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 <[e-mail]>
Cc: Andrew Morton <[e-mail]>
Cc: Andy Lutomirski <[e-mail]>
Cc: Borislav Petkov <[e-mail]>
Cc: Borislav Petkov <[e-mail]>
Cc: Brian Gerst <[e-mail]>
Cc: Dave Hansen <[e-mail]>
Cc: Denys Vlasenko <[e-mail]>
Cc: Elliott, Robert (Persistent Memory) <[e-mail]>
Cc: H. Peter Anvin <[e-mail]>
Cc: Josh Poimboeuf <[e-mail]>
Cc: Linus Torvalds <[e-mail]>
Cc: Naoya Horiguchi <[e-mail]>
Cc: Peter Zijlstra <[e-mail]>
Cc: Thomas Gleixner <[e-mail]>
Cc: [e-mail]
Cc: [e-mail]
Link: http://lkml.kernel.org/r/[e-mail]
Signed-off-by: Ingo Molnar <[e-mail]>

arch/x86/include/asm/page_64.h

···
 
 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

arch/x86/kernel/cpu/mcheck/mce.c

···
 #include <asm/mce.h>
 #include <asm/msr.h>
 #include <asm/reboot.h>
+#include <asm/set_memory.h>
 
 #include "mce-internal.h"
 
···
 		pr_err("Memory error not recovered");
 	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

include/linux/mm_inline.h

···
 
 #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

mm/memory-failure.c

···
 		return 0;
 	}
 
+	arch_unmap_kpfn(pfn);
+
 	orig_head = hpage = compound_head(p);
 	num_poisoned_pages_inc();