From: Eric Sandeen <sandeen@redhat.com>
To: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Cc: Andrew Morton <akpm@osdl.org>
Subject: [PATCH] reduce large do_mount stack usage with noinlines
Date: Wed, 06 Feb 2008 16:13:58 -0600 [thread overview]
Message-ID: <47AA3126.8090102@redhat.com> (raw)
do_mount() uses a whopping 616 bytes of stack on x86_64 in
2.6.24-mm1, largely thanks to gcc inlining the various helper
functions.
noinlining these can slim it down a lot; on my box this patch
gets it down to 168, which is mostly the struct nameidata nd;
left on the stack.
These functions are called only as do_mount() helpers;
none of them should be in any path that would see a performance
benefit from inlining...
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
---
Index: linux-2.6.24-mm1/fs/namespace.c
===================================================================
--- linux-2.6.24-mm1.orig/fs/namespace.c
+++ linux-2.6.24-mm1/fs/namespace.c
@@ -1296,7 +1296,7 @@ out_unlock:
/*
* recursively change the type of the mountpoint.
*/
-static int do_change_type(struct nameidata *nd, int flag)
+static noinline int do_change_type(struct nameidata *nd, int flag)
{
struct vfsmount *m, *mnt = nd->path.mnt;
int recurse = flag & MS_REC;
@@ -1320,7 +1320,7 @@ static int do_change_type(struct nameida
/*
* do loopback mount.
*/
-static int do_loopback(struct nameidata *nd, char *old_name, int recurse)
+static noinline int do_loopback(struct nameidata *nd, char *old_name, int recurse)
{
struct nameidata old_nd;
struct vfsmount *mnt = NULL;
@@ -1387,7 +1387,7 @@ static int change_mount_flags(struct vfs
* If you've mounted a non-root directory somewhere and want to do remount
* on it - tough luck.
*/
-static int do_remount(struct nameidata *nd, int flags, int mnt_flags,
+static noinline int do_remount(struct nameidata *nd, int flags, int mnt_flags,
void *data)
{
int err;
@@ -1425,7 +1425,7 @@ static inline int tree_contains_unbindab
return 0;
}
-static int do_move_mount(struct nameidata *nd, char *old_name)
+static noinline int do_move_mount(struct nameidata *nd, char *old_name)
{
struct nameidata old_nd, parent_nd;
struct vfsmount *p;
@@ -1505,7 +1505,7 @@ out:
* create a new mount for userspace and request it to be added into the
* namespace's tree
*/
-static int do_new_mount(struct nameidata *nd, char *type, int flags,
+static noinline int do_new_mount(struct nameidata *nd, char *type, int flags,
int mnt_flags, char *name, void *data)
{
struct vfsmount *mnt;
next reply other threads:[~2008-02-06 22:14 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-02-06 22:13 Eric Sandeen [this message]
2008-02-06 22:34 ` [PATCH] reduce large do_mount stack usage with noinlines Andrew Morton
2008-02-06 22:54 ` Arjan van de Ven
2008-02-06 23:01 ` Eric Sandeen
2008-02-06 22:55 ` Eric Sandeen
2008-02-06 23:11 ` Eric Sandeen
2008-02-06 23:22 ` Andrew Morton
2008-02-06 23:34 ` Eric Sandeen
2008-02-06 23:46 ` Andrew Morton
2008-02-07 23:08 ` Eric Sandeen
2008-02-07 23:23 ` Arjan van de Ven
2008-02-07 23:26 ` Andrew Morton
2008-02-08 16:50 ` Andi Kleen
2008-02-08 16:54 ` Eric Sandeen
2008-02-08 17:23 ` Al Viro
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=47AA3126.8090102@redhat.com \
--to=sandeen@redhat.com \
--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.