ENOSPC delalloc flushing fixes

v2->v3:
- Reordered the patches to have "btrfs: wake up async_delalloc_pages waiters after submit" come first, as it's a general fix that we need.
- Fixed the nit that Nikolay pointed out in "btrfs: handle shrink_delalloc pages calculation differently".
- Added "btrfs: include delalloc related info in dump space info tracepoint" to include the delalloc_bytes and ordered_bytes in the dump_space_info tracepoint.

v1->v2:
- Two extra patches to remove the last user of sync_inode() (p9fs) and then remove sync_inode() itself, as per hch's request

--- Original email ---

Hello,

I've been debugging and fixing a problem we hit in production related to getting ENOSPC when we still should have had space to use. A variety of different things were going wrong, but one of them was sometimes we wouldn't wait for all of delalloc to be flushed. This series of patches fixes a few problems in this area, starting with

btrfs: handle shrink_delalloc pages calculation differently

When we switched to writing pages instead of full inodes for flushing we didn't adjust the counters to give us pages, instead using the "items" amount. This is incorrect as we'd just not flush that much delalloc, leaving a lot laying around when we finally ENOSPC'd.

The next bit are related to compression, as we have compression on everywhere in production.

btrfs: wait on async extents when flushing delalloc
btrfs: wake up async_delalloc_pages waiters after submit

I ripped this code out because I added a second sync_inode() if we had async extents in order to do the proper waiting. However sync_inode() could skip writeout if writeout had begun already on the inode, so we still need this waiting in order to make sure we don't try to wait on ordered extents until all ordered extents have been created.

And finally these two patches

fs: add a filemap_fdatawrite_wbc helper
btrfs: use the filemap_fdatawrite_wbc helper for delalloc shrinking

We need a writeback helper that will take a wbc and not try to do anything fancy other than write out the inode we want. sync_inode() has the drawback that it will skip writeout if the inode is currently under writeback, and thus we won't wait properly.

I ran this series through fsperf, the results ar eposted below. I'm still printing the %diff, but I'm also printing the stdev so you can see the variance in teh values we expect. Generally there is no change or it's within the normal range. These patches really only affect anything when we're very full on space.

I also ran this through the enospc stress test and saw no early enospc. Thanks,

Josef

dbench60 results
metric baseline current stdev diff
==============================================================
qpathinfo 11.26 11.87 0.64 5.35%

throughput     652.90     628.99      65.68    -3.66%
flush        22292.41   35005.49   15197.94    57.03%
qfileinfo        1.04       1.24       0.17    19.02%
ntcreatex     3965.31    6167.38    7790.74    55.53%
qfsinfo          1.79       1.41       0.40   -20.88%
close            1.81       1.95       0.39     7.67%
sfileinfo        4.96       5.50       1.13    10.70%

rename 2640.98 5844.86 5873.86 121.31%
find 12.29 12.83 1.00 4.36%
unlink 3310.42 4809.88 7179.77 45.30%
writex 12521.22 37992.79 7567.72 203.43%
deltree 409.86 363.46 224.63 -11.32%

readx            2.37       4.18       0.66    76.01%
mkdir            0.03       0.11       0.01   271.06%
lockx            0.44       0.17       0.27   -61.83%
unlockx          0.16       0.64       0.11   291.70%

emptyfiles500k results
metric baseline current stdev diff
======================================================================

write_io_kbytes       125000     125000            0    0.00%
read_clat_ns_p99           0          0            0    0.00%
write_bw_bytes      1.83e+08   1.77e+08   5516543.33   -3.19%
read_iops                  0          0            0    0.00%
write_clat_ns_p50      17536      17792       273.68    1.46%
read_io_kbytes             0          0            0    0.00%
read_io_bytes              0          0            0    0.00%
write_clat_ns_p99      72576      74240      2677.97    2.29%
read_bw_bytes              0          0            0    0.00%
elapsed                    1          1            0    0.00%
write_lat_ns_min           0          0            0    0.00%
sys_cpu                91.86      91.29         0.66   -0.63%
write_lat_ns_max           0          0            0    0.00%
read_lat_ns_min            0          0            0    0.00%
write_iops          44583.10   43162.98      1346.81   -3.19%
read_lat_ns_max            0          0            0    0.00%
read_clat_ns_p50           0          0            0    0.00%

smallfiles100k results
metric baseline current stdev diff
========================================================================

write_io_kbytes     2.04e+08   2.04e+08             0     0.00%
read_clat_ns_p99           0          0             0     0.00%
write_bw_bytes      1.33e+08   1.40e+08   15033768.39     5.20%
read_iops                  0          0             0     0.00%
write_clat_ns_p50       6424       6688         79.77     4.11%
read_io_kbytes             0          0             0     0.00%
read_io_bytes              0          0             0     0.00%
write_clat_ns_p99      15960      16320        251.24     2.26%
read_bw_bytes              0          0             0     0.00%
elapsed              1592.50       1492        234.98    -6.31%
write_lat_ns_min     2730.75       2768         45.73     1.36%
sys_cpu                 5.75       6.25          0.65     8.68%
write_lat_ns_max    5.17e+08   1.52e+08      1.05e+09   -70.54%
read_lat_ns_min            0          0             0     0.00%
write_iops          32545.54   34239.34       3670.35     5.20%
read_lat_ns_max            0          0             0     0.00%
read_clat_ns_p50           0          0             0     0.00%

bufferedrandwrite16g results
metric baseline current stdev diff
===========================================================================

write_io_kbytes        16777216   16777216             0     0.00%
read_clat_ns_p99              0          0             0     0.00%
write_bw_bytes      91729211.62   94071836   13859731.84     2.55%
read_iops                     0          0             0     0.00%
write_clat_ns_p50         12704      11840       1227.26    -6.80%
read_io_kbytes                0          0             0     0.00%
read_io_bytes                 0          0             0     0.00%
write_clat_ns_p99         31136      32384       2925.86     4.01%
read_bw_bytes                 0          0             0     0.00%

elapsed 191.38 183 25.98 -4.38%
write_lat_ns_min 3961.25 4068 95.29 2.69%
sys_cpu 31.83 29.77 6.85 -6.47%
write_lat_ns_max 3.05e+10 1.70e+10 1.95e+10 -44.34%

read_lat_ns_min               0          0             0     0.00%
write_iops             22394.83   22966.76       3383.72     2.55%
read_lat_ns_max               0          0             0     0.00%
read_clat_ns_p50              0          0             0     0.00%

dio4kbs16threads results
metric baseline current stdev diff
===========================================================================
write_io_kbytes 4187092 4949612 599817.04 18.21%
read_clat_ns_p99 0 0 0 0.00%
write_bw_bytes 71448568.62 84467746 10233293.68 18.22%
read_iops 0 0 0 0.00%
write_clat_ns_p50 245504 240640 4370.25 -1.98%

read_io_kbytes                0          0             0     0.00%
read_io_bytes                 0          0             0     0.00%
write_clat_ns_p99      22249472   20054016    1120427.41    -9.87%
read_bw_bytes                 0          0             0     0.00%
elapsed                      61         61             0     0.00%
write_lat_ns_min          38440      38571        225.10     0.34%

sys_cpu 3.89 4.57 0.46 17.47%
write_lat_ns_max 1.23e+09 9.48e+08 7.00e+08 -23.22%
read_lat_ns_min 0 0 0 0.00%
write_iops 17443.50 20622.01 2498.36 18.22%

read_lat_ns_max               0          0             0     0.00%
read_clat_ns_p50              0          0             0     0.00%

randwrite2xram results
metric baseline current stdev diff
========================================================================

write_io_kbytes     33948247   34359528    4793805.63     1.21%
read_clat_ns_p99           0          0             0     0.00%
write_bw_bytes      1.15e+08   1.17e+08   17455355.14     1.47%
read_iops                  0          0             0     0.00%
write_clat_ns_p50      15232      14400       1043.81    -5.46%
read_io_kbytes             0          0             0     0.00%
read_io_bytes              0          0             0     0.00%
write_clat_ns_p99      95264      67072      25379.05   -29.59%
read_bw_bytes              0          0             0     0.00%
elapsed               313.62        314          5.10     0.12%
write_lat_ns_min     5399.38       5658        126.19     4.79%
sys_cpu                11.61      11.06          2.04    -4.69%
write_lat_ns_max    3.16e+10   2.91e+10      1.46e+10    -8.05%
read_lat_ns_min            0          0             0     0.00%
write_iops          28099.86   28511.67       4261.56     1.47%
read_lat_ns_max            0          0             0     0.00%
read_clat_ns_p50           0          0             0     0.00%

untarfirefox results
metric baseline current stdev diff
======================================================
elapsed 46.89 46.75 0.18 -0.29%

Josef Bacik (9):
btrfs: wake up async_delalloc_pages waiters after submit
btrfs: include delalloc related info in dump space info tracepoint
btrfs: enable a tracepoint when we fail tickets
btrfs: handle shrink_delalloc pages calculation differently
btrfs: wait on async extents when flushing delalloc
fs: add a filemap_fdatawrite_wbc helper
btrfs: use the filemap_fdatawrite_wbc helper for delalloc shrinking 9p: migrate from sync_inode to filemap_fdatawrite_wbc
fs: kill sync_inode

 fs/9p/vfs_file.c             |  7 +--
 fs/btrfs/ctree.h             |  9 ++--
 fs/btrfs/inode.c             | 16 +++----
 fs/btrfs/space-info.c        | 82 ++++++++++++++++++++++++++++++------
 fs/fs-writeback.c            | 19 +--------
 include/linux/fs.h           |  3 +-
 include/trace/events/btrfs.h | 21 +++++++--
 mm/filemap.c                 | 35 +++++++++++----
 8 files changed, 128 insertions(+), 64 deletions(-)

2.26.3

We use the async_delalloc_pages mechanism to make sure that we've completed our async work before trying to continue our delalloc flushing. The reason for this is we need to see any ordered extents that were created by our delalloc flushing. However we're waking up before we do the submit work, which is before we create the ordered extents. This is a pretty wide race window where we could potentially think there are no ordered extents and thus exit shrink_delalloc prematurely. Fix this by waking us up after we've done the work to create ordered extents.

---
 fs/btrfs/inode.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

index e6eb20987351..0f42864d5dd4 100644
@@ -1290,11 +1290,6 @@ static noinline void async_cow_submit(struct btrfs_work *work)
 	nr_pages = (async_chunk->end - async_chunk->start + PAGE_SIZE) >>
 		PAGE_SHIFT;
 
-	/* atomic_sub_return implies a barrier */
-	if (atomic_sub_return(nr_pages, &fs_info->async_delalloc_pages) <
-	    5 * SZ_1M)
-		cond_wake_up_nomb(&fs_info->async_submit_wait);
-
 	/*
 	 * ->inode could be NULL if async_chunk_start has failed to compress,
 	 * in which case we don't have anything to submit, yet we need to

@@ -1303,6 +1298,11 @@ static noinline void async_cow_submit(struct btrfs_work *work)
 	 */
 	if (async_chunk->inode)
 		submit_compressed_extents(async_chunk);
+
+	/* atomic_sub_return implies a barrier */
+	if (atomic_sub_return(nr_pages, &fs_info->async_delalloc_pages) <
+	    5 * SZ_1M)
+		cond_wake_up_nomb(&fs_info->async_submit_wait);
 }
 
 static noinline void async_cow_free(struct btrfs_work *work)

2.26.3

In order to debug delalloc flushing issues I added delalloc_bytes and ordered_bytes to this tracepoint to see if they were non-zero when we were ENOSPC'ing. This was valuable for me and showed me cases where we weren't waiting on ordered extents properly. In order to add this to the tracepoint we need to take away the const modifier for fs_info, as percpu_sum_counter_positive() will change the counter when it adds up the percpu buckets. This is needed to make sure we're getting accurate information at these tracepoints, as the wrong information could send us down the wrong path when debugging problems.

---
 include/trace/events/btrfs.h | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

index c7237317a8b9..a87953e74573 100644
@@ -2037,7 +2037,7 @@ TRACE_EVENT(btrfs_convert_extent_bit,
 );
 
 DECLARE_EVENT_CLASS(btrfs_dump_space_info,
-	TP_PROTO(const struct btrfs_fs_info *fs_info,
+	TP_PROTO(struct btrfs_fs_info *fs_info,
 		 const struct btrfs_space_info *sinfo),
 
 	TP_ARGS(fs_info, sinfo),

@@ -2057,6 +2057,8 @@ DECLARE_EVENT_CLASS(btrfs_dump_space_info,
 		__field(	u64,	delayed_refs_reserved	)
 		__field(	u64,	delayed_reserved	)
 		__field(	u64,	free_chunk_space	)
+		__field(	u64,	delalloc_bytes		)
+		__field(	u64,	ordered_bytes		)
 	),
 
 	TP_fast_assign_btrfs(fs_info,

@@ -2074,6 +2076,8 @@ DECLARE_EVENT_CLASS(btrfs_dump_space_info,
 		__entry->delayed_refs_reserved	=	fs_info->delayed_refs_rsv.reserved;
 		__entry->delayed_reserved	=	fs_info->delayed_block_rsv.reserved;
 		__entry->free_chunk_space	=	atomic64_read(&fs_info->free_chunk_space);
+		__entry->delalloc_bytes		=	percpu_counter_sum_positive(&fs_info->delalloc_bytes);
+		__entry->ordered_bytes		=	percpu_counter_sum_positive(&fs_info->ordered_bytes);
 	),
 
 	TP_printk_btrfs("flags=%s total_bytes=%llu bytes_used=%llu "

@@ -2081,7 +2085,8 @@ DECLARE_EVENT_CLASS(btrfs_dump_space_info,
 			"bytes_may_use=%llu bytes_readonly=%llu "
 			"reclaim_size=%llu clamp=%d global_reserved=%llu "
 			"trans_reserved=%llu delayed_refs_reserved=%llu "
-			"delayed_reserved=%llu chunk_free_space=%llu",
+			"delayed_reserved=%llu chunk_free_space=%llu "
+			"delalloc_bytes=%llu ordered_bytes=%llu",
 			__print_flags(__entry->flags, "|", BTRFS_GROUP_FLAGS),
 			__entry->total_bytes, __entry->bytes_used,
 			__entry->bytes_pinned, __entry->bytes_reserved,

@@ -2089,11 +2094,12 @@ DECLARE_EVENT_CLASS(btrfs_dump_space_info,
 			__entry->reclaim_size, __entry->clamp,
 			__entry->global_reserved, __entry->trans_reserved,
 			__entry->delayed_refs_reserved,
-			__entry->delayed_reserved, __entry->free_chunk_space)
+			__entry->delayed_reserved, __entry->free_chunk_space,
+			__entry->delalloc_bytes, __entry->ordered_bytes)
 );
 
 DEFINE_EVENT(btrfs_dump_space_info, btrfs_done_preemptive_reclaim,
-	TP_PROTO(const struct btrfs_fs_info *fs_info,
+	TP_PROTO(struct btrfs_fs_info *fs_info,
 		 const struct btrfs_space_info *sinfo),
 	TP_ARGS(fs_info, sinfo)
 );

2.26.3

When debugging early enospc problems it was useful to have a tracepoint where we failed all tickets so I could check the state of the enospc counters at failure time to validate my fixes. This adds the tracpoint so you can easily get that information.

---
 fs/btrfs/space-info.c        | 2 ++
 include/trace/events/btrfs.h | 6 ++++++
 2 files changed, 8 insertions(+)

index 392997376a1c..af161eb808a2 100644
@@ -825,6 +825,8 @@ static bool maybe_fail_all_tickets(struct btrfs_fs_info *fs_info,
 	struct reserve_ticket *ticket;
 	u64 tickets_id = space_info->tickets_id;
 
+	trace_btrfs_fail_all_tickets(fs_info, space_info);
+
 	if (btrfs_test_opt(fs_info, ENOSPC_DEBUG)) {
 		btrfs_info(fs_info, "cannot satisfy tickets, dumping space info");
 		__btrfs_dump_space_info(fs_info, space_info);

index a87953e74573..d754c6ec1797 100644
@@ -2104,6 +2104,12 @@ DEFINE_EVENT(btrfs_dump_space_info, btrfs_done_preemptive_reclaim,
 	TP_ARGS(fs_info, sinfo)
 );
 
+DEFINE_EVENT(btrfs_dump_space_info, btrfs_fail_all_tickets,
+	TP_PROTO(struct btrfs_fs_info *fs_info,
+		 const struct btrfs_space_info *sinfo),
+	TP_ARGS(fs_info, sinfo)
+);
+
 TRACE_EVENT(btrfs_reserve_ticket,
 	TP_PROTO(const struct btrfs_fs_info *fs_info, u64 flags, u64 bytes,
 		 u64 start_ns, int flush, int error),

2.26.3

We have been hitting some early ENOSPC issues in production with more recent kernels, and I tracked it down to us simply not flushing delalloc as aggressively as we should be. With tracing I was seeing us failing all tickets with all of the block rsvs at or around 0, with very little pinned space, but still around 120MiB of outstanding bytes_may_used. Upon further investigation I saw that we were flushing around 14 pages per shrink call for delalloc, despite having around 2GiB of delalloc outstanding.

Consider the example of a 8 way machine, all CPUs trying to create a file in parallel, which at the time of this commit requires 5 items to do. Assuming a 16k leaf size, we have 10MiB of total metadata reclaim size waiting on reservations. Now assume we have 128MiB of delalloc outstanding. With our current math we would set items to 20, and then set to_reclaim to 20 * 256k, or 5MiB.

Assuming that we went through this loop all 3 times, for both FLUSH_DELALLOC and FLUSH_DELALLOC_WAIT, and then did the full loop twice, we'd only flush 60MiB of the 128MiB delalloc space. This could leave a fair bit of delalloc reservations still hanging around by the time we go to ENOSPC out all the remaining tickets.

Fix this two ways. First, change the calculations to be a fraction of the total delalloc bytes on the system. Prior to this change we were calculating based on dirty inodes so our math made more sense, now it's just completely unrelated to what we're actually doing.

Second add a FLUSH_DELALLOC_FULL state, that we hold off until we've gone through the flush states at least once. This will empty the system of all delalloc so we're sure to be truly out of space when we start failing tickets.

I'm tagging stable 5.10 and forward, because this is where we started using the page stuff heavily again. This affects earlier kernel versions as well, but would be a pain to backport to them as the flushing mechanisms aren't the same.

---
 fs/btrfs/ctree.h             |  9 ++++----
 fs/btrfs/space-info.c        | 40 +++++++++++++++++++++++++-----------
 include/trace/events/btrfs.h |  1 +
 3 files changed, 34 insertions(+), 16 deletions(-)

index d7ef4d7d2c1a..232ff1a49ca6 100644
@@ -2783,10 +2783,11 @@ enum btrfs_flush_state {
 	FLUSH_DELAYED_REFS	=	4,
 	FLUSH_DELALLOC		=	5,
 	FLUSH_DELALLOC_WAIT	=	6,
-	ALLOC_CHUNK		=	7,
-	ALLOC_CHUNK_FORCE	=	8,
-	RUN_DELAYED_IPUTS	=	9,
-	COMMIT_TRANS		=	10,
+	FLUSH_DELALLOC_FULL	=	7,
+	ALLOC_CHUNK		=	8,
+	ALLOC_CHUNK_FORCE	=	9,
+	RUN_DELAYED_IPUTS	=	10,
+	COMMIT_TRANS		=	11,
 };
 
 int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,

index af161eb808a2..1a7e44dd0ba7 100644
@@ -494,6 +494,11 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info,
 	long time_left;
 	int loops;
 
+	delalloc_bytes = percpu_counter_sum_positive(&fs_info->delalloc_bytes);
+	ordered_bytes = percpu_counter_sum_positive(&fs_info->ordered_bytes);
+	if (delalloc_bytes == 0 && ordered_bytes == 0)
+		return;
+
 	/* Calc the number of the pages we need flush for space reservation */
 	if (to_reclaim == U64_MAX) {
 		items = U64_MAX;

@@ -501,22 +506,21 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info,
 		/*
 		 * to_reclaim is set to however much metadata we need to
 		 * reclaim, but reclaiming that much data doesn't really track
-		 * exactly, so increase the amount to reclaim by 2x in order to
-		 * make sure we're flushing enough delalloc to hopefully reclaim
-		 * some metadata reservations.
+		 * exactly.  What we really want to do is reclaim full inode's
+		 * worth of reservations, however that's not available to us
+		 * here.  We will take a fraction of the delalloc bytes for our
+		 * flushing loops and hope for the best.  Delalloc will expand
+		 * the amount we write to cover an entire dirty extent, which
+		 * will reclaim the metadata reservation for that range.  If
+		 * it's not enough subsequent flush stages will be more
+		 * aggressive.
 		 */
+		to_reclaim = max(to_reclaim, delalloc_bytes >> 3);
 		items = calc_reclaim_items_nr(fs_info, to_reclaim) * 2;
-		to_reclaim = items * EXTENT_SIZE_PER_ITEM;
 	}
 
 	trans = (struct btrfs_trans_handle *)current->journal_info;
 
-	delalloc_bytes = percpu_counter_sum_positive(
-						&fs_info->delalloc_bytes);
-	ordered_bytes = percpu_counter_sum_positive(&fs_info->ordered_bytes);
-	if (delalloc_bytes == 0 && ordered_bytes == 0)
-		return;
-
 	/*
 	 * If we are doing more ordered than delalloc we need to just wait on
 	 * ordered extents, otherwise we'll waste time trying to flush delalloc

@@ -596,8 +600,11 @@ static void flush_space(struct btrfs_fs_info *fs_info,
 		break;
 	case FLUSH_DELALLOC:
 	case FLUSH_DELALLOC_WAIT:
+	case FLUSH_DELALLOC_FULL:
+		if (state == FLUSH_DELALLOC_FULL)
+			num_bytes = U64_MAX;
 		shrink_delalloc(fs_info, space_info, num_bytes,
-				state == FLUSH_DELALLOC_WAIT, for_preempt);
+				state != FLUSH_DELALLOC, for_preempt);
 		break;
 	case FLUSH_DELAYED_REFS_NR:
 	case FLUSH_DELAYED_REFS:

@@ -907,6 +914,14 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work)
 				commit_cycles--;
 		}
 
+		/*
+		 * We do not want to empty the system of delalloc unless we're
+		 * under heavy pressure, so allow one trip through the flushing
+		 * logic before we start doing a FLUSH_DELALLOC_FULL.
+		 */
+		if (flush_state == FLUSH_DELALLOC_FULL && !commit_cycles)
+			flush_state++;
+
 		/*
 		 * We don't want to force a chunk allocation until we've tried
 		 * pretty hard to reclaim space.  Think of the case where we

@@ -1070,7 +1085,7 @@ static void btrfs_preempt_reclaim_metadata_space(struct work_struct *work)
  *   so if we now have space to allocate do the force chunk allocation.
  */
 static const enum btrfs_flush_state data_flush_states[] = {
-	FLUSH_DELALLOC_WAIT,
+	FLUSH_DELALLOC_FULL,
 	RUN_DELAYED_IPUTS,
 	COMMIT_TRANS,
 	ALLOC_CHUNK_FORCE,

@@ -1159,6 +1174,7 @@ static const enum btrfs_flush_state evict_flush_states[] = {
 	FLUSH_DELAYED_REFS,
 	FLUSH_DELALLOC,
 	FLUSH_DELALLOC_WAIT,
+	FLUSH_DELALLOC_FULL,
 	ALLOC_CHUNK,
 	COMMIT_TRANS,
 };

index d754c6ec1797..e21744a3cdb6 100644
@@ -94,6 +94,7 @@ struct btrfs_space_info;
 	EM( FLUSH_DELAYED_ITEMS,	"FLUSH_DELAYED_ITEMS")		\
 	EM( FLUSH_DELALLOC,		"FLUSH_DELALLOC")		\
 	EM( FLUSH_DELALLOC_WAIT,	"FLUSH_DELALLOC_WAIT")		\
+	EM( FLUSH_DELALLOC_FULL,	"FLUSH_DELALLOC_FULL")		\
 	EM( FLUSH_DELAYED_REFS_NR,	"FLUSH_DELAYED_REFS_NR")	\
 	EM( FLUSH_DELAYED_REFS,		"FLUSH_ELAYED_REFS")		\
 	EM( ALLOC_CHUNK,		"ALLOC_CHUNK")			\

2.26.3

I've been debugging an early ENOSPC problem in production and finally root caused it to this problem. When we switched to the per-inode in 38d715f494f2 ("btrfs: use btrfs_start_delalloc_roots in shrink_delalloc") I pulled out the async extent handling, because we were doing the correct thing by calling filemap_flush() if we had async extents set. This would properly wait on any async extents by locking the page in the second flush, thus making sure our ordered extents were properly set up.

However when I switched us back to page based flushing, I used sync_inode(), which allows us to pass in our own wbc. The problem here is that sync_inode() is smarter than the filemap_* helpers, it tries to avoid calling writepages at all. This means that our second call could skip calling do_writepages altogether, and thus not wait on the pagelock for the async helpers. This means we could come back before any ordered extents were created and then simply continue on in our flushing mechanisms and ENOSPC out when we have plenty of space to use.

Fix this by putting back the async pages logic in shrink_delalloc. This allows us to bulk write out everything that we need to, and then we can wait in one place for the async helpers to catch up, and then wait on any ordered extents that are created.

---
 fs/btrfs/inode.c      |  4 ----
 fs/btrfs/space-info.c | 40 ++++++++++++++++++++++++++++++++++++++++
 2 files changed, 40 insertions(+), 4 deletions(-)

index 0f42864d5dd4..e388153c4ae4 100644
@@ -9714,10 +9714,6 @@ static int start_delalloc_inodes(struct btrfs_root *root,
 					 &work->work);
 		} else {
 			ret = sync_inode(inode, wbc);
-			if (!ret &&
-			    test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
-				     &BTRFS_I(inode)->runtime_flags))
-				ret = sync_inode(inode, wbc);
 			btrfs_add_delayed_iput(inode);
 			if (ret || wbc->nr_to_write <= 0)
 				goto out;

index 1a7e44dd0ba7..9dcde8c97c43 100644
@@ -533,9 +533,49 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info,
 	while ((delalloc_bytes || ordered_bytes) && loops < 3) {
 		u64 temp = min(delalloc_bytes, to_reclaim) >> PAGE_SHIFT;
 		long nr_pages = min_t(u64, temp, LONG_MAX);
+		int async_pages;
 
 		btrfs_start_delalloc_roots(fs_info, nr_pages, true);
 
+		/*
+		 * We need to make sure any outstanding async pages are now
+		 * processed before we continue.  This is because things like
+		 * sync_inode() try to be smart and skip writing if the inode is
+		 * marked clean.  We don't use filemap_fwrite for flushing
+		 * because we want to control how many pages we write out at a
+		 * time, thus this is the only safe way to make sure we've
+		 * waited for outstanding compressed workers to have started
+		 * their jobs and thus have ordered extents set up properly.
+		 *
+		 * This exists because we do not want to wait for each
+		 * individual inode to finish its async work, we simply want to
+		 * start the IO on everybody, and then come back here and wait
+		 * for all of the async work to catch up.  Once we're done with
+		 * that we know we'll have ordered extents for everything and we
+		 * can decide if we wait for that or not.
+		 *
+		 * If we choose to replace this in the future, make absolutely
+		 * sure that the proper waiting is being done in the async case,
+		 * as there have been bugs in that area before.
+		 */
+		async_pages = atomic_read(&fs_info->async_delalloc_pages);
+		if (!async_pages)
+			goto skip_async;
+
+		/*
+		 * We don't want to wait forever, if we wrote less pages in this
+		 * loop than we have outstanding, only wait for that number of
+		 * pages, otherwise we can wait for all async pages to finish
+		 * before continuing.
+		 */
+		if (async_pages > nr_pages)
+			async_pages -= nr_pages;
+		else
+			async_pages = 0;
+		wait_event(fs_info->async_submit_wait,
+			   atomic_read(&fs_info->async_delalloc_pages) <=
+			   async_pages);
+skip_async:
 		loops++;
 		if (wait_ordered && !trans) {
 			btrfs_wait_ordered_roots(fs_info, items, 0, (u64)-1);

2.26.3

Btrfs sometimes needs to flush dirty pages on a bunch of dirty inodes in order to reclaim metadata reservations. Unfortunately most helpers in this area are too smart for us

1) The normal filemap_fdata* helpers only take range and sync modes, and don't give any indication of how much was written, so we can only flush full inodes, which isn't what we want in most cases.
2) The normal writeback path requires us to have the s_umount sem held, but we can't unconditionally take it in this path because we could deadlock.
3) The normal writeback path also skips inodes with I_SYNC set if we write with WB_SYNC_NONE. This isn't the behavior we want under heavy ENOSPC pressure, we want to actually make sure the pages are under writeback before returning, and if another thread is in the middle of writing the file we may return before they're under writeback and miss our ordered extents and not properly wait for completion.
4) sync_inode() uses the normal writeback path and has the same problem as #3.

What we really want is to call do_writepages() with our wbc. This way we can make sure that writeback is actually started on the pages, and we can control how many pages are written as a whole as we write many inodes using the same wbc. Accomplish this with a new helper that does just that so we can use it for our ENOSPC flushing infrastructure.

---
 include/linux/fs.h |  2 ++
 mm/filemap.c       | 35 ++++++++++++++++++++++++++---------
 2 files changed, 28 insertions(+), 9 deletions(-)

index c3c88fdb9b2a..aace07f88b73 100644
@@ -2886,6 +2886,8 @@ extern int filemap_fdatawrite_range(struct address_space *mapping,
 				loff_t start, loff_t end);
 extern int filemap_check_errors(struct address_space *mapping);
 extern void __filemap_set_wb_err(struct address_space *mapping, int err);
+extern int filemap_fdatawrite_wbc(struct address_space *mapping,
+				  struct writeback_control *wbc);
 
 static inline int filemap_write_and_wait(struct address_space *mapping)
 {

index 66f7e9fdfbc4..8395eafc178b 100644
@@ -376,6 +376,31 @@ static int filemap_check_and_keep_errors(struct address_space *mapping)
 		return -ENOSPC;
 	return 0;
 }
+/**
+ * filemap_fdatawrite_wbc - start writeback on mapping dirty pages in range
+ * @mapping:	address space structure to write
+ * @wbc:	the writeback_control controlling the writeout
+ *
+ * Call writepages on the mapping using the provided wbc to control the
+ * writeout.
+ *
+ * Return: %0 on success, negative error code otherwise.
+ */
+int filemap_fdatawrite_wbc(struct address_space *mapping,
+			   struct writeback_control *wbc)
+{
+	int ret;
+
+	if (!mapping_can_writeback(mapping) ||
+	    !mapping_tagged(mapping, PAGECACHE_TAG_DIRTY))
+		return 0;
+
+	wbc_attach_fdatawrite_inode(wbc, mapping->host);
+	ret = do_writepages(mapping, wbc);
+	wbc_detach_inode(wbc);
+	return ret;
+}
+EXPORT_SYMBOL(filemap_fdatawrite_wbc);
 
 /**
  * __filemap_fdatawrite_range - start writeback on mapping dirty pages in range

@@ -397,7 +422,6 @@ static int filemap_check_and_keep_errors(struct address_space *mapping)
 int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start,
 				loff_t end, int sync_mode)
 {
-	int ret;
 	struct writeback_control wbc = {
 		.sync_mode = sync_mode,
 		.nr_to_write = LONG_MAX,

@@ -405,14 +429,7 @@ int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start,
 		.range_end = end,
 	};
 
-	if (!mapping_can_writeback(mapping) ||
-	    !mapping_tagged(mapping, PAGECACHE_TAG_DIRTY))
-		return 0;
-
-	wbc_attach_fdatawrite_inode(&wbc, mapping->host);
-	ret = do_writepages(mapping, &wbc);
-	wbc_detach_inode(&wbc);
-	return ret;
+	return filemap_fdatawrite_wbc(mapping, &wbc);
 }
 
 static inline int __filemap_fdatawrite(struct address_space *mapping,

2.26.3

I'll remove it when applying the patch. There are many externs in the file but also declaration without it so this one won't stand out.

sync_inode() has some holes that can cause problems if we're under heavy ENOSPC pressure. If there's writeback running on a separate thread sync_inode() will skip writing the inode altogether. What we really want is to make sure writeback has been started on all the pages to make sure we can see the ordered extents and wait on them if appropriate. Switch to this new helper which will allow us to accomplish this and avoid ENOSPC'ing early.

---
 fs/btrfs/inode.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index e388153c4ae4..b25c84aba743 100644
@@ -9713,7 +9713,7 @@ static int start_delalloc_inodes(struct btrfs_root *root,
 			btrfs_queue_work(root->fs_info->flush_workers,
 					 &work->work);
 		} else {
-			ret = sync_inode(inode, wbc);
+			ret = filemap_fdatawrite_wbc(inode->i_mapping, wbc);
 			btrfs_add_delayed_iput(inode);
 			if (ret || wbc->nr_to_write <= 0)
 				goto out;

2.26.3

We're going to remove sync_inode, so migrate to filemap_fdatawrite_wbc instead.

---
 fs/9p/vfs_file.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

index 59c32c9b799f..6b64e8391f30 100644
@@ -625,12 +625,7 @@ static void v9fs_mmap_vm_close(struct vm_area_struct *vma)
 	p9_debug(P9_DEBUG_VFS, "9p VMA close, %p, flushing", vma);
 
 	inode = file_inode(vma->vm_file);
-
-	if (!mapping_can_writeback(inode->i_mapping))
-		wbc.nr_to_write = 0;
-
-	might_sleep();
-	sync_inode(inode, &wbc);
+	filemap_fdatawrite_wbc(inode->i_mapping, &wbc);
 }
 
 

2.26.3

Looks good,

I don't see any 9p people in CC, adding them. The cover letter is
https://lore.kernel.org/linux-btrfs/[email protected]/

Now that all users of sync_inode() have been deleted, remove sync_inode().

---
 fs/fs-writeback.c  | 19 +------------------
 include/linux/fs.h |  1 -
 2 files changed, 1 insertion(+), 19 deletions(-)

index e91980f49388..706dad22f735 100644
@@ -2608,23 +2608,6 @@ int write_inode_now(struct inode *inode, int sync)
 }
 EXPORT_SYMBOL(write_inode_now);
 
-/**
- * sync_inode - write an inode and its pages to disk.
- * @inode: the inode to sync
- * @wbc: controls the writeback mode
- *
- * sync_inode() will write an inode and its pages to disk.  It will also
- * correctly update the inode on its superblock's dirty inode lists and will
- * update inode->i_state.
- *
- * The caller must have a ref on the inode.
- */
-int sync_inode(struct inode *inode, struct writeback_control *wbc)
-{
-	return writeback_single_inode(inode, wbc);
-}
-EXPORT_SYMBOL(sync_inode);
-
 /**
  * sync_inode_metadata - write an inode to disk
  * @inode: the inode to sync

@@ -2641,6 +2624,6 @@ int sync_inode_metadata(struct inode *inode, int wait)
 		.nr_to_write = 0, /* metadata-only */
 	};
 
-	return sync_inode(inode, &wbc);
+	return writeback_single_inode(inode, &wbc);
 }
 EXPORT_SYMBOL(sync_inode_metadata);

index aace07f88b73..7c33e5414747 100644
@@ -2458,7 +2458,6 @@ static inline void file_accessed(struct file *file)
 
 extern int file_modified(struct file *file);
 
-int sync_inode(struct inode *inode, struct writeback_control *wbc);
 int sync_inode_metadata(struct inode *inode, int wait);
 
 struct file_system_type {

2.26.3

Looks good,

For the whole series:

I'll put it to my tree and it will appear in linux-next soon. The last two patches are not related to btrfs so I can send them separately or leave it in the series if nobody objects.