Re: Lavapipe performance

2025-01-30 Thread Jose Fonseca
Thanks for refreshing my memory, Erik. Indeed OpenSWR was it. George, the geometry is quite smooth and it doesn't look complex enough to justify 130k triangles. If the application has some sort of tunable NURBS tesselator, I'd tune it down so it spits out fewer triangles. That would increase llv

Re: Lavapipe performance

2025-01-30 Thread George Karpathios
Thank you all, for the tips. Using a smaller window doesn't change anything, neither does tweaking LP_NUM_THREADS (between 8/12/16). I am also not using MSAA. I'll have to dig some more regarding your other suggestions. I've shared a screenshot of what is being rendered at https://imgur.com/a/6P6OA

Re: Lavapipe performance

2025-01-30 Thread Erik Faye-Lund
On Thu, 2025-01-30 at 13:49 +, Jose Fonseca wrote: > At one point there was another SW renderer gallium driver by a team > at Intel that was specifically optimized for huge number of > primitives with no texturing (like the sort used by https://vtk.org/ > ), but I can't remember its name or fin

Re: Lavapipe performance

2025-01-30 Thread Brian Paul
In addition to what Jose listed, one simple experiment is to try rendering to a smaller window. How does the frame rate change? If it's basically inversely proportional to the window size, that tells us that the rendering is limited by fragment generation/processing. Otherwise, perhaps you're ge

Re: Lavapipe performance

2025-01-30 Thread Jose Fonseca
That's a relatively large geometry count. llvmpipe (and consequently lavapipe) was optimized for desktop compositors and UIs: large textured quads with blending, where most of the time is spent in the fragment shader, particular texture sampling. Not much effort went into optimizing llvmpipe for

Lavapipe performance

2025-01-30 Thread George Karpathios
Hi, I'm using Lavapipe (24.2.2, Windows 10) and trying to evaluate its performance, but I don't know if I can get more performance out of it or if I'm hitting its expected limits. For example, rendering a scene with ~300k triangles and ~150k lines, I get 5 fps on an i7 11800H (8 cores). Of cours

Re: Lavapipe performance question during interactive drawing

2023-05-08 Thread George Karpathios
Hi Dave, Indeed after setting LP_DEBUG=fs I get a lot of repetitive printing during the interactive actions. After some more searching it finally seems that the problem is on our side, which forces some pipelines to be continuously re-created. Thanks everyone for your help, and I apologize for the

Re: Lavapipe performance question during interactive drawing

2023-05-05 Thread Dave Airlie
On Fri, 5 May 2023 at 21:30, George Karpathios wrote: > > Hi list, > > I'm using Lavapipe for Vulkan software rendering support in a modeling > application. I notice a large performance hit (with any Mesa version) in the > following scenario: The user clicks & drags the mouse in order to create

Re: Lavapipe performance question during interactive drawing

2023-05-05 Thread Mike Blumenkrantz
Anything as long as it exhibits the issue. On Fri, May 5, 2023 at 11:52 AM George Karpathios wrote: > Not sure, I may have to get approval first. Will see what I can do. If so, > are there any specific options you'd like enabled? Thanks > > Best regards, > George > > On Fri, May 5, 2023 at 5:33 

Re: Lavapipe performance question during interactive drawing

2023-05-05 Thread George Karpathios
Not sure, I may have to get approval first. Will see what I can do. If so, are there any specific options you'd like enabled? Thanks Best regards, George On Fri, May 5, 2023 at 5:33 PM Mike Blumenkrantz < michael.blumenkra...@gmail.com> wrote: > Can you provide a gfxreconstruct of the scenario?

Re: Lavapipe performance question during interactive drawing

2023-05-05 Thread Mike Blumenkrantz
Can you provide a gfxreconstruct of the scenario? On Fri, May 5, 2023 at 10:32 AM George Karpathios wrote: > Hi Mike, > > Thanks for responding. I commented the call > to lvp_find_inlinable_uniforms() in lvp_pipeline.c but unfortunately it > didn't help this time. It may have gotten a bit worse

Re: Lavapipe performance question during interactive drawing

2023-05-05 Thread George Karpathios
Hi Mike, Thanks for responding. I commented the call to lvp_find_inlinable_uniforms() in lvp_pipeline.c but unfortunately it didn't help this time. It may have gotten a bit worse actually. Best regards, George On Fri, May 5, 2023 at 3:08 PM Mike Blumenkrantz < michael.blumenkra...@gmail.com> wr

Re: Lavapipe performance question during interactive drawing

2023-05-05 Thread Mike Blumenkrantz
Hi, Can you try commenting out the same lines from last time and see whether that affects anything? Mike On Fri, May 5, 2023 at 7:30 AM George Karpathios wrote: > Hi list, > > I'm using Lavapipe for Vulkan software rendering support in a modeling > application. I notice a large performance hi

Lavapipe performance question during interactive drawing

2023-05-05 Thread George Karpathios
Hi list, I'm using Lavapipe for Vulkan software rendering support in a modeling application. I notice a large performance hit (with any Mesa version) in the following scenario: The user clicks & drags the mouse in order to create a simple shape (e.g. plane, cube, sphere) dynamically and the result