Messages in this thread | | | Date | Thu, 19 Jul 2012 23:08:55 +0100 | From | Mel Gorman <> | Subject | Re: [PATCH 30/34] mm: vmscan: Do not force kswapd to scan small targets |
| |
On Thu, Jul 19, 2012 at 03:37:43PM -0500, Jonathan Nieder wrote: > Mel Gorman wrote: > > > commit ad2b8e601099a23dffffb53f91c18d874fe98854 upstream - WARNING: partial backport only > > > > Stable note: Not tracked in Bugzilla. This patch is very controversial for > > -stable. The upstream patch is addressing a completely different > > issue but accidentally contained an important fix. The workload > > in question was running memcached and then started IO in the > > background. memcached should stay resident but without this patch > > it gets swapped. Sometimes this manifests as a drop in throughput > > but mostly it was observed through /proc/vmstat. > > I had thought partial backports like this were pretty normal and ok, > actually, as long as they are well explained.
I was not aware of that and it's not something I had noticed for any other MM backports to -stable. I wanted to call it out instead of seeing if reviewers happened to spot the upstream commit did not resemble this patch :)
> Otherwise there would > be no way to fix bugs in stable in the above situation. >
If there is no problem with this then the note can change slightly
Stable note: Not tracked in Bugzilla. This is a partial backport of an upstream commit addressing a completely different issue that accidentally contained an important fix. The workload this patch helps was memcached when IO is started in the background. memcached should stay resident but without this patch it gets swapped. Sometimes this manifests as a drop in throughput but mostly it was observed through /proc/vmstat.
-- Mel Gorman SUSE Labs
| |