spi: spi-fsl-dspi: Target mode improvements

Merge series from James Clark <james.clark@linaro.org>:

Improve usability of target mode by reporting FIFO errors and increasing
the buffer size when DMA is used. While we're touching DMA stuff also
switch to non-coherent memory, although this is unrelated to target
mode.

With the combination of the commit to increase the DMA buffer size and
the commit to use non-coherent memory, the host mode performance figures
are as follows on S32G3:

  # spidev_test --device /dev/spidev1.0 --bpw 8 --size <test_size> --cpha --iter 10000000 --speed 10000000

  Coherent (4096 byte transfers): 6534 kbps
  Non-coherent:                   7347 kbps

  Coherent (16 byte transfers):    447 kbps
  Non-coherent:                    448 kbps

Just for comparison running the same test in XSPI mode:

  4096 byte transfers:            2143 kbps
  16 byte transfers:               637 kbps

These tests required hacking S32G3 to use DMA in host mode, although
the figures should be representative of target mode too where DMA is
used. And the other devices that use DMA in host mode should see similar
improvements.

1 files changed, 19 insertions, 5 deletions

diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index f23d75986947..c953297aa89a 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -1512,7 +1512,7 @@ out:
 
 /*
  * Return 0 if we have submitted or queued the sector for submission.
- * Return <0 for critical errors.
+ * Return <0 for critical errors, and the sector will have its dirty flag cleared.
  *
  * Caller should make sure filepos < i_size and handle filepos >= i_size case.
  */
@@ -1535,8 +1535,17 @@ static int submit_one_sector(struct btrfs_inode *inode,
 	ASSERT(filepos < i_size);
 
 	em = btrfs_get_extent(inode, NULL, filepos, sectorsize);
-	if (IS_ERR(em))
+	if (IS_ERR(em)) {
+		/*
+		 * When submission failed, we should still clear the folio dirty.
+		 * Or the folio will be written back again but without any
+		 * ordered extent.
+		 */
+		btrfs_folio_clear_dirty(fs_info, folio, filepos, sectorsize);
+		btrfs_folio_set_writeback(fs_info, folio, filepos, sectorsize);
+		btrfs_folio_clear_writeback(fs_info, folio, filepos, sectorsize);
 		return PTR_ERR(em);
+	}
 
 	extent_offset = filepos - em->start;
 	em_end = btrfs_extent_map_end(em);
@@ -1609,8 +1618,12 @@ static noinline_for_stack int extent_writepage_io(struct btrfs_inode *inode,
 		folio_unlock(folio);
 		return 1;
 	}
-	if (ret < 0)
+	if (ret < 0) {
+		btrfs_folio_clear_dirty(fs_info, folio, start, len);
+		btrfs_folio_set_writeback(fs_info, folio, start, len);
+		btrfs_folio_clear_writeback(fs_info, folio, start, len);
 		return ret;
+	}
 
 	for (cur = start; cur < start + len; cur += fs_info->sectorsize)
 		set_bit((cur - folio_start) >> fs_info->sectorsize_bits, &range_bitmap);
@@ -1666,8 +1679,8 @@ static noinline_for_stack int extent_writepage_io(struct btrfs_inode *inode,
 	 * Here we set writeback and clear for the range. If the full folio
 	 * is no longer dirty then we clear the PAGECACHE_TAG_DIRTY tag.
 	 *
-	 * If we hit any error, the corresponding sector will still be dirty
-	 * thus no need to clear PAGECACHE_TAG_DIRTY.
+	 * If we hit any error, the corresponding sector will have its dirty
+	 * flag cleared and writeback finished, thus no need to handle the error case.
 	 */
 	if (!submitted_io && !error) {
 		btrfs_folio_set_writeback(fs_info, folio, start, len);
@@ -1813,6 +1826,7 @@ static noinline_for_stack bool lock_extent_buffer_for_io(struct extent_buffer *e
 		xas_load(&xas);
 		xas_set_mark(&xas, PAGECACHE_TAG_WRITEBACK);
 		xas_clear_mark(&xas, PAGECACHE_TAG_DIRTY);
+		xas_clear_mark(&xas, PAGECACHE_TAG_TOWRITE);
 		xas_unlock_irqrestore(&xas, flags);
 
 		btrfs_set_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN);