Commit-ID: ce0fa3e56ad20f04d8252353dcd24e924abdafca Gitweb: http://git.kernel.org/tip/ce0fa3e56ad20f04d8252353dcd24e924abdafca Author: Tony Luck <tony.l...@intel.com> AuthorDate: Wed, 16 Aug 2017 10:18:03 -0700 Committer: Ingo Molnar <mi...@kernel.org> CommitDate: Thu, 17 Aug 2017 10:30:49 +0200
x86/mm, mm/hwpoison: Clear PRESENT bit for kernel 1:1 mappings of poison pages Speculative processor accesses may reference any memory that has a valid page table entry. While a speculative access won't generate a machine check, it will log the error in a machine check bank. That could cause escalation of a subsequent error since the overflow bit will be then set in the machine check bank status register. Code has to be double-plus-tricky to avoid mentioning the 1:1 virtual address of the page we want to map out otherwise we may trigger the very problem we are trying to avoid. We use a non-canonical address that passes through the usual Linux table walking code to get to the same "pte". Thanks to Dave Hansen for reviewing several iterations of this. Also see: http://marc.info/?l=linux-mm&m=149860136413338&w=2 Signed-off-by: Tony Luck <tony.l...@intel.com> Cc: Andrew Morton <a...@linux-foundation.org> Cc: Andy Lutomirski <l...@kernel.org> Cc: Borislav Petkov <b...@alien8.de> Cc: Borislav Petkov <b...@suse.de> Cc: Brian Gerst <brge...@gmail.com> Cc: Dave Hansen <dave.han...@intel.com> Cc: Denys Vlasenko <dvlas...@redhat.com> Cc: Elliott, Robert (Persistent Memory) <elli...@hpe.com> Cc: H. Peter Anvin <h...@zytor.com> Cc: Josh Poimboeuf <jpoim...@redhat.com> Cc: Linus Torvalds <torva...@linux-foundation.org> Cc: Naoya Horiguchi <n-horigu...@ah.jp.nec.com> Cc: Peter Zijlstra <pet...@infradead.org> Cc: Thomas Gleixner <t...@linutronix.de> Cc: linux...@kvack.org Cc: sta...@vger.kernel.org Link: http://lkml.kernel.org/r/20170816171803.28342-1-tony.l...@intel.com Signed-off-by: Ingo Molnar <mi...@kernel.org> --- arch/x86/include/asm/page_64.h | 4 ++++ arch/x86/kernel/cpu/mcheck/mce.c | 43 ++++++++++++++++++++++++++++++++++++++++ include/linux/mm_inline.h | 6 ++++++ mm/memory-failure.c | 2 ++ 4 files changed, 55 insertions(+) diff --git a/arch/x86/include/asm/page_64.h b/arch/x86/include/asm/page_64.h index b4a0d43..b50df06 100644 --- a/arch/x86/include/asm/page_64.h +++ b/arch/x86/include/asm/page_64.h @@ -51,6 +51,10 @@ static inline void clear_page(void *page) void copy_page(void *to, void *from); +#ifdef CONFIG_X86_MCE +#define arch_unmap_kpfn arch_unmap_kpfn +#endif + #endif /* !__ASSEMBLY__ */ #ifdef CONFIG_X86_VSYSCALL_EMULATION diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 6dde049..3b413065 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -51,6 +51,7 @@ #include <asm/mce.h> #include <asm/msr.h> #include <asm/reboot.h> +#include <asm/set_memory.h> #include "mce-internal.h" @@ -1051,6 +1052,48 @@ static int do_memory_failure(struct mce *m) return ret; } +#if defined(arch_unmap_kpfn) && defined(CONFIG_MEMORY_FAILURE) + +void arch_unmap_kpfn(unsigned long pfn) +{ + unsigned long decoy_addr; + + /* + * Unmap this page from the kernel 1:1 mappings to make sure + * we don't log more errors because of speculative access to + * the page. + * We would like to just call: + * set_memory_np((unsigned long)pfn_to_kaddr(pfn), 1); + * but doing that would radically increase the odds of a + * speculative access to the posion page because we'd have + * the virtual address of the kernel 1:1 mapping sitting + * around in registers. + * Instead we get tricky. We create a non-canonical address + * that looks just like the one we want, but has bit 63 flipped. + * This relies on set_memory_np() not checking whether we passed + * a legal address. + */ + +/* + * Build time check to see if we have a spare virtual bit. Don't want + * to leave this until run time because most developers don't have a + * system that can exercise this code path. This will only become a + * problem if/when we move beyond 5-level page tables. + * + * Hard code "9" here because cpp doesn't grok ilog2(PTRS_PER_PGD) + */ +#if PGDIR_SHIFT + 9 < 63 + decoy_addr = (pfn << PAGE_SHIFT) + (PAGE_OFFSET ^ BIT(63)); +#else +#error "no unused virtual bit available" +#endif + + if (set_memory_np(decoy_addr, 1)) + pr_warn("Could not invalidate pfn=0x%lx from 1:1 map\n", pfn); + +} +#endif + /* * The actual machine check handler. This only handles real * exceptions when something got corrupted coming in through int 18. diff --git a/include/linux/mm_inline.h b/include/linux/mm_inline.h index e030a68..25438b2 100644 --- a/include/linux/mm_inline.h +++ b/include/linux/mm_inline.h @@ -126,4 +126,10 @@ static __always_inline enum lru_list page_lru(struct page *page) #define lru_to_page(head) (list_entry((head)->prev, struct page, lru)) +#ifdef arch_unmap_kpfn +extern void arch_unmap_kpfn(unsigned long pfn); +#else +static __always_inline void arch_unmap_kpfn(unsigned long pfn) { } +#endif + #endif diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 1cd3b35..8836662 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -1146,6 +1146,8 @@ int memory_failure(unsigned long pfn, int trapno, int flags) return 0; } + arch_unmap_kpfn(pfn); + orig_head = hpage = compound_head(p); num_poisoned_pages_inc();
