This test seems to be unnecessary since we always have rootfs mounted before
calling a usermodehelper.
Signed-off-by: Andreas Gruenbacher <[email protected]>
Signed-off-by: Jan Blunck <[email protected]>
---
kernel/kmod.c | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
Index: b/kernel/kmod.c
===================================================================
--- a/kernel/kmod.c
+++ b/kernel/kmod.c
@@ -173,10 +173,7 @@ static int ____call_usermodehelper(void
*/
set_user_nice(current, 0);
- retval = -EPERM;
- if (current->fs->root)
- retval = kernel_execve(sub_info->path,
- sub_info->argv, sub_info->envp);
+ retval = kernel_execve(sub_info->path, sub_info->argv, sub_info->envp);
/* Exec failed? */
sub_info->retval = retval;
--
On Thu, Sep 27, 2007 at 04:12:02PM +0200, [email protected] wrote:
> This test seems to be unnecessary since we always have rootfs mounted before
> calling a usermodehelper.
Are you sure this is true? I thought we called the usermode helper for
hotplug _very_ early in the boot sequence when the device tree starts to
get populated.
But things could have changed from the 2.5 days when this was first
added, if so, nevermind me :)
thanks,
greg k-h
On Thu, Sep 27, 2007 at 10:46:04AM -0700, Greg KH wrote:
> On Thu, Sep 27, 2007 at 04:12:02PM +0200, [email protected] wrote:
> > This test seems to be unnecessary since we always have rootfs mounted before
> > calling a usermodehelper.
>
> Are you sure this is true? I thought we called the usermode helper for
> hotplug _very_ early in the boot sequence when the device tree starts to
> get populated.
rootfs is mounted by init_mount_tree, and curret->fs is set up for init
there aswell. This is called by mnt_init, which is called by
vfs_caches_init, which is called by start_kernel far before we go to
rest_init which finally creates a thread to call kernel_init which then
calls do_basic_setup which calls do_initcalls to initialize drivers and
afterwards runs the initrd/initramfs.
While the actual function names in main.c changed quite a bit we've
initialized the initial namespace very early on since the 2.5 days.
On Thu, Sep 27, 2007 at 10:39:22PM +0200, Christoph Hellwig wrote:
> On Thu, Sep 27, 2007 at 10:46:04AM -0700, Greg KH wrote:
> > On Thu, Sep 27, 2007 at 04:12:02PM +0200, [email protected] wrote:
> > > This test seems to be unnecessary since we always have rootfs mounted before
> > > calling a usermodehelper.
> >
> > Are you sure this is true? I thought we called the usermode helper for
> > hotplug _very_ early in the boot sequence when the device tree starts to
> > get populated.
>
> rootfs is mounted by init_mount_tree, and curret->fs is set up for init
> there aswell. This is called by mnt_init, which is called by
> vfs_caches_init, which is called by start_kernel far before we go to
> rest_init which finally creates a thread to call kernel_init which then
> calls do_basic_setup which calls do_initcalls to initialize drivers and
> afterwards runs the initrd/initramfs.
>
> While the actual function names in main.c changed quite a bit we've
> initialized the initial namespace very early on since the 2.5 days.
Ah, ok, great, thanks for correcting me. I have no objection to this
patch then.
thanks,
greg k-h
On Thu, Sep 27, 2007 at 04:12:02PM +0200, [email protected] wrote:
> This test seems to be unnecessary since we always have rootfs mounted before
> calling a usermodehelper.
Ok.