From 215dcc0f98913e0b26b328472505569b1b3957e9 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 14 Jul 2022 15:35:24 -0700 Subject: [PATCH] Revert "vf/remap: return the amount of bytes actually deduplicated" MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit This reverts commit f89377f8ca5189bb7c017cbc2f612026d150a768. Dave Chinner reports: "As I suspected would occur, this change causes test failures. e.g generic/517 in fstests fails with: generic/517 1s ... - output mismatch [..] -deduped 131172/131172 bytes at offset 65536 +deduped 131072/131172 bytes at offset 65536" can you please revert this commit for the 5.19 series to give us more time to investigate and consider the impact of the the API change on userspace applications before we commit to changing the API" That changed return value seems to reflect reality, but with the fstest change, let's revert for now. Requested-by: Dave Chinner Link: https://lore.kernel.org/all/20220714223238.GH3600936@dread.disaster.area/ Cc: Ansgar Lößer Signed-off-by: Linus Torvalds --- fs/remap_range.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/remap_range.c b/fs/remap_range.c index 5e0d97e02f968..881a306ee2473 100644 --- a/fs/remap_range.c +++ b/fs/remap_range.c @@ -547,7 +547,7 @@ int vfs_dedupe_file_range(struct file *file, struct file_dedupe_range *same) else if (deduped < 0) info->status = deduped; else - info->bytes_deduped = deduped; + info->bytes_deduped = len; next_fdput: fdput(dst_fd); -- 2.39.5