From: Wu Fengguang <wfg@mail.ustc.edu.cn>
To: Andrew Morton <akpm@osdl.org>
Cc: linux-kernel@vger.kernel.org, Wu Fengguang <wfg@mail.ustc.edu.cn>
Subject: [PATCH 11/28] readahead: min/max sizes
Date: Wed, 15 Nov 2006 15:50:18 +0800 [thread overview]
Message-ID: <363577024.11779@ustc.edu.cn> (raw)
Message-ID: <20061115075027.832896629@localhost.localdomain> (raw)
In-Reply-To: 20061115075007.832957580@localhost.localdomain
[-- Attachment #1: readahead-min-max-sizes.patch --]
[-- Type: text/plain, Size: 2096 bytes --]
- Enlarge VM_MAX_READAHEAD to 1024 if new read-ahead code is compiled in.
This value is no longer tightly coupled with the thrashing problem,
therefore constrained by it. The adaptive read-ahead logic merely takes
it as an upper bound, and will not stick to it under memory pressure.
- Slightly enlarge minimal/initial read-ahead size on big memory systems.
Memory bounty systems are less likely to suffer from thrashing on small
read-ahead sizes. A bigger initial value helps the ra_size scaling up
progress.
Signed-off-by: Wu Fengguang <wfg@mail.ustc.edu.cn>
Signed-off-by: Andrew Morton <akpm@osdl.org>
--- linux-2.6.19-rc5-mm2.orig/include/linux/mm.h
+++ linux-2.6.19-rc5-mm2/include/linux/mm.h
@@ -1046,7 +1046,11 @@ extern int filemap_populate(struct vm_ar
int write_one_page(struct page *page, int wait);
/* readahead.c */
+#ifdef CONFIG_ADAPTIVE_READAHEAD
+#define VM_MAX_READAHEAD 1024 /* kbytes */
+#else
#define VM_MAX_READAHEAD 128 /* kbytes */
+#endif
#define VM_MIN_READAHEAD 16 /* kbytes (includes current page) */
#define VM_MAX_CACHE_HIT 256 /* max pages in a row in cache before
* turning readahead off */
--- linux-2.6.19-rc5-mm2.orig/mm/readahead.c
+++ linux-2.6.19-rc5-mm2/mm/readahead.c
@@ -814,6 +814,30 @@ out:
return nr_pages;
}
+/*
+ * ra_min is mainly determined by the size of cache memory. Reasonable?
+ *
+ * Table of concrete numbers for 4KB page size:
+ * inactive + free (MB): 4 8 16 32 64 128 256 512 1024
+ * ra_min (KB): 16 16 16 16 20 24 32 48 64
+ */
+static inline void get_readahead_bounds(struct file_ra_state *ra,
+ unsigned long *ra_min,
+ unsigned long *ra_max)
+{
+ unsigned long active;
+ unsigned long inactive;
+ unsigned long free;
+
+ __get_zone_counts(&active, &inactive, &free, NODE_DATA(numa_node_id()));
+
+ free += inactive;
+ *ra_max = min(min(ra->ra_pages, 0xFFFFUL), free / 2);
+ *ra_min = min(min(MIN_RA_PAGES + (free >> 14),
+ DIV_ROUND_UP(64*1024, PAGE_CACHE_SIZE)),
+ *ra_max / 8);
+}
+
#endif /* CONFIG_ADAPTIVE_READAHEAD */
/*
--
next prev parent reply other threads:[~2006-11-15 7:50 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-11-15 7:50 [PATCH 00/28] Adaptive readahead V16 Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 01/28] readahead: kconfig options Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 02/28] radixtree: introduce scan hole/data functions Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 03/28] mm: introduce probe_page() Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 04/28] mm: introduce PG_readahead Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 06/28] readahead: insert cond_resched() calls Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 09/28] readahead: rescue_pages() Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang [this message]
2006-11-15 7:50 ` [PATCH 11/28] readahead: min/max sizes Wu Fengguang
2006-11-15 7:50 ` [PATCH 12/28] readahead: state based method - aging accounting Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 16:54 ` Christoph Lameter
2006-11-16 13:39 ` Wu Fengguang
2006-11-16 13:39 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 13/28] readahead: state based method - routines Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 14/28] readahead: state based method Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 15/28] readahead: context " Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 16/28] readahead: initial method - guiding sizes Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 17/28] readahead: initial method - thrashing guard size Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 18/28] readahead: initial method - user recommended size Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 20/28] readahead: backward prefetching method Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 21/28] readahead: thrashing recovery method Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 22/28] readahead: call scheme Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 23/28] readahead: laptop mode Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 24/28] readahead: loop case Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 25/28] readahead: nfsd case Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
2006-11-15 7:50 ` [PATCH 26/28] readahead: turn on by default Wu Fengguang
2006-11-15 7:50 ` Wu Fengguang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=363577024.11779@ustc.edu.cn \
--to=wfg@mail.ustc.edu.cn \
--cc=akpm@osdl.org \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.