erofs: fix up inplace decompression success rate

Message ID 20221014064915.8103-1-hsiangkao@linux.alibaba.com
State New
Headers
Series erofs: fix up inplace decompression success rate |

Commit Message

Gao Xiang Oct. 14, 2022, 6:49 a.m. UTC
  Partial decompression should be checked after updating length.
It's a new regression when introducing multi-reference pclusters.

Fixes: 2bfab9c0edac ("erofs: record the longest decompressed size in this round")
Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
---
 fs/erofs/zdata.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)
  

Comments

Chao Yu Oct. 15, 2022, 2:51 a.m. UTC | #1
On 2022/10/14 14:49, Gao Xiang wrote:
> Partial decompression should be checked after updating length.
> It's a new regression when introducing multi-reference pclusters.
> 
> Fixes: 2bfab9c0edac ("erofs: record the longest decompressed size in this round")
> Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>

Reviewed-by: Chao Yu <chao@kernel.org>

Thanks,
  

Patch

diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c
index cce56dde135c..2417d7b73622 100644
--- a/fs/erofs/zdata.c
+++ b/fs/erofs/zdata.c
@@ -812,15 +812,14 @@  static int z_erofs_do_read_page(struct z_erofs_decompress_frontend *fe,
 	++spiltted;
 	if (fe->pcl->pageofs_out != (map->m_la & ~PAGE_MASK))
 		fe->pcl->multibases = true;
-
-	if ((map->m_flags & EROFS_MAP_FULL_MAPPED) &&
-	    !(map->m_flags & EROFS_MAP_PARTIAL_REF) &&
-	    fe->pcl->length == map->m_llen)
-		fe->pcl->partial = false;
 	if (fe->pcl->length < offset + end - map->m_la) {
 		fe->pcl->length = offset + end - map->m_la;
 		fe->pcl->pageofs_out = map->m_la & ~PAGE_MASK;
 	}
+	if ((map->m_flags & EROFS_MAP_FULL_MAPPED) &&
+	    !(map->m_flags & EROFS_MAP_PARTIAL_REF) &&
+	    fe->pcl->length == map->m_llen)
+		fe->pcl->partial = false;
 next_part:
 	/* shorten the remaining extent to update progress */
 	map->m_llen = offset + cur - map->m_la;