>>> On 14.04.17 at 17:37, <boris.ostrov...@oracle.com> wrote:
> Instead of scrubbing pages while holding heap lock we can mark
> buddy's head as being scrubbed and drop the lock temporarily.
> If someone (most likely alloc_heap_pages()) tries to access
> this chunk it will signal the scrubber to abort scrub by setting
> head's PAGE_SCRUB_ABORT bit. The scrubber checks this bit after
> processing each page and stops its work as soon as it sees it.

Explanation sounds fine, but didn't you earlier indicate you think
yourself that the title is not really suitable (anymore)? I'm of
that opinion, at least, as pages are always available now, it's just
that the latency to get hold of the heap lock is higher before this
change than what we want it to be.

Looking at the patch itself is not overly useful as it seems without
the disposition of patch 5 being clear.

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Reply via email to