summaryrefslogtreecommitdiffstats
path: root/mm/fadvise.c
diff options
context:
space:
mode:
authorMel Gorman <mgorman@suse.de>2014-12-13 01:56:33 +0100
committerLinus Torvalds <torvalds@linux-foundation.org>2014-12-13 21:42:49 +0100
commit441c228f817f7597e090d84aca74bdb7c2bd5040 (patch)
tree2b03cb17475aeb302d230e4c36dcd18a374a05f6 /mm/fadvise.c
parentmm/vmalloc.c: fix memory ordering bug (diff)
downloadlinux-441c228f817f7597e090d84aca74bdb7c2bd5040.tar.xz
linux-441c228f817f7597e090d84aca74bdb7c2bd5040.zip
mm: fadvise: document the fadvise(FADV_DONTNEED) behaviour for partial pages
A random seek IO benchmark appeared to regress because of a change to readahead but the real problem was the benchmark. To ensure the IO request accesssed disk, it used fadvise(FADV_DONTNEED) on a block boundary (512K) but the hint is ignored by the kernel. This is correct but not necessarily obvious behaviour. As much as I dislike comment patches, the explanation for this behaviour predates current git history. Clarify why it behaves like this in case someone "fixes" fadvise or readahead for the wrong reasons. Signed-off-by: Mel Gorman <mgorman@suse.de> Cc: Michael Kerrisk <mtk.manpages@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to '')
-rw-r--r--mm/fadvise.c6
1 files changed, 5 insertions, 1 deletions
diff --git a/mm/fadvise.c b/mm/fadvise.c
index 3bcfd81db45e..2ad7adf4f0a4 100644
--- a/mm/fadvise.c
+++ b/mm/fadvise.c
@@ -117,7 +117,11 @@ SYSCALL_DEFINE4(fadvise64_64, int, fd, loff_t, offset, loff_t, len, int, advice)
__filemap_fdatawrite_range(mapping, offset, endbyte,
WB_SYNC_NONE);
- /* First and last FULL page! */
+ /*
+ * First and last FULL page! Partial pages are deliberately
+ * preserved on the expectation that it is better to preserve
+ * needed memory than to discard unneeded memory.
+ */
start_index = (offset+(PAGE_CACHE_SIZE-1)) >> PAGE_CACHE_SHIFT;
end_index = (endbyte >> PAGE_CACHE_SHIFT);