support sparse opnorm #1466
Merged
support sparse opnorm #1466
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
A side question: maybe GPUArrays should have some kind of sparse abstraction so that this code can be generic? |
|
I realize somehow this kernel doesn't work on large sparse array, and I got the following n = 15
A = sprand(1<<n, 1<<n, 0.01)
dA = cu(A)Any idea why it errors in this way? julia> sum_dim1(CuSparseMatrixCSR(dA))
ERROR: CUDA error: invalid argument (code 1, ERROR_INVALID_VALUE)
Stacktrace:
[1] throw_api_error(res::CUDA.cudaError_enum)
@ CUDA ~/.julia/packages/CUDA/5jdFl/lib/cudadrv/error.jl:91
[2] macro expansion
@ ~/.julia/packages/CUDA/5jdFl/lib/cudadrv/error.jl:101 [inlined]
[3] cuLaunchKernel(f::CuFunction, gridDimX::UInt32, gridDimY::UInt32, gridDimZ::UInt32, blockDimX::UInt32, blockDimY::UInt32, blockDimZ::UInt32, sharedMemBytes::Int64, hStream::CuStream, kernelParams::Vector{Ptr{Nothing}}, extra::Ptr{Nothing})
@ CUDA ~/.julia/packages/CUDA/5jdFl/lib/utils/call.jl:26
[4] #39
@ ~/.julia/packages/CUDA/5jdFl/lib/cudadrv/execution.jl:69 [inlined]
[5] macro expansion
@ ~/.julia/packages/CUDA/5jdFl/lib/cudadrv/execution.jl:33 [inlined]
[6] macro expansion
@ ./none:0 [inlined]
[7] pack_arguments(::CUDA.var"#39#40"{Bool, Int64, CuStream, CuFunction, CuDim3, CuDim3}, ::CUDA.KernelState, ::CuDeviceVector{Float32, 1}, ::CuSparseDeviceMatrixCSR{Float32, Int32, 1})
@ CUDA ./none:0
[8] #launch#38
@ ~/.julia/packages/CUDA/5jdFl/lib/cudadrv/execution.jl:62 [inlined]
[9] #44
@ ~/.julia/packages/CUDA/5jdFl/lib/cudadrv/execution.jl:136 [inlined]
[10] macro expansion
@ ~/.julia/packages/CUDA/5jdFl/lib/cudadrv/execution.jl:95 [inlined]
[11] macro expansion
@ ./none:0 [inlined]
[12] convert_arguments
@ ./none:0 [inlined]
[13] #cudacall#43
@ ~/.julia/packages/CUDA/5jdFl/lib/cudadrv/execution.jl:135 [inlined]
[14] macro expansion
@ ~/.julia/packages/CUDA/5jdFl/src/compiler/execution.jl:204 [inlined]
[15] macro expansion
@ ./none:0 [inlined]
[16] #call#241
@ ./none:0 [inlined]
[17] #_#262
@ ~/.julia/packages/CUDA/5jdFl/src/compiler/execution.jl:462 [inlined]
[18] macro expansion
@ ~/.julia/packages/CUDA/5jdFl/src/compiler/execution.jl:104 [inlined]
[19] sum_dim1(A::CuSparseMatrixCSR{Float32, Int32})
@ EaRydCUDA ~/EaRyd/lib/EaRydCUDA/src/opnorm.jl:17
[20] top-level scope
@ ~/EaRyd/lib/EaRydCUDA/benchmark/expmv.jl:29 |
|
Ah nvm, I should div the threads |
Codecov Report
@@ Coverage Diff @@
## master #1466 +/- ##
==========================================
- Coverage 77.31% 76.57% -0.74%
==========================================
Files 120 121 +1
Lines 9265 9290 +25
==========================================
- Hits 7163 7114 -49
- Misses 2102 2176 +74
Continue to review full report at Codecov.
|
maleadt
reviewed
Apr 11, 2022
maleadt
reviewed
May 4, 2022
|
Thanks! |
simonbyrne
pushed a commit
to simonbyrne/CUDA.jl
that referenced
this pull request
Nov 13, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I only supported Inf opnorm, which is relatively simple, and is used in Krylov subspace methods. I guess it's fine to just error on other norms so if anyone hit it they could submit PR for it.