[Bf-committers] Cycles CUDA Toolkit 8 transition

[Sergey Sharybin]

Hi,

There are now updated builds on the buildbot. Please give them a try.

On Wed, Aug 3, 2016 at 4:58 PM, PerfectionCat <sindra1961reborn at yahoo.co.jp>
wrote:

> Hi.
>
> I tried SHA-1-70100b4ec7ba299912bf999f15914c0a29125ffd., but this revision
> is slow, too.
> With best regards, PerfectionCat.
>
> ----- Original Message -----
> *From:* PerfectionCat <sindra1961reborn at yahoo.co.jp>
> *To:* PerfectionCat <sindra1961reborn at yahoo.co.jp>; Sergey Sharybin <
> sergey.vfx at gmail.com>; bf-blender developers <bf-committers at blender.org>;
> bf-blender developers <bf-committers at blender.org>
> *Date:* 2016/8/3, Wed 23:49
> *Subject:* Re: [Bf-committers] Cycles CUDA Toolkit 8 transition
>
> I'm sorry.
> I got a wrong hash number.
> I have not yet tried SHA-1-70100b4ec7ba299912bf999f15914c0a29125ffd.
> With best regards, PerfectionCat.
>
> ----- Original Message -----
> *From:* PerfectionCat <sindra1961reborn at yahoo.co.jp>
> *To:* Sergey Sharybin <sergey.vfx at gmail.com>; bf-blender developers <
> bf-committers at blender.org>
> *Date:* 2016/8/3, Wed 23:28
> *Subject:* Re: [Bf-committers] Cycles CUDA Toolkit 8 transition
>
> Hi.
>
> I built in SHA-1: 6353ecb996898b4ce2fe8065130ed1f5ea3b6989 and tried it.
> The build did not become slow.
> However, blender which built in SHA-1:
> 70100b4ec7ba299912bf999f15914c0a29125ffd is slow.
>
>
> With best regards, PerfectionCat.
>
>
> ----- Original Message -----
> >From: Sergey Sharybin <sergey.vfx at gmail.com>
> >To: PerfectionCat <sindra1961reborn at yahoo.co.jp>; bf-blender developers <
> bf-committers at blender.org>
> >Date: 2016/8/3, Wed 17:09
> >Subject: Re: [Bf-committers] Cycles CUDA Toolkit 8 transition
> >
> >
> >Interesting.
> >
> >
> >Can you test revision 6353ecb and the one prior to it to see if it's this
> revision caused the problem?
> >
> >
> >On Wed, Aug 3, 2016 at 12:27 AM, PerfectionCat <
> sindra1961reborn at yahoo.co.jp> wrote:
> >
> >Hi.
> >>
> >>blender which built using a source file at the time of 467e12514e60 in
> CUDA8 has a shorter time required for rendering of BMW for around 20
> seconds.
> >>
> >>SHA-1-2f5db2e45b7a8af51b15c45bb41f5dcb49ae9d4d is slow.
> >>SHA-1-467e12514e60b4130cd0a33b59f53aa4cf7056b6 is fast.
> >>
> >>Windows 10 pro 64bits
> >>
> >>Intel i7 6700K 4.4GHz
> >>NVIDIA GTX 1070
> >>MEMORY 16GB
> >>
> >>
> >>With best regards, PerfectionCat.
> >>
> >>
> >>
> >>
> >>----- Original Message -----
> >>>From: Sergey Sharybin <sergey.vfx at gmail.com>
> >>>To: Blender Developers <bf-committers at blender.org>
> >>>Date: 2016/8/2, Tue 23:44
> >>>Subject: Re: [Bf-committers] Cycles CUDA Toolkit 8 transition
> >>>
> >>>Hey again,
> >>>
> >>>Spent majority of the day trying to solve the regression, without much
> >>>success. Even simplest kernel needed for BMW scene is about 10% slower.
> >>>This is mainly coming from bump nodes. Enabling all other features makes
> >>>things even worse performance wise.
> >>>
> >>>I did some tweaks again to make sure all functions are inlined in the
> same
> >>>manner by CUDA 8.0 as they used to be before. So now PTax output shows
> >>>exactly same function, but for some reason spills are just higher with
> new
> >>>toolkit and at the same time stack usage is reasonably slower. Not sure
> yet
> >>>what's going on here and think we'd better leave this alone for until
> >>>official toolkit is released.
> >>>
> >>>For the time being i've switched buildbots to use more complicated
> setup,
> >>>using CUDA 7.5 for all kernels except sm_60 and sm_61 (new generation
> >>>cards) and using new toolkit only for new kernels.
> >>>
> >>>So hopefully now all maxwell and lower crads have same performance as
> >>>before. And yet users of new cards can have some degree of GPU
> rendering.
> >>>
> >>>
> >>>On Mon, Aug 1, 2016 at 6:15 PM, Sergey Sharybin <sergey.vfx at gmail.com>
> >>>wrote:
> >>>
> >>>> Hey everyone,
> >>>>
> >>>> In order to make Cycles officially support new Pascal cards (GTX
> 10x0) it
> >>>> is crucial to use CUDA Toolkit 8. While it is still in the RC stage we
> >>>> started looking into making it official one for Cycles.
> >>>>
> >>>> There is already some initial work done:
> >>>>
> >>>> - Tweaks are done to Cycles kernel to have same level of inlined
> functions
> >>>> (new toolkit barely used inlined functions, which resulted in poor
> >>>> performance on all cards)
> >>>> - New toolkit was installed onto all release environments and
> buildbots.
> >>>>
> >>>> So on a positive side, next buildbot's builds will support Pascal
> cards,
> >>>> yay! :)
> >>>>
> >>>> But unfortunately, while i've managed to keep Kepler cards (GTX760)
> same
> >>>> performance as before, we had a report from Jens in IRC about Maxwell
> cards
> >>>> having a poor performance yet again (in his case, up to 40% worse).
> >>>>
> >>>> This is really unacceptable, we'll keep working on solution for this
> >>>> issue. It might take some time, so just be aware and please be
> patient :)
> >>>>
> >>>> I will keep posted here with our progress.
> >>>>
> >>>> --
> >>>> With best regards, Sergey Sharybin
> >>>>
> >>>
> >>>
> >>>
> >>>--
> >>>With best regards, Sergey Sharybin
> >>
> >>>_______________________________________________
> >>>Bf-committers mailing list
> >>>Bf-committers at blender.org
> >>>https://lists.blender.org/mailman/listinfo/bf-committers
> >>>
> >>>
> >>>
> >>_______________________________________________
> >>Bf-committers mailing list
> >>Bf-committers at blender.org
> >>https://lists.blender.org/mailman/listinfo/bf-committers
> >>
> >
> >
> >
> >--
> >
> >With best regards, Sergey Sharybin
> >
> >
> _______________________________________________
> Bf-committers mailing list
> Bf-committers at blender.org
> https://lists.blender.org/mailman/listinfo/bf-committers
>
>
>
>
>


-- 
With best regards, Sergey Sharybin