From: Will McVicker <willmcvicker@google.com>
To: Andrew Morton <akpm@linux-foundation.org>,
John Hubbard <jhubbard@nvidia.com>
Cc: kernel-team@android.com, Will McVicker <willmcvicker@google.com>,
Minchan Kim <minchan@google.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: [PATCH v1 1/1] mm/gup: skip pinnable check for refs==1
Date: Mon, 31 Jan 2022 20:35:04 +0000 [thread overview]
Message-ID: <20220131203504.3458775-1-willmcvicker@google.com> (raw)
This fixes commit 54d516b1d62f ("mm/gup: small refactoring: simplify
try_grab_page()") which refactors try_grab_page() to call
try_grab_compound_head() with refs=1. The refactor commit is causing
pin_user_pages() to return -ENOMEM when we try to pin one user page that
is migratable and not in the movable zone. Previously, try_grab_page()
didn't check if the page was pinnable for FOLL_PIN. To match the same
functionality, this fix adds the check `refs > 1 &&` to skip the call to
is_pinnable_page().
This issue is reproducible with the Pixel 6 on the 5.15 LTS kernel. Here
is the call stack to reproduce the -ENOMEM error:
Call trace:
: dump_backtrace.cfi_jt+0x0/0x8
: show_stack+0x1c/0x2c
: dump_stack_lvl+0x68/0x98
: try_grab_compound_head+0x298/0x3c4
: follow_page_pte+0x1dc/0x330
: follow_page_mask+0x174/0x340
: __get_user_pages+0x158/0x34c
: __gup_longterm_locked+0xfc/0x194
: __gup_longterm_unlocked+0x80/0xf4
: internal_get_user_pages_fast+0xf0/0x15c
: pin_user_pages_fast+0x24/0x40
: edgetpu_device_group_map+0x130/0x584 [abrolhos]
: edgetpu_ioctl_map_buffer+0x110/0x3b4 [abrolhos]
: edgetpu_ioctl+0x238/0x408 [abrolhos]
: edgetpu_fs_ioctl+0x14/0x24 [abrolhos]
Fixes: 54d516b1d62f ("mm/gup: small refactoring: simplify try_grab_page()")
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Minchan Kim <minchan@google.com>
Signed-off-by: Will McVicker <willmcvicker@google.com>
---
mm/gup.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/mm/gup.c b/mm/gup.c
index f0af462ac1e2..0509c49c46a3 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -135,7 +135,7 @@ struct page *try_grab_compound_head(struct page *page,
* right zone, so fail and let the caller fall back to the slow
* path.
*/
- if (unlikely((flags & FOLL_LONGTERM) &&
+ if (refs > 1 && unlikely((flags & FOLL_LONGTERM) &&
!is_pinnable_page(page)))
return NULL;
--
2.35.0.rc2.247.g8bbb082509-goog
next reply other threads:[~2022-01-31 20:35 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-01-31 20:35 Will McVicker [this message]
2022-01-31 20:49 ` John Hubbard
2022-01-31 21:37 ` Will McVicker
2022-02-01 7:28 ` Minchan Kim
2022-02-01 7:35 ` John Hubbard
2022-02-01 7:43 ` John Hubbard
2022-02-01 8:33 ` John Hubbard
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=20220131203504.3458775-1-willmcvicker@google.com \
--to=willmcvicker@google.com \
--cc=akpm@linux-foundation.org \
--cc=jhubbard@nvidia.com \
--cc=kernel-team@android.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=minchan@google.com \
/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