Create your own GitHub profile
Sign up for your own profile on GitHub, the best place to host code, manage projects, and build software alongside 50 million developers.
Sign up
Pinned
5,278 contributions in the last year
Contribution activity
October 2020
- tkf/CodeX.jl Julia
- tkf/compiler-explorer Assembly
- tkf/opencilk-project C++
- tkf/DebuggableCompiler.jl Julia
- tkf/LiveServer.jl Julia
- tkf/zulip-archive Python
- tkf/zulip-archive-test
Created a pull request in JuliaDebug/Cthulhu.jl that received 9 comments
Support descending into deoptimized and non-deoptimized calls
Update: In response to @vchuravy's comment #98 (comment), this PR now lets the user decide the method to descend into. So, it's possible to look at…
- Add a FAQ page
- Note that the input sizes of the examples are small
- Fix a typo
- Update juliafolds-logo.ico
- Add a link to top-level
- Generate a table for parallel processing libraries
- Some more tweaks in quick-introduction.md
- Workaround type instability of collatz_histogram
- Tweak github button for mobile
- Some more tweaks in quick-introduction.md
- Suppress an error from webfactory/ssh-agent
- Really avoid failing in PRs from forks
- Avoid failing in PRs from forks
- xfail: Introduce bugs for testing #8
- Always include test results (but hide it via CSS)
- Some more tweaks in quick-introduction.md
- Note that the git repository has the full environments
- Format .github/workflows/deploy.yml
- Fix plotting (vanished x/y labels) on GitHub Actions
- Setup Mergify
- Setup preview on PR
- Add badges
- Set up ~/.julia/registries/General as a Git repository
- Add an example: estimating pi using Monte-Carlo method
- Create ToC page for examples
- Demo with lazy mapping (and fix a bug)
- Setup Documenter + Literate
- [DON'T MERGE] Triggering CI
- Tweak/fix README
- Post code coverage to Codecov
- Revert "Workaround "Out of GPU memory" error"
- Revert "Workaround "Out of GPU memory" error"
- Setup CI on gitlab.com/JuliaGPU
- Rename CUDAFolds -> FoldsCUDA
- Remove tmp.jl
- Add an example using transducer
- Remove non-generic debug statements
- Skip GPU tests on PkgEval
- Run doctests
- Setup Mergify
- Setup tests w/o GPU
Created an issue in JuliaGPU/CUDA.jl that received 5 comments
mapreduce assumes commutative op
Describe the bug
It looks like CUDA.jl's mapreduce assumes commutativity of op here:
CUDA.jl/src/mapreduce.jl Lines 67 to 73 in bc2d5f4
a…