[lustre-devel] [PATCH 10/18] lustre: readahead: fix to reserve min pages
James Simmons
jsimmons at infradead.org
Mon Jul 19 05:32:05 PDT 2021
From: Wang Shilong <wshilong at ddn.com>
@pages_min might be larger than @pages which indicate
more pages should be read, and it will cause a warning
later.
WC-bug-id: https://jira.whamcloud.com/browse/LU-14778
Lustre-commit: 4fc127428f00d6a3 ("LU-14778 readahead: fix to reserve min pages")
Signed-off-by: Wang Shilong <wshilong at ddn.com>
Reviewed-on: https://review.whamcloud.com/44050
Reviewed-by: Andreas Dilger <adilger at whamcloud.com>
Reviewed-by: Bobi Jam <bobijam at hotmail.com>
Reviewed-by: Oleg Drokin <green at whamcloud.com>
Signed-off-by: James Simmons <jsimmons at infradead.org>
---
fs/lustre/llite/rw.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/fs/lustre/llite/rw.c b/fs/lustre/llite/rw.c
index 184e5e8..4de77f6 100644
--- a/fs/lustre/llite/rw.c
+++ b/fs/lustre/llite/rw.c
@@ -85,8 +85,9 @@ static unsigned long ll_ra_count_get(struct ll_sb_info *sbi,
struct ll_ra_info *ra = &sbi->ll_ra_info;
long ret;
+ WARN_ON_ONCE(pages_min > pages);
/**
- * Don't try readahead agreesively if we are limited
+ * Don't try readahead aggresively if we are limited
* LRU pages, otherwise, it could cause deadlock.
*/
pages = min(sbi->ll_cache->ccc_lru_max >> 2, pages);
@@ -95,7 +96,7 @@ static unsigned long ll_ra_count_get(struct ll_sb_info *sbi,
* this will make us leak @ra_cur_pages, because
* ll_ra_count_put() acutally freed @pages.
*/
- if (WARN_ON_ONCE(pages_min > pages))
+ if (unlikely(pages_min > pages))
pages_min = pages;
/*
@@ -829,7 +830,8 @@ static int ll_readahead(const struct lu_env *env, struct cl_io *io,
/* don't over reserved for mmap range read */
if (skip_index)
pages_min = 0;
-
+ if (pages_min > pages)
+ pages = pages_min;
ria->ria_reserved = ll_ra_count_get(ll_i2sbi(inode), ria, pages,
pages_min);
if (ria->ria_reserved < pages)
--
1.8.3.1
More information about the lustre-devel
mailing list