[ Upstream commit 471de20303dda0b67981e06d59cc6c4a83fd2a3c ]
We flush the data device cache before we issue external log IO. If the flush fails, we shut down the log immediately and return. However, the iclog->ic_sema is left in a decremented state so let's add an up(). Prior to this patch, xfs/438 would fail consistently when running with an external log device:
sync -> xfs_log_force -> xlog_write_iclog -> down(&iclog->ic_sema) -> blkdev_issue_flush (fail causes us to intiate shutdown) -> xlog_force_shutdown -> return
unmount -> xfs_log_umount -> xlog_wait_iclog_completion -> down(&iclog->ic_sema) --------> HANG
There is a second early return / shutdown. Make sure the up() happens for it as well. Also make sure we cleanup the iclog state, xlog_state_done_syncing, before dropping the iclog lock.
Fixes: b5d721eaae47 ("xfs: external logs need to flush data device") Fixes: 842a42d126b4 ("xfs: shutdown on failure to add page to log bio") Fixes: 7d839e325af2 ("xfs: check return codes when flushing block devices") Signed-off-by: Leah Rumancik leah.rumancik@gmail.com Reviewed-by: "Darrick J. Wong" djwong@kernel.org Signed-off-by: Chandan Babu R chandanbabu@kernel.org Signed-off-by: Leah Rumancik leah.rumancik@gmail.com --- fs/xfs/xfs_log.c | 23 ++++++++++++----------- 1 file changed, 12 insertions(+), 11 deletions(-)
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 59c982297503..ce6b303484cf 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1889,13 +1889,11 @@ xlog_write_iclog( * the log state machine to propagate I/O errors instead of * doing it here. We kick of the state machine and unlock * the buffer manually, the code needs to be kept in sync * with the I/O completion path. */ - xlog_state_done_syncing(iclog); - up(&iclog->ic_sema); - return; + goto sync; }
/* * We use REQ_SYNC | REQ_IDLE here to tell the block layer the are more * IOs coming immediately after this one. This prevents the block layer @@ -1921,44 +1919,47 @@ xlog_write_iclog( * writeback from the log succeeded. Repeating the flush is * not possible, hence we must shut down with log IO error to * avoid shutdown re-entering this path and erroring out again. */ if (log->l_targ != log->l_mp->m_ddev_targp && - blkdev_issue_flush(log->l_mp->m_ddev_targp->bt_bdev)) { - xlog_force_shutdown(log, SHUTDOWN_LOG_IO_ERROR); - return; - } + blkdev_issue_flush(log->l_mp->m_ddev_targp->bt_bdev)) + goto shutdown; } if (iclog->ic_flags & XLOG_ICL_NEED_FUA) iclog->ic_bio.bi_opf |= REQ_FUA;
iclog->ic_flags &= ~(XLOG_ICL_NEED_FLUSH | XLOG_ICL_NEED_FUA);
- if (xlog_map_iclog_data(&iclog->ic_bio, iclog->ic_data, count)) { - xlog_force_shutdown(log, SHUTDOWN_LOG_IO_ERROR); - return; - } + if (xlog_map_iclog_data(&iclog->ic_bio, iclog->ic_data, count)) + goto shutdown; + if (is_vmalloc_addr(iclog->ic_data)) flush_kernel_vmap_range(iclog->ic_data, count);
/* * If this log buffer would straddle the end of the log we will have * to split it up into two bios, so that we can continue at the start. */ if (bno + BTOBB(count) > log->l_logBBsize) { struct bio *split;
split = bio_split(&iclog->ic_bio, log->l_logBBsize - bno, GFP_NOIO, &fs_bio_set); bio_chain(split, &iclog->ic_bio); submit_bio(split);
/* restart at logical offset zero for the remainder */ iclog->ic_bio.bi_iter.bi_sector = log->l_logBBstart; }
submit_bio(&iclog->ic_bio); + return; +shutdown: + xlog_force_shutdown(log, SHUTDOWN_LOG_IO_ERROR); +sync: + xlog_state_done_syncing(iclog); + up(&iclog->ic_sema); }
/* * We need to bump cycle number for the part of the iclog that is * written to the start of the log. Watch out for the header magic