x86/mm/64: Stop using CR3.PCID == 0 in ASID-aware code

Putting the logical ASID into CR3's PCID bits directly means that we
have two cases to consider separately: ASID == 0 and ASID != 0.
This means that bugs that only hit in one of these cases trigger
nondeterministically.

There were some bugs like this in the past, and I think there's
still one in current kernels.  In particular, we have a number of
ASID-unware code paths that save CR3, write some special value, and
then restore CR3.  This includes suspend/resume, hibernate, kexec,
EFI, and maybe other things I've missed.  This is currently
dangerous: if ASID != 0, then this code sequence will leave garbage
in the TLB tagged for ASID 0.  We could potentially see corruption
when switching back to ASID 0.  In principle, an
initialize_tlbstate_and_flush() call after these sequences would
solve the problem, but EFI, at least, does not call this.  (And it
probably shouldn't -- initialize_tlbstate_and_flush() is rather
expensive.)

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bpetkov@suse.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/cdc14bbe5d3c3ef2a562be09a6368ffe9bd947a6.1505663533.git.luto@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>

1 files changed, 19 insertions, 2 deletions

diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h
index a999ba6..c120b5d 100644
--- a/arch/x86/include/asm/mmu_context.h
+++ b/arch/x86/include/asm/mmu_context.h
@@ -286,14 +286,31 @@ static inline bool arch_vma_access_permitted(struct vm_area_struct *vma,
 	return __pkru_allows_pkey(vma_pkey(vma), write);
 }
 
+/*
+ * If PCID is on, ASID-aware code paths put the ASID+1 into the PCID
+ * bits.  This serves two purposes.  It prevents a nasty situation in
+ * which PCID-unaware code saves CR3, loads some other value (with PCID
+ * == 0), and then restores CR3, thus corrupting the TLB for ASID 0 if
+ * the saved ASID was nonzero.  It also means that any bugs involving
+ * loading a PCID-enabled CR3 with CR4.PCIDE off will trigger
+ * deterministically.
+ */
+
 static inline unsigned long build_cr3(struct mm_struct *mm, u16 asid)
 {
-	return __sme_pa(mm->pgd) | asid;
+	if (static_cpu_has(X86_FEATURE_PCID)) {
+		VM_WARN_ON_ONCE(asid > 4094);
+		return __sme_pa(mm->pgd) | (asid + 1);
+	} else {
+		VM_WARN_ON_ONCE(asid != 0);
+		return __sme_pa(mm->pgd);
+	}
 }
 
 static inline unsigned long build_cr3_noflush(struct mm_struct *mm, u16 asid)
 {
-	return __sme_pa(mm->pgd) | asid | CR3_NOFLUSH;
+	VM_WARN_ON_ONCE(asid > 4094);
+	return __sme_pa(mm->pgd) | (asid + 1) | CR3_NOFLUSH;
 }
 
 /*