[alsa-devel] [PATCH v2] ALSA: hda: Abort capability probe on invalid capability

Takashi Iwai tiwai at suse.de
Thu Oct 19 09:15:41 CEST 2017


On Wed, 18 Oct 2017 18:52:54 +0200,
Ughreja, Rakesh A wrote:
> 
> 
> 
> >-----Original Message-----
> >From: Koul, Vinod
> >Sent: Wednesday, October 18, 2017 9:57 PM
> >To: Ughreja, Rakesh A <rakesh.a.ughreja at intel.com>
> >Cc: alsa-devel at alsa-project.org; tiwai at suse.de
> >Subject: Re: [PATCH v2] ALSA: hda: Abort capability probe on invalid capability
> >
> >On Thu, Oct 19, 2017 at 05:15:25AM +0530, Ughreja, Rakesh A wrote:
> >> From: Rakesh Ughreja <rakesh.a.ughreja at intel.com>
> >>
> >> Due to bugs in BIOS it's possible that the HDA capability link
> >
> >It maybe due to BIOS bug or some other issue, lets not quantify that please.
> >
> >I would say "On reading wrong capablity pointer values we may crash...
> >
> >> list is not constructed properly. This may lead to driver going
> >> into unknown state.
> >
> >I dont think we maintain driver states which are going wrong here
> >
> >> So whenever driver discovers unknown HDA
> >> capability, log it as error and stop traversing the link list
> >> further.
> >
> >that sound good
> 
> Hi Vinod, Takashi,
> 
> Are you both ok with following commit message ?
> 
> On reading wrong capability pointer values driver may crash, so 
> whenever driver discovers unknown HDA capability, log it as error 
> and stop traversing the link list further.

Looks good to me.


thanks,

Takashi


More information about the Alsa-devel mailing list