Repository navigation
Build the thread ID buffer of wait workers concretely - #36
Merged
Merged
Conversation
`ntuple` with a literal count above 10 returns a tuple of unknown length, which
was then converted to `NTuple{32, UInt8}` with generic methods (`ntuple` over a
`Val` of unknown type, comparisons with `Any`) that packages like ForwardDiff
invalidate. When such a package was loaded first, that discarded the
precompiled code that starts wait workers in packages using GPUToolbox.
This file contains hidden or 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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
create_wait_workerbuilt the buffer for the new thread's ID asntuple(i -> 0, 32), which returns a tuple of unknown length (ntupleonly unrolls up to 10 elements), and then converted that toNTuple{32, UInt8}. That conversion was compiled with edges to generic methods (ntupleover aValof unknown type, comparisons withAny) that packages like ForwardDiff invalidate, so with such a package loaded first, Julia discarded the precompiled code that starts wait workers in the packages using GPUToolbox. This builds the buffer with aValcount instead.Found while surveying invalidations in the JuliaGPU stack (JuliaGPU/GPUCompiler.jl#1048). Tested with the test suite.