https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
Rick Macklem changed:
What|Removed |Added
Assignee|b...@freebsd.org|rmack...@freebsd.org
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
--- Comment #16 from commit-h...@freebsd.org ---
A commit references this bug:
Author: rmacklem
Date: Tue Aug 25 00:58:14 UTC 2020
New revision: 364744
URL: https://svnweb.freebsd.org/changeset/base/364744
Log:
Fix hangs with processes
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
--- Comment #15 from Ryan Libby ---
You can add me to the reviewed-by too. _NOFREE for
non-UMA_MD_SMALL_ALLOC should be safe, and no worse than before
r358097. Please reference that revision in the commit log for
archaeology.
--
You
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
--- Comment #14 from Rick Macklem ---
Well, I know nothing about the VM. However, if you think that putting
back the _NOFREE is safe to do for the non-UMA_MD_SMALL_ALLOC case,
I will retest the patch on the current head and then commit it
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
--- Comment #13 from Mark Johnston ---
(In reply to Rick Macklem from comment #12)
No, that revision is unrelated.
I'm not sure why the patch isn't working, I have to think about it some more.
If you wish to commit the original patch in
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
--- Comment #12 from Rick Macklem ---
It looks like the keg_drain patch is what was committed
to head as r364302?
If so, it does not fix the problem. A post-r364302 kernel
still hangs.
--
You are receiving this mail because:
You are the
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
Mark Johnston changed:
What|Removed |Added
Attachment #217196|0 |1
is obsolete|
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
--- Comment #10 from Mark Johnston ---
(In reply to Rick Macklem from comment #9)
My apologies, I see another bug: I'm decrementing by uk_ppera, not uk_ipers, so
the logic isn't correct and keg_drain() will free more items than it should.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
--- Comment #9 from Rick Macklem ---
The updated keg_drain patch hangs in the same way as an unpatched
kernel from head.
Lots of processes stuck in btalloc.
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
Mark Johnston changed:
What|Removed |Added
Attachment #217173|0 |1
is obsolete|
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
--- Comment #7 from Rick Macklem ---
I tested the keg_drain patch and got a panic.
The panic string appears to be just "page fault"?
It happens in:
zone_reclaim(0) at zone_reclaim+0x1ee/frame 0xe596a9c
Does this mean that the "struct
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
--- Comment #6 from Mark Johnston ---
Created attachment 217173
--> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=217173=edit
keg_drain patch
Here is a lightly-tested patch that will fix the hangs without adding _NOFREE
back,
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
Mark Johnston changed:
What|Removed |Added
CC||ma...@freebsd.org
--- Comment #5
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
--- Comment #4 from Rick Macklem ---
Just fyi, the email discussion that preceeded this PR is here:
http://docs.FreeBSD.org/cgi/mid.cgi?QB1PR01MB3364709D9D16CD0E550DAC7CDD6A0
--
You are receiving this mail because:
You are the assignee
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248008
Kubilay Kocak changed:
What|Removed |Added
Status|New |Open
Summary|Post
15 matches
Mail list logo
