2024-02-12 07:14:12

by Christoph Hellwig

[permalink] [raw]
Subject: convert write_cache_pages() to an iterator v7

Hi all,

this is an evolution of the series Matthew Wilcox originally sent in June
2023, which has changed quite a bit since and now has a while based
iterator.

Note that in this version two patches are so different from the previous
version that I've not kept any Reviews or Acks for them, even if the
final result look almost the same as the previous patches with the
incremental patch on the list.

Changes since v6:
- don't access folio->index after releasing the folio batch
- add a new patch to fix a pre-existing bug where a positive value is
passed to mapping_set_error

Changes since v5:
- completely reshuffle the series to directly prepare for the
writeback_iter() style.
- don't require *error to be initialized on first call
- improve various comments
- fix a bisection hazard where write_cache_pages don't return delayed
error for a few commits
- fix a whitespace error
- drop the iomap patch again for now as the iomap map multiple blocks
series isn't in mainline yet

Changes since v4:
- added back the (rebased) iomap conversion now that the conflict is in
mainline
- add a new patch to change the iterator

Changes since v3:
- various commit log spelling fixes
- remove a statement from a commit log that isn't true any more with the
changes in v3
- rename a function
- merge two helpers

Diffstat:
include/linux/pagevec.h | 18 ++
include/linux/writeback.h | 12 +
mm/page-writeback.c | 389 ++++++++++++++++++++++++++--------------------
3 files changed, 249 insertions(+), 170 deletions(-)


2024-02-12 07:16:10

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 06/14] writeback: rework the loop termination condition in write_cache_pages

Rework the way we deal with the cleanup after the writepage call.

First handle the magic AOP_WRITEPAGE_ACTIVATE separately from real error
returns to get it out of the way of the actual error handling path.

The split the handling on intgrity vs non-integrity branches first,
and return early using a goto for the non-ingegrity early loop condition
to remove the need for the done and done_index local variables, and for
assigning the error to ret when we can just return error directly.

Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Brian Foster <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
---
mm/page-writeback.c | 84 ++++++++++++++++++---------------------------
1 file changed, 33 insertions(+), 51 deletions(-)

diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 4d862f196d1f05..b49ee15a863e99 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2396,13 +2396,12 @@ int write_cache_pages(struct address_space *mapping,
void *data)
{
int ret = 0;
- int done = 0;
int error;
struct folio_batch fbatch;
+ struct folio *folio;
int nr_folios;
pgoff_t index;
pgoff_t end; /* Inclusive */
- pgoff_t done_index;
xa_mark_t tag;

folio_batch_init(&fbatch);
@@ -2419,8 +2418,7 @@ int write_cache_pages(struct address_space *mapping,
} else {
tag = PAGECACHE_TAG_DIRTY;
}
- done_index = index;
- while (!done && (index <= end)) {
+ while (index <= end) {
int i;

nr_folios = filemap_get_folios_tag(mapping, &index, end,
@@ -2430,11 +2428,7 @@ int write_cache_pages(struct address_space *mapping,
break;

for (i = 0; i < nr_folios; i++) {
- struct folio *folio = fbatch.folios[i];
- unsigned long nr;
-
- done_index = folio->index;
-
+ folio = fbatch.folios[i];
folio_lock(folio);

/*
@@ -2469,45 +2463,32 @@ int write_cache_pages(struct address_space *mapping,

trace_wbc_writepage(wbc, inode_to_bdi(mapping->host));
error = writepage(folio, wbc, data);
- nr = folio_nr_pages(folio);
- wbc->nr_to_write -= nr;
- if (unlikely(error)) {
- /*
- * Handle errors according to the type of
- * writeback. There's no need to continue for
- * background writeback. Just push done_index
- * past this page so media errors won't choke
- * writeout for the entire file. For integrity
- * writeback, we must process the entire dirty
- * set regardless of errors because the fs may
- * still have state to clear for each page. In
- * that case we continue processing and return
- * the first error.
- */
- if (error == AOP_WRITEPAGE_ACTIVATE) {
- folio_unlock(folio);
- error = 0;
- } else if (wbc->sync_mode != WB_SYNC_ALL) {
- ret = error;
- done_index = folio->index + nr;
- done = 1;
- break;
- }
- if (!ret)
- ret = error;
+ wbc->nr_to_write -= folio_nr_pages(folio);
+
+ if (error == AOP_WRITEPAGE_ACTIVATE) {
+ folio_unlock(folio);
+ error = 0;
}

/*
- * We stop writing back only if we are not doing
- * integrity sync. In case of integrity sync we have to
- * keep going until we have written all the pages
- * we tagged for writeback prior to entering this loop.
+ * For integrity writeback we have to keep going until
+ * we have written all the folios we tagged for
+ * writeback above, even if we run past wbc->nr_to_write
+ * or encounter errors.
+ * We stash away the first error we encounter in
+ * wbc->saved_err so that it can be retrieved when we're
+ * done. This is because the file system may still have
+ * state to clear for each folio.
+ *
+ * For background writeback we exit as soon as we run
+ * past wbc->nr_to_write or encounter the first error.
*/
- done_index = folio->index + nr;
- if (wbc->nr_to_write <= 0 &&
- wbc->sync_mode == WB_SYNC_NONE) {
- done = 1;
- break;
+ if (wbc->sync_mode == WB_SYNC_ALL) {
+ if (error && !ret)
+ ret = error;
+ } else {
+ if (error || wbc->nr_to_write <= 0)
+ goto done;
}
}
folio_batch_release(&fbatch);
@@ -2524,14 +2505,15 @@ int write_cache_pages(struct address_space *mapping,
* of the file if we are called again, which can only happen due to
* -ENOMEM from the file system.
*/
- if (wbc->range_cyclic) {
- if (done)
- mapping->writeback_index = done_index;
- else
- mapping->writeback_index = 0;
- }
-
+ if (wbc->range_cyclic)
+ mapping->writeback_index = 0;
return ret;
+
+done:
+ if (wbc->range_cyclic)
+ mapping->writeback_index = folio->index + folio_nr_pages(folio);
+ folio_batch_release(&fbatch);
+ return error;
}
EXPORT_SYMBOL(write_cache_pages);

--
2.39.2


2024-02-12 07:17:03

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 08/14] writeback: Factor writeback_get_batch() out of write_cache_pages()

From: "Matthew Wilcox (Oracle)" <[email protected]>

This simple helper will be the basis of the writeback iterator.
To make this work, we need to remember the current index
and end positions in writeback_control.

Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
[hch: heavily rebased, add helpers to get the tag and end index,
don't keep the end index in struct writeback_control]
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Brian Foster <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Acked-by: Dave Chinner <[email protected]>
---
include/linux/writeback.h | 6 ++++
mm/page-writeback.c | 60 +++++++++++++++++++++++++--------------
2 files changed, 44 insertions(+), 22 deletions(-)

diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index 4b8cf9e4810bad..f67b3ea866a0fb 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -11,6 +11,7 @@
#include <linux/flex_proportions.h>
#include <linux/backing-dev-defs.h>
#include <linux/blk_types.h>
+#include <linux/pagevec.h>

struct bio;

@@ -40,6 +41,7 @@ enum writeback_sync_modes {
* in a manner such that unspecified fields are set to zero.
*/
struct writeback_control {
+ /* public fields that can be set and/or consumed by the caller: */
long nr_to_write; /* Write this many pages, and decrement
this for each page written */
long pages_skipped; /* Pages which were not written */
@@ -77,6 +79,10 @@ struct writeback_control {
*/
struct swap_iocb **swap_plug;

+ /* internal fields used by the ->writepages implementation: */
+ struct folio_batch fbatch;
+ pgoff_t index;
+
#ifdef CONFIG_CGROUP_WRITEBACK
struct bdi_writeback *wb; /* wb this writeback is issued under */
struct inode *inode; /* inode being written out */
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 20ff00c8be9d90..045ca252c0423d 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2392,6 +2392,29 @@ static bool folio_prepare_writeback(struct address_space *mapping,
return true;
}

+static xa_mark_t wbc_to_tag(struct writeback_control *wbc)
+{
+ if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
+ return PAGECACHE_TAG_TOWRITE;
+ return PAGECACHE_TAG_DIRTY;
+}
+
+static pgoff_t wbc_end(struct writeback_control *wbc)
+{
+ if (wbc->range_cyclic)
+ return -1;
+ return wbc->range_end >> PAGE_SHIFT;
+}
+
+static void writeback_get_batch(struct address_space *mapping,
+ struct writeback_control *wbc)
+{
+ folio_batch_release(&wbc->fbatch);
+ cond_resched();
+ filemap_get_folios_tag(mapping, &wbc->index, wbc_end(wbc),
+ wbc_to_tag(wbc), &wbc->fbatch);
+}
+
/**
* write_cache_pages - walk the list of dirty pages of the given address space and write all of them.
* @mapping: address space structure to write
@@ -2429,38 +2452,32 @@ int write_cache_pages(struct address_space *mapping,
{
int ret = 0;
int error;
- struct folio_batch fbatch;
struct folio *folio;
- int nr_folios;
- pgoff_t index;
pgoff_t end; /* Inclusive */
- xa_mark_t tag;

- folio_batch_init(&fbatch);
if (wbc->range_cyclic) {
- index = mapping->writeback_index; /* prev offset */
+ wbc->index = mapping->writeback_index; /* prev offset */
end = -1;
} else {
- index = wbc->range_start >> PAGE_SHIFT;
+ wbc->index = wbc->range_start >> PAGE_SHIFT;
end = wbc->range_end >> PAGE_SHIFT;
}
- if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages) {
- tag_pages_for_writeback(mapping, index, end);
- tag = PAGECACHE_TAG_TOWRITE;
- } else {
- tag = PAGECACHE_TAG_DIRTY;
- }
- while (index <= end) {
+ if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
+ tag_pages_for_writeback(mapping, wbc->index, end);
+
+ folio_batch_init(&wbc->fbatch);
+
+ while (wbc->index <= end) {
int i;

- nr_folios = filemap_get_folios_tag(mapping, &index, end,
- tag, &fbatch);
+ writeback_get_batch(mapping, wbc);

- if (nr_folios == 0)
+ if (wbc->fbatch.nr == 0)
break;

- for (i = 0; i < nr_folios; i++) {
- folio = fbatch.folios[i];
+ for (i = 0; i < wbc->fbatch.nr; i++) {
+ folio = wbc->fbatch.folios[i];
+
folio_lock(folio);
if (!folio_prepare_writeback(mapping, wbc, folio)) {
folio_unlock(folio);
@@ -2498,8 +2515,6 @@ int write_cache_pages(struct address_space *mapping,
goto done;
}
}
- folio_batch_release(&fbatch);
- cond_resched();
}

/*
@@ -2512,6 +2527,7 @@ int write_cache_pages(struct address_space *mapping,
* of the file if we are called again, which can only happen due to
* -ENOMEM from the file system.
*/
+ folio_batch_release(&wbc->fbatch);
if (wbc->range_cyclic)
mapping->writeback_index = 0;
return ret;
@@ -2519,7 +2535,7 @@ int write_cache_pages(struct address_space *mapping,
done:
if (wbc->range_cyclic)
mapping->writeback_index = folio->index + folio_nr_pages(folio);
- folio_batch_release(&fbatch);
+ folio_batch_release(&wbc->fbatch);
return error;
}
EXPORT_SYMBOL(write_cache_pages);
--
2.39.2


2024-02-12 07:18:12

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 11/14] writeback: Use the folio_batch queue iterator

From: "Matthew Wilcox (Oracle)" <[email protected]>

Instead of keeping our own local iterator variable, use the one just
added to folio_batch.

Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Brian Foster <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Acked-by: Dave Chinner <[email protected]>
---
mm/page-writeback.c | 28 +++++++++++++++-------------
1 file changed, 15 insertions(+), 13 deletions(-)

diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index a94a77b1805969..62b663debe713b 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2406,13 +2406,21 @@ static pgoff_t wbc_end(struct writeback_control *wbc)
return wbc->range_end >> PAGE_SHIFT;
}

-static void writeback_get_batch(struct address_space *mapping,
+static struct folio *writeback_get_folio(struct address_space *mapping,
struct writeback_control *wbc)
{
- folio_batch_release(&wbc->fbatch);
- cond_resched();
- filemap_get_folios_tag(mapping, &wbc->index, wbc_end(wbc),
- wbc_to_tag(wbc), &wbc->fbatch);
+ struct folio *folio;
+
+ folio = folio_batch_next(&wbc->fbatch);
+ if (!folio) {
+ folio_batch_release(&wbc->fbatch);
+ cond_resched();
+ filemap_get_folios_tag(mapping, &wbc->index, wbc_end(wbc),
+ wbc_to_tag(wbc), &wbc->fbatch);
+ folio = folio_batch_next(&wbc->fbatch);
+ }
+
+ return folio;
}

/**
@@ -2454,7 +2462,6 @@ int write_cache_pages(struct address_space *mapping,
int error;
struct folio *folio;
pgoff_t end; /* Inclusive */
- int i = 0;

if (wbc->range_cyclic) {
wbc->index = mapping->writeback_index; /* prev offset */
@@ -2469,15 +2476,10 @@ int write_cache_pages(struct address_space *mapping,
folio_batch_init(&wbc->fbatch);

for (;;) {
- if (i == wbc->fbatch.nr) {
- writeback_get_batch(mapping, wbc);
- i = 0;
- }
- if (wbc->fbatch.nr == 0)
+ folio = writeback_get_folio(mapping, wbc);
+ if (!folio)
break;

- folio = wbc->fbatch.folios[i++];
-
folio_lock(folio);
if (!folio_prepare_writeback(mapping, wbc, folio)) {
folio_unlock(folio);
--
2.39.2


2024-02-12 07:18:59

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 13/14] writeback: add a writeback iterator

Refactor the code left in write_cache_pages into an iterator that the
file system can call to get the next folio for a writeback operation:

struct folio *folio = NULL;

while ((folio = writeback_iter(mapping, wbc, folio, &error))) {
error = <do per-folio writeback>;
}

The twist here is that the error value is passed by reference, so that
the iterator can restore it when breaking out of the loop.

Handling of the magic AOP_WRITEPAGE_ACTIVATE value stays outside the
iterator and needs is just kept in the write_cache_pages legacy wrapper.
in preparation for eventually killing it off.

Heavily based on a for_each* based iterator from Matthew Wilcox.

Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Brian Foster <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
---
include/linux/writeback.h | 4 +
mm/page-writeback.c | 192 ++++++++++++++++++++++----------------
2 files changed, 118 insertions(+), 78 deletions(-)

diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index f67b3ea866a0fb..9845cb62e40b2d 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -82,6 +82,7 @@ struct writeback_control {
/* internal fields used by the ->writepages implementation: */
struct folio_batch fbatch;
pgoff_t index;
+ int saved_err;

#ifdef CONFIG_CGROUP_WRITEBACK
struct bdi_writeback *wb; /* wb this writeback is issued under */
@@ -366,6 +367,9 @@ int balance_dirty_pages_ratelimited_flags(struct address_space *mapping,

bool wb_over_bg_thresh(struct bdi_writeback *wb);

+struct folio *writeback_iter(struct address_space *mapping,
+ struct writeback_control *wbc, struct folio *folio, int *error);
+
typedef int (*writepage_t)(struct folio *folio, struct writeback_control *wbc,
void *data);

diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 01f076db4f2118..1996200849e577 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2325,18 +2325,18 @@ void __init page_writeback_init(void)
}

/**
- * tag_pages_for_writeback - tag pages to be written by write_cache_pages
+ * tag_pages_for_writeback - tag pages to be written by writeback
* @mapping: address space structure to write
* @start: starting page index
* @end: ending page index (inclusive)
*
* This function scans the page range from @start to @end (inclusive) and tags
- * all pages that have DIRTY tag set with a special TOWRITE tag. The idea is
- * that write_cache_pages (or whoever calls this function) will then use
- * TOWRITE tag to identify pages eligible for writeback. This mechanism is
- * used to avoid livelocking of writeback by a process steadily creating new
- * dirty pages in the file (thus it is important for this function to be quick
- * so that it can tag pages faster than a dirtying process can create them).
+ * all pages that have DIRTY tag set with a special TOWRITE tag. The caller
+ * can then use the TOWRITE tag to identify pages eligible for writeback.
+ * This mechanism is used to avoid livelocking of writeback by a process
+ * steadily creating new dirty pages in the file (thus it is important for this
+ * function to be quick so that it can tag pages faster than a dirtying process
+ * can create them).
*/
void tag_pages_for_writeback(struct address_space *mapping,
pgoff_t start, pgoff_t end)
@@ -2434,69 +2434,68 @@ static struct folio *writeback_get_folio(struct address_space *mapping,
}

/**
- * write_cache_pages - walk the list of dirty pages of the given address space and write all of them.
+ * writeback_iter - iterate folio of a mapping for writeback
* @mapping: address space structure to write
- * @wbc: subtract the number of written pages from *@wbc->nr_to_write
- * @writepage: function called for each page
- * @data: data passed to writepage function
+ * @wbc: writeback context
+ * @folio: previously iterated folio (%NULL to start)
+ * @error: in-out pointer for writeback errors (see below)
*
- * If a page is already under I/O, write_cache_pages() skips it, even
- * if it's dirty. This is desirable behaviour for memory-cleaning writeback,
- * but it is INCORRECT for data-integrity system calls such as fsync(). fsync()
- * and msync() need to guarantee that all the data which was dirty at the time
- * the call was made get new I/O started against them. If wbc->sync_mode is
- * WB_SYNC_ALL then we were called for data integrity and we must wait for
- * existing IO to complete.
- *
- * To avoid livelocks (when other process dirties new pages), we first tag
- * pages which should be written back with TOWRITE tag and only then start
- * writing them. For data-integrity sync we have to be careful so that we do
- * not miss some pages (e.g., because some other process has cleared TOWRITE
- * tag we set). The rule we follow is that TOWRITE tag can be cleared only
- * by the process clearing the DIRTY tag (and submitting the page for IO).
- *
- * To avoid deadlocks between range_cyclic writeback and callers that hold
- * pages in PageWriteback to aggregate IO until write_cache_pages() returns,
- * we do not loop back to the start of the file. Doing so causes a page
- * lock/page writeback access order inversion - we should only ever lock
- * multiple pages in ascending page->index order, and looping back to the start
- * of the file violates that rule and causes deadlocks.
+ * This function returns the next folio for the writeback operation described by
+ * @wbc on @mapping and should be called in a while loop in the ->writepages
+ * implementation.
*
- * Return: %0 on success, negative error code otherwise
+ * To start the writeback operation, %NULL is passed in the @folio argument, and
+ * for every subsequent iteration the folio returned previously should be passed
+ * back in.
+ *
+ * If there was an error in the per-folio writeback inside the writeback_iter()
+ * loop, @error should be set to the error value.
+ *
+ * Once the writeback described in @wbc has finished, this function will return
+ * %NULL and if there was an error in any iteration restore it to @error.
+ *
+ * Note: callers should not manually break out of the loop using break or goto
+ * but must keep calling writeback_iter() until it returns %NULL.
+ *
+ * Return: the folio to write or %NULL if the loop is done.
*/
-int write_cache_pages(struct address_space *mapping,
- struct writeback_control *wbc, writepage_t writepage,
- void *data)
+struct folio *writeback_iter(struct address_space *mapping,
+ struct writeback_control *wbc, struct folio *folio, int *error)
{
- int ret = 0;
- int error;
- struct folio *folio;
- pgoff_t end; /* Inclusive */
-
- if (wbc->range_cyclic) {
- wbc->index = mapping->writeback_index; /* prev offset */
- end = -1;
- } else {
- wbc->index = wbc->range_start >> PAGE_SHIFT;
- end = wbc->range_end >> PAGE_SHIFT;
- }
- if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
- tag_pages_for_writeback(mapping, wbc->index, end);
-
- folio_batch_init(&wbc->fbatch);
+ if (!folio) {
+ folio_batch_init(&wbc->fbatch);
+ wbc->saved_err = *error = 0;

- for (;;) {
- folio = writeback_get_folio(mapping, wbc);
- if (!folio)
- break;
+ /*
+ * For range cyclic writeback we remember where we stopped so
+ * that we can continue where we stopped.
+ *
+ * For non-cyclic writeback we always start at the beginning of
+ * the passed in range.
+ */
+ if (wbc->range_cyclic)
+ wbc->index = mapping->writeback_index;
+ else
+ wbc->index = wbc->range_start >> PAGE_SHIFT;

- error = writepage(folio, wbc, data);
+ /*
+ * To avoid livelocks when other processes dirty new pages, we
+ * first tag pages which should be written back and only then
+ * start writing them.
+ *
+ * For data-integrity writeback we have to be careful so that we
+ * do not miss some pages (e.g., because some other process has
+ * cleared the TOWRITE tag we set). The rule we follow is that
+ * TOWRITE tag can be cleared only by the process clearing the
+ * DIRTY tag (and submitting the page for I/O).
+ */
+ if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
+ tag_pages_for_writeback(mapping, wbc->index,
+ wbc_end(wbc));
+ } else {
wbc->nr_to_write -= folio_nr_pages(folio);

- if (error == AOP_WRITEPAGE_ACTIVATE) {
- folio_unlock(folio);
- error = 0;
- }
+ WARN_ON_ONCE(*error > 0);

/*
* For integrity writeback we have to keep going until we have
@@ -2510,33 +2509,70 @@ int write_cache_pages(struct address_space *mapping,
* wbc->nr_to_write or encounter the first error.
*/
if (wbc->sync_mode == WB_SYNC_ALL) {
- if (error && !ret)
- ret = error;
+ if (*error && !wbc->saved_err)
+ wbc->saved_err = *error;
} else {
- if (error || wbc->nr_to_write <= 0)
+ if (*error || wbc->nr_to_write <= 0)
goto done;
}
}

- /*
- * For range cyclic writeback we need to remember where we stopped so
- * that we can continue there next time we are called. If we hit the
- * last page and there is more work to be done, wrap back to the start
- * of the file.
- *
- * For non-cyclic writeback we always start looking up at the beginning
- * of the file if we are called again, which can only happen due to
- * -ENOMEM from the file system.
- */
- folio_batch_release(&wbc->fbatch);
- if (wbc->range_cyclic)
- mapping->writeback_index = 0;
- return ret;
+ folio = writeback_get_folio(mapping, wbc);
+ if (!folio) {
+ /*
+ * To avoid deadlocks between range_cyclic writeback and callers
+ * that hold pages in PageWriteback to aggregate I/O until
+ * the writeback iteration finishes, we do not loop back to the
+ * start of the file. Doing so causes a page lock/page
+ * writeback access order inversion - we should only ever lock
+ * multiple pages in ascending page->index order, and looping
+ * back to the start of the file violates that rule and causes
+ * deadlocks.
+ */
+ if (wbc->range_cyclic)
+ mapping->writeback_index = 0;
+
+ /*
+ * Return the first error we encountered (if there was any) to
+ * the caller.
+ */
+ *error = wbc->saved_err;
+ }
+ return folio;

done:
if (wbc->range_cyclic)
mapping->writeback_index = folio->index + folio_nr_pages(folio);
folio_batch_release(&wbc->fbatch);
+ return NULL;
+}
+
+/**
+ * write_cache_pages - walk the list of dirty pages of the given address space and write all of them.
+ * @mapping: address space structure to write
+ * @wbc: subtract the number of written pages from *@wbc->nr_to_write
+ * @writepage: function called for each page
+ * @data: data passed to writepage function
+ *
+ * Return: %0 on success, negative error code otherwise
+ *
+ * Note: please use writeback_iter() instead.
+ */
+int write_cache_pages(struct address_space *mapping,
+ struct writeback_control *wbc, writepage_t writepage,
+ void *data)
+{
+ struct folio *folio = NULL;
+ int error;
+
+ while ((folio = writeback_iter(mapping, wbc, folio, &error))) {
+ error = writepage(folio, wbc, data);
+ if (error == AOP_WRITEPAGE_ACTIVATE) {
+ folio_unlock(folio);
+ error = 0;
+ }
+ }
+
return error;
}
EXPORT_SYMBOL(write_cache_pages);
--
2.39.2


2024-02-12 08:19:32

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 03/14] writeback: fix done_index when hitting the wbc->nr_to_write

When write_cache_pages finishes writing out a folio, it fails to update
done_index to account for the number of pages in the folio just written.
That means when range_cyclic writeback is restarted, it will be
restarted at this folio instead of after it as it should. Fix that
by updating done_index before breaking out of the loop.

Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Brian Foster <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Acked-by: Dave Chinner <[email protected]>
---
mm/page-writeback.c | 1 +
1 file changed, 1 insertion(+)

diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 62901fa905f01e..aa3b432f77e37a 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2505,6 +2505,7 @@ int write_cache_pages(struct address_space *mapping,
* keep going until we have written all the pages
* we tagged for writeback prior to entering this loop.
*/
+ done_index = folio->index + nr;
wbc->nr_to_write -= nr;
if (wbc->nr_to_write <= 0 &&
wbc->sync_mode == WB_SYNC_NONE) {
--
2.39.2


2024-02-12 08:19:55

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 04/14] writeback: also update wbc->nr_to_write on writeback failure

When exiting write_cache_pages early due to a non-integrity write
failure, wbc->nr_to_write currently doesn't account for the folio
we just failed to write. This doesn't matter because the callers
always ingore the value on a failure, but moving the update to
common code will allow to simplify the code, so do it.

Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Brian Foster <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Acked-by: Dave Chinner <[email protected]>
---
mm/page-writeback.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index aa3b432f77e37a..06afba8f078515 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2473,6 +2473,7 @@ int write_cache_pages(struct address_space *mapping,
trace_wbc_writepage(wbc, inode_to_bdi(mapping->host));
error = writepage(folio, wbc, data);
nr = folio_nr_pages(folio);
+ wbc->nr_to_write -= nr;
if (unlikely(error)) {
/*
* Handle errors according to the type of
@@ -2506,7 +2507,6 @@ int write_cache_pages(struct address_space *mapping,
* we tagged for writeback prior to entering this loop.
*/
done_index = folio->index + nr;
- wbc->nr_to_write -= nr;
if (wbc->nr_to_write <= 0 &&
wbc->sync_mode == WB_SYNC_NONE) {
done = 1;
--
2.39.2


2024-02-12 08:28:01

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 01/14] writeback: don't call mapping_set_error in writepage_cb

writepage_cb is the iterator callback for write_cache_pages, which
already tracks all errors and returns them to the caller. There is
no need to additionally cal mapping_set_error which is intended
for contexts where the error can't be directly returned (e.g. the
I/O completion handlers).

Remove the mapping_set_error call in writepage_cb which is not only
superfluous but also buggy as it can be called with the error argument
set to AOP_WRITEPAGE_ACTIVATE, which is not actually an error but a
magic return value asking the caller to unlock the page.

Signed-off-by: Christoph Hellwig <[email protected]>
---
mm/page-writeback.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 3f255534986a2f..62901fa905f01e 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2534,9 +2534,8 @@ static int writepage_cb(struct folio *folio, struct writeback_control *wbc,
void *data)
{
struct address_space *mapping = data;
- int ret = mapping->a_ops->writepage(&folio->page, wbc);
- mapping_set_error(mapping, ret);
- return ret;
+
+ return mapping->a_ops->writepage(&folio->page, wbc);
}

int do_writepages(struct address_space *mapping, struct writeback_control *wbc)
--
2.39.2


2024-02-12 09:15:37

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 12/14] writeback: Move the folio_prepare_writeback loop out of write_cache_pages()

From: "Matthew Wilcox (Oracle)" <[email protected]>

Move the loop for should-we-write-this-folio to writeback_get_folio.

Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
[hch: folded the loop into the existing helper instead of a separate one
as suggested by Jan Kara]
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Brian Foster <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Acked-by: Dave Chinner <[email protected]>
---
mm/page-writeback.c | 18 ++++++++++--------
1 file changed, 10 insertions(+), 8 deletions(-)

diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 62b663debe713b..01f076db4f2118 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2411,6 +2411,7 @@ static struct folio *writeback_get_folio(struct address_space *mapping,
{
struct folio *folio;

+retry:
folio = folio_batch_next(&wbc->fbatch);
if (!folio) {
folio_batch_release(&wbc->fbatch);
@@ -2418,8 +2419,17 @@ static struct folio *writeback_get_folio(struct address_space *mapping,
filemap_get_folios_tag(mapping, &wbc->index, wbc_end(wbc),
wbc_to_tag(wbc), &wbc->fbatch);
folio = folio_batch_next(&wbc->fbatch);
+ if (!folio)
+ return NULL;
+ }
+
+ folio_lock(folio);
+ if (unlikely(!folio_prepare_writeback(mapping, wbc, folio))) {
+ folio_unlock(folio);
+ goto retry;
}

+ trace_wbc_writepage(wbc, inode_to_bdi(mapping->host));
return folio;
}

@@ -2480,14 +2490,6 @@ int write_cache_pages(struct address_space *mapping,
if (!folio)
break;

- folio_lock(folio);
- if (!folio_prepare_writeback(mapping, wbc, folio)) {
- folio_unlock(folio);
- continue;
- }
-
- trace_wbc_writepage(wbc, inode_to_bdi(mapping->host));
-
error = writepage(folio, wbc, data);
wbc->nr_to_write -= folio_nr_pages(folio);

--
2.39.2


2024-02-12 09:53:02

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 05/14] writeback: only update ->writeback_index for range_cyclic writeback

mapping->writeback_index is only [1] used as the starting point for
range_cyclic writeback, so there is no point in updating it for other
types of writeback.

[1] except for btrfs_defrag_file which does really odd things with
mapping->writeback_index. But btrfs doesn't use write_cache_pages at
all, so this isn't relevant here.

Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Brian Foster <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Acked-by: Dave Chinner <[email protected]>
---
mm/page-writeback.c | 24 ++++++++++++++----------
1 file changed, 14 insertions(+), 10 deletions(-)

diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 06afba8f078515..4d862f196d1f05 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2403,7 +2403,6 @@ int write_cache_pages(struct address_space *mapping,
pgoff_t index;
pgoff_t end; /* Inclusive */
pgoff_t done_index;
- int range_whole = 0;
xa_mark_t tag;

folio_batch_init(&fbatch);
@@ -2413,8 +2412,6 @@ int write_cache_pages(struct address_space *mapping,
} else {
index = wbc->range_start >> PAGE_SHIFT;
end = wbc->range_end >> PAGE_SHIFT;
- if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX)
- range_whole = 1;
}
if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages) {
tag_pages_for_writeback(mapping, index, end);
@@ -2518,14 +2515,21 @@ int write_cache_pages(struct address_space *mapping,
}

/*
- * If we hit the last page and there is more work to be done: wrap
- * back the index back to the start of the file for the next
- * time we are called.
+ * For range cyclic writeback we need to remember where we stopped so
+ * that we can continue there next time we are called. If we hit the
+ * last page and there is more work to be done, wrap back to the start
+ * of the file.
+ *
+ * For non-cyclic writeback we always start looking up at the beginning
+ * of the file if we are called again, which can only happen due to
+ * -ENOMEM from the file system.
*/
- if (wbc->range_cyclic && !done)
- done_index = 0;
- if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0))
- mapping->writeback_index = done_index;
+ if (wbc->range_cyclic) {
+ if (done)
+ mapping->writeback_index = done_index;
+ else
+ mapping->writeback_index = 0;
+ }

return ret;
}
--
2.39.2


2024-02-12 10:05:17

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 07/14] writeback: Factor folio_prepare_writeback() out of write_cache_pages()

From: "Matthew Wilcox (Oracle)" <[email protected]>

Reduce write_cache_pages() by about 30 lines; much of it is commentary,
but it all bundles nicely into an obvious function.

Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
[hch: rename should_writeback_folio to folio_prepare_writeback based on
a comment from Jan Kara]
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Brian Foster <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Acked-by: Dave Chinner <[email protected]>
---
mm/page-writeback.c | 61 +++++++++++++++++++++++++--------------------
1 file changed, 34 insertions(+), 27 deletions(-)

diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index b49ee15a863e99..20ff00c8be9d90 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2360,6 +2360,38 @@ void tag_pages_for_writeback(struct address_space *mapping,
}
EXPORT_SYMBOL(tag_pages_for_writeback);

+static bool folio_prepare_writeback(struct address_space *mapping,
+ struct writeback_control *wbc, struct folio *folio)
+{
+ /*
+ * Folio truncated or invalidated. We can freely skip it then,
+ * even for data integrity operations: the folio has disappeared
+ * concurrently, so there could be no real expectation of this
+ * data integrity operation even if there is now a new, dirty
+ * folio at the same pagecache index.
+ */
+ if (unlikely(folio->mapping != mapping))
+ return false;
+
+ /*
+ * Did somebody else write it for us?
+ */
+ if (!folio_test_dirty(folio))
+ return false;
+
+ if (folio_test_writeback(folio)) {
+ if (wbc->sync_mode == WB_SYNC_NONE)
+ return false;
+ folio_wait_writeback(folio);
+ }
+ BUG_ON(folio_test_writeback(folio));
+
+ if (!folio_clear_dirty_for_io(folio))
+ return false;
+
+ return true;
+}
+
/**
* write_cache_pages - walk the list of dirty pages of the given address space and write all of them.
* @mapping: address space structure to write
@@ -2430,38 +2462,13 @@ int write_cache_pages(struct address_space *mapping,
for (i = 0; i < nr_folios; i++) {
folio = fbatch.folios[i];
folio_lock(folio);
-
- /*
- * Page truncated or invalidated. We can freely skip it
- * then, even for data integrity operations: the page
- * has disappeared concurrently, so there could be no
- * real expectation of this data integrity operation
- * even if there is now a new, dirty page at the same
- * pagecache address.
- */
- if (unlikely(folio->mapping != mapping)) {
-continue_unlock:
+ if (!folio_prepare_writeback(mapping, wbc, folio)) {
folio_unlock(folio);
continue;
}

- if (!folio_test_dirty(folio)) {
- /* someone wrote it for us */
- goto continue_unlock;
- }
-
- if (folio_test_writeback(folio)) {
- if (wbc->sync_mode != WB_SYNC_NONE)
- folio_wait_writeback(folio);
- else
- goto continue_unlock;
- }
-
- BUG_ON(folio_test_writeback(folio));
- if (!folio_clear_dirty_for_io(folio))
- goto continue_unlock;
-
trace_wbc_writepage(wbc, inode_to_bdi(mapping->host));
+
error = writepage(folio, wbc, data);
wbc->nr_to_write -= folio_nr_pages(folio);

--
2.39.2


2024-02-12 11:14:54

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 02/14] writeback: remove a duplicate prototype for tag_pages_for_writeback

From: "Matthew Wilcox (Oracle)" <[email protected]>

Signed-off-by: Christoph Hellwig <[email protected]>
[hch: split from a larger patch]
Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
Reviewed-by: Brian Foster <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Acked-by: Dave Chinner <[email protected]>
---
include/linux/writeback.h | 2 --
1 file changed, 2 deletions(-)

diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index 453736fd1d23ce..4b8cf9e4810bad 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -363,8 +363,6 @@ bool wb_over_bg_thresh(struct bdi_writeback *wb);
typedef int (*writepage_t)(struct folio *folio, struct writeback_control *wbc,
void *data);

-void tag_pages_for_writeback(struct address_space *mapping,
- pgoff_t start, pgoff_t end);
int write_cache_pages(struct address_space *mapping,
struct writeback_control *wbc, writepage_t writepage,
void *data);
--
2.39.2


2024-02-12 11:30:12

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 10/14] pagevec: Add ability to iterate a queue

From: "Matthew Wilcox (Oracle)" <[email protected]>

Add a loop counter inside the folio_batch to let us iterate from 0-nr
instead of decrementing nr and treating the batch as a stack. It would
generate some very weird and suboptimal I/O patterns for page writeback
to iterate over the batch as a stack.

Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Brian Foster <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Acked-by: Dave Chinner <[email protected]>
---
include/linux/pagevec.h | 18 ++++++++++++++++++
1 file changed, 18 insertions(+)

diff --git a/include/linux/pagevec.h b/include/linux/pagevec.h
index 87cc678adc850b..fcc06c300a72c3 100644
--- a/include/linux/pagevec.h
+++ b/include/linux/pagevec.h
@@ -27,6 +27,7 @@ struct folio;
*/
struct folio_batch {
unsigned char nr;
+ unsigned char i;
bool percpu_pvec_drained;
struct folio *folios[PAGEVEC_SIZE];
};
@@ -40,12 +41,14 @@ struct folio_batch {
static inline void folio_batch_init(struct folio_batch *fbatch)
{
fbatch->nr = 0;
+ fbatch->i = 0;
fbatch->percpu_pvec_drained = false;
}

static inline void folio_batch_reinit(struct folio_batch *fbatch)
{
fbatch->nr = 0;
+ fbatch->i = 0;
}

static inline unsigned int folio_batch_count(struct folio_batch *fbatch)
@@ -75,6 +78,21 @@ static inline unsigned folio_batch_add(struct folio_batch *fbatch,
return folio_batch_space(fbatch);
}

+/**
+ * folio_batch_next - Return the next folio to process.
+ * @fbatch: The folio batch being processed.
+ *
+ * Use this function to implement a queue of folios.
+ *
+ * Return: The next folio in the queue, or NULL if the queue is empty.
+ */
+static inline struct folio *folio_batch_next(struct folio_batch *fbatch)
+{
+ if (fbatch->i == fbatch->nr)
+ return NULL;
+ return fbatch->folios[fbatch->i++];
+}
+
void __folio_batch_release(struct folio_batch *pvec);

static inline void folio_batch_release(struct folio_batch *fbatch)
--
2.39.2


2024-02-12 11:41:55

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 09/14] writeback: Simplify the loops in write_cache_pages()

From: "Matthew Wilcox (Oracle)" <[email protected]>

Collapse the two nested loops into one. This is needed as a step
towards turning this into an iterator.

Note that this drops the "index <= end" check in the previous outer loop
and just relies on filemap_get_folios_tag() to return 0 entries when
index > end. This actually has a subtle implication when end == -1
because then the returned index will be -1 as well and thus if there is
page present on index -1, we could be looping indefinitely. But as the
comment in filemap_get_folios_tag documents this as already broken anyway
we should not worry about it here either. The fix for that would
probably a change to the filemap_get_folios_tag() calling convention.

Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
[hch: updated the commit log based on feedback from Jan Kara]
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Brian Foster <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Acked-by: Dave Chinner <[email protected]>
---
mm/page-writeback.c | 75 ++++++++++++++++++++++-----------------------
1 file changed, 36 insertions(+), 39 deletions(-)

diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 045ca252c0423d..a94a77b1805969 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2454,6 +2454,7 @@ int write_cache_pages(struct address_space *mapping,
int error;
struct folio *folio;
pgoff_t end; /* Inclusive */
+ int i = 0;

if (wbc->range_cyclic) {
wbc->index = mapping->writeback_index; /* prev offset */
@@ -2467,53 +2468,49 @@ int write_cache_pages(struct address_space *mapping,

folio_batch_init(&wbc->fbatch);

- while (wbc->index <= end) {
- int i;
-
- writeback_get_batch(mapping, wbc);
-
+ for (;;) {
+ if (i == wbc->fbatch.nr) {
+ writeback_get_batch(mapping, wbc);
+ i = 0;
+ }
if (wbc->fbatch.nr == 0)
break;

- for (i = 0; i < wbc->fbatch.nr; i++) {
- folio = wbc->fbatch.folios[i];
+ folio = wbc->fbatch.folios[i++];

- folio_lock(folio);
- if (!folio_prepare_writeback(mapping, wbc, folio)) {
- folio_unlock(folio);
- continue;
- }
+ folio_lock(folio);
+ if (!folio_prepare_writeback(mapping, wbc, folio)) {
+ folio_unlock(folio);
+ continue;
+ }

- trace_wbc_writepage(wbc, inode_to_bdi(mapping->host));
+ trace_wbc_writepage(wbc, inode_to_bdi(mapping->host));

- error = writepage(folio, wbc, data);
- wbc->nr_to_write -= folio_nr_pages(folio);
+ error = writepage(folio, wbc, data);
+ wbc->nr_to_write -= folio_nr_pages(folio);

- if (error == AOP_WRITEPAGE_ACTIVATE) {
- folio_unlock(folio);
- error = 0;
- }
+ if (error == AOP_WRITEPAGE_ACTIVATE) {
+ folio_unlock(folio);
+ error = 0;
+ }

- /*
- * For integrity writeback we have to keep going until
- * we have written all the folios we tagged for
- * writeback above, even if we run past wbc->nr_to_write
- * or encounter errors.
- * We stash away the first error we encounter in
- * wbc->saved_err so that it can be retrieved when we're
- * done. This is because the file system may still have
- * state to clear for each folio.
- *
- * For background writeback we exit as soon as we run
- * past wbc->nr_to_write or encounter the first error.
- */
- if (wbc->sync_mode == WB_SYNC_ALL) {
- if (error && !ret)
- ret = error;
- } else {
- if (error || wbc->nr_to_write <= 0)
- goto done;
- }
+ /*
+ * For integrity writeback we have to keep going until we have
+ * written all the folios we tagged for writeback above, even if
+ * we run past wbc->nr_to_write or encounter errors.
+ * We stash away the first error we encounter in wbc->saved_err
+ * so that it can be retrieved when we're done. This is because
+ * the file system may still have state to clear for each folio.
+ *
+ * For background writeback we exit as soon as we run past
+ * wbc->nr_to_write or encounter the first error.
+ */
+ if (wbc->sync_mode == WB_SYNC_ALL) {
+ if (error && !ret)
+ ret = error;
+ } else {
+ if (error || wbc->nr_to_write <= 0)
+ goto done;
}
}

--
2.39.2


2024-02-12 12:17:02

by Christoph Hellwig

[permalink] [raw]
Subject: [PATCH 14/14] writeback: Remove a use of write_cache_pages() from do_writepages()

From: "Matthew Wilcox (Oracle)" <[email protected]>

Use the new writeback_iter() directly instead of indirecting
through a callback.

Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
[hch: ported to the while based iter style]
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Brian Foster <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
---
mm/page-writeback.c | 27 +++++++++++++++++----------
1 file changed, 17 insertions(+), 10 deletions(-)

diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 1996200849e577..2fd83d438f92bd 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2577,12 +2577,24 @@ int write_cache_pages(struct address_space *mapping,
}
EXPORT_SYMBOL(write_cache_pages);

-static int writepage_cb(struct folio *folio, struct writeback_control *wbc,
- void *data)
+static int writeback_use_writepage(struct address_space *mapping,
+ struct writeback_control *wbc)
{
- struct address_space *mapping = data;
+ struct folio *folio = NULL;
+ struct blk_plug plug;
+ int err;

- return mapping->a_ops->writepage(&folio->page, wbc);
+ blk_start_plug(&plug);
+ while ((folio = writeback_iter(mapping, wbc, folio, &err))) {
+ err = mapping->a_ops->writepage(&folio->page, wbc);
+ if (err == AOP_WRITEPAGE_ACTIVATE) {
+ folio_unlock(folio);
+ err = 0;
+ }
+ }
+ blk_finish_plug(&plug);
+
+ return err;
}

int do_writepages(struct address_space *mapping, struct writeback_control *wbc)
@@ -2598,12 +2610,7 @@ int do_writepages(struct address_space *mapping, struct writeback_control *wbc)
if (mapping->a_ops->writepages) {
ret = mapping->a_ops->writepages(mapping, wbc);
} else if (mapping->a_ops->writepage) {
- struct blk_plug plug;
-
- blk_start_plug(&plug);
- ret = write_cache_pages(mapping, wbc, writepage_cb,
- mapping);
- blk_finish_plug(&plug);
+ ret = writeback_use_writepage(mapping, wbc);
} else {
/* deal with chardevs and other special files */
ret = 0;
--
2.39.2


2024-02-13 13:07:48

by Jan Kara

[permalink] [raw]
Subject: Re: [PATCH 01/14] writeback: don't call mapping_set_error in writepage_cb

On Mon 12-02-24 08:13:35, Christoph Hellwig wrote:
> writepage_cb is the iterator callback for write_cache_pages, which
> already tracks all errors and returns them to the caller. There is
> no need to additionally cal mapping_set_error which is intended
^^^ call

> for contexts where the error can't be directly returned (e.g. the
> I/O completion handlers).
>
> Remove the mapping_set_error call in writepage_cb which is not only
> superfluous but also buggy as it can be called with the error argument
> set to AOP_WRITEPAGE_ACTIVATE, which is not actually an error but a
> magic return value asking the caller to unlock the page.
>
> Signed-off-by: Christoph Hellwig <[email protected]>

Our error handling in writeback has always been ... spotty. E.g.
block_write_full_page() and iomap_writepage_map() call mapping_set_error()
as well so this seems to be a common way to do things, OTOH ext4 calls
mapping_set_error() only on IO completion. I guess the question is how
an error in ->writepages from background writeback should propagate to
eventual fsync(2) caller? Because currently such error propagates all the
way up to writeback_sb_inodes() where it is silently dropped...

Honza

> ---
> mm/page-writeback.c | 5 ++---
> 1 file changed, 2 insertions(+), 3 deletions(-)
>
> diff --git a/mm/page-writeback.c b/mm/page-writeback.c
> index 3f255534986a2f..62901fa905f01e 100644
> --- a/mm/page-writeback.c
> +++ b/mm/page-writeback.c
> @@ -2534,9 +2534,8 @@ static int writepage_cb(struct folio *folio, struct writeback_control *wbc,
> void *data)
> {
> struct address_space *mapping = data;
> - int ret = mapping->a_ops->writepage(&folio->page, wbc);
> - mapping_set_error(mapping, ret);
> - return ret;
> +
> + return mapping->a_ops->writepage(&folio->page, wbc);
> }
>
> int do_writepages(struct address_space *mapping, struct writeback_control *wbc)
> --
> 2.39.2
>
--
Jan Kara <[email protected]>
SUSE Labs, CR

2024-02-13 13:14:18

by Brian Foster

[permalink] [raw]
Subject: Re: [PATCH 01/14] writeback: don't call mapping_set_error in writepage_cb

On Tue, Feb 13, 2024 at 02:07:13PM +0100, Jan Kara wrote:
> On Mon 12-02-24 08:13:35, Christoph Hellwig wrote:
> > writepage_cb is the iterator callback for write_cache_pages, which
> > already tracks all errors and returns them to the caller. There is
> > no need to additionally cal mapping_set_error which is intended
> ^^^ call
>
> > for contexts where the error can't be directly returned (e.g. the
> > I/O completion handlers).
> >
> > Remove the mapping_set_error call in writepage_cb which is not only
> > superfluous but also buggy as it can be called with the error argument
> > set to AOP_WRITEPAGE_ACTIVATE, which is not actually an error but a
> > magic return value asking the caller to unlock the page.
> >
> > Signed-off-by: Christoph Hellwig <[email protected]>
>
> Our error handling in writeback has always been ... spotty. E.g.
> block_write_full_page() and iomap_writepage_map() call mapping_set_error()
> as well so this seems to be a common way to do things, OTOH ext4 calls
> mapping_set_error() only on IO completion. I guess the question is how
> an error in ->writepages from background writeback should propagate to
> eventual fsync(2) caller? Because currently such error propagates all the
> way up to writeback_sb_inodes() where it is silently dropped...
>

A couple related notes from skimming around:

- Things like iomap might make this call in I/O completion paths, but
then invoke bio completion paths on submission side errors (i.e.
iomap_submit_ioend() -> bio_endio()).
- __writeback_single_inode() calls filemap_fdatawait() shortly after
do_writepages(), which basically looks like it relies on mapping error
state to propagate error within the writeback path.

The call removed by this path only seems to apply to contexts that don't
define their own .writepages, so it's not clear to me how much this
really matters. It just seems like it's a little hard to quantify
whether this is an undesireable change in behavior or not.

Brian

> Honza
>
> > ---
> > mm/page-writeback.c | 5 ++---
> > 1 file changed, 2 insertions(+), 3 deletions(-)
> >
> > diff --git a/mm/page-writeback.c b/mm/page-writeback.c
> > index 3f255534986a2f..62901fa905f01e 100644
> > --- a/mm/page-writeback.c
> > +++ b/mm/page-writeback.c
> > @@ -2534,9 +2534,8 @@ static int writepage_cb(struct folio *folio, struct writeback_control *wbc,
> > void *data)
> > {
> > struct address_space *mapping = data;
> > - int ret = mapping->a_ops->writepage(&folio->page, wbc);
> > - mapping_set_error(mapping, ret);
> > - return ret;
> > +
> > + return mapping->a_ops->writepage(&folio->page, wbc);
> > }
> >
> > int do_writepages(struct address_space *mapping, struct writeback_control *wbc)
> > --
> > 2.39.2
> >
> --
> Jan Kara <[email protected]>
> SUSE Labs, CR
>