[Mesa-dev] [Mesa-stable] [PATCH] clover: Fix bug with computing hard_event status

Emil Velikov emil.l.velikov at gmail.com
Mon Aug 3 05:36:07 PDT 2015


On 3 August 2015 at 13:17, Marek Olšák <maraeo at gmail.com> wrote:
> On Mon, Aug 3, 2015 at 1:33 PM, Emil Velikov <emil.l.velikov at gmail.com> wrote:
>> On 1 August 2015 at 19:18, Marek Olšák <maraeo at gmail.com> wrote:
>>> On Sat, Aug 1, 2015 at 7:29 PM, Emil Velikov <emil.l.velikov at gmail.com> wrote:
>>>> On 1 August 2015 at 17:57, Marek Olšák <maraeo at gmail.com> wrote:
>>>>> No, we don't need this. radeonsi doesn't return a NULL fence anymore.
>>>>>
>>>> Glad to hear. On the other hand is that fix in stable ? I believe the
>>>> following commit resolves the issue, yet it's a bit short of
>>>> nomination.
>>>> Can anyone confirm if we need it for 10.6 ?
>>>>
>>>> commit 800efb0690e962750b9a072bcbab279fdaae24a1
>>>> Author: Michel Dänzer <michel.daenzer at amd.com>
>>>> Date:   Wed Jul 22 16:11:39 2015 +0900
>>>>
>>>>     radeonsi: Flush when we're asked to return a fence but don't have one yet
>>>>
>>>>     Reviewed-by: Marek Olšák <marek.olsak at amd.com>
>>>
>>> This and Tom's patch fix a bug introduced in:
>>>
>>> http://cgit.freedesktop.org/mesa/mesa/commit/?id=f1be3d8cdde17a9b9ae283e1bab2f46b992d3bf3
>>>
>>> which is not in 10.6 as far as I know, so none of these fixes need to
>>> be in 10.6.
>>>
>> Indeed it is. Considering the commit wasn't mentioned previously (plus
>> Tom's commit was aimed at 10.6) I've assumed a long standing bug. A
>> kernel style Fixes tag will be amazing. Any chance that we can get
>> something like that, please ?
>
> Yes, but that wouldn't help in this case, because the original poster
> didn't know where the bug was coming from.
>
Bummer. Thanks for the information !

-Emil


More information about the mesa-dev mailing list