[PATCH 2.6.32 08/55] x86/mm: Improve switch_mm() barrier comments

[Willy Tarreau]

2.6.32-longterm review patch.  If anyone has any objections, please let me know.

------------------

From: Andy Lutomirski <luto@xxxxxxxxxx>

commit 4eaffdd5a5fe6ff9f95e1ab4de1ac904d5e0fa8b upstream.

My previous comments were still a bit confusing and there was a
typo. Fix it up.

Reported-by: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Signed-off-by: Andy Lutomirski <luto@xxxxxxxxxx>
Cc: Andy Lutomirski <luto@xxxxxxxxxxxxxx>
Cc: Borislav Petkov <bp@xxxxxxxxx>
Cc: Brian Gerst <brgerst@xxxxxxxxx>
Cc: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>
Cc: Denys Vlasenko <dvlasenk@xxxxxxxxxx>
Cc: H. Peter Anvin <hpa@xxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Rik van Riel <riel@xxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Fixes: 71b3c126e611 ("x86/mm: Add barriers and document switch_mm()-vs-flush synchronization")
Link: http://lkml.kernel.org/r/0a0b43cdcdd241c5faaaecfbcc91a155ddedc9a1.1452631609.git.luto@xxxxxxxxxx
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
Signed-off-by: Ben Hutchings <ben@xxxxxxxxxxxxxxx>
Signed-off-by: Willy Tarreau <w@xxxxxx>
---
 arch/x86/include/asm/mmu_context.h | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h
index b2c847a..a897c3b 100644
--- a/arch/x86/include/asm/mmu_context.h
+++ b/arch/x86/include/asm/mmu_context.h
@@ -58,14 +58,16 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 		 * be sent, and CPU 0's TLB will contain a stale entry.)
 		 *
 		 * The bad outcome can occur if either CPU's load is
-		 * reordered before that CPU's store, so both CPUs much
+		 * reordered before that CPU's store, so both CPUs must
 		 * execute full barriers to prevent this from happening.
 		 *
 		 * Thus, switch_mm needs a full barrier between the
 		 * store to mm_cpumask and any operation that could load
-		 * from next->pgd.  This barrier synchronizes with
-		 * remote TLB flushers.  Fortunately, load_cr3 is
-		 * serializing and thus acts as a full barrier.
+		 * from next->pgd.  TLB fills are special and can happen
+		 * due to instruction fetches or for no reason at all,
+		 * and neither LOCK nor MFENCE orders them.
+		 * Fortunately, load_cr3() is serializing and gives the
+		 * ordering guarantee we need.
 		 *
 		 */
 		load_cr3(next->pgd);
@@ -89,9 +91,8 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 			 * tlb flush IPI delivery. We must reload CR3
 			 * to make sure to use no freed page tables.
 			 *
-			 * As above, this is a barrier that forces
-			 * TLB repopulation to be ordered after the
-			 * store to mm_cpumask.
+			 * As above, load_cr3() is serializing and orders TLB
+			 * fills with respect to the mm_cpumask write.
 			 */
 			load_cr3(next->pgd);
 			load_LDT_nolock(&next->context);
-- 
1.7.12.2.21.g234cd45.dirty