AcoSail/kernel_optimize_test
8
0
Fork 0
forked from luck/tmp_suning_uos_patched
Code Pull Requests Activity

gfs2: Disable page faults during lockless buffered reads

Browse Source 
[ Upstream commit 52f3f033a5dbd023307520af1ff551cadfd7f037 ]

During lockless buffered reads, filemap_read() holds page cache page
references while trying to copy data to the user-space buffer.  The
calling process isn't holding the inode glock, but the page references
it holds prevent those pages from being removed from the page cache, and
that prevents the underlying inode glock from being moved to another
node.  Thus, we can end up in the same kinds of distributed deadlock
situations as with normal (non-lockless) buffered reads.

Fix that by disabling page faults during lockless reads as well.

Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
This commit is contained in:
Andreas Gruenbacher 2022-03-14 18:32:02 +01:00 committed by Greg Kroah-Hartman
parent e803f12ea2
commit bab037ebbe
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
fs/gfs2/file.c
View File

@ -858,14 +858,16 @@ static ssize_t gfs2_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
return ret; return ret;
iocb->ki_flags &= ~IOCB_DIRECT; iocb->ki_flags &= ~IOCB_DIRECT;
} }
pagefault_disable();
iocb->ki_flags |= IOCB_NOIO; iocb->ki_flags |= IOCB_NOIO;
ret = generic_file_read_iter(iocb, to); ret = generic_file_read_iter(iocb, to);
iocb->ki_flags &= ~IOCB_NOIO; iocb->ki_flags &= ~IOCB_NOIO;
pagefault_enable();
if (ret >= 0) { if (ret >= 0) {
if (!iov_iter_count(to)) if (!iov_iter_count(to))
return ret; return ret;
written = ret; written = ret;
} else { } else if (ret != -EFAULT) {
if (ret != -EAGAIN) if (ret != -EAGAIN)
return ret; return ret;
if (iocb->ki_flags & IOCB_NOWAIT) if (iocb->ki_flags & IOCB_NOWAIT)