Re: [PATCH 4.2 41/54] drm/nouveau/fbcon: take runpm reference when userspace has an open fd
From: Ben Hutchings
Date: Sat Oct 24 2015 - 02:13:16 EST
On Fri, 2015-10-23 at 10:45 -0700, Greg Kroah-Hartman wrote:
> 4.2-stable review patch.ÂÂIf anyone has any objections, please let me know.
> From: Ben Skeggs <bskeggs@xxxxxxxxxx>
> commit f231976c2e8964ceaa9250e57d27c35ff03825c2 upstream.
> We need to do this in order to prevent accesses to the device while it's
> powered down.ÂÂUserspace may have an mmap of the fb, and there's no good
> way (that I know of) to prevent it from touching the device otherwise.
> This fixes some nasty races between runpm and plymouth on some systems,
> which result in the GPU getting very upset and hanging the boot.
> Signed-off-by: Ben Skeggs <bskeggs@xxxxxxxxxx>
> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
> Âdrivers/gpu/drm/nouveau/nouveau_fbcon.c |ÂÂÂ24 ++++++++++++++++++++++++
> Â1 file changed, 24 insertions(+)
> --- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
> +++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
> @@ -178,8 +178,30 @@ nouveau_fbcon_sync(struct fb_info *info)
> Â> > return 0;
> +static int
> +nouveau_fbcon_open(struct fb_info *info, int user)
> +> > struct nouveau_fbdev *fbcon = info->par;
> +> > struct nouveau_drm *drm = nouveau_drm(fbcon->dev);
> +> > int ret = pm_runtime_get_sync(drm->dev->dev);
> + if (ret < 0 && ret != -EACCES)
> + return ret;
> +> > return 0;
As pm_runtime_get_sync() always bumps the usage count, the error path
here should call pm_runtime_put().
Beware of bugs in the above code;
I have only proved it correct, not tried it. - Donald Knuth
Description: This is a digitally signed message part