On Sat, Mar 17, 2018 at 10:32 PM, Dylan Baker wrote:
> Quoting Jason Ekstrand (2018-03-17 09:53:13)
> > On March 16, 2018 23:36:50 Dylan Baker wrote:
> >
> > > Quoting Rob Clark (2018-03-16 14:25:09)
> > >> On Fri, Mar 16, 2018 at 4:30 PM, Dylan Baker
> wrote:
> > >> > Quoting Rob Clark (2018-0
Quoting Jason Ekstrand (2018-03-17 09:53:13)
> On March 16, 2018 23:36:50 Dylan Baker wrote:
>
> > Quoting Rob Clark (2018-03-16 14:25:09)
> >> On Fri, Mar 16, 2018 at 4:30 PM, Dylan Baker wrote:
> >> > Quoting Rob Clark (2018-03-16 12:20:10)
> >> >> On Fri, Mar 16, 2018 at 3:13 PM, Jason Ekstra
The index bounds are computed only when they are needed for uploading
vertices that are passed via a CPU pointer (user_buffer). In all other
cases, computing the index bounds has a performance cost, which can be very
significant.
If you rely on u_vbuf to upload vertices for you, you shouldn't need
@ least, here fromm Germany.
https://patchwork.freedesktop.org/
Server Error (500)
emeril.freedesktop.org
is still alive.
Dieter
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
On Fri, Mar 16, 2018 at 2:28 PM, Dylan Baker wrote:
> Quoting Rob Clark (2018-03-15 18:43:28)
>> I threatened to do this a long time ago.. I probably *should* have done
>> it a long time ago when there where many fewer intrinsics. But the
>> system of macro/#include magic for dealing with intrins
Make a simple worklist by basically just wrapping u_vector.
This is intended used in nir_opt_dce to reduce the number of calls
to ralloc, as we are currenlty spamming ralloc quite bad. It should
also give better cache locality and much lower memory usage.
---
src/compiler/nir/nir_worklist.h | 69 +
Shader-db runtime change avarage of five runs:
Before 125,77 seconds (+/- 0,09%)
After 124,48 seconds (+/- 0,07%)
---
src/compiler/nir/nir_opt_dce.c | 53 ++---
src/compiler/nir/nir_worklist.h | 1 +
2 files changed, 19 insertions(+), 35 deletions(-)
d
This is take two on reducing ralloc overhead in nir_opt_dce.
I've ditched the previous solution with a freelist, and instead
gone for a wrapper on u_vector. That should remove the need for a
freelist alltogether, and at the same time lower our memory usage.
CC: Eric Anholt
Thomas Helland (2):
Nice. Thanks!
Reviewed-by: Matt Turner
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
On Sat, Mar 17, 2018 at 2:12 PM, Erico Nunes wrote:
> Hi all,
>
> I have been working to add indexed drawing/glDrawElements support to
> the mesa-lima driver currently in development
> (https://github.com/yuq/mesa-lima).
> For that implementation, it seems that we need to have the minimum and
> ma
Hi all,
I have been working to add indexed drawing/glDrawElements support to
the mesa-lima driver currently in development
(https://github.com/yuq/mesa-lima).
For that implementation, it seems that we need to have the minimum and
maximum index values from the index buffer available in order to set
On March 16, 2018 18:25:14 Caio Marcelo de Oliveira Filho
wrote:
Generalize the code for remove dead loops to also remove dead if
nodes. The conditions are the same in both cases, if the node (and
it's children) don't have side-effects AND the nodes after it don't
use the values produced by th
On March 16, 2018 23:36:50 Dylan Baker wrote:
Quoting Rob Clark (2018-03-16 14:25:09)
On Fri, Mar 16, 2018 at 4:30 PM, Dylan Baker wrote:
> Quoting Rob Clark (2018-03-16 12:20:10)
>> On Fri, Mar 16, 2018 at 3:13 PM, Jason Ekstrand
wrote:
>> > On Fri, Mar 16, 2018 at 11:53 AM, Dylan Baker
https://bugs.freedesktop.org/show_bug.cgi?id=105412
Michel Dänzer changed:
What|Removed |Added
Component|Drivers/Vulkan/radeon |DRM/Radeon
QA Contact|mesa-dev@
https://bugs.freedesktop.org/show_bug.cgi?id=105442
mirh changed:
What|Removed |Added
CC||m...@protonmail.ch
--
You are receiving this ma
https://bugs.freedesktop.org/show_bug.cgi?id=105551
Michelle Sullivan changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
https://bugs.freedesktop.org/show_bug.cgi?id=105412
--- Comment #7 from Vasya ---
Created attachment 138165
--> https://bugs.freedesktop.org/attachment.cgi?id=138165&action=edit
cs source
--
You are receiving this mail because:
You are the assignee for the bug.
You are the QA Contact for the
17 matches
Mail list logo