Commits

Breadcrumbs

History for

onmain

User selector

All users

Datepicker

All time

Commit history

Commits on Jul 19, 2024

• add cmake builds for examples

  austinvhuang

  committed
  5e030ebView commit details

  Copy full SHA for 5e030eb

  View code at this pointBrowse 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

  View code at this pointBrowse repository at this point

• temporarily revert build refactor in main, staging build refactor in the dev branch

  austinvhuang

  committed
  60b1b7dView commit details

  Copy full SHA for 60b1b7d

  View code at this pointBrowse repository at this point

• Unify the example makefiles!

  MichealReed

  committed
  427ea77View commit details

  Copy full SHA for 427ea77

  View code at this pointBrowse repository at this point

• Merge remote-tracking branch 'origin/main' into build_refactor_windows

  MichealReed

  committed
  ebf4565View commit details

  Copy full SHA for ebf4565

  View code at this pointBrowse repository at this point

Commits on Jul 18, 2024

• Increase version and fix min max usage for Windows

  MichealReed

  committed
  3e09549View commit details

  Copy full SHA for 3e09549

  View code at this pointBrowse repository at this point

• fix OS detection

  MichealReed

  committed
  7d1c255View commit details

  Copy full SHA for 7d1c255

  View code at this pointBrowse repository at this point

• abstract to example.cmake

  MichealReed

  committed
  1730a65View commit details

  Copy full SHA for 1730a65

  View code at this pointBrowse repository at this point

• Windows_NT not Windows

  MichealReed

  committed
  4d55073View commit details

  Copy full SHA for 4d55073

  View code at this pointBrowse repository at this point

• fix debug makefiles

  MichealReed

  committed
  e02b241View commit details

  Copy full SHA for e02b241

  View code at this pointBrowse repository at this point

• no need to copy object other than windows

  MichealReed

  committed
  36f031dView commit details

  Copy full SHA for 36f031d

  View code at this pointBrowse repository at this point

• windows support

  MichealReed

  committed
  3b7c8e6View commit details

  Copy full SHA for 3b7c8e6

  View code at this pointBrowse repository at this point

• Adds windows support

  MichealReed

  committed
  2c4a9b9View commit details

  Copy full SHA for 2c4a9b9

  View code at this pointBrowse repository at this point

Commits on Jul 16, 2024

• ShaderCode -> KernelCode, remove createShader in favor of KernelCode constructor, fix reloading handling in shadertui, reset cursor position instead of clearing screen to reduce flickering, add lib…

  Show description for 0d40ed1
  austinvhuang

  committed
  0d40ed1View commit details

  Copy full SHA for 0d40ed1

  View code at this pointBrowse repository at this point

Commits on Jul 14, 2024

• Add libc++ check to example Makefiles, remove entr check from top-level Makefile

  austinvhuang

  committed
  e567c2aView commit details

  Copy full SHA for e567c2a

  View code at this pointBrowse repository at this point

Commits on Jul 12, 2024

• remove deprecated developers.md and tutorial.md, which has been consolidated in the readme. fix ssl issue with downloading the shared library (see https://stackoverflow.com/questions/68275857/urlli…

  Show description for 0c76223
  austinvhuang

  committed
  0c76223View commit details

  Copy full SHA for 0c76223

  View code at this pointBrowse repository at this point

• add README images

  austinvhuang

  committed
  dbf0e78View commit details

  Copy full SHA for dbf0e78

  View code at this pointBrowse repository at this point

Commits on Jul 10, 2024

• add -stdlib=libc++ to build flags, minor cleanup

  austinvhuang

  committed
  f983bcfView commit details

  Copy full SHA for f983bcf

  View code at this pointBrowse repository at this point

Commits on Jul 4, 2024

• update readme to use nWorkgroups

  austinvhuang

  committed
  3e7943aView commit details

  Copy full SHA for 3e7943a

  View code at this pointBrowse repository at this point

• Specify xyz of number of workgroups for createKernel (ala cuda) instead of specifying total # threads (and automatically dividing each dimension by workgroup size). tiled matmul shader now produces…

  Show description for f72c857
  austinvhuang

  committed
  f72c857View commit details

  Copy full SHA for f72c857

  View code at this pointBrowse repository at this point

Commits on Jul 2, 2024

• TensorList -> Bindings to signal intent and generalize to handling TensorViews, remove unnecessary nThreads parameter from resetCommandBuffer, add cdiv + cdiv support for shapes, preparing to refac…

  Show description for 1835f4d
  austinvhuang

  committed
  1835f4dView commit details

  Copy full SHA for 1835f4d

  View code at this pointBrowse repository at this point

Commits on Jun 30, 2024

• refine TensorList view functionality, start matmul example, tweaks to examples

  austinvhuang

  committed
  0a6a977View commit details

  Copy full SHA for 0a6a977

  View code at this pointBrowse repository at this point

Commits on Jun 28, 2024

• use wgpu style convention for functions

  austinvhuang

  committed
  fae185fView commit details

  Copy full SHA for fae185f

  View code at this pointBrowse repository at this point

Commits on Jun 26, 2024

• move kernel tests to transformers, tweak readme and builds. move superfluous files out of utils/

  austinvhuang

  committed
  e635ec4View commit details

  Copy full SHA for e635ec4

  View code at this pointBrowse repository at this point

• update readme to focus on prebuilt binary builds

  austinvhuang

  committed
  6398611View commit details

  Copy full SHA for 6398611

  View code at this pointBrowse repository at this point

Commits on Jun 25, 2024

• add more docstrings to gpu.h, add sonnet fluid sim to experimental, fix makefile dependency chain for examples

  austinvhuang

  committed
  59c7bf6View commit details

  Copy full SHA for 59c7bf6

  View code at this pointBrowse repository at this point

• clean up top level makefile to default to hello world example build

  austinvhuang

  committed
  a4f5b4fView commit details

  Copy full SHA for a4f5b4f

  View code at this pointBrowse repository at this point

• fast builds working for linux + examples

  austinvhuang

  committed
  2b6154bView commit details

  Copy full SHA for 2b6154b

  View code at this pointBrowse repository at this point

• linux shared library build working for hello world example

  austinvhuang

  committed
  3ecb5a9View commit details

  Copy full SHA for 3ecb5a9

  View code at this pointBrowse repository at this point

• linux shared library build (WIP - not working yet)

  austinvhuang

  committed
  f6f4694View commit details

  Copy full SHA for f6f4694

  View code at this pointBrowse repository at this point

Commits on Jun 24, 2024

• Migrate to prebuilt libdawn binary. MacOS+apple silicon is working, will check linux next. Add setup.cpp script for downloading the shared library and other setup tasks.

  austinvhuang

  committed
  d285702View commit details

  Copy full SHA for d285702

  View code at this pointBrowse repository at this point

Commits on Jun 23, 2024

• minor cleaup + release prep

  austinvhuang

  committed
  34c17a9View commit details

  Copy full SHA for 34c17a9

  View code at this pointBrowse repository at this point

Commits on Jun 19, 2024

• Refactor - fold output into one of the TensorList arguments, get rid of MultiKernel (for now), update examples accordingly

  austinvhuang

  committed
  cb77adcView commit details

  Copy full SHA for cb77adc

  View code at this pointBrowse repository at this point

Commits on Jun 18, 2024

• API refinement in preparation for release (GPUTensor -> Tensor, GPUContext -> Context for Create* consistency), fix example builds, start docstrings (WIP)

  austinvhuang

  committed
  32828d5View commit details

  Copy full SHA for 32828d5

  View code at this pointBrowse repository at this point

Commits on Jun 17, 2024

• handle 3D nthreads -> workgroup mappings for CreateKernel, fix GELU bug for inputs > 10.0, get rid of inferred # threads from output from output dimensions (too easy to introduce bugs), refine rend…

  Show description for 0735132
  austinvhuang

  committed
  0735132View commit details

  Copy full SHA for 0735132

  View code at this pointBrowse repository at this point