Commits

Commit history

1. Commits on Jul 30, 2024

   • Merge pull request #28 from ichowenyu/main

     Show description for e636de3
     austinvhuang

     authored
     e636de3View commit details

     Copy full SHA for e636de3

     Browse repository at this point
2. Commits on Jul 29, 2024

   • experimental first implementation of browser shim for wasm builds

     austinvhuang

     committed
     b6d0f8eView commit details

     Copy full SHA for b6d0f8e

     Browse repository at this point

   • Merge pull request #22 from junjihashimoto/feature/transpose

     Show description for 47a85b7
     austinvhuang

     authored
     47a85b7View commit details

     Copy full SHA for 47a85b7

     Browse repository at this point

   • Merge pull request #30 from fadichahineOS/patch-1

     Show description for bd8192b
     austinvhuang

     authored
     bd8192bView commit details

     Copy full SHA for bd8192b

     Browse repository at this point

   • Merge pull request #27 from junjihashimoto/fix/typo

     Show description for e9d7db3
     austinvhuang

     authored
     e9d7db3View commit details

     Copy full SHA for e9d7db3

     Browse repository at this point
3. Commits on Jul 28, 2024

   • Update README.md

     Show description for bfd8680
     fadichahineOS

     authored
     bfd8680View commit details

     Copy full SHA for bfd8680

     Browse repository at this point

   • Update README.md --fix typos

     Show description for d460c1f
     soyowenyu

     authored
     d460c1fView commit details

     Copy full SHA for d460c1f

     Browse repository at this point

   • Fix typo of Puzzle 13

     junjihashimoto

     committed
     39471a6View commit details

     Copy full SHA for 39471a6

     Browse repository at this point

   • Add transpose

     junjihashimoto

     committed
     b1da604View commit details

     Copy full SHA for b1da604

     Browse repository at this point
4. Commits on Jul 27, 2024

   • puzzles 13 + 14

     sarahpannn

     committed
     9586723View commit details

     Copy full SHA for 9586723

     Browse repository at this point

   • Merge pull request #25 from arozrai/update-readme

     Show description for 88ea60a
     austinvhuang

     authored
     88ea60aView commit details

     Copy full SHA for 88ea60a

     Browse repository at this point

   • fix typo in README.md

     arozrai

     committed
     fd76b8eView commit details

     Copy full SHA for fd76b8e

     Browse repository at this point
5. Commits on Jul 26, 2024

   • fix transformer build

     austinvhuang

     committed
     0a2d734View commit details

     Copy full SHA for 0a2d734

     Browse repository at this point

   • multimodal audio test

     austinvhuang

     committed
     e6a3dcdView commit details

     Copy full SHA for e6a3dcd

     Browse repository at this point
6. Commits on Jul 22, 2024

   • remove debug flag for matmul

     austinvhuang

     committed
     93e2103View commit details

     Copy full SHA for 93e2103

     Browse repository at this point

   • increase logging buffer size limit to avoid failing on large logging messages

     austinvhuang

     committed
     15142bdView commit details

     Copy full SHA for 15142bd

     Browse repository at this point

   • avoid serially blocking for async matmul dispatch, adjust BK subtile size based on m1 performance

     austinvhuang

     committed
     9819edeView commit details

     Copy full SHA for 9819ede

     Browse repository at this point
7. Commits on Jul 21, 2024

   • add float16 example

     austinvhuang

     committed
     a89d309View commit details

     Copy full SHA for a89d309

     Browse repository at this point

   • f16 WGSL computation works

     austinvhuang

     committed
     4b96087View commit details

     Copy full SHA for 4b96087

     Browse repository at this point

   • test-half build portability tweaks

     austinvhuang

     committed
     b2dd545View commit details

     Copy full SHA for b2dd545

     Browse repository at this point

   • generalize createTensor to work for arbitrary numeric types, add a cpu/gpu round trip test for f16

     austinvhuang

     committed
     456c475View commit details

     Copy full SHA for 456c475

     Browse repository at this point

   • change toGPU to allow ahead-of-time creation of the staging buffer, fix out-of-date docstrings

     austinvhuang

     committed
     d6a02d5View commit details

     Copy full SHA for d6a02d5

     Browse repository at this point

   • get rid of redundant kDebug flag since NDEBUG turns off the logger already in logging.h

     austinvhuang

     committed
     4c6a6acView commit details

     Copy full SHA for 4c6a6ac

     Browse repository at this point

   • assignment and casting operators, clean up test tcode

     austinvhuang

     committed
     b46b213View commit details

     Copy full SHA for b46b213

     Browse repository at this point

   • make intermediate values for half conversions const

     austinvhuang

     committed
     3e3602cView commit details

     Copy full SHA for 3e3602c

     Browse repository at this point

   • half round-trip tests pass

     austinvhuang

     committed
     1dac2a3View commit details

     Copy full SHA for 1dac2a3

     Browse repository at this point
8. Commits on Jul 20, 2024

   • Initial work towards support for fp16 (WIP)

     austinvhuang

     committed
     a266082View commit details

     Copy full SHA for a266082

     Browse repository at this point
9. Commits on Jul 19, 2024

   • add cmake builds for examples

     austinvhuang

     committed
     5e030ebView commit details

     Copy full SHA for 5e030eb

     Browse repository at this point

   • pull in top-level cmakelists from micheals PR

     austinvhuang

     committed
     1125966View commit details

     Copy full SHA for 1125966

     Browse repository at this point

   • merge

     austinvhuang

     committed
     420e460View commit details

     Copy full SHA for 420e460

     Browse repository at this point

   • cherrypick cmake/*.cmake scripts from https://github.com/AnswerDotAI/gpu.cpp/pull/16

     austinvhuang

     committed
     d9705edView commit details

     Copy full SHA for d9705ed

     Browse repository at this point

   • fix stale cmakelists

     austinvhuang

     committed
     0bcc8d0View commit details

     Copy full SHA for 0bcc8d0

     Browse repository at this point

   • mm target -> matul for consistency, reduce progress report log spamming of setup.py

     austinvhuang

     committed
     1a654f1View commit details

     Copy full SHA for 1a654f1

     Browse repository at this point

   • add unified top-level examples/Makefile based on https://github.com/AnswerDotAI/gpu.cpp/pull/16 w/o windows (except for wsl2) native branching code, provide feedback on file load failure in shadert…

     austinvhuang

     committed
     354e76aView commit details

     Copy full SHA for 354e76a

     Browse repository at this point

   • remove revert.patch from repo

     austinvhuang

     committed
     45e640dView commit details

     Copy full SHA for 45e640d

     Browse repository at this point