linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Hugh Dickins <hughd@google.com>
To: Mel Gorman <mgorman@techsingularity.net>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	Vlastimil Babka <vbabka@suse.cz>, Li Wang <liwang@redhat.com>,
	Alex Shi <alex.shi@linux.alibaba.com>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: [PATCH] mm, page_alloc: capture page in task context only
Date: Wed, 10 Jun 2020 13:48:59 -0700 (PDT)	[thread overview]
Message-ID: <alpine.LSU.2.11.2006101342250.4607@eggly.anvils> (raw)

While stressing compaction, one run oopsed on NULL capc->cc in
__free_one_page()'s task_capc(zone): compact_zone_order() had been
interrupted, and a page was being freed in the return from interrupt.

Though you would not expect it from the source, both gccs I was using
(a 4.8.1 and a 7.5.0) had chosen to compile compact_zone_order() with
the ".cc = &cc" implemented by mov %rbx,-0xb0(%rbp) immediately before
callq compact_zone - long after the "current->capture_control = &capc".
An interrupt in between those finds capc->cc NULL (zeroed by an earlier
rep stos).

This could presumably be fixed by a barrier() before setting
current->capture_control in compact_zone_order(); but would also need
more care on return from compact_zone(), in order not to risk leaking
a page captured by interrupt just before capture_control is reset.

Maybe that is the preferable fix, but I felt safer for task_capc() to
exclude the rather surprising possibility of capture at interrupt time.

Fixes: 5e1f0f098b46 ("mm, compaction: capture a page under direct compaction")
Cc: stable@vger.kernel.org # 5.1+
Signed-off-by: Hugh Dickins <hughd@google.com>
---

 mm/page_alloc.c |    1 +
 1 file changed, 1 insertion(+)

--- 5.8-rc0/mm/page_alloc.c	2020-06-08 14:38:47.298625588 -0700
+++ linux/mm/page_alloc.c	2020-06-10 12:12:34.982950441 -0700
@@ -814,6 +814,7 @@ static inline struct capture_control *ta
 	struct capture_control *capc = current->capture_control;
 
 	return capc &&
+		in_task() &&
 		!(current->flags & PF_KTHREAD) &&
 		!capc->page &&
 		capc->cc->zone == zone &&

             reply	other threads:[~2020-06-10 20:49 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-10 20:48 Hugh Dickins [this message]
2020-06-11 15:43 ` Mel Gorman
2020-06-12 10:30 ` Vlastimil Babka
2020-06-15 21:03   ` Hugh Dickins
2020-06-16  7:45     ` Vlastimil Babka
2020-06-16  8:26       ` [PATCH 1/2] mm, compaction: make capture control handling safe wrt interrupts Vlastimil Babka
2020-06-16  8:26         ` [PATCH 2/2] mm, page_alloc: use unlikely() in task_capc() Vlastimil Babka
2020-06-16 20:29           ` Hugh Dickins
2020-06-17  9:55             ` Vlastimil Babka
2020-06-22  8:58               ` Mel Gorman
2020-06-16 20:18         ` [PATCH 1/2] mm, compaction: make capture control handling safe wrt interrupts Hugh Dickins

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=alpine.LSU.2.11.2006101342250.4607@eggly.anvils \
    --to=hughd@google.com \
    --cc=akpm@linux-foundation.org \
    --cc=alex.shi@linux.alibaba.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=liwang@redhat.com \
    --cc=mgorman@techsingularity.net \
    --cc=vbabka@suse.cz \
    --subject='Re: [PATCH] mm, page_alloc: capture page in task context only' \
    /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

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).