From: Andrea Arcangeli <aarcange@redhat.com>
To: Pravin Shelar <pshelar@nicira.com>
Cc: Khalid Aziz <khalid.aziz@oracle.com>,
gregkh@linuxfoundation.org,
Ben Hutchings <bhutchings@solarflare.com>,
Christoph Lameter <cl@linux.com>,
hannes@cmpxchg.org, mel@csn.ul.ie, riel@redhat.com,
minchan@kernel.org, andi@firstfloor.org,
Andrew Morton <akpm@linux-foundation.org>,
torvalds@linux-foundation.org, linux-mm@kvack.org
Subject: Re: [PATCH] mm: hugetlbfs: fix hugetlbfs optimization
Date: Fri, 15 Nov 2013 18:00:57 +0100 [thread overview]
Message-ID: <20131115170057.GE23559@redhat.com> (raw)
In-Reply-To: <20131113161022.GI15985@redhat.com>
Hi,
so while optimizing away the _mapcount tail page refcounting for slab
and hugetlbfs pages incremental with the fix for the hugetlbfs
optimization I just sent, I also noticed another bug in the current
code (already fixed by the patch). gup_fast is still increasing
_mapcount for hugetlbfs, but it's not decreased in put_page.
It's only noticeable if you do:
echo 0 >/sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages
On current upstream I get:
BUG: Bad page state in process bash pfn:59a01
page:ffffea000139b038 count:0 mapcount:10 mapping: (null) index:0x0
page flags: 0x1c00000000008000(tail)
Modules linked in:
CPU: 6 PID: 2018 Comm: bash Not tainted 3.12.0+ #25
Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
0000000000000009 ffff880079cb5cc8 ffffffff81640e8b 0000000000000006
ffffea000139b038 ffff880079cb5ce8 ffffffff8115bb15 00000000000002c1
ffffea000139b038 ffff880079cb5d48 ffffffff8115bd83 ffff880079cb5de8
Call Trace:
[<ffffffff81640e8b>] dump_stack+0x55/0x76
[<ffffffff8115bb15>] bad_page+0xd5/0x130
[<ffffffff8115bd83>] free_pages_prepare+0x213/0x280
[<ffffffff8115df16>] __free_pages+0x36/0x80
[<ffffffff8119b011>] update_and_free_page+0xc1/0xd0
[<ffffffff8119b512>] free_pool_huge_page+0xc2/0xe0
[<ffffffff8119b8cc>] set_max_huge_pages.part.58+0x14c/0x220
[<ffffffff81308a8c>] ? _kstrtoull+0x2c/0x90
[<ffffffff8119ba70>] nr_hugepages_store_common.isra.60+0xd0/0xf0
[<ffffffff8119bac3>] nr_hugepages_store+0x13/0x20
[<ffffffff812f763f>] kobj_attr_store+0xf/0x20
[<ffffffff812354e9>] sysfs_write_file+0x189/0x1e0
[<ffffffff811baff5>] vfs_write+0xc5/0x1f0
[<ffffffff811bb505>] SyS_write+0x55/0xb0
[<ffffffff81651712>] system_call_fastpath+0x16/0x1b
So good thing I stopped the hugetlbfs optimization from going into
stable.
I'll send a v2 of this work as a patchset of 3 patches where the first
is the same identical patch I already sent but incremental to upstream
and it contains all the fixes needed including for the above
problem.
Patch 1/3 should be applied more urgently as it fixes all those
various bugs. The 2/3 and 3/3 can be deferred.
The patch 3/3 especially should be benchmarked in the usual 8GB/sec
setup before being applied, unless it makes a real difference I
wouldn't apply it because it tends to slowdown the THP case a bit and
it complicates things a bit more.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
prev parent reply other threads:[~2013-11-15 17:01 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-11-05 22:10 [PATCH] mm: hugetlbfs: fix hugetlbfs optimization Andrea Arcangeli
2013-11-07 17:18 ` Khalid Aziz
2013-11-12 19:22 ` Pravin Shelar
2013-11-13 16:10 ` Andrea Arcangeli
2013-11-15 17:00 ` Andrea Arcangeli [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20131115170057.GE23559@redhat.com \
--to=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=bhutchings@solarflare.com \
--cc=cl@linux.com \
--cc=gregkh@linuxfoundation.org \
--cc=hannes@cmpxchg.org \
--cc=khalid.aziz@oracle.com \
--cc=linux-mm@kvack.org \
--cc=mel@csn.ul.ie \
--cc=minchan@kernel.org \
--cc=pshelar@nicira.com \
--cc=riel@redhat.com \
--cc=torvalds@linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).