[PATCH ]Re: hal doesn't recognize mounting of devices

[Jon Nettleton]

On Tue, 2005-10-25 at 14:56 +0200, Danny Kukawka wrote:
> On Tuesday 25 October 2005 00:41, Jon Nettleton wrote:
> > I have already found this lone post to the kernel about the problem,
> > http://lkml.org/lkml/2005/9/14/138 .  I have done some of my own testing
> > and basically come to the conclusion, this is less likely to happen when
> > not using preemption, however it can be triggered.  The ridiculously
> > simple patch I am using right now which seems to work fine is this.
> 
> If I see this rigth it's really a kernel bug if you use voluntary preempt 
> kernel. Is this in any distro-default kernel? Not really sure if we should 
> workaround this defaultly.
> 
> Btw. I attached a patch as workaround. Maybe someone could test this and check 
> the sleep (maybe also higher and lower values). I added a retry instead of a 
> sleep before check /proc/mounts to prevent delay on systems where this work 
> correct. Unfortunately we have now a delay (0.2 seconds) with relly unmounted 
> devices. The only solution would be add a key to enable/disable the retry 
> (disabled by default and only enable with fdi file).
> 
> Please review and comment.
> 
> Cheers,
> 
> Danny
> 
>  ChangeLog              |    8 ++++++
>  hald/linux2/blockdev.c |   52 +++++++++++++++++++++++++--------------
>  2 files changed, 42 insertions(+), 18 deletions(-)
> _______________________________________________
> hal mailing list
> hal at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/hal
> 

Thanks for the patch.  I had to bump the delay to 0.3 seconds to get
everything to reliably mount and unmount on my machines.  Other than
that it works as expected.  

I was wondering if this wouldn't be the perfect place to use inotify.
Setup an inotify watch on /proc/mounts and block until we get the write
finished event.  Just a thought.

-Jon