On Thu, Nov 10, 2016 at 1:16 AM, Aaron Lu <aaron.lu@xxxxxxxxx> wrote:
> Prior to 3.15, there was a race between zap_pte_range() and
> page_mkclean() where writes to a page could be lost. Dave Hansen
> discovered by inspection that there is a similar race between
> move_ptes() and page_mkclean().
Ok, patch applied.
I'm not entirely happy with the force_flush vs need_flush games, and I
really think this code should be updated to use the same "struct
mmu_gather" that we use for the other TLB flushing cases (no need for
the page freeing batching, but the tlb_flush_mmu_tlbonly() logic
should be the same).
But I guess that's a bigger change, so that wouldn't be approriate for
rc5 or stable back-porting anyway. But it would be lovely if somebody
could look at that. Hint hint.
Hmm?
Linus
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>