Subject: + readahead-fix-sequential-read-cache-miss-detection.patch added to -mm tree
To: damien.ramonda@xxxxxxxxx,david.a.cohen@xxxxxxxxxxxxxxx,fengguang.wu@xxxxxxxxx,pierre.tardy@xxxxxxxxx
From: akpm@xxxxxxxxxxxxxxxxxxxx
Date: Thu, 17 Oct 2013 13:23:01 -0700
The patch titled
Subject: readahead: fix sequential read cache miss detection
has been added to the -mm tree. Its filename is
readahead-fix-sequential-read-cache-miss-detection.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/readahead-fix-sequential-read-cache-miss-detection.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/readahead-fix-sequential-read-cache-miss-detection.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Damien Ramonda <damien.ramonda@xxxxxxxxx>
Subject: readahead: fix sequential read cache miss detection
The kernel's readahead algorithm sometimes interprets random read accesses
as sequential and triggers unnecessary data prefecthing from storage
device (impacting random read average latency).
In order to identify sequential cache read misses, the readahead algorithm
intends to check whether offset - previous offset == 1 (trivial sequential
reads) or offset - previous offset == 0 (sequential reads not aligned on
page boundary):
if (offset - (ra->prev_pos >> PAGE_CACHE_SHIFT) <= 1UL)
The current offset is stored in the "offset" variable of type "pgoff_t"
(unsigned long), while previous offset is stored in "ra->prev_pos" of type
"loff_t" (long long). Therefore, operands of the if statement are
implicitly converted to type long long. Consequently, when previous
offset > current offset (which happens on random pattern), the if
condition is true and access is wrongly interpeted as sequential. An
unnecessary data prefetching is triggered, impacting the average random
read latency.
Storing the previous offset value in a "pgoff_t" variable (unsigned long)
fixes the sequential read detection logic.
Signed-off-by: Damien Ramonda <damien.ramonda@xxxxxxxxx>
Reviewed-by: Fengguang Wu <fengguang.wu@xxxxxxxxx>
Acked-by: Pierre Tardy <pierre.tardy@xxxxxxxxx>
Acked-by: David Cohen <david.a.cohen@xxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/readahead.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff -puN mm/readahead.c~readahead-fix-sequential-read-cache-miss-detection mm/readahead.c
--- a/mm/readahead.c~readahead-fix-sequential-read-cache-miss-detection
+++ a/mm/readahead.c
@@ -401,6 +401,7 @@ ondemand_readahead(struct address_space
unsigned long req_size)
{
unsigned long max = max_sane_readahead(ra->ra_pages);
+ pgoff_t prev_offset;
/*
* start of file
@@ -452,8 +453,11 @@ ondemand_readahead(struct address_space
/*
* sequential cache miss
+ * trivial case: (offset - prev_offset) == 1
+ * unaligned reads: (offset - prev_offset) == 0
*/
- if (offset - (ra->prev_pos >> PAGE_CACHE_SHIFT) <= 1UL)
+ prev_offset = (unsigned long long)ra->prev_pos >> PAGE_CACHE_SHIFT;
+ if (offset - prev_offset <= 1UL)
goto initial_readahead;
/*
_
Patches currently in -mm which might be from damien.ramonda@xxxxxxxxx are
readahead-fix-sequential-read-cache-miss-detection.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
