From: Andrew Morton <akpm@osdl.org>
To: Al Viro <viro@ftp.linux.org.uk>,
Andrew MChuck Ebbert <76306.1226@compuserve.com>,
linux-kernel <linux-kernel@vger.kernel.org>,
"Linus Torvalds orton <akpm@osdl.org>" <torvalds@osdl.org>
Subject: Re: [patch] pipe: Don't oops when pipe filesystem isn't mounted
Date: Mon, 11 Dec 2006 02:27:46 -0800 [thread overview]
Message-ID: <20061211022746.9ec80c03.akpm@osdl.org> (raw)
In-Reply-To: <20061211021718.a6954106.akpm@osdl.org>
On Mon, 11 Dec 2006 02:17:18 -0800
Andrew Morton <akpm@osdl.org> wrote:
> > Said that, I think that pipes should be initialized early.
>
> Judging by the comment there, the only reason we prepare the rootfs prior
> to running initcalls is for firmware. So the sequence
>
> run initcalls
> populate rootfs
> run initcalls which want to access files
>
> fixes everything?
Like this... The other part is hunting down all those drivers which want
to access the filesystem at init time.
include/asm-generic/vmlinux.lds.h | 6 +++++-
include/linux/init.h | 5 +++++
init/initramfs.c | 3 ++-
init/main.c | 7 -------
4 files changed, 12 insertions(+), 9 deletions(-)
diff -puN init/main.c~dont-run-userspace-until-initcalls-have-completed init/main.c
--- a/init/main.c~dont-run-userspace-until-initcalls-have-completed
+++ a/init/main.c
@@ -94,7 +94,6 @@ extern void pidmap_init(void);
extern void prio_tree_init(void);
extern void radix_tree_init(void);
extern void free_initmem(void);
-extern void populate_rootfs(void);
extern void driver_init(void);
extern void prepare_namespace(void);
#ifdef CONFIG_ACPI
@@ -739,12 +738,6 @@ static int init(void * unused)
cpuset_init_smp();
- /*
- * Do this before initcalls, because some drivers want to access
- * firmware files.
- */
- populate_rootfs();
-
do_basic_setup();
/*
diff -puN include/linux/init.h~dont-run-userspace-until-initcalls-have-completed include/linux/init.h
--- a/include/linux/init.h~dont-run-userspace-until-initcalls-have-completed
+++ a/include/linux/init.h
@@ -115,6 +115,11 @@ extern void setup_arch(char **);
#define device_initcall_sync(fn) __define_initcall("6s",fn,6s)
#define late_initcall(fn) __define_initcall("7",fn,7)
#define late_initcall_sync(fn) __define_initcall("7s",fn,7s)
+#define populate_rootfs_initcall(fn) __define_initcall("8",fn,8)
+#define populate_rootfs_initcall_sync(fn) __define_initcall("8s",fn,8s)
+#define rootfs_neeeded_initcall(fn) __define_initcall("9",fn,9)
+#define rootfs_neeeded_initcall_sync(fn) __define_initcall("9s",fn,9s)
+
#define __initcall(fn) device_initcall(fn)
diff -puN include/asm-generic/vmlinux.lds.h~dont-run-userspace-until-initcalls-have-completed include/asm-generic/vmlinux.lds.h
--- a/include/asm-generic/vmlinux.lds.h~dont-run-userspace-until-initcalls-have-completed
+++ a/include/asm-generic/vmlinux.lds.h
@@ -245,5 +245,9 @@
*(.initcall6.init) \
*(.initcall6s.init) \
*(.initcall7.init) \
- *(.initcall7s.init)
+ *(.initcall7s.init) \
+ *(.initcall8.init) \
+ *(.initcall8s.init) \
+ *(.initcall9.init) \
+ *(.initcall9s.init)
diff -puN init/initramfs.c~dont-run-userspace-until-initcalls-have-completed init/initramfs.c
--- a/init/initramfs.c~dont-run-userspace-until-initcalls-have-completed
+++ a/init/initramfs.c
@@ -526,7 +526,7 @@ static void __init free_initrd(void)
#endif
-void __init populate_rootfs(void)
+static int __init populate_rootfs(void)
{
char *err = unpack_to_rootfs(__initramfs_start,
__initramfs_end - __initramfs_start, 0);
@@ -566,3 +566,4 @@ void __init populate_rootfs(void)
}
#endif
}
+populate_rootfs_initcall(populate_rootfs);
_
next prev parent reply other threads:[~2006-12-11 10:28 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-12-11 8:27 [patch] pipe: Don't oops when pipe filesystem isn't mounted Chuck Ebbert
2006-12-11 8:55 ` Andrew Morton
2006-12-11 9:13 ` Andrew Morton
2006-12-11 9:21 ` Al Viro
2006-12-11 9:25 ` Andrew Morton
2006-12-11 9:33 ` Al Viro
2006-12-11 9:47 ` Andrew Morton
2006-12-11 10:03 ` Al Viro
2006-12-11 10:17 ` Andrew Morton
2006-12-11 10:22 ` Al Viro
2006-12-11 10:34 ` Andrew Morton
2006-12-11 10:47 ` Al Viro
2006-12-11 10:57 ` Jeff Garzik
2006-12-11 10:27 ` Andrew Morton [this message]
2006-12-11 10:45 ` Al Viro
2006-12-11 16:01 ` Linus Torvalds
2006-12-11 16:12 ` Al Viro
2006-12-11 16:40 ` Linus Torvalds
2006-12-12 0:33 ` Benjamin Herrenschmidt
2006-12-12 2:08 ` Andrew Morton
2006-12-12 2:17 ` Dominik Brodowski
2006-12-12 2:23 ` Linus Torvalds
2006-12-11 10:52 ` [PATCH] get rid of ARCH_HAVE_XTIME_LOCK Eric Dumazet
2006-12-11 13:50 ` [PATCH] Optimize calc_load() Eric Dumazet
2006-12-11 19:04 ` [PATCH] constify pipe_buf_operations Eric Dumazet
2006-12-11 9:13 ` [patch] pipe: Don't oops when pipe filesystem isn't mounted Al Viro
2006-12-11 9:40 ` Michael Tokarev
2006-12-11 15:44 ` Linus Torvalds
2006-12-12 12:20 ` Michael Tokarev
-- strict thread matches above, loose matches on Subject: below --
2006-12-11 9:33 Chuck Ebbert
2006-12-11 10:48 ` Olaf Hering
2006-12-11 10:52 ` 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=20061211022746.9ec80c03.akpm@osdl.org \
--to=akpm@osdl.org \
--cc=76306.1226@compuserve.com \
--cc=linux-kernel@vger.kernel.org \
--cc=torvalds@osdl.org \
--cc=viro@ftp.linux.org.uk \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox