dd2f34c777
This improves performance of llama.cpp. fossil-db (navi21): Totals from 1598 (2.00% of 79825) affected shaders: MaxWaves: 30182 -> 29278 (-3.00%); split: +0.04%, -3.03% Instrs: 1013136 -> 1013065 (-0.01%); split: -0.07%, +0.07% CodeSize: 5275876 -> 5274948 (-0.02%); split: -0.06%, +0.04% VGPRs: 86176 -> 88016 (+2.14%); split: -0.22%, +2.36% SpillVGPRs: 0 -> 11 (+inf%) Scratch: 0 -> 4096 (+inf%) Latency: 7954289 -> 7824742 (-1.63%); split: -1.64%, +0.01% InvThroughput: 1511429 -> 1510912 (-0.03%); split: -0.89%, +0.86% VClause: 26503 -> 26460 (-0.16%); split: -0.23%, +0.07% SClause: 19032 -> 19039 (+0.04%); split: -0.01%, +0.05% Copies: 74577 -> 74329 (-0.33%); split: -0.79%, +0.46% Branches: 20278 -> 20279 (+0.00%) VALU: 665079 -> 664831 (-0.04%); split: -0.09%, +0.05% SALU: 124899 -> 124818 (-0.06%); split: -0.08%, +0.01% VMEM: 46141 -> 46163 (+0.05%) fossil-db (navi31): Totals from 1609 (2.02% of 79825) affected shaders: MaxWaves: 39724 -> 38880 (-2.12%) Instrs: 1147767 -> 1147595 (-0.01%); split: -0.04%, +0.03% CodeSize: 5777072 -> 5776376 (-0.01%); split: -0.03%, +0.02% VGPRs: 91752 -> 93132 (+1.50%); split: -0.03%, +1.53% Latency: 7526930 -> 7396201 (-1.74%); split: -1.74%, +0.00% InvThroughput: 1083131 -> 1088328 (+0.48%); split: -0.45%, +0.93% VClause: 25864 -> 25789 (-0.29%); split: -0.33%, +0.04% SClause: 19136 -> 19135 (-0.01%); split: -0.02%, +0.01% Copies: 80797 -> 80501 (-0.37%); split: -0.42%, +0.05% VALU: 674455 -> 674160 (-0.04%); split: -0.05%, +0.01% SALU: 123849 -> 123806 (-0.03%) fossil-db (gfx1201): Totals from 1614 (2.02% of 79839) affected shaders: MaxWaves: 40140 -> 39296 (-2.10%) Instrs: 1183227 -> 1183102 (-0.01%); split: -0.04%, +0.03% CodeSize: 6091060 -> 6090636 (-0.01%); split: -0.03%, +0.03% VGPRs: 90708 -> 92040 (+1.47%); split: -0.01%, +1.48% Latency: 7588683 -> 7425866 (-2.15%); split: -2.15%, +0.00% InvThroughput: 1070469 -> 1075700 (+0.49%); split: -0.50%, +0.99% VClause: 25691 -> 25597 (-0.37%); split: -0.37%, +0.00% SClause: 19095 -> 19086 (-0.05%); split: -0.05%, +0.01% Copies: 80753 -> 80452 (-0.37%); split: -0.42%, +0.05% VALU: 665218 -> 664922 (-0.04%); split: -0.05%, +0.01% SALU: 144059 -> 144011 (-0.03%) Signed-off-by: Rhys Perry <pendingchaos02@gmail.com> Reviewed-by: Georg Lehmann <dadschoorse@gmail.com> Reviewed-by: Daniel Schürmann <daniel@schuermann.dev> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37791>
`Mesa <https://mesa3d.org>`_ - The 3D Graphics Library ====================================================== Source ------ This repository lives at https://gitlab.freedesktop.org/mesa/mesa. Other repositories are likely forks, and code found there is not supported. Build & install --------------- You can find more information in our documentation (`docs/install.rst <https://docs.mesa3d.org/install.html>`_), but the recommended way is to use Meson (`docs/meson.rst <https://docs.mesa3d.org/meson.html>`_): .. code-block:: sh $ meson setup build $ ninja -C build/ $ sudo ninja -C build/ install Support ------- Many Mesa devs hang on IRC; if you're not sure which channel is appropriate, you should ask your question on `OFTC's #dri-devel <irc://irc.oftc.net/dri-devel>`_, someone will redirect you if necessary. Remember that not everyone is in the same timezone as you, so it might take a while before someone qualified sees your question. To figure out who you're talking to, or which nick to ping for your question, check out `Who's Who on IRC <https://dri.freedesktop.org/wiki/WhosWho/>`_. The next best option is to ask your question in an email to the mailing lists: `mesa-dev\@lists.freedesktop.org <https://lists.freedesktop.org/mailman/listinfo/mesa-dev>`_ Bug reports ----------- If you think something isn't working properly, please file a bug report (`docs/bugs.rst <https://docs.mesa3d.org/bugs.html>`_). Contributing ------------ Contributions are welcome, and step-by-step instructions can be found in our documentation (`docs/submittingpatches.rst <https://docs.mesa3d.org/submittingpatches.html>`_). Note that Mesa uses gitlab for patches submission, review and discussions.