On Tue, Jul 17, 2018 at 3:06 AM, Michel Dänzer wrote:
> On 2018-07-16 08:07 PM, Marek Olšák wrote:
>> On Mon, Jul 16, 2018 at 12:06 PM, Michel Dänzer wrote:
>>>
>>> It might be even better to compare the total number of frames rendered
>>> in a fixed amount of time, with multiple runs for each te
On 2018-07-16 08:07 PM, Marek Olšák wrote:
> On Mon, Jul 16, 2018 at 12:06 PM, Michel Dänzer wrote:
>>
>> It might be even better to compare the total number of frames rendered
>> in a fixed amount of time, with multiple runs for each test
>> configuration, in case there is systemic variance betwe
On Mon, Jul 16, 2018 at 12:06 PM, Michel Dänzer wrote:
> On 2018-07-13 08:35 PM, Marek Olšák wrote:
>> https://pastebin.com/raw/itNDpPv0
>>
>> Cleanups only: +10.85%
>> Total: +11.8%
>
> Thanks, that's more useful.
>
> If you pass all 3 data files to a single ministat -s invocation, it
> combines
On 2018-07-13 08:35 PM, Marek Olšák wrote:
> https://pastebin.com/raw/itNDpPv0
>
> Cleanups only: +10.85%
> Total: +11.8%
Thanks, that's more useful.
If you pass all 3 data files to a single ministat -s invocation, it
combines all of them in a single graph, and prints information about
whether o
https://pastebin.com/raw/itNDpPv0
Cleanups only: +10.85%
Total: +11.8%
Marek
On Fri, Jul 13, 2018 at 11:05 AM, Michel Dänzer wrote:
> On 2018-07-12 07:26 AM, Marek Olšák wrote:
>>
>> The maximum glxgears FPS improves as follows.
>>
>> Initially: 13285
>> 8 patches: 14403 (+8.4% vs initial)
>> +
On 2018-07-12 07:26 AM, Marek Olšák wrote:
>
> The maximum glxgears FPS improves as follows.
>
> Initially: 13285
> 8 patches: 14403 (+8.4% vs initial)
> + patch 9: 15498 (+16.6% vs initial)
Those are nice numbers, but I'm afraid they're almost meaningless
without more information about how you
Hi,
This series improves RadeonSI performance for trivial CPU-bound
benchmarks. Other CPU-bound benchmarks may be affected marginally.
The first 8 patches are cleanups that surprisingly increase
performance.
The last patch uses a new chunk type in the CS ioctl for passing
the array of buffer han
