomley Signed-off-by: Ingo Molnar --- arch/x86/kernel/irqinit_32.c | 12 ------------ arch/x86/mach-default/setup.c | 12 ++++++++++++ arch/x86/mach-voyager/setup.c | 12 +++++++++++- arch/x86/mach-voyager/voyager_smp.c | 25 ++++++++++++------------- 4 files changed, 35 insertions(+), 26 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/irqinit_32.c b/arch/x86/kernel/irqinit_32.c index 1507ad4e674..10a09c2f182 100644 --- a/arch/x86/kernel/irqinit_32.c +++ b/arch/x86/kernel/irqinit_32.c @@ -78,15 +78,6 @@ void __init init_ISA_irqs(void) } } -/* - * IRQ2 is cascade interrupt to second interrupt controller - */ -static struct irqaction irq2 = { - .handler = no_action, - .mask = CPU_MASK_NONE, - .name = "cascade", -}; - DEFINE_PER_CPU(vector_irq_t, vector_irq) = { [0 ... IRQ0_VECTOR - 1] = -1, [IRQ0_VECTOR] = 0, @@ -178,9 +169,6 @@ void __init native_init_IRQ(void) alloc_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt); #endif - if (!acpi_ioapic) - setup_irq(2, &irq2); - /* setup after call gates are initialised (usually add in * the architecture specific gates) */ diff --git a/arch/x86/mach-default/setup.c b/arch/x86/mach-default/setup.c index df167f26562..a265a7c6319 100644 --- a/arch/x86/mach-default/setup.c +++ b/arch/x86/mach-default/setup.c @@ -38,6 +38,15 @@ void __init pre_intr_init_hook(void) init_ISA_irqs(); } +/* + * IRQ2 is cascade interrupt to second interrupt controller + */ +static struct irqaction irq2 = { + .handler = no_action, + .mask = CPU_MASK_NONE, + .name = "cascade", +}; + /** * intr_init_hook - post gate setup interrupt initialisation * @@ -53,6 +62,9 @@ void __init intr_init_hook(void) if (x86_quirks->arch_intr_init()) return; } + if (!acpi_ioapic) + setup_irq(2, &irq2); + } /** diff --git a/arch/x86/mach-voyager/setup.c b/arch/x86/mach-voyager/setup.c index a580b9562e7..d914a7996a6 100644 --- a/arch/x86/mach-voyager/setup.c +++ b/arch/x86/mach-voyager/setup.c @@ -33,13 +33,23 @@ void __init intr_init_hook(void) setup_irq(2, &irq2); } -void __init pre_setup_arch_hook(void) +static void voyager_disable_tsc(void) { /* Voyagers run their CPUs from independent clocks, so disable * the TSC code because we can't sync them */ setup_clear_cpu_cap(X86_FEATURE_TSC); } +void __init pre_setup_arch_hook(void) +{ + voyager_disable_tsc(); +} + +void __init pre_time_init_hook(void) +{ + voyager_disable_tsc(); +} + void __init trap_init_hook(void) { } diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c index 9840b7ec749..7ffcdeec463 100644 --- a/arch/x86/mach-voyager/voyager_smp.c +++ b/arch/x86/mach-voyager/voyager_smp.c @@ -81,7 +81,7 @@ static void enable_local_vic_irq(unsigned int irq); static void disable_local_vic_irq(unsigned int irq); static void before_handle_vic_irq(unsigned int irq); static void after_handle_vic_irq(unsigned int irq); -static void set_vic_irq_affinity(unsigned int irq, cpumask_t mask); +static void set_vic_irq_affinity(unsigned int irq, const struct cpumask *mask); static void ack_vic_irq(unsigned int irq); static void vic_enable_cpi(void); static void do_boot_cpu(__u8 cpuid); @@ -211,8 +211,6 @@ static __u32 cpu_booted_map; static cpumask_t smp_commenced_mask = CPU_MASK_NONE; /* This is for the new dynamic CPU boot code */ -cpumask_t cpu_callin_map = CPU_MASK_NONE; -cpumask_t cpu_callout_map = CPU_MASK_NONE; /* The per processor IRQ masks (these are usually kept in sync) */ static __u16 vic_irq_mask[NR_CPUS] __cacheline_aligned; @@ -378,7 +376,7 @@ void __init find_smp_config(void) cpus_addr(phys_cpu_present_map)[0] |= voyager_extended_cmos_read(VOYAGER_PROCESSOR_PRESENT_MASK + 3) << 24; - cpu_possible_map = phys_cpu_present_map; + init_cpu_possible(&phys_cpu_present_map); printk("VOYAGER SMP: phys_cpu_present_map = 0x%lx\n", cpus_addr(phys_cpu_present_map)[0]); /* Here we set up the VIC to enable SMP */ @@ -1599,16 +1597,16 @@ static void after_handle_vic_irq(unsigned int irq) * change the mask and then do an interrupt enable CPI to re-enable on * the selected processors */ -void set_vic_irq_affinity(unsigned int irq, cpumask_t mask) +void set_vic_irq_affinity(unsigned int irq, const struct cpumask *mask) { /* Only extended processors handle interrupts */ unsigned long real_mask; unsigned long irq_mask = 1 << irq; int cpu; - real_mask = cpus_addr(mask)[0] & voyager_extended_vic_processors; + real_mask = cpus_addr(*mask)[0] & voyager_extended_vic_processors; - if (cpus_addr(mask)[0] == 0) + if (cpus_addr(*mask)[0] == 0) /* can't have no CPUs to accept the interrupt -- extremely * bad things will happen */ return; @@ -1750,10 +1748,11 @@ static void __cpuinit voyager_smp_prepare_boot_cpu(void) init_gdt(smp_processor_id()); switch_to_new_gdt(); - cpu_set(smp_processor_id(), cpu_online_map); - cpu_set(smp_processor_id(), cpu_callout_map); - cpu_set(smp_processor_id(), cpu_possible_map); - cpu_set(smp_processor_id(), cpu_present_map); + cpu_online_map = cpumask_of_cpu(smp_processor_id()); + cpu_callout_map = cpumask_of_cpu(smp_processor_id()); + cpu_callin_map = CPU_MASK_NONE; + cpu_present_map = cpumask_of_cpu(smp_processor_id()); + } static int __cpuinit voyager_cpu_up(unsigned int cpu) @@ -1783,9 +1782,9 @@ void __init smp_setup_processor_id(void) x86_write_percpu(cpu_number, hard_smp_processor_id()); } -static void voyager_send_call_func(cpumask_t callmask) +static void voyager_send_call_func(const struct cpumask *callmask) { - __u32 mask = cpus_addr(callmask)[0] & ~(1 << smp_processor_id()); + __u32 mask = cpus_addr(*callmask)[0] & ~(1 << smp_processor_id()); send_CPI(mask, VIC_CALL_FUNCTION_CPI); } -- cgit v1.2.3 From 9a8ecae87a2b698964b1db9ea504ba1099f479fc Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Sat, 31 Jan 2009 20:12:14 -0500 Subject: x86: add cache descriptors for Intel Core i7 Signed-off-by: Dave Jones Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/intel_cacheinfo.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index 48533d77be7..da299eb85fc 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -36,8 +36,11 @@ static struct _cache_table cache_table[] __cpuinitdata = { { 0x06, LVL_1_INST, 8 }, /* 4-way set assoc, 32 byte line size */ { 0x08, LVL_1_INST, 16 }, /* 4-way set assoc, 32 byte line size */ + { 0x09, LVL_1_INST, 32 }, /* 4-way set assoc, 64 byte line size */ { 0x0a, LVL_1_DATA, 8 }, /* 2 way set assoc, 32 byte line size */ { 0x0c, LVL_1_DATA, 16 }, /* 4-way set assoc, 32 byte line size */ + { 0x0d, LVL_1_DATA, 16 }, /* 4-way set assoc, 64 byte line size */ + { 0x21, LVL_2, 256 }, /* 8-way set assoc, 64 byte line size */ { 0x22, LVL_3, 512 }, /* 4-way set assoc, sectored cache, 64 byte line size */ { 0x23, LVL_3, 1024 }, /* 8-way set assoc, sectored cache, 64 byte line size */ { 0x25, LVL_3, 2048 }, /* 8-way set assoc, sectored cache, 64 byte line size */ @@ -85,6 +88,18 @@ static struct _cache_table cache_table[] __cpuinitdata = { 0x85, LVL_2, 2048 }, /* 8-way set assoc, 32 byte line size */ { 0x86, LVL_2, 512 }, /* 4-way set assoc, 64 byte line size */ { 0x87, LVL_2, 1024 }, /* 8-way set assoc, 64 byte line size */ + { 0xd0, LVL_3, 512 }, /* 4-way set assoc, 64 byte line size */ + { 0xd1, LVL_3, 1024 }, /* 4-way set assoc, 64 byte line size */ + { 0xd2, LVL_3, 2048 }, /* 4-way set assoc, 64 byte line size */ + { 0xd6, LVL_3, 1024 }, /* 8-way set assoc, 64 byte line size */ + { 0xd7, LVL_3, 2038 }, /* 8-way set assoc, 64 byte line size */ + { 0xd8, LVL_3, 4096 }, /* 12-way set assoc, 64 byte line size */ + { 0xdc, LVL_3, 2048 }, /* 12-way set assoc, 64 byte line size */ + { 0xdd, LVL_3, 4096 }, /* 12-way set assoc, 64 byte line size */ + { 0xde, LVL_3, 8192 }, /* 12-way set assoc, 64 byte line size */ + { 0xe2, LVL_3, 2048 }, /* 16-way set assoc, 64 byte line size */ + { 0xe3, LVL_3, 4096 }, /* 16-way set assoc, 64 byte line size */ + { 0xe4, LVL_3, 8192 }, /* 16-way set assoc, 64 byte line size */ { 0x00, 0, 0} }; -- cgit v1.2.3 From 10b888d6cec2688e65e9e128b14bf98ecd199da2 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sat, 31 Jan 2009 14:50:07 -0800 Subject: irq, x86: fix lock status with numa_migrate_irq_desc Eric Paris reported: > I have an hp dl785g5 which is unable to successfully run > 2.6.29-0.66.rc3.fc11.x86_64 or 2.6.29-rc2-next-20090126. During bootup > (early in userspace daemons starting) I get the below BUG, which quickly > renders the machine dead. I assume it is because sparse_irq_lock never > gets released when the BUG kills that task. Adjust lock sequence when migrating a descriptor with CONFIG_NUMA_MIGRATE_IRQ_DESC enabled. Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/kernel/io_apic.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c index 1c4a1302536..9b0c480c383 100644 --- a/arch/x86/kernel/io_apic.c +++ b/arch/x86/kernel/io_apic.c @@ -2528,14 +2528,15 @@ static void irq_complete_move(struct irq_desc **descp) vector = ~get_irq_regs()->orig_ax; me = smp_processor_id(); + + if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain)) { #ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC *descp = desc = move_irq_desc(desc, me); /* get the new one */ cfg = desc->chip_data; #endif - - if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain)) send_cleanup_vector(cfg); + } } #else static inline void irq_complete_move(struct irq_desc **descp) {} -- cgit v1.2.3 From a67798cd7bb130bf37f5ffb28f3260f4c10232db Mon Sep 17 00:00:00 2001 From: Martin Hicks Date: Fri, 30 Jan 2009 10:50:54 -0600 Subject: x86: push old stack address on irqstack for unwinder Impact: Fixes dumpstack and KDB on 64 bits This re-adds the old stack pointer to the top of the irqstack to help with unwinding. It was removed in commit d99015b1abbad743aa049b439c1e1dede6d0fa49 as part of the save_args out-of-line work. Both dumpstack and KDB require this information. Signed-off-by: Martin Hicks Signed-off-by: H. Peter Anvin --- arch/x86/kernel/entry_64.S | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index e28c7a98779..a1346217e43 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -346,6 +346,7 @@ ENTRY(save_args) popq_cfi %rax /* move return address... */ mov %gs:pda_irqstackptr,%rsp EMPTY_FRAME 0 + pushq_cfi %rbp /* backlink for unwinder */ pushq_cfi %rax /* ... to the new stack */ /* * We entered an interrupt context - irqs are off: -- cgit v1.2.3 From 06fc732c33a7ff5e4c91bcf4a6ca86b5e335ad9a Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Tue, 3 Feb 2009 16:01:46 -0800 Subject: xen: disable interrupts before saving in percpu Impact: Fix race condition xen_mc_batch has a small preempt race where it takes the address of a percpu variable immediately before disabling interrupts, thereby leaving a small window in which we may migrate to another cpu and save the flags in the wrong percpu variable. Disable interrupts before saving the old flags in a percpu. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: H. Peter Anvin --- arch/x86/xen/multicalls.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/xen/multicalls.h b/arch/x86/xen/multicalls.h index 85893824161..fa3e10725d9 100644 --- a/arch/x86/xen/multicalls.h +++ b/arch/x86/xen/multicalls.h @@ -19,8 +19,10 @@ DECLARE_PER_CPU(unsigned long, xen_mc_irq_flags); paired with xen_mc_issue() */ static inline void xen_mc_batch(void) { + unsigned long flags; /* need to disable interrupts until this entry is complete */ - local_irq_save(__get_cpu_var(xen_mc_irq_flags)); + local_irq_save(flags); + __get_cpu_var(xen_mc_irq_flags) = flags; } static inline struct multicall_space xen_mc_entry(size_t args) -- cgit v1.2.3 From 858770619debfb9269add63e4ba8b7c6b5538dd1 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Tue, 3 Feb 2009 16:24:22 +0100 Subject: x86: APIC: enable workaround on AMD Fam10h CPUs Impact: fix to enable APIC for AMD Fam10h on chipsets with a missing/b0rked ACPI MP table (MADT) Booting a 32bit kernel on an AMD Fam10h CPU running on chipsets with missing/b0rked MP table leads to a hang pretty early in the boot process due to the APIC not being initialized. Fix that by falling back to the default APIC base address in 32bit code, as it is done in the 64bit codepath. Signed-off-by: Borislav Petkov Signed-off-by: H. Peter Anvin --- arch/x86/kernel/apic.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic.c b/arch/x86/kernel/apic.c index 4b6df2469fe..115449f869e 100644 --- a/arch/x86/kernel/apic.c +++ b/arch/x86/kernel/apic.c @@ -1436,7 +1436,7 @@ static int __init detect_init_APIC(void) switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: if ((boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model > 1) || - (boot_cpu_data.x86 == 15)) + (boot_cpu_data.x86 >= 15)) break; goto no_apic; case X86_VENDOR_INTEL: -- cgit v1.2.3 From 62663ea8220366472fe20462831f2d69d7987439 Mon Sep 17 00:00:00 2001 From: Thomas Renninger Date: Tue, 3 Feb 2009 17:46:46 +0100 Subject: ACPI: cpufreq: Remove deprecated /proc/acpi/processor/../performance proc entries They were long enough set deprecated... Update Documentation/cpu-freq/users-guide.txt: The deprecated files listed there seen not to exist for some time anymore already. Signed-off-by: Thomas Renninger Signed-off-by: Len Brown --- arch/x86/kernel/cpu/cpufreq/Kconfig | 11 ----------- 1 file changed, 11 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/cpufreq/Kconfig b/arch/x86/kernel/cpu/cpufreq/Kconfig index efae3b22a0f..65792c2cc46 100644 --- a/arch/x86/kernel/cpu/cpufreq/Kconfig +++ b/arch/x86/kernel/cpu/cpufreq/Kconfig @@ -245,17 +245,6 @@ config X86_E_POWERSAVER comment "shared options" -config X86_ACPI_CPUFREQ_PROC_INTF - bool "/proc/acpi/processor/../performance interface (deprecated)" - depends on PROC_FS - depends on X86_ACPI_CPUFREQ || X86_POWERNOW_K7_ACPI || X86_POWERNOW_K8_ACPI - help - This enables the deprecated /proc/acpi/processor/../performance - interface. While it is helpful for debugging, the generic, - cross-architecture cpufreq interfaces should be used. - - If in doubt, say N. - config X86_SPEEDSTEP_LIB tristate default (X86_SPEEDSTEP_ICH || X86_SPEEDSTEP_SMI || X86_P4_CLOCKMOD) -- cgit v1.2.3 From 48ec4d9537282a55d602136724f069faafcac8c8 Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Wed, 4 Feb 2009 15:54:45 -0500 Subject: x86, 64-bit: print DMI info in the oops trace This patch echoes what we already do on 32-bit since 90f7d25c6b672137344f447a30a9159945ffea72, and prints the DMI product name in show_regs, so that system specific problems can be easily identified. Signed-off-by: Kyle McMartin Signed-off-by: Ingo Molnar --- arch/x86/kernel/process_64.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 416fb9282f4..85b4cb5c198 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -40,6 +40,7 @@ #include #include #include +#include #include #include @@ -151,14 +152,18 @@ void __show_regs(struct pt_regs *regs, int all) unsigned long d0, d1, d2, d3, d6, d7; unsigned int fsindex, gsindex; unsigned int ds, cs, es; + const char *board; printk("\n"); print_modules(); - printk(KERN_INFO "Pid: %d, comm: %.20s %s %s %.*s\n", + board = dmi_get_system_info(DMI_PRODUCT_NAME); + if (!board) + board = ""; + printk(KERN_INFO "Pid: %d, comm: %.20s %s %s %.*s %s\n", current->pid, current->comm, print_tainted(), init_utsname()->release, (int)strcspn(init_utsname()->version, " "), - init_utsname()->version); + init_utsname()->version, board); printk(KERN_INFO "RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->ip); printk_address(regs->ip, 1); printk(KERN_INFO "RSP: %04lx:%016lx EFLAGS: %08lx\n", regs->ss, -- cgit v1.2.3 From 36723bfe81f374fd8abc05a46f10a7cac2f01a75 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Wed, 4 Feb 2009 21:44:04 +0100 Subject: x86/Kconfig.cpu: make Kconfig help readable in the console Impact: cleanup Some lines exceed the 80 char width making them unreadable. Signed-off-by: Borislav Petkov Signed-off-by: Ingo Molnar --- arch/x86/Kconfig.cpu | 28 +++++++++++++++------------- 1 file changed, 15 insertions(+), 13 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu index 8078955845a..c98d52e8296 100644 --- a/arch/x86/Kconfig.cpu +++ b/arch/x86/Kconfig.cpu @@ -167,9 +167,9 @@ config MK7 config MK8 bool "Opteron/Athlon64/Hammer/K8" help - Select this for an AMD Opteron or Athlon64 Hammer-family processor. Enables - use of some extended instructions, and passes appropriate optimization - flags to GCC. + Select this for an AMD Opteron or Athlon64 Hammer-family processor. + Enables use of some extended instructions, and passes appropriate + optimization flags to GCC. config MCRUSOE bool "Crusoe" @@ -256,9 +256,11 @@ config MPSC config MCORE2 bool "Core 2/newer Xeon" help - Select this for Intel Core 2 and newer Core 2 Xeons (Xeon 51xx and 53xx) - CPUs. You can distinguish newer from older Xeons by the CPU family - in /proc/cpuinfo. Newer ones have 6 and older ones 15 (not a typo) + + Select this for Intel Core 2 and newer Core 2 Xeons (Xeon 51xx and + 53xx) CPUs. You can distinguish newer from older Xeons by the CPU + family in /proc/cpuinfo. Newer ones have 6 and older ones 15 + (not a typo) config GENERIC_CPU bool "Generic-x86-64" @@ -320,14 +322,14 @@ config X86_PPRO_FENCE bool "PentiumPro memory ordering errata workaround" depends on M686 || M586MMX || M586TSC || M586 || M486 || M386 || MGEODEGX1 help - Old PentiumPro multiprocessor systems had errata that could cause memory - operations to violate the x86 ordering standard in rare cases. Enabling this - option will attempt to work around some (but not all) occurances of - this problem, at the cost of much heavier spinlock and memory barrier - operations. + Old PentiumPro multiprocessor systems had errata that could cause + memory operations to violate the x86 ordering standard in rare cases. + Enabling this option will attempt to work around some (but not all) + occurances of this problem, at the cost of much heavier spinlock and + memory barrier operations. - If unsure, say n here. Even distro kernels should think twice before enabling - this: there are few systems, and an unlikely bug. + If unsure, say n here. Even distro kernels should think twice before + enabling this: there are few systems, and an unlikely bug. config X86_F00F_BUG def_bool y -- cgit v1.2.3 From a6a95406c676ffe4f9dee708eb404a17c69f7fdd Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Wed, 4 Feb 2009 13:40:31 +0300 Subject: x86: fix hpet timer reinit for x86_64 There's a small problem with hpet_rtc_reinit function - it checks for the: hpet_readl(HPET_COUNTER) - hpet_t1_cmp > 0 to continue increasing both the HPET_T1_CMP (register) and the hpet_t1_cmp (variable). But since the HPET_COUNTER is always 32-bit, if the hpet_t1_cmp is 64-bit this condition will always be FALSE once the latter hits the 32-bit boundary, and we can have a situation, when we don't increase the HPET_T1_CMP register high enough. The result - timer stops ticking, since HPET_T1_CMP becomes less, than the COUNTER and never increased again. The solution is (based on Linus's suggestion) to not compare 64-bits (on 64-bit x86), but to do the comparison on 32-bit signed integers. Reported-by: Kirill Korotaev Signed-off-by: Pavel Emelyanov Signed-off-by: Ingo Molnar --- arch/x86/kernel/hpet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 64d5ad0b8ad..c761f914430 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -1075,7 +1075,7 @@ static void hpet_rtc_timer_reinit(void) hpet_t1_cmp += delta; hpet_writel(hpet_t1_cmp, HPET_T1_CMP); lost_ints++; - } while ((long)(hpet_readl(HPET_COUNTER) - hpet_t1_cmp) > 0); + } while ((s32)(hpet_readl(HPET_COUNTER) - hpet_t1_cmp) > 0); if (lost_ints) { if (hpet_rtc_flags & RTC_PIE) -- cgit v1.2.3 From 4560839939f4b4a96e21e80584f87308ac93c1da Mon Sep 17 00:00:00 2001 From: Alex Chiang Date: Wed, 4 Feb 2009 16:44:01 -0700 Subject: x86: fix grammar in user-visible BIOS warning Fix user-visible grammo. Signed-off-by: Alex Chiang Signed-off-by: Ingo Molnar --- arch/x86/kernel/setup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index ae0d8042cf6..c461f6d6907 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -607,7 +607,7 @@ struct x86_quirks *x86_quirks __initdata = &default_x86_quirks; static int __init dmi_low_memory_corruption(const struct dmi_system_id *d) { printk(KERN_NOTICE - "%s detected: BIOS may corrupt low RAM, working it around.\n", + "%s detected: BIOS may corrupt low RAM, working around it.\n", d->ident); e820_update_range(0, 0x10000, E820_RAM, E820_RESERVED); -- cgit v1.2.3 From b534816b552d35bbd3c60702139ed5c7da2f55c2 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 4 Feb 2009 18:33:38 -0800 Subject: x86: don't apply __supported_pte_mask to non-present ptes On an x86 system which doesn't support global mappings, __supported_pte_mask has _PAGE_GLOBAL clear, to make sure it never appears in the PTE. pfn_pte() and so on will enforce it with: static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot) { return __pte((((phys_addr_t)page_nr << PAGE_SHIFT) | pgprot_val(pgprot)) & __supported_pte_mask); } However, we overload _PAGE_GLOBAL with _PAGE_PROTNONE on non-present ptes to distinguish them from swap entries. However, applying __supported_pte_mask indiscriminately will clear the bit and corrupt the pte. I guess the best fix is to only apply __supported_pte_mask to present ptes. This seems like the right solution to me, as it means we can completely ignore the issue of overlaps between the present pte bits and the non-present pte-as-swap entry use of the bits. __supported_pte_mask contains the set of flags we support on the current hardware. We also use bits in the pte for things like logically present ptes with no permissions, and swap entries for swapped out pages. We should only apply __supported_pte_mask to present ptes, because otherwise we may destroy other information being stored in the ptes. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/pgtable.h | 26 ++++++++++++++++++++------ arch/x86/include/asm/xen/page.h | 2 +- 2 files changed, 21 insertions(+), 7 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index 06bbcbd66e9..4f5af8447d5 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -302,16 +302,30 @@ static inline pte_t pte_mkspecial(pte_t pte) extern pteval_t __supported_pte_mask; +/* + * Mask out unsupported bits in a present pgprot. Non-present pgprots + * can use those bits for other purposes, so leave them be. + */ +static inline pgprotval_t massage_pgprot(pgprot_t pgprot) +{ + pgprotval_t protval = pgprot_val(pgprot); + + if (protval & _PAGE_PRESENT) + protval &= __supported_pte_mask; + + return protval; +} + static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot) { - return __pte((((phys_addr_t)page_nr << PAGE_SHIFT) | - pgprot_val(pgprot)) & __supported_pte_mask); + return __pte(((phys_addr_t)page_nr << PAGE_SHIFT) | + massage_pgprot(pgprot)); } static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot) { - return __pmd((((phys_addr_t)page_nr << PAGE_SHIFT) | - pgprot_val(pgprot)) & __supported_pte_mask); + return __pmd(((phys_addr_t)page_nr << PAGE_SHIFT) | + massage_pgprot(pgprot)); } static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) @@ -323,7 +337,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) * the newprot (if present): */ val &= _PAGE_CHG_MASK; - val |= pgprot_val(newprot) & (~_PAGE_CHG_MASK) & __supported_pte_mask; + val |= massage_pgprot(newprot) & ~_PAGE_CHG_MASK; return __pte(val); } @@ -339,7 +353,7 @@ static inline pgprot_t pgprot_modify(pgprot_t oldprot, pgprot_t newprot) #define pte_pgprot(x) __pgprot(pte_flags(x) & PTE_FLAGS_MASK) -#define canon_pgprot(p) __pgprot(pgprot_val(p) & __supported_pte_mask) +#define canon_pgprot(p) __pgprot(massage_pgprot(p)) static inline int is_new_memtype_allowed(unsigned long flags, unsigned long new_flags) diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h index 7ef617ef1df..4bd990ee43d 100644 --- a/arch/x86/include/asm/xen/page.h +++ b/arch/x86/include/asm/xen/page.h @@ -137,7 +137,7 @@ static inline pte_t mfn_pte(unsigned long page_nr, pgprot_t pgprot) pte_t pte; pte.pte = ((phys_addr_t)page_nr << PAGE_SHIFT) | - (pgprot_val(pgprot) & __supported_pte_mask); + massage_pgprot(pgprot); return pte; } -- cgit v1.2.3 From 0cd5c3c80a0ebd68c08312fa7d8c13149cc61c4c Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Wed, 4 Feb 2009 14:29:19 -0800 Subject: x86: disable intel_iommu support by default Due to recurring issues with DMAR support on certain platforms. There's a number of filesystem corruption incidents reported: https://bugzilla.redhat.com/show_bug.cgi?id=479996 http://bugzilla.kernel.org/show_bug.cgi?id=12578 Provide a Kconfig option to change whether it is enabled by default. If disabled, it can still be reenabled by passing intel_iommu=on to the kernel. Keep the .config option off by default. Signed-off-by: Kyle McMartin Signed-off-by: Andrew Morton Acked-By: David Woodhouse Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 73f7fe8fd4d..9c39095b33f 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1802,6 +1802,17 @@ config DMAR and include PCI device scope covered by these DMA remapping devices. +config DMAR_DEFAULT_ON + def_bool n + prompt "Enable DMA Remapping Devices by default" + depends on DMAR + help + Selecting this option will enable a DMAR device at boot time if + one is found. If this option is not selected, DMAR support can + be enabled by passing intel_iommu=on to the kernel. It is + recommended you say N here while the DMAR code remains + experimental. + config DMAR_GFX_WA def_bool y prompt "Support for Graphics workaround" -- cgit v1.2.3 From 732553e567c2700ba5b9bccc6ec885c75779a94b Mon Sep 17 00:00:00 2001 From: Mark Langsdorf Date: Tue, 3 Feb 2009 17:46:43 +0100 Subject: [CPUFREQ] powernow-k8: Get transition latency from ACPI _PSS table At this time, the PowerNow! driver for K8 uses an experimentally derived formula to calculate transition latency. The value it provides is orders of magnitude too large on modern systems. This patch replaces the formula with ACPI _PSS latency values for more accuracy and better performance. I've tested it on two 2nd generation Opteron systems, a 3rd generation Operton system, and a Turion X2 without seeing any stability problems. Signed-off-by: Mark Langsdorf Signed-off-by: Thomas Renninger Signed-off-by: Dave Jones --- arch/x86/kernel/cpu/cpufreq/powernow-k8.c | 28 ++++++++++++++++++++++------ 1 file changed, 22 insertions(+), 6 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c index 5c28b37dea1..fb039cd345d 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c @@ -939,10 +939,25 @@ static void powernow_k8_cpu_exit_acpi(struct powernow_k8_data *data) free_cpumask_var(data->acpi_data.shared_cpu_map); } +static int get_transition_latency(struct powernow_k8_data *data) +{ + int max_latency = 0; + int i; + for (i = 0; i < data->acpi_data.state_count; i++) { + int cur_latency = data->acpi_data.states[i].transition_latency + + data->acpi_data.states[i].bus_master_latency; + if (cur_latency > max_latency) + max_latency = cur_latency; + } + /* value in usecs, needs to be in nanoseconds */ + return 1000 * max_latency; +} + #else static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) { return -ENODEV; } static void powernow_k8_cpu_exit_acpi(struct powernow_k8_data *data) { return; } static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned int index) { return; } +static int get_transition_latency(struct powernow_k8_data *data) { return 0; } #endif /* CONFIG_X86_POWERNOW_K8_ACPI */ /* Take a frequency, and issue the fid/vid transition command */ @@ -1173,7 +1188,13 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) if (rc) { goto err_out; } - } + /* Take a crude guess here. + * That guess was in microseconds, so multiply with 1000 */ + pol->cpuinfo.transition_latency = ( + ((data->rvo + 8) * data->vstable * VST_UNITS_20US) + + ((1 << data->irt) * 30)) * 1000; + } else /* ACPI _PSS objects available */ + pol->cpuinfo.transition_latency = get_transition_latency(data); /* only run on specific CPU from here on */ oldmask = current->cpus_allowed; @@ -1204,11 +1225,6 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) cpumask_copy(pol->cpus, &per_cpu(cpu_core_map, pol->cpu)); data->available_cores = pol->cpus; - /* Take a crude guess here. - * That guess was in microseconds, so multiply with 1000 */ - pol->cpuinfo.transition_latency = (((data->rvo + 8) * data->vstable * VST_UNITS_20US) - + (3 * (1 << data->irt) * 10)) * 1000; - if (cpu_family == CPU_HW_PSTATE) pol->cur = find_khz_freq_from_pstate(data->powernow_table, data->currpstate); else -- cgit v1.2.3 From 9be260a646bf76fa418ee519afa10196b3164681 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Thu, 5 Feb 2009 17:12:39 -0500 Subject: prevent kprobes from catching spurious page faults Prevent kprobes from catching spurious faults which will cause infinite recursive page-fault and memory corruption by stack overflow. Signed-off-by: Masami Hiramatsu Cc: [2.6.28.x] Signed-off-by: Linus Torvalds --- arch/x86/mm/fault.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 90dfae511a4..c76ef1d701c 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -603,8 +603,6 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code) si_code = SEGV_MAPERR; - if (notify_page_fault(regs)) - return; if (unlikely(kmmio_fault(regs, address))) return; @@ -634,6 +632,9 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code) if (spurious_fault(address, error_code)) return; + /* kprobes don't want to hook the spurious faults. */ + if (notify_page_fault(regs)) + return; /* * Don't take the mm semaphore here. If we fixup a prefetch * fault we could otherwise deadlock. @@ -641,6 +642,9 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code) goto bad_area_nosemaphore; } + /* kprobes don't want to hook the spurious faults. */ + if (notify_page_fault(regs)) + return; /* * It's safe to allow irq's after cr2 has been saved and the -- cgit v1.2.3 From ff08f76d738d0ec0f334b187f61e160caa321d54 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Wed, 4 Feb 2009 13:40:31 +0300 Subject: x86: clean up hpet timer reinit Implement Linus's suggestion: introduce the hpet_cnt_ahead() helper function to compare hpet time values - like other wrapping counter comparisons are abstracted away elsewhere. (jiffies, ktime_t, etc.) Reported-by: Kirill Korotaev Signed-off-by: Pavel Emelyanov Signed-off-by: Ingo Molnar --- arch/x86/kernel/hpet.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index c761f914430..388254f69a2 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -897,13 +897,21 @@ static unsigned long hpet_rtc_flags; static int hpet_prev_update_sec; static struct rtc_time hpet_alarm_time; static unsigned long hpet_pie_count; -static unsigned long hpet_t1_cmp; +static u32 hpet_t1_cmp; static unsigned long hpet_default_delta; static unsigned long hpet_pie_delta; static unsigned long hpet_pie_limit; static rtc_irq_handler irq_handler; +/* + * Check that the hpet counter c1 is ahead of the c2 + */ +static inline int hpet_cnt_ahead(u32 c1, u32 c2) +{ + return (s32)(c2 - c1) < 0; +} + /* * Registers a IRQ handler. */ @@ -1075,7 +1083,7 @@ static void hpet_rtc_timer_reinit(void) hpet_t1_cmp += delta; hpet_writel(hpet_t1_cmp, HPET_T1_CMP); lost_ints++; - } while ((s32)(hpet_readl(HPET_COUNTER) - hpet_t1_cmp) > 0); + } while (!hpet_cnt_ahead(hpet_t1_cmp, hpet_readl(HPET_COUNTER))); if (lost_ints) { if (hpet_rtc_flags & RTC_PIE) -- cgit v1.2.3 From c09249f8d1b84344eca882547afdbffee8c09d14 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 6 Feb 2009 18:15:18 -0800 Subject: x86-64: fix int $0x80 -ENOSYS return One of my past fixes to this code introduced a different new bug. When using 32-bit "int $0x80" entry for a bogus syscall number, the return value is not correctly set to -ENOSYS. This only happens when neither syscall-audit nor syscall tracing is enabled (i.e., never seen if auditd ever started). Test program: /* gcc -o int80-badsys -m32 -g int80-badsys.c Run on x86-64 kernel. Note to reproduce the bug you need auditd never to have started. */ #include #include int main (void) { long res; asm ("int $0x80" : "=a" (res) : "0" (99999)); printf ("bad syscall returns %ld\n", res); return res != -ENOSYS; } The fix makes the int $0x80 path match the sysenter and syscall paths. Reported-by: Dmitry V. Levin Signed-off-by: Roland McGrath --- arch/x86/ia32/ia32entry.S | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index 256b00b6189..5a0d76dc56a 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -418,9 +418,9 @@ ENTRY(ia32_syscall) orl $TS_COMPAT,TI_status(%r10) testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10) jnz ia32_tracesys -ia32_do_syscall: cmpl $(IA32_NR_syscalls-1),%eax - ja int_ret_from_sys_call /* ia32_tracesys has set RAX(%rsp) */ + ja ia32_badsys +ia32_do_call: IA32_ARG_FIXUP call *ia32_sys_call_table(,%rax,8) # xxx: rip relative ia32_sysret: @@ -435,7 +435,9 @@ ia32_tracesys: call syscall_trace_enter LOAD_ARGS32 ARGOFFSET /* reload args from stack in case ptrace changed it */ RESTORE_REST - jmp ia32_do_syscall + cmpl $(IA32_NR_syscalls-1),%eax + ja int_ret_from_sys_call /* ia32_tracesys has set RAX(%rsp) */ + jmp ia32_do_call END(ia32_syscall) ia32_badsys: -- cgit v1.2.3 From e736ad548db152776de61d7a26805cfae77ce5ce Mon Sep 17 00:00:00 2001 From: "Pallipadi, Venkatesh" Date: Fri, 6 Feb 2009 16:52:05 -0800 Subject: x86: add clflush before monitor for Intel 7400 series For Intel 7400 series CPUs, the recommendation is to use a clflush on the monitored address just before monitor and mwait pair [1]. This clflush makes sure that there are no false wakeups from mwait when the monitored address was recently written to. [1] "MONITOR/MWAIT Recommendations for Intel Xeon Processor 7400 series" section in specification update document of 7400 series http://download.intel.com/design/xeon/specupdt/32033601.pdf Signed-off-by: Venkatesh Pallipadi Signed-off-by: Ingo Molnar --- arch/x86/include/asm/cpufeature.h | 1 + arch/x86/kernel/cpu/intel.c | 3 +++ arch/x86/kernel/process.c | 6 ++++++ 3 files changed, 10 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index ea408dcba51..7301e60dc4a 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -93,6 +93,7 @@ #define X86_FEATURE_XTOPOLOGY (3*32+22) /* cpu topology enum extensions */ #define X86_FEATURE_TSC_RELIABLE (3*32+23) /* TSC is known to be reliable */ #define X86_FEATURE_NONSTOP_TSC (3*32+24) /* TSC does not stop in C states */ +#define X86_FEATURE_CLFLUSH_MONITOR (3*32+25) /* "" clflush reqd with monitor */ /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ #define X86_FEATURE_XMM3 (4*32+ 0) /* "pni" SSE-3 */ diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 430e5c38a54..24ff26a38ad 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -291,6 +291,9 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) ds_init_intel(c); } + if (c->x86 == 6 && c->x86_model == 29 && cpu_has_clflush) + set_cpu_cap(c, X86_FEATURE_CLFLUSH_MONITOR); + #ifdef CONFIG_X86_64 if (c->x86 == 15) c->x86_cache_alignment = c->x86_clflush_size * 2; diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index e68bb9e3086..6d12f7e37f8 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -180,6 +180,9 @@ void mwait_idle_with_hints(unsigned long ax, unsigned long cx) trace_power_start(&it, POWER_CSTATE, (ax>>4)+1); if (!need_resched()) { + if (cpu_has(¤t_cpu_data, X86_FEATURE_CLFLUSH_MONITOR)) + clflush((void *)¤t_thread_info()->flags); + __monitor((void *)¤t_thread_info()->flags, 0, 0); smp_mb(); if (!need_resched()) @@ -194,6 +197,9 @@ static void mwait_idle(void) struct power_trace it; if (!need_resched()) { trace_power_start(&it, POWER_CSTATE, 1); + if (cpu_has(¤t_cpu_data, X86_FEATURE_CLFLUSH_MONITOR)) + clflush((void *)¤t_thread_info()->flags); + __monitor((void *)¤t_thread_info()->flags, 0, 0); smp_mb(); if (!need_resched()) -- cgit v1.2.3 From 3f4a739c6accd651a11fcf3c7a20ec8147c42660 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sun, 8 Feb 2009 16:18:03 -0800 Subject: x86: find nr_irqs_gsi with mp_ioapic_routing Impact: find right nr_irqs_gsi on some systems. One test-system has gap between gsi's: [ 0.000000] ACPI: IOAPIC (id[0x04] address[0xfec00000] gsi_base[0]) [ 0.000000] IOAPIC[0]: apic_id 4, version 0, address 0xfec00000, GSI 0-23 [ 0.000000] ACPI: IOAPIC (id[0x05] address[0xfeafd000] gsi_base[48]) [ 0.000000] IOAPIC[1]: apic_id 5, version 0, address 0xfeafd000, GSI 48-54 [ 0.000000] ACPI: IOAPIC (id[0x06] address[0xfeafc000] gsi_base[56]) [ 0.000000] IOAPIC[2]: apic_id 6, version 0, address 0xfeafc000, GSI 56-62 ... [ 0.000000] nr_irqs_gsi: 38 So nr_irqs_gsi is not right. some irq for MSI will overwrite with io_apic. need to get that with acpi_probe_gsi when acpi io_apic is used Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/include/asm/mpspec.h | 6 ++++++ arch/x86/kernel/acpi/boot.c | 23 +++++++++++++++++++++++ arch/x86/kernel/io_apic.c | 20 +++++++++++++++----- 3 files changed, 44 insertions(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/mpspec.h b/arch/x86/include/asm/mpspec.h index 62d14ce3cd0..bd22f2a3713 100644 --- a/arch/x86/include/asm/mpspec.h +++ b/arch/x86/include/asm/mpspec.h @@ -60,6 +60,7 @@ extern void mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi); extern void mp_config_acpi_legacy_irqs(void); extern int mp_register_gsi(u32 gsi, int edge_level, int active_high_low); +extern int acpi_probe_gsi(void); #ifdef CONFIG_X86_IO_APIC extern int mp_config_acpi_gsi(unsigned char number, unsigned int devfn, u8 pin, u32 gsi, int triggering, int polarity); @@ -71,6 +72,11 @@ mp_config_acpi_gsi(unsigned char number, unsigned int devfn, u8 pin, return 0; } #endif +#else /* !CONFIG_ACPI: */ +static inline int acpi_probe_gsi(void) +{ + return 0; +} #endif /* CONFIG_ACPI */ #define PHYSID_ARRAY_SIZE BITS_TO_LONGS(MAX_APICS) diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index d37593c2f43..7678f10c456 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -973,6 +973,29 @@ void __init mp_register_ioapic(int id, u32 address, u32 gsi_base) nr_ioapics++; } +int __init acpi_probe_gsi(void) +{ + int idx; + int gsi; + int max_gsi = 0; + + if (acpi_disabled) + return 0; + + if (!acpi_ioapic) + return 0; + + max_gsi = 0; + for (idx = 0; idx < nr_ioapics; idx++) { + gsi = mp_ioapic_routing[idx].gsi_end; + + if (gsi > max_gsi) + max_gsi = gsi; + } + + return max_gsi + 1; +} + static void assign_to_mp_irq(struct mp_config_intsrc *m, struct mp_config_intsrc *mp_irq) { diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c index 9b0c480c383..bc7ac4da90d 100644 --- a/arch/x86/kernel/io_apic.c +++ b/arch/x86/kernel/io_apic.c @@ -3841,14 +3841,24 @@ int __init io_apic_get_redir_entries (int ioapic) void __init probe_nr_irqs_gsi(void) { - int idx; int nr = 0; - for (idx = 0; idx < nr_ioapics; idx++) - nr += io_apic_get_redir_entries(idx) + 1; - - if (nr > nr_irqs_gsi) + nr = acpi_probe_gsi(); + if (nr > nr_irqs_gsi) { nr_irqs_gsi = nr; + } else { + /* for acpi=off or acpi is not compiled in */ + int idx; + + nr = 0; + for (idx = 0; idx < nr_ioapics; idx++) + nr += io_apic_get_redir_entries(idx) + 1; + + if (nr > nr_irqs_gsi) + nr_irqs_gsi = nr; + } + + printk(KERN_DEBUG "nr_irqs_gsi: %d\n", nr_irqs_gsi); } /* -------------------------------------------------------------------------- -- cgit v1.2.3 From 55a8ba4b7f76bebd7e8ce3f74c04b140627a1bad Mon Sep 17 00:00:00 2001 From: Alok Kataria Date: Fri, 6 Feb 2009 10:29:35 -0800 Subject: x86, vmi: put a missing paravirt_release_pmd in pgd_dtor Commit 6194ba6ff6ccf8d5c54c857600843c67aa82c407 ("x86: don't special-case pmd allocations as much") made changes to the way we handle pmd allocations, and while doing that it dropped a call to paravirt_release_pd on the pgd page from the pgd_dtor code path. As a result of this missing release, the hypervisor is now unaware of the pgd page being freed, and as a result it ends up tracking this page as a page table page. After this the guest may start using the same page for other purposes, and depending on what use the page is put to, it may result in various performance and/or functional issues ( hangs, reboots). Since this release is only required for VMI, I now release the pgd page from the (vmi)_pgd_free hook. Signed-off-by: Alok N Kataria Acked-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar Cc: --- arch/x86/kernel/vmi_32.c | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c index 1d3302cc2dd..bef58b4982d 100644 --- a/arch/x86/kernel/vmi_32.c +++ b/arch/x86/kernel/vmi_32.c @@ -320,6 +320,16 @@ static void vmi_release_pmd(unsigned long pfn) vmi_ops.release_page(pfn, VMI_PAGE_L2); } +/* + * We use the pgd_free hook for releasing the pgd page: + */ +static void vmi_pgd_free(struct mm_struct *mm, pgd_t *pgd) +{ + unsigned long pfn = __pa(pgd) >> PAGE_SHIFT; + + vmi_ops.release_page(pfn, VMI_PAGE_L2); +} + /* * Helper macros for MMU update flags. We can defer updates until a flush * or page invalidation only if the update is to the current address space @@ -762,6 +772,7 @@ static inline int __init activate_vmi(void) if (vmi_ops.release_page) { pv_mmu_ops.release_pte = vmi_release_pte; pv_mmu_ops.release_pmd = vmi_release_pmd; + pv_mmu_ops.pgd_free = vmi_pgd_free; } /* Set linear is needed in all cases */ -- cgit v1.2.3 From 914c3d630b29b07d04908eab1b246812dadd5bd6 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 9 Feb 2009 22:17:39 +0900 Subject: x86: include correct %gs in a.out core dump Impact: dump the correct %gs into a.out core dump aout_dump_thread() read %gs but didn't include it in core dump. Fix it. Signed-off-by: Tejun Heo Signed-off-by: Ingo Molnar --- arch/x86/include/asm/a.out-core.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/a.out-core.h b/arch/x86/include/asm/a.out-core.h index 37822206083..3c601f8224b 100644 --- a/arch/x86/include/asm/a.out-core.h +++ b/arch/x86/include/asm/a.out-core.h @@ -23,8 +23,6 @@ */ static inline void aout_dump_thread(struct pt_regs *regs, struct user *dump) { - u16 gs; - /* changed the size calculations - should hopefully work better. lbt */ dump->magic = CMAGIC; dump->start_code = 0; @@ -57,7 +55,7 @@ static inline void aout_dump_thread(struct pt_regs *regs, struct user *dump) dump->regs.ds = (u16)regs->ds; dump->regs.es = (u16)regs->es; dump->regs.fs = (u16)regs->fs; - savesegment(gs, gs); + savesegment(gs, dump->regs.gs); dump->regs.orig_ax = regs->orig_ax; dump->regs.ip = regs->ip; dump->regs.cs = (u16)regs->cs; -- cgit v1.2.3 From ae6af41f5a4841f06eb92bc86ad020ad44ae2a30 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 9 Feb 2009 22:17:39 +0900 Subject: x86: math_emu info cleanup Impact: cleanup * Come on, struct info? s/struct info/struct math_emu_info/ * Use struct pt_regs and kernel_vm86_regs instead of defining its own register frame structure. Signed-off-by: Tejun Heo Signed-off-by: Ingo Molnar --- arch/x86/include/asm/math_emu.h | 29 +++++------------- arch/x86/include/asm/processor.h | 2 +- arch/x86/math-emu/fpu_entry.c | 2 +- arch/x86/math-emu/fpu_proto.h | 2 +- arch/x86/math-emu/fpu_system.h | 14 ++++----- arch/x86/math-emu/get_address.c | 63 +++++++++++++++++++--------------------- 6 files changed, 48 insertions(+), 64 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/math_emu.h b/arch/x86/include/asm/math_emu.h index 5a65b107ad5..302492c7795 100644 --- a/arch/x86/include/asm/math_emu.h +++ b/arch/x86/include/asm/math_emu.h @@ -1,31 +1,18 @@ #ifndef _ASM_X86_MATH_EMU_H #define _ASM_X86_MATH_EMU_H +#include +#include + /* This structure matches the layout of the data saved to the stack following a device-not-present interrupt, part of it saved automatically by the 80386/80486. */ -struct info { +struct math_emu_info { long ___orig_eip; - long ___ebx; - long ___ecx; - long ___edx; - long ___esi; - long ___edi; - long ___ebp; - long ___eax; - long ___ds; - long ___es; - long ___fs; - long ___orig_eax; - long ___eip; - long ___cs; - long ___eflags; - long ___esp; - long ___ss; - long ___vm86_es; /* This and the following only in vm86 mode */ - long ___vm86_ds; - long ___vm86_fs; - long ___vm86_gs; + union { + struct pt_regs regs; + struct kernel_vm86_regs vm86; + }; }; #endif /* _ASM_X86_MATH_EMU_H */ diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index 091cd8855f2..3bfd5235a9e 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -353,7 +353,7 @@ struct i387_soft_struct { u8 no_update; u8 rm; u8 alimit; - struct info *info; + struct math_emu_info *info; u32 entry_eip; }; diff --git a/arch/x86/math-emu/fpu_entry.c b/arch/x86/math-emu/fpu_entry.c index c7b06feb139..c268abe7225 100644 --- a/arch/x86/math-emu/fpu_entry.c +++ b/arch/x86/math-emu/fpu_entry.c @@ -659,7 +659,7 @@ static int valid_prefix(u_char *Byte, u_char __user **fpu_eip, } } -void math_abort(struct info *info, unsigned int signal) +void math_abort(struct math_emu_info *info, unsigned int signal) { FPU_EIP = FPU_ORIG_EIP; current->thread.trap_no = 16; diff --git a/arch/x86/math-emu/fpu_proto.h b/arch/x86/math-emu/fpu_proto.h index aa49b6a0d85..51bfbb61c5b 100644 --- a/arch/x86/math-emu/fpu_proto.h +++ b/arch/x86/math-emu/fpu_proto.h @@ -52,7 +52,7 @@ extern void fst_i_(void); extern void fstp_i(void); /* fpu_entry.c */ asmlinkage extern void math_emulate(long arg); -extern void math_abort(struct info *info, unsigned int signal); +extern void math_abort(struct math_emu_info *info, unsigned int signal); /* fpu_etc.c */ extern void FPU_etc(void); /* fpu_tags.c */ diff --git a/arch/x86/math-emu/fpu_system.h b/arch/x86/math-emu/fpu_system.h index 13488fa153e..6729c6a3134 100644 --- a/arch/x86/math-emu/fpu_system.h +++ b/arch/x86/math-emu/fpu_system.h @@ -18,7 +18,7 @@ /* This sets the pointer FPU_info to point to the argument part of the stack frame of math_emulate() */ -#define SETUP_DATA_AREA(arg) FPU_info = (struct info *) &arg +#define SETUP_DATA_AREA(arg) FPU_info = (struct math_emu_info *) &arg /* s is always from a cpu register, and the cpu does bounds checking * during register load --> no further bounds checks needed */ @@ -38,12 +38,12 @@ #define I387 (current->thread.xstate) #define FPU_info (I387->soft.info) -#define FPU_CS (*(unsigned short *) &(FPU_info->___cs)) -#define FPU_SS (*(unsigned short *) &(FPU_info->___ss)) -#define FPU_DS (*(unsigned short *) &(FPU_info->___ds)) -#define FPU_EAX (FPU_info->___eax) -#define FPU_EFLAGS (FPU_info->___eflags) -#define FPU_EIP (FPU_info->___eip) +#define FPU_CS (*(unsigned short *) &(FPU_info->regs.cs)) +#define FPU_SS (*(unsigned short *) &(FPU_info->regs.ss)) +#define FPU_DS (*(unsigned short *) &(FPU_info->regs.ds)) +#define FPU_EAX (FPU_info->regs.ax) +#define FPU_EFLAGS (FPU_info->regs.flags) +#define FPU_EIP (FPU_info->regs.ip) #define FPU_ORIG_EIP (FPU_info->___orig_eip) #define FPU_lookahead (I387->soft.lookahead) diff --git a/arch/x86/math-emu/get_address.c b/arch/x86/math-emu/get_address.c index d701e2b39e4..62daa7fcc44 100644 --- a/arch/x86/math-emu/get_address.c +++ b/arch/x86/math-emu/get_address.c @@ -29,42 +29,39 @@ #define FPU_WRITE_BIT 0x10 static int reg_offset[] = { - offsetof(struct info, ___eax), - offsetof(struct info, ___ecx), - offsetof(struct info, ___edx), - offsetof(struct info, ___ebx), - offsetof(struct info, ___esp), - offsetof(struct info, ___ebp), - offsetof(struct info, ___esi), - offsetof(struct info, ___edi) + offsetof(struct math_emu_info, regs.ax), + offsetof(struct math_emu_info, regs.cx), + offsetof(struct math_emu_info, regs.dx), + offsetof(struct math_emu_info, regs.bx), + offsetof(struct math_emu_info, regs.sp), + offsetof(struct math_emu_info, regs.bp), + offsetof(struct math_emu_info, regs.si), + offsetof(struct math_emu_info, regs.di) }; #define REG_(x) (*(long *)(reg_offset[(x)]+(u_char *) FPU_info)) static int reg_offset_vm86[] = { - offsetof(struct info, ___cs), - offsetof(struct info, ___vm86_ds), - offsetof(struct info, ___vm86_es), - offsetof(struct info, ___vm86_fs), - offsetof(struct info, ___vm86_gs), - offsetof(struct info, ___ss), - offsetof(struct info, ___vm86_ds) + offsetof(struct math_emu_info, regs.cs), + offsetof(struct math_emu_info, vm86.ds), + offsetof(struct math_emu_info, vm86.es), + offsetof(struct math_emu_info, vm86.fs), + offsetof(struct math_emu_info, vm86.gs), + offsetof(struct math_emu_info, regs.ss), + offsetof(struct math_emu_info, vm86.ds) }; #define VM86_REG_(x) (*(unsigned short *) \ (reg_offset_vm86[((unsigned)x)]+(u_char *) FPU_info)) -/* This dummy, gs is not saved on the stack. */ -#define ___GS ___ds - static int reg_offset_pm[] = { - offsetof(struct info, ___cs), - offsetof(struct info, ___ds), - offsetof(struct info, ___es), - offsetof(struct info, ___fs), - offsetof(struct info, ___GS), - offsetof(struct info, ___ss), - offsetof(struct info, ___ds) + offsetof(struct math_emu_info, regs.cs), + offsetof(struct math_emu_info, regs.ds), + offsetof(struct math_emu_info, regs.es), + offsetof(struct math_emu_info, regs.fs), + offsetof(struct math_emu_info, regs.ds), /* dummy, not saved on stack */ + offsetof(struct math_emu_info, regs.ss), + offsetof(struct math_emu_info, regs.ds) }; #define PM_REG_(x) (*(unsigned short *) \ @@ -349,34 +346,34 @@ void __user *FPU_get_address_16(u_char FPU_modrm, unsigned long *fpu_eip, } switch (rm) { case 0: - address += FPU_info->___ebx + FPU_info->___esi; + address += FPU_info->regs.bx + FPU_info->regs.si; break; case 1: - address += FPU_info->___ebx + FPU_info->___edi; + address += FPU_info->regs.bx + FPU_info->regs.di; break; case 2: - address += FPU_info->___ebp + FPU_info->___esi; + address += FPU_info->regs.bp + FPU_info->regs.si; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 3: - address += FPU_info->___ebp + FPU_info->___edi; + address += FPU_info->regs.bp + FPU_info->regs.di; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 4: - address += FPU_info->___esi; + address += FPU_info->regs.si; break; case 5: - address += FPU_info->___edi; + address += FPU_info->regs.di; break; case 6: - address += FPU_info->___ebp; + address += FPU_info->regs.bp; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 7: - address += FPU_info->___ebx; + address += FPU_info->regs.bx; break; } -- cgit v1.2.3 From a5ef7ca0e2636bad0ccd07b996d775348ae2b65e Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Sun, 8 Feb 2009 17:39:58 -0500 Subject: x86: spinlocks: define dummy __raw_spin_is_contended Architectures other than mips and x86 are not using ticket spinlocks. Therefore, the contention on the lock is meaningless, since there is nobody known to be waiting on it (arguably /fairly/ unfair locks). Dummy it out to return 0 on other architectures. Signed-off-by: Kyle McMartin Acked-by: Ralf Baechle Acked-by: Ingo Molnar Signed-off-by: Linus Torvalds --- arch/x86/include/asm/paravirt.h | 1 + arch/x86/include/asm/spinlock.h | 1 + 2 files changed, 2 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index ba3e2ff6aed..c09a1412758 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -1402,6 +1402,7 @@ static inline int __raw_spin_is_contended(struct raw_spinlock *lock) { return PVOP_CALL1(int, pv_lock_ops.spin_is_contended, lock); } +#define __raw_spin_is_contended __raw_spin_is_contended static __always_inline void __raw_spin_lock(struct raw_spinlock *lock) { diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h index d17c91981da..8247e94ac6b 100644 --- a/arch/x86/include/asm/spinlock.h +++ b/arch/x86/include/asm/spinlock.h @@ -245,6 +245,7 @@ static inline int __raw_spin_is_contended(raw_spinlock_t *lock) { return __ticket_spin_is_contended(lock); } +#define __raw_spin_is_contended __raw_spin_is_contended static __always_inline void __raw_spin_lock(raw_spinlock_t *lock) { -- cgit v1.2.3 From d315760ffa261c15ff92699ac6f514112543d7ca Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 9 Feb 2009 22:17:39 +0900 Subject: x86: fix math_emu register frame access do_device_not_available() is the handler for #NM and it declares that it takes a unsigned long and calls math_emu(), which takes a long argument and surprisingly expects the stack frame starting at the zero argument would match struct math_emu_info, which isn't true regardless of configuration in the current code. This patch makes do_device_not_available() take struct pt_regs like other exception handlers and initialize struct math_emu_info with pointer to it and pass pointer to the math_emu_info to math_emulate() like normal C functions do. This way, unless gcc makes a copy of struct pt_regs in do_device_not_available(), the register frame is correctly accessed regardless of kernel configuration or compiler used. This doesn't fix all math_emu problems but it at least gets it somewhat working. Signed-off-by: Tejun Heo Signed-off-by: Ingo Molnar --- arch/x86/include/asm/math_emu.h | 4 +-- arch/x86/include/asm/traps.h | 4 +-- arch/x86/kernel/traps.c | 15 ++++++---- arch/x86/math-emu/fpu_entry.c | 4 +-- arch/x86/math-emu/fpu_proto.h | 2 +- arch/x86/math-emu/fpu_system.h | 16 ++++------ arch/x86/math-emu/get_address.c | 66 ++++++++++++++++++++--------------------- 7 files changed, 55 insertions(+), 56 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/math_emu.h b/arch/x86/include/asm/math_emu.h index 302492c7795..031f6266f42 100644 --- a/arch/x86/include/asm/math_emu.h +++ b/arch/x86/include/asm/math_emu.h @@ -11,8 +11,8 @@ struct math_emu_info { long ___orig_eip; union { - struct pt_regs regs; - struct kernel_vm86_regs vm86; + struct pt_regs *regs; + struct kernel_vm86_regs *vm86; }; }; #endif /* _ASM_X86_MATH_EMU_H */ diff --git a/arch/x86/include/asm/traps.h b/arch/x86/include/asm/traps.h index 2ee0a3bceed..cf3bb053da0 100644 --- a/arch/x86/include/asm/traps.h +++ b/arch/x86/include/asm/traps.h @@ -41,7 +41,7 @@ dotraplinkage void do_int3(struct pt_regs *, long); dotraplinkage void do_overflow(struct pt_regs *, long); dotraplinkage void do_bounds(struct pt_regs *, long); dotraplinkage void do_invalid_op(struct pt_regs *, long); -dotraplinkage void do_device_not_available(struct pt_regs *, long); +dotraplinkage void do_device_not_available(struct pt_regs); dotraplinkage void do_coprocessor_segment_overrun(struct pt_regs *, long); dotraplinkage void do_invalid_TSS(struct pt_regs *, long); dotraplinkage void do_segment_not_present(struct pt_regs *, long); @@ -77,7 +77,7 @@ extern int panic_on_unrecovered_nmi; extern int kstack_depth_to_print; void math_error(void __user *); -asmlinkage void math_emulate(long); +void math_emulate(struct math_emu_info *); #ifdef CONFIG_X86_32 unsigned long patch_espfix_desc(unsigned long, unsigned long); #else diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 98c2d055284..7932338d7cb 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -896,7 +896,7 @@ asmlinkage void math_state_restore(void) EXPORT_SYMBOL_GPL(math_state_restore); #ifndef CONFIG_MATH_EMULATION -asmlinkage void math_emulate(long arg) +void math_emulate(struct math_emu_info *info) { printk(KERN_EMERG "math-emulation not enabled and no coprocessor found.\n"); @@ -906,16 +906,19 @@ asmlinkage void math_emulate(long arg) } #endif /* CONFIG_MATH_EMULATION */ -dotraplinkage void __kprobes -do_device_not_available(struct pt_regs *regs, long error) +dotraplinkage void __kprobes do_device_not_available(struct pt_regs regs) { #ifdef CONFIG_X86_32 if (read_cr0() & X86_CR0_EM) { - conditional_sti(regs); - math_emulate(0); + struct math_emu_info info = { }; + + conditional_sti(®s); + + info.regs = ®s; + math_emulate(&info); } else { math_state_restore(); /* interrupts still off */ - conditional_sti(regs); + conditional_sti(®s); } #else math_state_restore(); diff --git a/arch/x86/math-emu/fpu_entry.c b/arch/x86/math-emu/fpu_entry.c index c268abe7225..5d87f586f8d 100644 --- a/arch/x86/math-emu/fpu_entry.c +++ b/arch/x86/math-emu/fpu_entry.c @@ -131,7 +131,7 @@ u_char emulating = 0; static int valid_prefix(u_char *Byte, u_char __user ** fpu_eip, overrides * override); -asmlinkage void math_emulate(long arg) +void math_emulate(struct math_emu_info *info) { u_char FPU_modrm, byte1; unsigned short code; @@ -161,7 +161,7 @@ asmlinkage void math_emulate(long arg) RE_ENTRANT_CHECK_ON; #endif /* RE_ENTRANT_CHECKING */ - SETUP_DATA_AREA(arg); + FPU_info = info; FPU_ORIG_EIP = FPU_EIP; diff --git a/arch/x86/math-emu/fpu_proto.h b/arch/x86/math-emu/fpu_proto.h index 51bfbb61c5b..9779df436b7 100644 --- a/arch/x86/math-emu/fpu_proto.h +++ b/arch/x86/math-emu/fpu_proto.h @@ -51,7 +51,7 @@ extern void ffreep(void); extern void fst_i_(void); extern void fstp_i(void); /* fpu_entry.c */ -asmlinkage extern void math_emulate(long arg); +extern void math_emulate(struct math_emu_info *info); extern void math_abort(struct math_emu_info *info, unsigned int signal); /* fpu_etc.c */ extern void FPU_etc(void); diff --git a/arch/x86/math-emu/fpu_system.h b/arch/x86/math-emu/fpu_system.h index 6729c6a3134..50fa0ec2c8a 100644 --- a/arch/x86/math-emu/fpu_system.h +++ b/arch/x86/math-emu/fpu_system.h @@ -16,10 +16,6 @@ #include #include -/* This sets the pointer FPU_info to point to the argument part - of the stack frame of math_emulate() */ -#define SETUP_DATA_AREA(arg) FPU_info = (struct math_emu_info *) &arg - /* s is always from a cpu register, and the cpu does bounds checking * during register load --> no further bounds checks needed */ #define LDT_DESCRIPTOR(s) (((struct desc_struct *)current->mm->context.ldt)[(s) >> 3]) @@ -38,12 +34,12 @@ #define I387 (current->thread.xstate) #define FPU_info (I387->soft.info) -#define FPU_CS (*(unsigned short *) &(FPU_info->regs.cs)) -#define FPU_SS (*(unsigned short *) &(FPU_info->regs.ss)) -#define FPU_DS (*(unsigned short *) &(FPU_info->regs.ds)) -#define FPU_EAX (FPU_info->regs.ax) -#define FPU_EFLAGS (FPU_info->regs.flags) -#define FPU_EIP (FPU_info->regs.ip) +#define FPU_CS (*(unsigned short *) &(FPU_info->regs->cs)) +#define FPU_SS (*(unsigned short *) &(FPU_info->regs->ss)) +#define FPU_DS (*(unsigned short *) &(FPU_info->regs->ds)) +#define FPU_EAX (FPU_info->regs->ax) +#define FPU_EFLAGS (FPU_info->regs->flags) +#define FPU_EIP (FPU_info->regs->ip) #define FPU_ORIG_EIP (FPU_info->___orig_eip) #define FPU_lookahead (I387->soft.lookahead) diff --git a/arch/x86/math-emu/get_address.c b/arch/x86/math-emu/get_address.c index 62daa7fcc44..420b3b6e391 100644 --- a/arch/x86/math-emu/get_address.c +++ b/arch/x86/math-emu/get_address.c @@ -29,43 +29,43 @@ #define FPU_WRITE_BIT 0x10 static int reg_offset[] = { - offsetof(struct math_emu_info, regs.ax), - offsetof(struct math_emu_info, regs.cx), - offsetof(struct math_emu_info, regs.dx), - offsetof(struct math_emu_info, regs.bx), - offsetof(struct math_emu_info, regs.sp), - offsetof(struct math_emu_info, regs.bp), - offsetof(struct math_emu_info, regs.si), - offsetof(struct math_emu_info, regs.di) + offsetof(struct pt_regs, ax), + offsetof(struct pt_regs, cx), + offsetof(struct pt_regs, dx), + offsetof(struct pt_regs, bx), + offsetof(struct pt_regs, sp), + offsetof(struct pt_regs, bp), + offsetof(struct pt_regs, si), + offsetof(struct pt_regs, di) }; -#define REG_(x) (*(long *)(reg_offset[(x)]+(u_char *) FPU_info)) +#define REG_(x) (*(long *)(reg_offset[(x)] + (u_char *)FPU_info->regs)) static int reg_offset_vm86[] = { - offsetof(struct math_emu_info, regs.cs), - offsetof(struct math_emu_info, vm86.ds), - offsetof(struct math_emu_info, vm86.es), - offsetof(struct math_emu_info, vm86.fs), - offsetof(struct math_emu_info, vm86.gs), - offsetof(struct math_emu_info, regs.ss), - offsetof(struct math_emu_info, vm86.ds) + offsetof(struct pt_regs, cs), + offsetof(struct kernel_vm86_regs, ds), + offsetof(struct kernel_vm86_regs, es), + offsetof(struct kernel_vm86_regs, fs), + offsetof(struct kernel_vm86_regs, gs), + offsetof(struct pt_regs, ss), + offsetof(struct kernel_vm86_regs, ds) }; #define VM86_REG_(x) (*(unsigned short *) \ - (reg_offset_vm86[((unsigned)x)]+(u_char *) FPU_info)) + (reg_offset_vm86[((unsigned)x)] + (u_char *)FPU_info->regs)) static int reg_offset_pm[] = { - offsetof(struct math_emu_info, regs.cs), - offsetof(struct math_emu_info, regs.ds), - offsetof(struct math_emu_info, regs.es), - offsetof(struct math_emu_info, regs.fs), - offsetof(struct math_emu_info, regs.ds), /* dummy, not saved on stack */ - offsetof(struct math_emu_info, regs.ss), - offsetof(struct math_emu_info, regs.ds) + offsetof(struct pt_regs, cs), + offsetof(struct pt_regs, ds), + offsetof(struct pt_regs, es), + offsetof(struct pt_regs, fs), + offsetof(struct pt_regs, ds), /* dummy, not saved on stack */ + offsetof(struct pt_regs, ss), + offsetof(struct pt_regs, ds) }; #define PM_REG_(x) (*(unsigned short *) \ - (reg_offset_pm[((unsigned)x)]+(u_char *) FPU_info)) + (reg_offset_pm[((unsigned)x)] + (u_char *)FPU_info->regs)) /* Decode the SIB byte. This function assumes mod != 0 */ static int sib(int mod, unsigned long *fpu_eip) @@ -346,34 +346,34 @@ void __user *FPU_get_address_16(u_char FPU_modrm, unsigned long *fpu_eip, } switch (rm) { case 0: - address += FPU_info->regs.bx + FPU_info->regs.si; + address += FPU_info->regs->bx + FPU_info->regs->si; break; case 1: - address += FPU_info->regs.bx + FPU_info->regs.di; + address += FPU_info->regs->bx + FPU_info->regs->di; break; case 2: - address += FPU_info->regs.bp + FPU_info->regs.si; + address += FPU_info->regs->bp + FPU_info->regs->si; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 3: - address += FPU_info->regs.bp + FPU_info->regs.di; + address += FPU_info->regs->bp + FPU_info->regs->di; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 4: - address += FPU_info->regs.si; + address += FPU_info->regs->si; break; case 5: - address += FPU_info->regs.di; + address += FPU_info->regs->di; break; case 6: - address += FPU_info->regs.bp; + address += FPU_info->regs->bp; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 7: - address += FPU_info->regs.bx; + address += FPU_info->regs->bx; break; } -- cgit v1.2.3 From b52af40923fc91a12e3c7152d833e0c0c6a508f6 Mon Sep 17 00:00:00 2001 From: Clemens Ladisch Date: Tue, 10 Feb 2009 09:21:07 +0100 Subject: i8327: fix outb() parameter order In i8237A_resume(), when resetting the DMA controller, the parameters to dma_outb() were mixed up. Signed-off-by: Clemens Ladisch [ cleaned up the file a tiny bit. ] Signed-off-by: Ingo Molnar --- arch/x86/kernel/i8237.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/i8237.c b/arch/x86/kernel/i8237.c index dbd6c1d1b63..b42ca694dc6 100644 --- a/arch/x86/kernel/i8237.c +++ b/arch/x86/kernel/i8237.c @@ -28,10 +28,10 @@ static int i8237A_resume(struct sys_device *dev) flags = claim_dma_lock(); - dma_outb(DMA1_RESET_REG, 0); - dma_outb(DMA2_RESET_REG, 0); + dma_outb(0, DMA1_RESET_REG); + dma_outb(0, DMA2_RESET_REG); - for (i = 0;i < 8;i++) { + for (i = 0; i < 8; i++) { set_dma_addr(i, 0x000000); /* DMA count is a bit weird so this is not 0 */ set_dma_count(i, 1); @@ -51,14 +51,14 @@ static int i8237A_suspend(struct sys_device *dev, pm_message_t state) } static struct sysdev_class i8237_sysdev_class = { - .name = "i8237", - .suspend = i8237A_suspend, - .resume = i8237A_resume, + .name = "i8237", + .suspend = i8237A_suspend, + .resume = i8237A_resume, }; static struct sys_device device_i8237A = { - .id = 0, - .cls = &i8237_sysdev_class, + .id = 0, + .cls = &i8237_sysdev_class, }; static int __init i8237A_init_sysfs(void) @@ -68,5 +68,4 @@ static int __init i8237A_init_sysfs(void) error = sysdev_register(&device_i8237A); return error; } - device_initcall(i8237A_init_sysfs); -- cgit v1.2.3 From e3944bfac961cd7fc82f3b3143c55dc375748569 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Tue, 10 Feb 2009 13:07:13 -0500 Subject: tracing, x86: fix fixup section to return to original code Impact: fix to prevent a kernel crash on fault If for some reason the pointer to the parent function on the stack takes a fault, the fix up code will not return back to the original faulting code. This can lead to unpredictable results and perhaps even a kernel panic. A fault should not happen, but if it does, we should simply disable the tracer, warn, and continue running the kernel. It should not lead to a kernel crash. Signed-off-by: Steven Rostedt --- arch/x86/kernel/ftrace.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index 1b43086b097..9d549e4fe88 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -491,13 +491,15 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr) "1: " _ASM_MOV " (%[parent_old]), %[old]\n" "2: " _ASM_MOV " %[return_hooker], (%[parent_replaced])\n" " movl $0, %[faulted]\n" + "3:\n" ".section .fixup, \"ax\"\n" - "3: movl $1, %[faulted]\n" + "4: movl $1, %[faulted]\n" + " jmp 3b\n" ".previous\n" - _ASM_EXTABLE(1b, 3b) - _ASM_EXTABLE(2b, 3b) + _ASM_EXTABLE(1b, 4b) + _ASM_EXTABLE(2b, 4b) : [parent_replaced] "=r" (parent), [old] "=r" (old), [faulted] "=r" (faulted) -- cgit v1.2.3 From f47a454db9129d2e61b224a40f4365cdd4f83042 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Tue, 10 Feb 2009 11:53:23 -0500 Subject: tracing, x86: fix constraint for parent variable The constraint used for retrieving and restoring the parent function pointer is incorrect. The parent variable is a pointer, and the address of the pointer is modified by the asm statement and not the pointer itself. It is incorrect to pass it in as an output constraint since the asm will never update the pointer. Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- arch/x86/kernel/ftrace.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index 9d549e4fe88..231bdd3c5b1 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -488,8 +488,8 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr) * ignore such a protection. */ asm volatile( - "1: " _ASM_MOV " (%[parent_old]), %[old]\n" - "2: " _ASM_MOV " %[return_hooker], (%[parent_replaced])\n" + "1: " _ASM_MOV " (%[parent]), %[old]\n" + "2: " _ASM_MOV " %[return_hooker], (%[parent])\n" " movl $0, %[faulted]\n" "3:\n" @@ -501,9 +501,8 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr) _ASM_EXTABLE(1b, 4b) _ASM_EXTABLE(2b, 4b) - : [parent_replaced] "=r" (parent), [old] "=r" (old), - [faulted] "=r" (faulted) - : [parent_old] "0" (parent), [return_hooker] "r" (return_hooker) + : [old] "=r" (old), [faulted] "=r" (faulted) + : [parent] "r" (parent), [return_hooker] "r" (return_hooker) : "memory" ); -- cgit v1.2.3 From 9f339e7028e2855717af3193c938f9960ad13b38 Mon Sep 17 00:00:00 2001 From: Markus Metzger Date: Wed, 11 Feb 2009 15:10:27 +0100 Subject: x86, ptrace, mm: fix double-free on race Ptrace_detach() races with __ptrace_unlink() if the traced task is reaped while detaching. This might cause a double-free of the BTS buffer. Change the ptrace_detach() path to only do the memory accounting in ptrace_bts_detach() and leave the buffer free to ptrace_bts_untrace() which will be called from __ptrace_unlink(). The fix follows a proposal from Oleg Nesterov. Reported-by: Oleg Nesterov Signed-off-by: Markus Metzger Signed-off-by: Ingo Molnar --- arch/x86/kernel/ptrace.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 0a5df5f82fb..5a4c23d8989 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -810,12 +810,16 @@ static void ptrace_bts_untrace(struct task_struct *child) static void ptrace_bts_detach(struct task_struct *child) { - if (unlikely(child->bts)) { - ds_release_bts(child->bts); - child->bts = NULL; - - ptrace_bts_free_buffer(child); - } + /* + * Ptrace_detach() races with ptrace_untrace() in case + * the child dies and is reaped by another thread. + * + * We only do the memory accounting at this point and + * leave the buffer deallocation and the bts tracer + * release to ptrace_bts_untrace() which will be called + * later on with tasklist_lock held. + */ + release_locked_buffer(child->bts_buffer, child->bts_size); } #else static inline void ptrace_bts_fork(struct task_struct *tsk) {} -- cgit v1.2.3 From 4f06b0436b2ddbd3b67b10e77098a6862787b3eb Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 11 Feb 2009 09:32:19 -0800 Subject: x86/cpa: make sure cpa is safe to call in lazy mmu mode Impact: fix race leading to crash under KVM and Xen The CPA code may be called while we're in lazy mmu update mode - for example, when using DEBUG_PAGE_ALLOC and doing a slab allocation in an interrupt handler which interrupted a lazy mmu update. In this case, the in-memory pagetable state may be out of date due to pending queued updates. We need to flush any pending updates before inspecting the page table. Similarly, we must explicitly flush any modifications CPA may have made (which comes down to flushing queued operations when flushing the TLB). Signed-off-by: Jeremy Fitzhardinge Acked-by: Marcelo Tosatti Cc: Stable Kernel Signed-off-by: Ingo Molnar --- arch/x86/mm/pageattr.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 84ba74820ad..f664bc1c409 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -576,6 +576,13 @@ static int __change_page_attr(struct cpa_data *cpa, int primary) else address = *cpa->vaddr; + /* + * If we're called with lazy mmu updates enabled, the + * in-memory pte state may be stale. Flush pending updates to + * bring them up to date. + */ + arch_flush_lazy_mmu_mode(); + repeat: kpte = lookup_address(address, &level); if (!kpte) @@ -854,6 +861,13 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, } else cpa_flush_all(cache); + /* + * If we've been called with lazy mmu updates enabled, then + * make sure that everything gets flushed out before we + * return. + */ + arch_flush_lazy_mmu_mode(); + out: return ret; } -- cgit v1.2.3 From be03d9e8022030c16abf534e33e185bfc3d40eef Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Wed, 11 Feb 2009 11:20:23 -0800 Subject: x86, pat: fix warn_on_once() while mapping 0-1MB range with /dev/mem Jeff Mahoney reported: > With Suse's hwinfo tool, on -tip: > WARNING: at arch/x86/mm/pat.c:637 reserve_pfn_range+0x5b/0x26d() reserve_pfn_range() is not tracking the memory range below 1MB as non-RAM and as such is inconsistent with similar checks in reserve_memtype() and free_memtype() Rename the pagerange_is_ram() to pat_pagerange_is_ram() and add the "track legacy 1MB region as non RAM" condition. And also, fix reserve_pfn_range() to return -EINVAL, when the pfn range is RAM. This is to be consistent with this API design. Reported-and-tested-by: Jeff Mahoney Signed-off-by: Suresh Siddha Signed-off-by: Venkatesh Pallipadi Signed-off-by: Ingo Molnar --- arch/x86/include/asm/page.h | 1 - arch/x86/mm/ioremap.c | 19 ----------- arch/x86/mm/pat.c | 83 ++++++++++++++++++++++++--------------------- 3 files changed, 45 insertions(+), 58 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/page.h b/arch/x86/include/asm/page.h index e9873a2e869..776579119a0 100644 --- a/arch/x86/include/asm/page.h +++ b/arch/x86/include/asm/page.h @@ -57,7 +57,6 @@ typedef struct { pgdval_t pgd; } pgd_t; typedef struct { pgprotval_t pgprot; } pgprot_t; extern int page_is_ram(unsigned long pagenr); -extern int pagerange_is_ram(unsigned long start, unsigned long end); extern int devmem_is_allowed(unsigned long pagenr); extern void map_devmem(unsigned long pfn, unsigned long size, pgprot_t vma_prot); diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index af750ab973b..f45d5e29a72 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -134,25 +134,6 @@ int page_is_ram(unsigned long pagenr) return 0; } -int pagerange_is_ram(unsigned long start, unsigned long end) -{ - int ram_page = 0, not_rampage = 0; - unsigned long page_nr; - - for (page_nr = (start >> PAGE_SHIFT); page_nr < (end >> PAGE_SHIFT); - ++page_nr) { - if (page_is_ram(page_nr)) - ram_page = 1; - else - not_rampage = 1; - - if (ram_page == not_rampage) - return -1; - } - - return ram_page; -} - /* * Fix up the linear direct mapping of the kernel to avoid cache attribute * conflicts. diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index 7b61036427d..aebbf67a79d 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -211,6 +211,33 @@ chk_conflict(struct memtype *new, struct memtype *entry, unsigned long *type) static struct memtype *cached_entry; static u64 cached_start; +static int pat_pagerange_is_ram(unsigned long start, unsigned long end) +{ + int ram_page = 0, not_rampage = 0; + unsigned long page_nr; + + for (page_nr = (start >> PAGE_SHIFT); page_nr < (end >> PAGE_SHIFT); + ++page_nr) { + /* + * For legacy reasons, physical address range in the legacy ISA + * region is tracked as non-RAM. This will allow users of + * /dev/mem to map portions of legacy ISA region, even when + * some of those portions are listed(or not even listed) with + * different e820 types(RAM/reserved/..) + */ + if (page_nr >= (ISA_END_ADDRESS >> PAGE_SHIFT) && + page_is_ram(page_nr)) + ram_page = 1; + else + not_rampage = 1; + + if (ram_page == not_rampage) + return -1; + } + + return ram_page; +} + /* * For RAM pages, mark the pages as non WB memory type using * PageNonWB (PG_arch_1). We allow only one set_memory_uc() or @@ -336,20 +363,12 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, if (new_type) *new_type = actual_type; - /* - * For legacy reasons, some parts of the physical address range in the - * legacy 1MB region is treated as non-RAM (even when listed as RAM in - * the e820 tables). So we will track the memory attributes of this - * legacy 1MB region using the linear memtype_list always. - */ - if (end >= ISA_END_ADDRESS) { - is_range_ram = pagerange_is_ram(start, end); - if (is_range_ram == 1) - return reserve_ram_pages_type(start, end, req_type, - new_type); - else if (is_range_ram < 0) - return -EINVAL; - } + is_range_ram = pat_pagerange_is_ram(start, end); + if (is_range_ram == 1) + return reserve_ram_pages_type(start, end, req_type, + new_type); + else if (is_range_ram < 0) + return -EINVAL; new = kmalloc(sizeof(struct memtype), GFP_KERNEL); if (!new) @@ -446,19 +465,11 @@ int free_memtype(u64 start, u64 end) if (is_ISA_range(start, end - 1)) return 0; - /* - * For legacy reasons, some parts of the physical address range in the - * legacy 1MB region is treated as non-RAM (even when listed as RAM in - * the e820 tables). So we will track the memory attributes of this - * legacy 1MB region using the linear memtype_list always. - */ - if (end >= ISA_END_ADDRESS) { - is_range_ram = pagerange_is_ram(start, end); - if (is_range_ram == 1) - return free_ram_pages_type(start, end); - else if (is_range_ram < 0) - return -EINVAL; - } + is_range_ram = pat_pagerange_is_ram(start, end); + if (is_range_ram == 1) + return free_ram_pages_type(start, end); + else if (is_range_ram < 0) + return -EINVAL; spin_lock(&memtype_lock); list_for_each_entry(entry, &memtype_list, nd) { @@ -626,17 +637,13 @@ static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t *vma_prot, unsigned long flags; unsigned long want_flags = (pgprot_val(*vma_prot) & _PAGE_CACHE_MASK); - is_ram = pagerange_is_ram(paddr, paddr + size); + is_ram = pat_pagerange_is_ram(paddr, paddr + size); - if (is_ram != 0) { - /* - * For mapping RAM pages, drivers need to call - * set_memory_[uc|wc|wb] directly, for reserve and free, before - * setting up the PTE. - */ - WARN_ON_ONCE(1); - return 0; - } + /* + * reserve_pfn_range() doesn't support RAM pages. + */ + if (is_ram != 0) + return -EINVAL; ret = reserve_memtype(paddr, paddr + size, want_flags, &flags); if (ret) @@ -693,7 +700,7 @@ static void free_pfn_range(u64 paddr, unsigned long size) { int is_ram; - is_ram = pagerange_is_ram(paddr, paddr + size); + is_ram = pat_pagerange_is_ram(paddr, paddr + size); if (is_ram == 0) free_memtype(paddr, paddr + size); } -- cgit v1.2.3 From d85cf93da66977dbc645352be1b2084a659d8a0b Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 12 Feb 2009 10:02:56 -0800 Subject: x86/paravirt: make arch_flush_lazy_mmu/cpu disable preemption Impact: avoid access to percpu vars in preempible context They are intended to be used whenever there's the possibility that there's some stale state which is going to be overwritten with a queued update, or to force a state change when we may be in lazy mode. Either way, we could end up calling it with preemption enabled, so wrap the functions in their own little preempt-disable section so they can be safely called in any context (though preemption should never be enabled if we're actually in a lazy state). (Move out of line to avoid #include dependencies.) Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Thomas Gleixner --- arch/x86/include/asm/paravirt.h | 17 ++--------------- arch/x86/kernel/paravirt.c | 24 ++++++++++++++++++++++++ 2 files changed, 26 insertions(+), 15 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index ba3e2ff6aed..a660eceaa27 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -1352,14 +1352,7 @@ static inline void arch_leave_lazy_cpu_mode(void) PVOP_VCALL0(pv_cpu_ops.lazy_mode.leave); } -static inline void arch_flush_lazy_cpu_mode(void) -{ - if (unlikely(paravirt_get_lazy_mode() == PARAVIRT_LAZY_CPU)) { - arch_leave_lazy_cpu_mode(); - arch_enter_lazy_cpu_mode(); - } -} - +void arch_flush_lazy_cpu_mode(void); #define __HAVE_ARCH_ENTER_LAZY_MMU_MODE static inline void arch_enter_lazy_mmu_mode(void) @@ -1372,13 +1365,7 @@ static inline void arch_leave_lazy_mmu_mode(void) PVOP_VCALL0(pv_mmu_ops.lazy_mode.leave); } -static inline void arch_flush_lazy_mmu_mode(void) -{ - if (unlikely(paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU)) { - arch_leave_lazy_mmu_mode(); - arch_enter_lazy_mmu_mode(); - } -} +void arch_flush_lazy_mmu_mode(void); static inline void __set_fixmap(unsigned /* enum fixed_addresses */ idx, unsigned long phys, pgprot_t flags) diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index e4c8fb60887..dcba6c567a2 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -268,6 +268,30 @@ enum paravirt_lazy_mode paravirt_get_lazy_mode(void) return __get_cpu_var(paravirt_lazy_mode); } +void arch_flush_lazy_mmu_mode(void) +{ + preempt_disable(); + + if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU) { + arch_leave_lazy_mmu_mode(); + arch_enter_lazy_mmu_mode(); + } + + preempt_enable(); +} + +void arch_flush_lazy_cpu_mode(void) +{ + preempt_disable(); + + if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_CPU) { + arch_leave_lazy_cpu_mode(); + arch_enter_lazy_cpu_mode(); + } + + preempt_enable(); +} + struct pv_info pv_info = { .name = "bare hardware", .paravirt_enabled = 0, -- cgit v1.2.3 From 34b0900d323122113683685b200aae9f9b75e63b Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 12 Feb 2009 21:30:48 +0100 Subject: x86: warn if arch_flush_lazy_mmu_cpu is called in preemptible context Impact: Catch cases where lazy MMU state is active in a preemtible context arch_flush_lazy_mmu_cpu() has been changed to disable preemption so the checks in enter/leave will never trigger. Put the preemtible() check into arch_flush_lazy_mmu_cpu() to catch such cases. Signed-off-by: Thomas Gleixner --- arch/x86/kernel/paravirt.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index dcba6c567a2..c6520a4e85d 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -273,6 +273,7 @@ void arch_flush_lazy_mmu_mode(void) preempt_disable(); if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU) { + WARN_ON(preempt_count() == 1); arch_leave_lazy_mmu_mode(); arch_enter_lazy_mmu_mode(); } @@ -285,6 +286,7 @@ void arch_flush_lazy_cpu_mode(void) preempt_disable(); if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_CPU) { + WARN_ON(preempt_count() == 1); arch_leave_lazy_cpu_mode(); arch_enter_lazy_cpu_mode(); } -- cgit v1.2.3 From 7ad9de6ac83bd825996d2de98c92e0f425c31050 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 12 Feb 2009 21:16:09 +0100 Subject: x86: CPA avoid repeated lazy mmu flush Impact: Flush the lazy MMU only once Pending mmu updates only need to be flushed once to bring the in-memory pagetable state up to date. Signed-off-by: Thomas Gleixner --- arch/x86/mm/pageattr.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index f664bc1c409..8ca0d8566fc 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -575,14 +575,6 @@ static int __change_page_attr(struct cpa_data *cpa, int primary) address = cpa->vaddr[cpa->curpage]; else address = *cpa->vaddr; - - /* - * If we're called with lazy mmu updates enabled, the - * in-memory pte state may be stale. Flush pending updates to - * bring them up to date. - */ - arch_flush_lazy_mmu_mode(); - repeat: kpte = lookup_address(address, &level); if (!kpte) @@ -819,6 +811,13 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, vm_unmap_aliases(); + /* + * If we're called with lazy mmu updates enabled, the + * in-memory pte state may be stale. Flush pending updates to + * bring them up to date. + */ + arch_flush_lazy_mmu_mode(); + cpa.vaddr = addr; cpa.numpages = numpages; cpa.mask_set = mask_set; -- cgit v1.2.3 From b13e24644c138d0ddbc451403c30a96b09bfd556 Mon Sep 17 00:00:00 2001 From: john stultz Date: Thu, 12 Feb 2009 18:48:53 -0800 Subject: x86, hpet: fix for LS21 + HPET = boot hang Between 2.6.23 and 2.6.24-rc1 a change was made that broke IBM LS21 systems that had the HPET enabled in the BIOS, resulting in boot hangs for x86_64. Specifically commit b8ce33590687888ebb900d09557b8807c4539022, which merges the i386 and x86_64 HPET code. Prior to this commit, when we setup the HPET timers in x86_64, we did the following: hpet_writel(HPET_TN_ENABLE | HPET_TN_PERIODIC | HPET_TN_SETVAL | HPET_TN_32BIT, HPET_T0_CFG); However after the i386/x86_64 HPET merge, we do the following: cfg = hpet_readl(HPET_Tn_CFG(timer)); cfg |= HPET_TN_ENABLE | HPET_TN_PERIODIC | HPET_TN_SETVAL | HPET_TN_32BIT; hpet_writel(cfg, HPET_Tn_CFG(timer)); However on LS21s with HPET enabled in the BIOS, the HPET_T0_CFG register boots with Level triggered interrupts (HPET_TN_LEVEL) enabled. This causes the periodic interrupt to be not so periodic, and that results in the boot time hang I reported earlier in the delay calibration. My fix: Always disable HPET_TN_LEVEL when setting up periodic mode. Signed-off-by: John Stultz Signed-off-by: Ingo Molnar --- arch/x86/kernel/hpet.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 64d5ad0b8ad..5c8da2c2c18 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -269,6 +269,8 @@ static void hpet_set_mode(enum clock_event_mode mode, now = hpet_readl(HPET_COUNTER); cmp = now + (unsigned long) delta; cfg = hpet_readl(HPET_Tn_CFG(timer)); + /* Make sure we use edge triggered interrupts */ + cfg &= ~HPET_TN_LEVEL; cfg |= HPET_TN_ENABLE | HPET_TN_PERIODIC | HPET_TN_SETVAL | HPET_TN_32BIT; hpet_writel(cfg, HPET_Tn_CFG(timer)); -- cgit v1.2.3 From e49590b6dd356f8ef10ba3531a29e5086f6f2e3a Mon Sep 17 00:00:00 2001 From: Chris Ball Date: Fri, 13 Feb 2009 20:56:18 -0500 Subject: x86, olpc: fix model detection without OFW Impact: fix "garbled display, laptop is unusable" bug Commit e51a1ac2dfca9ad869471e88f828281db7e810c0 ("x86, olpc: fix endian bug in openfirmware workaround") breaks model comparison on OLPC; the value 0xc2 needs to be scaled up by olpc_board(). The pre-patch version was wrong, but accidentally worked anyway (big-endian 0xc2 is big enough to satisfy all other board revisions, but little endian 0xc2 is not). Signed-off-by: Chris Ball Cc: Andrew Morton Acked-by: Andres Salomon Cc: Harvey Harrison Signed-off-by: Ingo Molnar --- arch/x86/kernel/olpc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/olpc.c b/arch/x86/kernel/olpc.c index 7a13fac63a1..4006c522adc 100644 --- a/arch/x86/kernel/olpc.c +++ b/arch/x86/kernel/olpc.c @@ -203,7 +203,7 @@ static void __init platform_detect(void) static void __init platform_detect(void) { /* stopgap until OFW support is added to the kernel */ - olpc_platform_info.boardrev = 0xc2; + olpc_platform_info.boardrev = olpc_board(0xc2); } #endif -- cgit v1.2.3 From 7a0eb1960e8ddcb68ea631caf16815485af0e228 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Mon, 19 Jan 2009 14:57:52 +0200 Subject: KVM: Avoid using CONFIG_ in userspace visible headers Kconfig symbols are not available in userspace, and are not stripped by headers-install. Avoid their use by adding #defines in to suit each architecture. Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/kvm.h b/arch/x86/include/asm/kvm.h index d2e3bf3608a..886c9402ec4 100644 --- a/arch/x86/include/asm/kvm.h +++ b/arch/x86/include/asm/kvm.h @@ -9,6 +9,13 @@ #include #include +/* Select x86 specific features in */ +#define __KVM_HAVE_PIT +#define __KVM_HAVE_IOAPIC +#define __KVM_HAVE_DEVICE_ASSIGNMENT +#define __KVM_HAVE_MSI +#define __KVM_HAVE_USER_NMI + /* Architectural interrupt line count. */ #define KVM_NR_INTERRUPTS 256 -- cgit v1.2.3 From ad8ba2cd44d4d39fb3fe55d5dcc565b19fc3a7fb Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Tue, 6 Jan 2009 10:03:02 +0800 Subject: KVM: Add kvm_arch_sync_events to sync with asynchronize events kvm_arch_sync_events is introduced to quiet down all other events may happen contemporary with VM destroy process, like IRQ handler and work struct for assigned device. For kvm_arch_sync_events is called at the very beginning of kvm_destroy_vm(), so the state of KVM here is legal and can provide a environment to quiet down other events. Signed-off-by: Sheng Yang Signed-off-by: Avi Kivity --- arch/x86/kvm/x86.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index cc17546a240..b0fc079f1be 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -4127,6 +4127,10 @@ static void kvm_free_vcpus(struct kvm *kvm) } +void kvm_arch_sync_events(struct kvm *kvm) +{ +} + void kvm_arch_destroy_vm(struct kvm *kvm) { kvm_free_all_assigned_devices(kvm); -- cgit v1.2.3 From ba4cef31d5a397b64ba6d3ff713ce06c62f0c597 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Tue, 6 Jan 2009 10:03:03 +0800 Subject: KVM: Fix racy in kvm_free_assigned_irq In the past, kvm_get_kvm() and kvm_put_kvm() was called in assigned device irq handler and interrupt_work, in order to prevent cancel_work_sync() in kvm_free_assigned_irq got a illegal state when waiting for interrupt_work done. But it's tricky and still got two problems: 1. A bug ignored two conditions that cancel_work_sync() would return true result in a additional kvm_put_kvm(). 2. If interrupt type is MSI, we would got a window between cancel_work_sync() and free_irq(), which interrupt would be injected again... This patch discard the reference count used for irq handler and interrupt_work, and ensure the legal state by moving the free function at the very beginning of kvm_destroy_vm(). And the patch fix the second bug by disable irq before cancel_work_sync(), which may result in nested disable of irq but OK for we are going to free it. Signed-off-by: Sheng Yang Signed-off-by: Avi Kivity --- arch/x86/kvm/x86.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index b0fc079f1be..fc3e329f6ad 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -4129,11 +4129,11 @@ static void kvm_free_vcpus(struct kvm *kvm) void kvm_arch_sync_events(struct kvm *kvm) { + kvm_free_all_assigned_devices(kvm); } void kvm_arch_destroy_vm(struct kvm *kvm) { - kvm_free_all_assigned_devices(kvm); kvm_iommu_unmap_guest(kvm); kvm_free_pit(kvm); kfree(kvm->arch.vpic); -- cgit v1.2.3 From d2a8284e8fca9e2a938bee6cd074064d23864886 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Tue, 30 Dec 2008 15:55:05 -0200 Subject: KVM: PIT: fix i8254 pending count read count_load_time assignment is bogus: its supposed to contain what it means, not the expiration time. Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/kvm/i8254.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c index e665d1c623c..72bd275a9b5 100644 --- a/arch/x86/kvm/i8254.c +++ b/arch/x86/kvm/i8254.c @@ -207,7 +207,7 @@ static int __pit_timer_fn(struct kvm_kpit_state *ps) hrtimer_add_expires_ns(&pt->timer, pt->period); pt->scheduled = hrtimer_get_expires_ns(&pt->timer); if (pt->period) - ps->channels[0].count_load_time = hrtimer_get_expires(&pt->timer); + ps->channels[0].count_load_time = ktime_get(); return (pt->period == 0 ? 0 : 1); } -- cgit v1.2.3 From abe6655dd699069b53bcccbc65b2717f60203b12 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Tue, 10 Feb 2009 20:59:45 -0200 Subject: KVM: x86: disable kvmclock on non constant TSC hosts This is better. Currently, this code path is posing us big troubles, and we won't have a decent patch in time. So, temporarily disable it. Signed-off-by: Glauber Costa Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/kvm/x86.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index fc3e329f6ad..758b7a155ae 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -967,7 +967,6 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_MMU_SHADOW_CACHE_CONTROL: case KVM_CAP_SET_TSS_ADDR: case KVM_CAP_EXT_CPUID: - case KVM_CAP_CLOCKSOURCE: case KVM_CAP_PIT: case KVM_CAP_NOP_IO_DELAY: case KVM_CAP_MP_STATE: @@ -992,6 +991,9 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_IOMMU: r = iommu_found(); break; + case KVM_CAP_CLOCKSOURCE: + r = boot_cpu_has(X86_FEATURE_CONSTANT_TSC); + break; default: r = 0; break; -- cgit v1.2.3 From 2aaf69dcee864f4fb6402638dd2f263324ac839f Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Wed, 21 Jan 2009 16:52:16 +0800 Subject: KVM: MMU: Map device MMIO as UC in EPT Software are not allow to access device MMIO using cacheable memory type, the patch limit MMIO region with UC and WC(guest can select WC using PAT and PCD/PWT). Signed-off-by: Sheng Yang Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 9 +++++++-- arch/x86/kvm/vmx.c | 3 +-- 2 files changed, 8 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 83f11c7474a..2d4477c7147 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -1698,8 +1698,13 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, if (largepage) spte |= PT_PAGE_SIZE_MASK; if (mt_mask) { - mt_mask = get_memory_type(vcpu, gfn) << - kvm_x86_ops->get_mt_mask_shift(); + if (!kvm_is_mmio_pfn(pfn)) { + mt_mask = get_memory_type(vcpu, gfn) << + kvm_x86_ops->get_mt_mask_shift(); + mt_mask |= VMX_EPT_IGMT_BIT; + } else + mt_mask = MTRR_TYPE_UNCACHABLE << + kvm_x86_ops->get_mt_mask_shift(); spte |= mt_mask; } diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 6259d746764..07491c9c6ed 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3687,8 +3687,7 @@ static int __init vmx_init(void) if (vm_need_ept()) { bypass_guest_pf = 0; kvm_mmu_set_base_ptes(VMX_EPT_READABLE_MASK | - VMX_EPT_WRITABLE_MASK | - VMX_EPT_IGMT_BIT); + VMX_EPT_WRITABLE_MASK); kvm_mmu_set_mask_ptes(0ull, 0ull, 0ull, 0ull, VMX_EPT_EXECUTABLE_MASK, VMX_EPT_DEFAULT_MT << VMX_EPT_MT_EPTE_SHIFT); -- cgit v1.2.3 From b682b814e3cc340f905c14dff87ce8bdba7c5eba Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Tue, 10 Feb 2009 20:41:41 -0200 Subject: KVM: x86: fix LAPIC pending count calculation Simplify LAPIC TMCCT calculation by using hrtimer provided function to query remaining time until expiration. Fixes host hang with nested ESX. Signed-off-by: Marcelo Tosatti Signed-off-by: Alexander Graf Signed-off-by: Avi Kivity --- arch/x86/kvm/irq.c | 7 ------ arch/x86/kvm/irq.h | 1 - arch/x86/kvm/lapic.c | 66 ++++++++++++---------------------------------------- arch/x86/kvm/lapic.h | 2 -- arch/x86/kvm/svm.c | 1 - arch/x86/kvm/vmx.c | 1 - 6 files changed, 15 insertions(+), 63 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c index c019b8edcdb..cf17ed52f6f 100644 --- a/arch/x86/kvm/irq.c +++ b/arch/x86/kvm/irq.c @@ -87,13 +87,6 @@ void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu) } EXPORT_SYMBOL_GPL(kvm_inject_pending_timer_irqs); -void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec) -{ - kvm_apic_timer_intr_post(vcpu, vec); - /* TODO: PIT, RTC etc. */ -} -EXPORT_SYMBOL_GPL(kvm_timer_intr_post); - void __kvm_migrate_timers(struct kvm_vcpu *vcpu) { __kvm_migrate_apic_timer(vcpu); diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h index 2bf32a03cee..82579ee538d 100644 --- a/arch/x86/kvm/irq.h +++ b/arch/x86/kvm/irq.h @@ -89,7 +89,6 @@ static inline int irqchip_in_kernel(struct kvm *kvm) void kvm_pic_reset(struct kvm_kpic_state *s); -void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec); void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu); void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu); void kvm_apic_nmi_wd_deliver(struct kvm_vcpu *vcpu); diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index afac68c0815..f0b67f2cdd6 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -35,6 +35,12 @@ #include "kvm_cache_regs.h" #include "irq.h" +#ifndef CONFIG_X86_64 +#define mod_64(x, y) ((x) - (y) * div64_u64(x, y)) +#else +#define mod_64(x, y) ((x) % (y)) +#endif + #define PRId64 "d" #define PRIx64 "llx" #define PRIu64 "u" @@ -511,52 +517,22 @@ static void apic_send_ipi(struct kvm_lapic *apic) static u32 apic_get_tmcct(struct kvm_lapic *apic) { - u64 counter_passed; - ktime_t passed, now; + ktime_t remaining; + s64 ns; u32 tmcct; ASSERT(apic != NULL); - now = apic->timer.dev.base->get_time(); - tmcct = apic_get_reg(apic, APIC_TMICT); - /* if initial count is 0, current count should also be 0 */ - if (tmcct == 0) + if (apic_get_reg(apic, APIC_TMICT) == 0) return 0; - if (unlikely(ktime_to_ns(now) <= - ktime_to_ns(apic->timer.last_update))) { - /* Wrap around */ - passed = ktime_add(( { - (ktime_t) { - .tv64 = KTIME_MAX - - (apic->timer.last_update).tv64}; } - ), now); - apic_debug("time elapsed\n"); - } else - passed = ktime_sub(now, apic->timer.last_update); - - counter_passed = div64_u64(ktime_to_ns(passed), - (APIC_BUS_CYCLE_NS * apic->timer.divide_count)); - - if (counter_passed > tmcct) { - if (unlikely(!apic_lvtt_period(apic))) { - /* one-shot timers stick at 0 until reset */ - tmcct = 0; - } else { - /* - * periodic timers reset to APIC_TMICT when they - * hit 0. The while loop simulates this happening N - * times. (counter_passed %= tmcct) would also work, - * but might be slower or not work on 32-bit?? - */ - while (counter_passed > tmcct) - counter_passed -= tmcct; - tmcct -= counter_passed; - } - } else { - tmcct -= counter_passed; - } + remaining = hrtimer_expires_remaining(&apic->timer.dev); + if (ktime_to_ns(remaining) < 0) + remaining = ktime_set(0, 0); + + ns = mod_64(ktime_to_ns(remaining), apic->timer.period); + tmcct = div64_u64(ns, (APIC_BUS_CYCLE_NS * apic->timer.divide_count)); return tmcct; } @@ -653,8 +629,6 @@ static void start_apic_timer(struct kvm_lapic *apic) { ktime_t now = apic->timer.dev.base->get_time(); - apic->timer.last_update = now; - apic->timer.period = apic_get_reg(apic, APIC_TMICT) * APIC_BUS_CYCLE_NS * apic->timer.divide_count; atomic_set(&apic->timer.pending, 0); @@ -1110,16 +1084,6 @@ void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu) } } -void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec) -{ - struct kvm_lapic *apic = vcpu->arch.apic; - - if (apic && apic_lvt_vector(apic, APIC_LVTT) == vec) - apic->timer.last_update = ktime_add_ns( - apic->timer.last_update, - apic->timer.period); -} - int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu) { int vector = kvm_apic_has_interrupt(vcpu); diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h index 81858881287..45ab6ee7120 100644 --- a/arch/x86/kvm/lapic.h +++ b/arch/x86/kvm/lapic.h @@ -12,7 +12,6 @@ struct kvm_lapic { atomic_t pending; s64 period; /* unit: ns */ u32 divide_count; - ktime_t last_update; struct hrtimer dev; } timer; struct kvm_vcpu *vcpu; @@ -42,7 +41,6 @@ void kvm_set_apic_base(struct kvm_vcpu *vcpu, u64 data); void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu); int kvm_lapic_enabled(struct kvm_vcpu *vcpu); int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu); -void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec); void kvm_lapic_set_vapic_addr(struct kvm_vcpu *vcpu, gpa_t vapic_addr); void kvm_lapic_sync_from_vapic(struct kvm_vcpu *vcpu); diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 1452851ae25..a9e769e4e25 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1600,7 +1600,6 @@ static void svm_intr_assist(struct kvm_vcpu *vcpu) /* Okay, we can deliver the interrupt: grab it and update PIC state. */ intr_vector = kvm_cpu_get_interrupt(vcpu); svm_inject_irq(svm, intr_vector); - kvm_timer_intr_post(vcpu, intr_vector); out: update_cr8_intercept(vcpu); } diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 07491c9c6ed..b1fe1422afb 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3285,7 +3285,6 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu) } if (vcpu->arch.interrupt.pending) { vmx_inject_irq(vcpu, vcpu->arch.interrupt.nr); - kvm_timer_intr_post(vcpu, vcpu->arch.interrupt.nr); if (kvm_cpu_has_interrupt(vcpu)) enable_irq_window(vcpu); } -- cgit v1.2.3 From 516a1a7e9dc80358030fe01aabb3bedf882db9e2 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Sun, 15 Feb 2009 02:32:07 +0200 Subject: KVM: VMX: Flush volatile msrs before emulating rdmsr Some msrs (notable MSR_KERNEL_GS_BASE) are held in the processor registers and need to be flushed to the vcpu struture before they can be read. This fixes cygwin longjmp() failure on Windows x64. Signed-off-by: Avi Kivity --- arch/x86/kvm/vmx.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86') diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index b1fe1422afb..7611af57682 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -903,6 +903,7 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata) data = vmcs_readl(GUEST_SYSENTER_ESP); break; default: + vmx_load_host_state(to_vmx(vcpu)); msr = find_msr_entry(to_vmx(vcpu), msr_index); if (msr) { data = msr->data; -- cgit v1.2.3 From be716615fe596ee117292dc615e95f707fb67fd1 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 13 Jan 2009 23:36:34 +0100 Subject: x86, vm86: fix preemption bug Commit 3d2a71a596bd9c761c8487a2178e95f8a61da083 ("x86, traps: converge do_debug handlers") changed the preemption disable logic of do_debug() so vm86_handle_trap() is called with preemption disabled resulting in: BUG: sleeping function called from invalid context at include/linux/kernel.h:155 in_atomic(): 1, irqs_disabled(): 0, pid: 3005, name: dosemu.bin Pid: 3005, comm: dosemu.bin Tainted: G W 2.6.29-rc1 #51 Call Trace: [] copy_to_user+0x33/0x108 [] save_v86_state+0x65/0x149 [] handle_vm86_trap+0x20/0x8f [] do_debug+0x15b/0x1a4 [] debug_stack_correct+0x27/0x2c [] sysenter_do_call+0x12/0x2f BUG: scheduling while atomic: dosemu.bin/3005/0x10000001 Restore the original calling convention and reenable preemption before calling handle_vm86_trap(). Reported-by: Michal Suchanek Cc: stable@kernel.org Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- arch/x86/kernel/traps.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 7932338d7cb..a9e7548e179 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -99,6 +99,12 @@ static inline void preempt_conditional_sti(struct pt_regs *regs) local_irq_enable(); } +static inline void conditional_cli(struct pt_regs *regs) +{ + if (regs->flags & X86_EFLAGS_IF) + local_irq_disable(); +} + static inline void preempt_conditional_cli(struct pt_regs *regs) { if (regs->flags & X86_EFLAGS_IF) @@ -626,8 +632,10 @@ clear_dr7: #ifdef CONFIG_X86_32 debug_vm86: + /* reenable preemption: handle_vm86_trap() might sleep */ + dec_preempt_count(); handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, 1); - preempt_conditional_cli(regs); + conditional_cli(regs); return; #endif -- cgit v1.2.3 From 6bc5c366b1a45ca18fba6851f62db5743b3f6db5 Mon Sep 17 00:00:00 2001 From: Pekka Paalanen Date: Sat, 3 Jan 2009 21:23:51 +0200 Subject: trace: mmiotrace to the tracer menu in Kconfig Impact: cosmetic change in Kconfig menu layout This patch was originally suggested by Peter Zijlstra, but seems it was forgotten. CONFIG_MMIOTRACE and CONFIG_MMIOTRACE_TEST were selectable directly under the Kernel hacking / debugging menu in the kernel configuration system. They were present only for x86 and x86_64. Other tracers that use the ftrace tracing framework are in their own sub-menu. This patch moves the mmiotrace configuration options there. Since the Kconfig file, where the tracer menu is, is not architecture specific, HAVE_MMIOTRACE_SUPPORT is introduced and provided only by x86/x86_64. CONFIG_MMIOTRACE now depends on it. Signed-off-by: Pekka Paalanen Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- arch/x86/Kconfig.debug | 24 ++---------------------- 1 file changed, 2 insertions(+), 22 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index 10d6cc3fd05..e1983fa025d 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -174,28 +174,8 @@ config IOMMU_LEAK Add a simple leak tracer to the IOMMU code. This is useful when you are debugging a buggy device driver that leaks IOMMU mappings. -config MMIOTRACE - bool "Memory mapped IO tracing" - depends on DEBUG_KERNEL && PCI - select TRACING - help - Mmiotrace traces Memory Mapped I/O access and is meant for - debugging and reverse engineering. It is called from the ioremap - implementation and works via page faults. Tracing is disabled by - default and can be enabled at run-time. - - See Documentation/tracers/mmiotrace.txt. - If you are not helping to develop drivers, say N. - -config MMIOTRACE_TEST - tristate "Test module for mmiotrace" - depends on MMIOTRACE && m - help - This is a dumb module for testing mmiotrace. It is very dangerous - as it will write garbage to IO memory starting at a given address. - However, it should be safe to use on e.g. unused portion of VRAM. - - Say N, unless you absolutely know what you are doing. +config HAVE_MMIOTRACE_SUPPORT + def_bool y # # IO delay types: -- cgit v1.2.3 From a0abd520fd69295f4a3735e29a9448a32e101d47 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Mon, 16 Feb 2009 17:31:58 -0600 Subject: cpumask: fix powernow-k8: partial revert of 2fdf66b491ac706657946442789ec644cc317e1a Impact: fix powernow-k8 when acpi=off (or other error). There was a spurious change introduced into powernow-k8 in this patch: so that we try to "restore" the cpus_allowed we never saved. We revert that file. See lkml "[PATCH] x86/powernow: fix cpus_allowed brokage when acpi=off" from Yinghai for the bug report. Cc: Mike Travis Cc: Yinghai Lu Signed-off-by: Rusty Russell Acked-by: Ingo Molnar --- arch/x86/kernel/cpu/cpufreq/powernow-k8.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c index fb039cd345d..6428aa17b40 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c @@ -1157,8 +1157,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) data->cpu = pol->cpu; data->currpstate = HW_PSTATE_INVALID; - rc = powernow_k8_cpu_init_acpi(data); - if (rc) { + if (powernow_k8_cpu_init_acpi(data)) { /* * Use the PSB BIOS structure. This is only availabe on * an UP version, and is deprecated by AMD. @@ -1176,17 +1175,20 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) "ACPI maintainers and complain to your BIOS " "vendor.\n"); #endif - goto err_out; + kfree(data); + return -ENODEV; } if (pol->cpu != 0) { printk(KERN_ERR FW_BUG PFX "No ACPI _PSS objects for " "CPU other than CPU0. Complain to your BIOS " "vendor.\n"); - goto err_out; + kfree(data); + return -ENODEV; } rc = find_psb_table(data); if (rc) { - goto err_out; + kfree(data); + return -ENODEV; } /* Take a crude guess here. * That guess was in microseconds, so multiply with 1000 */ -- cgit v1.2.3 From bf51935f3e988e0ed6f34b55593e5912f990750a Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 17 Feb 2009 06:01:30 -0800 Subject: x86, rcu: fix strange load average and ksoftirqd behavior Damien Wyart reported high ksoftirqd CPU usage (20%) on an otherwise idle system. The function-graph trace Damien provided: > 799.521187 | 1) -0 | | rcu_check_callbacks() { > 799.521371 | 1) -0 | | rcu_check_callbacks() { > 799.521555 | 1) -0 | | rcu_check_callbacks() { > 799.521738 | 1) -0 | | rcu_check_callbacks() { > 799.521934 | 1) -0 | | rcu_check_callbacks() { > 799.522068 | 1) ksoftir-2324 | | rcu_check_callbacks() { > 799.522208 | 1) -0 | | rcu_check_callbacks() { > 799.522392 | 1) -0 | | rcu_check_callbacks() { > 799.522575 | 1) -0 | | rcu_check_callbacks() { > 799.522759 | 1) -0 | | rcu_check_callbacks() { > 799.522956 | 1) -0 | | rcu_check_callbacks() { > 799.523074 | 1) ksoftir-2324 | | rcu_check_callbacks() { > 799.523214 | 1) -0 | | rcu_check_callbacks() { > 799.523397 | 1) -0 | | rcu_check_callbacks() { > 799.523579 | 1) -0 | | rcu_check_callbacks() { > 799.523762 | 1) -0 | | rcu_check_callbacks() { > 799.523960 | 1) -0 | | rcu_check_callbacks() { > 799.524079 | 1) ksoftir-2324 | | rcu_check_callbacks() { > 799.524220 | 1) -0 | | rcu_check_callbacks() { > 799.524403 | 1) -0 | | rcu_check_callbacks() { > 799.524587 | 1) -0 | | rcu_check_callbacks() { > 799.524770 | 1) -0 | | rcu_check_callbacks() { > [ . . . ] Shows rcu_check_callbacks() being invoked way too often. It should be called once per jiffy, and here it is called no less than 22 times in about 3.5 milliseconds, meaning one call every 160 microseconds or so. Why do we need to call rcu_pending() and rcu_check_callbacks() from the idle loop of 32-bit x86, especially given that no other architecture does this? The following patch removes the call to rcu_pending() and rcu_check_callbacks() from the x86 32-bit idle loop in order to reduce the softirq load on idle systems. Reported-by: Damien Wyart Signed-off-by: Paul E. McKenney Signed-off-by: Ingo Molnar --- arch/x86/kernel/process_32.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index a546f55c77b..bd4da2af08a 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -104,9 +104,6 @@ void cpu_idle(void) check_pgt_cache(); rmb(); - if (rcu_pending(cpu)) - rcu_check_callbacks(cpu, 0); - if (cpu_is_offline(cpu)) play_dead(); -- cgit v1.2.3 From 6ec68bff3c81e776a455f6aca95c8c5f1d630198 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 12 Feb 2009 13:39:26 +0100 Subject: x86, mce: reinitialize per cpu features on resume Impact: Bug fix This fixes a long standing bug in the machine check code. On resume the boot CPU wouldn't get its vendor specific state like thermal handling reinitialized. This means the boot cpu wouldn't ever get any thermal events reported again. Call the respective initialization functions on resume v2: Remove ancient init because they don't have a resume device anyways. Pointed out by Thomas Gleixner. v3: Now fix the Subject too to reflect v2 change Signed-off-by: Andi Kleen Acked-by: Thomas Gleixner Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce_64.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c index 1c838032fd3..1f184efb6bc 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_64.c @@ -734,6 +734,7 @@ __setup("mce=", mcheck_enable); static int mce_resume(struct sys_device *dev) { mce_init(NULL); + mce_cpu_features(¤t_cpu_data); return 0; } -- cgit v1.2.3 From 380851bc6b1b4107c61dfa2997f9095dcf779336 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 12 Feb 2009 13:39:33 +0100 Subject: x86, mce: use force_sig_info to kill process in machine check Impact: bug fix (with tolerant == 3) do_exit cannot be called directly from the exception handler because it can sleep and the exception handler runs on the exception stack. Use force_sig() instead. Based on a earlier patch by Ying Huang who debugged the problem. Signed-off-by: Andi Kleen Acked-by: Thomas Gleixner Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce_64.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c index 1f184efb6bc..25cf624eccb 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_64.c @@ -295,11 +295,11 @@ void do_machine_check(struct pt_regs * regs, long error_code) * If we know that the error was in user space, send a * SIGBUS. Otherwise, panic if tolerance is low. * - * do_exit() takes an awful lot of locks and has a slight + * force_sig() takes an awful lot of locks and has a slight * risk of deadlocking. */ if (user_space) { - do_exit(SIGBUS); + force_sig(SIGBUS, current); } else if (panic_on_oops || tolerant < 2) { mce_panic("Uncorrected machine check", &panicm, mcestart); -- cgit v1.2.3 From 07db1c140eb233971341396e492cc73d4280e698 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 12 Feb 2009 13:39:35 +0100 Subject: x86, mce: fix ifdef for 64bit thermal apic vector clear on shutdown Impact: Bugfix The ifdef for the apic clear on shutdown for the 64bit intel thermal vector was incorrect and never triggered. Fix that. Signed-off-by: Andi Kleen Acked-by: Thomas Gleixner Signed-off-by: H. Peter Anvin --- arch/x86/kernel/apic.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic.c b/arch/x86/kernel/apic.c index 115449f869e..570f36e44e5 100644 --- a/arch/x86/kernel/apic.c +++ b/arch/x86/kernel/apic.c @@ -862,7 +862,7 @@ void clear_local_APIC(void) } /* lets not touch this if we didn't frob it */ -#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(X86_MCE_INTEL) +#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL) if (maxlvt >= 5) { v = apic_read(APIC_LVTTHMR); apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED); -- cgit v1.2.3 From f2dbcfa738368c8a40d4a5f0b65dc9879577cb21 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Wed, 18 Feb 2009 14:48:32 -0800 Subject: mm: clean up for early_pfn_to_nid() What's happening is that the assertion in mm/page_alloc.c:move_freepages() is triggering: BUG_ON(page_zone(start_page) != page_zone(end_page)); Once I knew this is what was happening, I added some annotations: if (unlikely(page_zone(start_page) != page_zone(end_page))) { printk(KERN_ERR "move_freepages: Bogus zones: " "start_page[%p] end_page[%p] zone[%p]\n", start_page, end_page, zone); printk(KERN_ERR "move_freepages: " "start_zone[%p] end_zone[%p]\n", page_zone(start_page), page_zone(end_page)); printk(KERN_ERR "move_freepages: " "start_pfn[0x%lx] end_pfn[0x%lx]\n", page_to_pfn(start_page), page_to_pfn(end_page)); printk(KERN_ERR "move_freepages: " "start_nid[%d] end_nid[%d]\n", page_to_nid(start_page), page_to_nid(end_page)); ... And here's what I got: move_freepages: Bogus zones: start_page[2207d0000] end_page[2207dffc0] zone[fffff8103effcb00] move_freepages: start_zone[fffff8103effcb00] end_zone[fffff8003fffeb00] move_freepages: start_pfn[0x81f600] end_pfn[0x81f7ff] move_freepages: start_nid[1] end_nid[0] My memory layout on this box is: [ 0.000000] Zone PFN ranges: [ 0.000000] Normal 0x00000000 -> 0x0081ff5d [ 0.000000] Movable zone start PFN for each node [ 0.000000] early_node_map[8] active PFN ranges [ 0.000000] 0: 0x00000000 -> 0x00020000 [ 0.000000] 1: 0x00800000 -> 0x0081f7ff [ 0.000000] 1: 0x0081f800 -> 0x0081fe50 [ 0.000000] 1: 0x0081fed1 -> 0x0081fed8 [ 0.000000] 1: 0x0081feda -> 0x0081fedb [ 0.000000] 1: 0x0081fedd -> 0x0081fee5 [ 0.000000] 1: 0x0081fee7 -> 0x0081ff51 [ 0.000000] 1: 0x0081ff59 -> 0x0081ff5d So it's a block move in that 0x81f600-->0x81f7ff region which triggers the problem. This patch: Declaration of early_pfn_to_nid() is scattered over per-arch include files, and it seems it's complicated to know when the declaration is used. I think it makes fix-for-memmap-init not easy. This patch moves all declaration to include/linux/mm.h After this, if !CONFIG_NODES_POPULATES_NODE_MAP && !CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID -> Use static definition in include/linux/mm.h else if !CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID -> Use generic definition in mm/page_alloc.c else -> per-arch back end function will be called. Signed-off-by: KAMEZAWA Hiroyuki Tested-by: KOSAKI Motohiro Reported-by: David Miller Cc: Mel Gorman Cc: Heiko Carstens Cc: [2.6.25.x, 2.6.26.x, 2.6.27.x, 2.6.28.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/include/asm/mmzone_32.h | 2 -- arch/x86/include/asm/mmzone_64.h | 2 -- arch/x86/mm/numa_64.c | 2 +- 3 files changed, 1 insertion(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/mmzone_32.h b/arch/x86/include/asm/mmzone_32.h index 07f1af494ca..105fb90a063 100644 --- a/arch/x86/include/asm/mmzone_32.h +++ b/arch/x86/include/asm/mmzone_32.h @@ -32,8 +32,6 @@ static inline void get_memcfg_numa(void) get_memcfg_numa_flat(); } -extern int early_pfn_to_nid(unsigned long pfn); - extern void resume_map_numa_kva(pgd_t *pgd); #else /* !CONFIG_NUMA */ diff --git a/arch/x86/include/asm/mmzone_64.h b/arch/x86/include/asm/mmzone_64.h index a5b3817d4b9..a29f48c2a32 100644 --- a/arch/x86/include/asm/mmzone_64.h +++ b/arch/x86/include/asm/mmzone_64.h @@ -40,8 +40,6 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr) #define node_end_pfn(nid) (NODE_DATA(nid)->node_start_pfn + \ NODE_DATA(nid)->node_spanned_pages) -extern int early_pfn_to_nid(unsigned long pfn); - #ifdef CONFIG_NUMA_EMU #define FAKE_NODE_MIN_SIZE (64 * 1024 * 1024) #define FAKE_NODE_MIN_HASH_MASK (~(FAKE_NODE_MIN_SIZE - 1UL)) diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index 71a14f89f89..f3516da035d 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -145,7 +145,7 @@ int __init compute_hash_shift(struct bootnode *nodes, int numnodes, return shift; } -int early_pfn_to_nid(unsigned long pfn) +int __meminit __early_pfn_to_nid(unsigned long pfn) { return phys_to_nid(pfn << PAGE_SHIFT); } -- cgit v1.2.3 From 48ffc70b675aa7798a52a2e92e20f6cce9140b3d Mon Sep 17 00:00:00 2001 From: Alok N Kataria Date: Wed, 18 Feb 2009 12:33:55 -0800 Subject: x86, vmi: TSC going backwards check in vmi clocksource Impact: fix time warps under vmware Similar to the check for TSC going backwards in the TSC clocksource, we also need this check for VMI clocksource. Signed-off-by: Alok N Kataria Cc: Zachary Amsden Signed-off-by: Ingo Molnar Cc: stable@kernel.org --- arch/x86/kernel/vmiclock_32.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c index c4c1f9e0940..bde106cae0a 100644 --- a/arch/x86/kernel/vmiclock_32.c +++ b/arch/x86/kernel/vmiclock_32.c @@ -283,10 +283,13 @@ void __devinit vmi_time_ap_init(void) #endif /** vmi clocksource */ +static struct clocksource clocksource_vmi; static cycle_t read_real_cycles(void) { - return vmi_timer_ops.get_cycle_counter(VMI_CYCLES_REAL); + cycle_t ret = (cycle_t)vmi_timer_ops.get_cycle_counter(VMI_CYCLES_REAL); + return ret >= clocksource_vmi.cycle_last ? + ret : clocksource_vmi.cycle_last; } static struct clocksource clocksource_vmi = { -- cgit v1.2.3 From 07a66d7c53a538e1a9759954a82bb6c07365eff9 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 20 Feb 2009 08:04:13 +0100 Subject: x86: use the right protections for split-up pagetables Steven Rostedt found a bug in where in his modified kernel ftrace was unable to modify the kernel text, due to the PMD itself having been marked read-only as well in split_large_page(). The fix, suggested by Linus, is to not try to 'clone' the reference protection of a huge-page, but to use the standard (and permissive) page protection bits of KERNPG_TABLE. The 'cloning' makes sense for the ptes but it's a confused and incorrect concept at the page table level - because the pagetable entry is a set of all ptes and hence cannot 'clone' any single protection attribute - the ptes can be any mixture of protections. With the permissive KERNPG_TABLE, even if the pte protections get changed after this point (due to ftrace doing code-patching or other similar activities like kprobes), the resulting combined protections will still be correct and the pte's restrictive (or permissive) protections will control it. Also update the comment. This bug was there for a long time but has not caused visible problems before as it needs a rather large read-only area to trigger. Steve possibly hacked his kernel with some really large arrays or so. Anyway, the bug is definitely worth fixing. [ Huang Ying also experienced problems in this area when writing the EFI code, but the real bug in split_large_page() was not realized back then. ] Reported-by: Steven Rostedt Reported-by: Huang Ying Acked-by: Linus Torvalds Signed-off-by: Ingo Molnar --- arch/x86/mm/pageattr.c | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 8ca0d8566fc..7be47d1a97e 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -508,18 +508,13 @@ static int split_large_page(pte_t *kpte, unsigned long address) #endif /* - * Install the new, split up pagetable. Important details here: + * Install the new, split up pagetable. * - * On Intel the NX bit of all levels must be cleared to make a - * page executable. See section 4.13.2 of Intel 64 and IA-32 - * Architectures Software Developer's Manual). - * - * Mark the entry present. The current mapping might be - * set to not present, which we preserved above. + * We use the standard kernel pagetable protections for the new + * pagetable protections, the actual ptes set above control the + * primary protection behavior: */ - ref_prot = pte_pgprot(pte_mkexec(pte_clrhuge(*kpte))); - pgprot_val(ref_prot) |= _PAGE_PRESENT; - __set_pmd_pte(kpte, address, mk_pte(base, ref_prot)); + __set_pmd_pte(kpte, address, mk_pte(base, __pgprot(_KERNPG_TABLE))); base = NULL; out_unlock: -- cgit v1.2.3 From cc3ca22063784076bd240fda87217387a8f2ae92 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Fri, 20 Feb 2009 23:35:51 -0800 Subject: x86, mce: remove incorrect __cpuinit for mce_cpu_features() Impact: Bug fix on UP Checkin 6ec68bff3c81e776a455f6aca95c8c5f1d630198: x86, mce: reinitialize per cpu features on resume introduced a call to mce_cpu_features() in the resume path, in order for the MCE machinery to get properly reinitialized after a resume. However, this function (and its successors) was flagged __cpuinit, which becomes __init on UP configurations (on SMP suspend/resume requires CPU hotplug and so this would not be seen.) Remove the offending __cpuinit annotations for mce_cpu_features() and its successor functions. Cc: Andi Kleen Cc: Linus Torvalds Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce_64.c | 2 +- arch/x86/kernel/cpu/mcheck/mce_amd_64.c | 2 +- arch/x86/kernel/cpu/mcheck/mce_intel_64.c | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c index 25cf624eccb..fe79985ce0f 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_64.c @@ -490,7 +490,7 @@ static void __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c) } -static void __cpuinit mce_cpu_features(struct cpuinfo_x86 *c) +static void mce_cpu_features(struct cpuinfo_x86 *c) { switch (c->x86_vendor) { case X86_VENDOR_INTEL: diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c index 8ae8c4ff094..f2ee0ae29bd 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c @@ -121,7 +121,7 @@ static long threshold_restart_bank(void *_tr) } /* cpu init entry point, called from mce.c with preempt off */ -void __cpuinit mce_amd_feature_init(struct cpuinfo_x86 *c) +void mce_amd_feature_init(struct cpuinfo_x86 *c) { unsigned int bank, block; unsigned int cpu = smp_processor_id(); diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c index 4b48f251fd3..f44c3662436 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c @@ -30,7 +30,7 @@ asmlinkage void smp_thermal_interrupt(void) irq_exit(); } -static void __cpuinit intel_init_thermal(struct cpuinfo_x86 *c) +static void intel_init_thermal(struct cpuinfo_x86 *c) { u32 l, h; int tm2 = 0; @@ -84,7 +84,7 @@ static void __cpuinit intel_init_thermal(struct cpuinfo_x86 *c) return; } -void __cpuinit mce_intel_feature_init(struct cpuinfo_x86 *c) +void mce_intel_feature_init(struct cpuinfo_x86 *c) { intel_init_thermal(c); } -- cgit v1.2.3