Re: [Patch] 2.6.10.rc1.bk6 /lib/kobject_uevent.c buffer issues

[Greg KH <greg@kroah.com>]

On Wed, Oct 27, 2004 at 10:29:25AM -0400, Andrew Duggan wrote:
> Hi,
> 
> In the lastest code I've seen 2.6.10 rc1 bk6 for lib/kobject_uevent.c
> the incrementing of the sequence number is held off until the call to
> hotplug_ops->hotplug ().  That function can consume addtional buffer
> and use addtional pointers in the envp array.  The problem is that the
> new value for i and the new logical end of buffer (held in scratch) do
> not get updated.  When the sequnce number is then processed the first
> item generated by hotplug_ops->hotplug () gets overwriten, and can
> potentially corrupt any env vars created by hotplug_ops->hotplug.

Ah, good catch.  Sorry about adding that bug.

> Here is a patch to handle that.  An element in envp is reserved for
> the seq num and a different buffer is sent to hotplug_ops->hotplug
> altogether. Using a different buffer there is OK since send_uevent
> wasn't seeing the part of the buffer filled in by hotplug_ops->hotplug
> anyway, and the older call_usermodehelper gets passed envp and not
> buffer. 

Why not just use the same buffer?  We should be able to do that.

> I used the same size for the buffer_pt2 as buffer for simplicity,
> since I didn't know if saving the memory was THAT important. If need
> be the kmalloc for buffer_pt2 could be held off until just before the
> call to hotplug_ops->hotplug so that it could be allocated only
> (BUFFERSIZE - (scratch - buffer)).  

I'd prefer we use the same buffer.  Care to respin your patch?

thanks,

greg k-h