kernel/git/stable/linux.git/fs/gfs2, branch linux-5.8.y

gfs2: add some much needed cleanup for log flushes that fail

2020-09-09T17:14:09+00:00

[ Upstream commit 462582b99b6079a6fbcdfc65bac49f5c2a27cfff ] When a log flush fails due to io errors, it signals the failure but does not clean up after itself very well. This is because buffers are added to the transaction tr_buf and tr_databuf queue, but the io error causes gfs2_log_flush to bypass the "after_commit" functions responsible for dequeueing the bd elements. If the bd elements are added to the ail list before the error, function ail_drain takes care of dequeueing them. But if they haven't gotten that far, the elements are forgotten and make the transactions unable to be freed. This patch introduces new function trans_drain which drains the bd elements from the transaction so they can be freed properly. Signed-off-by: Bob Peterson Signed-off-by: Andreas Gruenbacher Signed-off-by: Sasha Levin

gfs2: Never call gfs2_block_zero_range with an open transaction

2020-08-21T11:15:02+00:00

commit 70499cdfeb3625c87eebe4f7a7ea06fa7447e5df upstream. Before this patch, some functions started transactions then they called gfs2_block_zero_range. However, gfs2_block_zero_range, like writes, can start transactions, which results in a recursive transaction error. For example: do_shrink trunc_start gfs2_trans_begin <------------------------------------------------ gfs2_block_zero_range iomap_zero_range(inode, from, length, NULL, &gfs2_iomap_ops); iomap_apply ... iomap_zero_range_actor iomap_begin gfs2_iomap_begin gfs2_iomap_begin_write actor (iomap_zero_range_actor) iomap_zero iomap_write_begin gfs2_iomap_page_prepare gfs2_trans_begin <------------------------ This patch reorders the callers of gfs2_block_zero_range so that they only start their transactions after the call. It also adds a BUG_ON to ensure this doesn't happen again. Fixes: 2257e468a63b ("gfs2: implement gfs2_block_zero_range using iomap_zero_range") Cc: stable@vger.kernel.org # v5.5+ Signed-off-by: Bob Peterson Signed-off-by: Andreas Gruenbacher Signed-off-by: Greg Kroah-Hartman

gfs2: Fix refcount leak in gfs2_glock_poke

2020-08-21T11:15:02+00:00

commit c07bfb4d8fa1ee11c6d18b093d0bb6c8832d3626 upstream. In gfs2_glock_poke, make sure gfs2_holder_uninit is called on the local glock holder. Without that, we're leaking a glock and a pid reference. Fixes: 9e8990dea926 ("gfs2: Smarter iopen glock waiting") Cc: stable@vger.kernel.org # v5.8+ Signed-off-by: Andreas Gruenbacher Signed-off-by: Greg Kroah-Hartman

gfs2: Rework read and page fault locking

2020-07-07T21:40:12+00:00

So far, gfs2 has taken the inode glocks inside the ->readpage and ->readahead address space operations. Since commit d4388340ae0b ("fs: convert mpage_readpages to mpage_readahead"), gfs2_readahead is passed the pages to read ahead locked. With that, the current holder of the inode glock may be trying to lock one of those pages while gfs2_readahead is trying to take the inode glock, resulting in a deadlock. Fix that by moving the lock taking to the higher-level ->read_iter file and ->fault vm operations. This also gets rid of an ugly lock inversion workaround in gfs2_readpage. The cache consistency model of filesystems like gfs2 is such that if data is found in the page cache, the data is up to date and can be used without taking any filesystem locks. If a page is not cached, filesystem locks must be taken before populating the page cache. To avoid taking the inode glock when the data is already cached, gfs2_file_read_iter first tries to read the data with the IOCB_NOIO flag set. If that fails, the inode glock is taken and the operation is retried with the IOCB_NOIO flag cleared. Signed-off-by: Andreas Gruenbacher

gfs2: The freeze glock should never be frozen

2020-07-03T10:05:35+00:00

Before this patch, some gfs2 code locked the freeze glock with LM_FLAG_NOEXP (Do not freeze) flag, and some did not. We never want to freeze the freeze glock, so this patch makes it consistently use LM_FLAG_NOEXP always. Signed-off-by: Bob Peterson

gfs2: When freezing gfs2, use GL_EXACT and not GL_NOCACHE

2020-07-03T10:05:35+00:00

Before this patch, the freeze code in gfs2 specified GL_NOCACHE in several places. That's wrong because we always want to know the state of whether the file system is frozen. There was also a problem with freeze/thaw transitioning the glock from frozen (EX) to thawed (SH) because gfs2 will normally grant glocks in EX to processes that request it in SH mode, unless GL_EXACT is specified. Therefore, the freeze/thaw code, which tried to reacquire the glock in SH mode would get the glock in EX mode, and miss the transition from EX to SH. That made it think the thaw had completed normally, but since the glock was still cached in EX, other nodes could not freeze again. This patch removes the GL_NOCACHE flag to allow the freeze glock to be cached. It also adds the GL_EXACT flag so the glock is fully transitioned from EX to SH, thereby allowing future freeze operations. Signed-off-by: Bob Peterson

gfs2: read-only mounts should grab the sd_freeze_gl glock

2020-07-03T10:05:35+00:00

Before this patch, only read-write mounts would grab the freeze glock in read-only mode, as part of gfs2_make_fs_rw. So the freeze glock was never initialized. That meant requests to freeze, which request the glock in EX, were granted without any state transition. That meant you could mount a gfs2 file system, which is currently frozen on a different cluster node, in read-only mode. This patch makes read-only mounts lock the freeze glock in SH mode, which will block for file systems that are frozen on another node. Signed-off-by: Bob Peterson

gfs2: freeze should work on read-only mounts

2020-07-03T10:05:35+00:00

Before this patch, function freeze_go_sync, called when promoting the freeze glock, was testing for the SDF_JOURNAL_LIVE superblock flag. That's only set for read-write mounts. Read-only mounts don't use a journal, so the bit is never set, so the freeze never happened. This patch removes the check for SDF_JOURNAL_LIVE for freeze requests but still checks it when deciding whether to flush a journal. Signed-off-by: Bob Peterson

gfs2: eliminate GIF_ORDERED in favor of list_empty

2020-07-03T10:05:34+00:00

In several places, we used the GIF_ORDERED inode flag to determine if an inode was on the ordered writes list. However, since we always held the sd_ordered_lock spin_lock during the manipulation, we can just as easily check list_empty(&ip->i_ordered) instead. This allows us to keep more than one ordered writes list to make journal writing improvements. This patch eliminates GIF_ORDERED in favor of checking list_empty. Signed-off-by: Bob Peterson

gfs2: Don't sleep during glock hash walk

2020-06-30T11:04:45+00:00

In flush_delete_work, instead of flushing each individual pending delayed work item, cancel and re-queue them for immediate execution. The waiting isn't needed here because we're already waiting for all queued work items to complete in gfs2_flush_delete_work. This makes the code more efficient, but more importantly, it avoids sleeping during a rhashtable walk, inside rcu_read_lock(). Signed-off-by: Andreas Gruenbacher