Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Replace DevicePtr with Core.LLVMPtr. #199

Merged
merged 4 commits into from
Aug 24, 2020
Merged

Replace DevicePtr with Core.LLVMPtr. #199

merged 4 commits into from
Aug 24, 2020

Conversation

maleadt
Copy link
Member

@maleadt maleadt commented Jun 3, 2020

Depends on JuliaLLVM/LLVM.jl#185

1.5 only, but I want to be sure this works before 1.5 is actually released.

@maleadt maleadt added cuda kernels Stuff about writing CUDA kernels. enhancement New feature or request labels Jun 3, 2020
@maleadt
Copy link
Member Author

maleadt commented Jun 3, 2020

cc @thomasfaingnaert: after removing the addrspacecasts, this also does away with the inttoptr's! Still does a bitcast, since the pointers we emit for LLVMPtr are typeless, but that shouldn't be a problem.

cc @jpsamaroo: this simplification probably also applies to the AMD back-end.

@codecov
Copy link

codecov bot commented Aug 24, 2020

Codecov Report

Merging #199 into master will decrease coverage by 0.32%.
The diff coverage is 66.66%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master     #199      +/-   ##
==========================================
- Coverage   79.34%   79.01%   -0.33%     
==========================================
  Files         165      165              
  Lines        8742     8730      -12     
==========================================
- Hits         6936     6898      -38     
- Misses       1806     1832      +26     
Impacted Files Coverage Δ
src/CUDA.jl 100.00% <ø> (ø)
src/compiler/execution.jl 92.37% <0.00%> (ø)
src/gpuarrays.jl 43.47% <0.00%> (ø)
test/codegen.jl 53.33% <ø> (-4.17%) ⬇️
test/device/wmma.jl 22.22% <ø> (ø)
test/execution.jl 76.11% <ø> (ø)
examples/wmma/low-level.jl 100.00% <100.00%> (ø)
src/array.jl 92.18% <100.00%> (ø)
test/device/array.jl 94.44% <100.00%> (-0.16%) ⬇️
test/device/ldg.jl 100.00% <100.00%> (ø)
... and 9 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update c5b4d14...2eeb8df. Read the comment docs.

@maleadt maleadt marked this pull request as ready for review August 24, 2020 14:56
@maleadt maleadt merged commit dbb5885 into master Aug 24, 2020
@maleadt maleadt deleted the tb/llvmptr branch August 24, 2020 14:56
marius311 added a commit to marius311/CUDA.jl that referenced this pull request Sep 8, 2020
[Diff since v1.3.2](JuliaGPU/CUDA.jl@v1.3.2...v1.3.3)

**Closed issues:**
- Type changing Array conversions give error when allowscalar(false) (JuliaGPU#344)
- getindex(::CuArray, ::Adjoint, ::Colon) fails (JuliaGPU#345)
- View with array indices causes memory copy before broadcast (JuliaGPU#384)
- Regression with Julia 1.5 (JuliaGPU#390)

**Merged pull requests:**
- Replace DevicePtr with Core.LLVMPtr. (JuliaGPU#199) (@maleadt)
- Make sure view indices reside on the GPU too. (JuliaGPU#388) (@maleadt)
- CompatHelper: Update DataStructures to v0.18 (JuliaGPU#389) (@ChrisRackauckas)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
cuda kernels Stuff about writing CUDA kernels. enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant