Skip to content

Fix KeyError in hipcompile when fetching early hostcall metadata #924

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.

Sign up for GitHub

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

Merged
luraess merged 2 commits into from
May 27, 2026
+7 −4
Merged

Fix KeyError in hipcompile when fetching early hostcall metadata #924

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
23d2813
Fixup
luraess May 26, 2026
2786537
Try improve
luraess May 26, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 7 additions & 4 deletions src/compiler/codegen.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,8 +16,6 @@ const _hip_compiler_cache = Dict{HIP.HIPDevice, Dict{Any, HIP.HIPFunction}}()

# hash(fun, hash(f, hash(tt))) => HIPKernel
const _kernel_instances = Dict{UInt, Runtime.HIPKernel}()
# UInt (hash(job)) => Vector{Symbol} (global hostcall names)
const _global_hostcalls = Dict{UInt, Vector{Symbol}}()

function compiler_cache(dev::HIP.HIPDevice)
get!(() -> Dict{UInt, Any}(), _hip_compiler_cache, dev)
Expand All @@ -34,7 +32,10 @@ function GPUCompiler.link_libraries!(@nospecialize(job::HIPCompilerJob), mod::LL
Tuple{CompilerJob{GCNCompilerTarget},typeof(mod)}, job, mod)

# Detect global hostcalls here, before optimizations & cleanup occur.
_global_hostcalls[hash(job)] = find_global_hostcalls(mod)
# Accumulate into task-local storage so hipcompile can retrieve them
# on the same task, without any global dict or hash-collision race.
tls_hostcalls = get!(task_local_storage(), :amdgpu_early_hostcalls, Symbol[])
append!(tls_hostcalls, find_global_hostcalls(mod))

link_device_libs!(
job.config.target, mod;
Expand Down Expand Up @@ -200,7 +201,9 @@ function hipcompile(@nospecialize(job::CompilerJob))
GPUCompiler.compile(:obj, job)
end

global_hostcalls = pop!(_global_hostcalls, hash(job))
# Collect early-detected hostcalls written by link_libraries! on this task.
# Falls back gracefully to empty if link_libraries! was not called.
global_hostcalls = pop!(task_local_storage(), :amdgpu_early_hostcalls, Symbol[])
# Late global hostcalls detection.
append!(global_hostcalls, find_global_hostcalls(meta.ir))

Expand Down
Loading