Commit e6c495a9 authored Jul 03, 2013 by Vineet Gupta Committed by Linus Torvalds Jul 03, 2013

mm: fix the TLB range flushed when __tlb_remove_page() runs out of slots

zap_pte_range loops from @addr to @end.  In the middle, if it runs out of
batching slots, TLB entries needs to be flushed for @start to @interim,
NOT @interim to @end.

Since ARC port doesn't use page free batching I can't test it myself but
this seems like the right thing to do.

Observed this when working on a fix for the issue at thread:
http://www.spinics.net/lists/linux-arch/msg21736.html



Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

parent f60e2a96

mm/memory.c

+6 −3

Original line number	Original line	Diff line number	Diff line
	@@ -1101,6 +1101,7 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb,
	spinlock_t *ptl;		spinlock_t *ptl;
	pte_t *start_pte;		pte_t *start_pte;
	pte_t *pte;		pte_t *pte;
			unsigned long range_start = addr;

	again:		again:
	init_rss_vec(rss);		init_rss_vec(rss);
	@@ -1206,13 +1207,15 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb,
	force_flush = 0;		force_flush = 0;

	#ifdef HAVE_GENERIC_MMU_GATHER		#ifdef HAVE_GENERIC_MMU_GATHER
	tlb->start = addr;		tlb->start = range_start;
	tlb->end = end;		tlb->end = addr;
	#endif		#endif
	tlb_flush_mmu(tlb);		tlb_flush_mmu(tlb);
	if (addr != end)		if (addr != end) {
			range_start = addr;
	goto again;		goto again;
	}		}
			}

	return addr;		return addr;
	}		}