| To: | Dave Chinner <david@xxxxxxxxxxxxx> |
|---|---|
| Subject: | Re: [PATCH 0/2] xfs: regression fixes for 3.5-rc7 |
| From: | Ben Myers <bpm@xxxxxxx> |
| Date: | Wed, 11 Jul 2012 17:48:22 -0500 |
| Cc: | xfs@xxxxxxxxxxx |
| In-reply-to: | <1342042843-1773-1-git-send-email-david@xxxxxxxxxxxxx> |
| References: | <1342042843-1773-1-git-send-email-david@xxxxxxxxxxxxx> |
| User-agent: | Mutt/1.5.20 (2009-06-14) |
Hey Dave,
On Thu, Jul 12, 2012 at 07:40:41AM +1000, Dave Chinner wrote:
> These two patches need to go to Linus before 3.5 is released. The
> first fixes the btree cursor leak properly, and the second fixes a
> significant performance regression reported by Mel Gorman. Can you
> review them and send them onwards to Linus?
I have some test exposure on the first one, but not on the second. I'd rather
have a known performance regression in 3.5 than start popping stacks again this
late toward the end of the rc cycle. I suggest that #2 go in for the 3.6 merge
window and back to 3.5 via -stable. Any other opinions?
I'll feel better about if after some testing, so I'll get tests running asap.
What testing have you and Mel done?
IMO we should also consider "xfs: prevent recursion in xfs_buf_iorequest", from
Christoph.
Regards,
Ben
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | [PATCH 1/2] xfs: really fix the cursor leak in xfs_alloc_ag_vextent_near, Dave Chinner |
|---|---|
| Next by Date: | Re: [PATCH 0/2] xfs: regression fixes for 3.5-rc7, Dave Chinner |
| Previous by Thread: | Re: [PATCH 1/2] xfs: really fix the cursor leak in xfs_alloc_ag_vextent_near, Ben Myers |
| Next by Thread: | Re: [PATCH 0/2] xfs: regression fixes for 3.5-rc7, Dave Chinner |
| Indexes: | [Date] [Thread] [Top] [All Lists] |