Re: [patch 1/6] v4l: yet another video-buf interface update

[Gerd Knorr <kraxel@bytesex.org>]

On Mon, Nov 08, 2004 at 01:52:22PM -0800, Andrew Morton wrote:
> Gerd Knorr <kraxel@bytesex.org> wrote:
> >
> > This is one more interface fix for the video-buf.c module, the first
> > attempt on that wasn't that clever.  Instead of passing the driver
> > private data through all function calls I've just made that an element
> > of the videobuf_queue struct which is passed around everythere _anyway_.
> 
> This patch throws a reject in videobuf_vm_close() because it's expecting
> 
> 	map->q->ops->buf_release(vma->vm_file,map->q->bufs[i]);
> and not
> 	map->q->ops->buf_release(vma->vm_file->private_data,map->q->bufs[i]);
> 
> I'll fix things up, but please check that.

Someone (Peter Osterlund IIRC) mailed a patch with the "vma->vm_file" =>
"vma->vm_file->private_data" fix to lkml.  Thats actually the place I
forgot on the first attempt and which causes the tvtime crashes
mentioned in the changelog.

My patches are built against a fresh linus bk tree, not against -mm, so
its probably just that patch being merged into -mm which causes the
rejects.  No problem, my patch obsoletes this one-liner fix.

  Gerd

-- 
#define printk(args...) fprintf(stderr, ## args)