Vasily Galkin

# UPDATE Mesa 23.1.0~rc1 removes this limitation! https://cgit.freedesktop.org/mesa/mesa/commit/?h=mesa-23.1.0-rc1&id=71fe9dfe07b53115a9e8f8b031a04f6f387937d6 I Installed it from this PPA https://launchpad.net/~ernstp/+archive/ubuntu/mesarc and after updating memtest_vulkan become able to allocate 10.5GB for integrated GPU. Example run (in...

In short: Thanks! Please install `vulkaninfo` tool and post the output of `vulkaninfo | grep -B 20 -i StorageBufferRange` Detailed: Thanks for the update. It seems that the problem is...

Thanks for trying! Which OS - windows or linux? Can you collect some debug info? To get debug info - rename `memtest_vulkan` to `memtest_vulkan_verbose`, run it after renaming, then take...

In short: @nikicat, please provide a console output too. Deatiled: it seems that your problem with Intel GPU on Linux is *different* from earlier known problems - the Windows problem...

@nikicat, thanks for detailed logging. Now the problem is more clear for me - for some reason the vulkan initialization takes ~12 seconds. It's ~100x times slower then normal behaviour!...

I'm glad the problem for intel GPU on linux is solved. However the behaviour of very slow loading is wierd, I want to add some diagnostics for such case of...

In short: this is a known problem that is not easy to fix without sacrificing other goals. Here is the more detailed description: initially the tool was targeted for testing...

About memory types - memtest_vulkan performs all allocations at start: one small buffer in CPU+GPU-visible memory type for control info and error counting and a huge buffer used for memory...

Since the amd-gfx answer was "It will get swapped back to VRAM if it makes sense for performance." - is is interesting to see how the "makes sense for performance"...

Hi, you are right - the testing is performed in GPU memory, since the initial idea was to create the memory stability testing utility, not the performance benchmark. The output...