Int64 atomics #347
Int64 atomics #347
Conversation
|
Your PR requires formatting changes to meet the project's style guidelines. Click here to view the suggested changes.diff --git a/lib/intrinsics/src/atomic.jl b/lib/intrinsics/src/atomic.jl
index 9bbbdbe..c3027bf 100644
--- a/lib/intrinsics/src/atomic.jl
+++ b/lib/intrinsics/src/atomic.jl
@@ -63,23 +63,35 @@ end
for as in atomic_memory_types
@eval begin
-# There is native support for atomic_xchg on Float32, but not for Float64,
-# so we always reinterpret for consistency.
+ # There is native support for atomic_xchg on Float32, but not for Float64,
+ # so we always reinterpret for consistency.
@device_function atomic_xchg!(p::LLVMPtr{Float32,$as}, val::Float32) =
- reinterpret(Float32, atomic_xchg!(reinterpret(LLVMPtr{UInt32,$as}, p),
- reinterpret(UInt32, val)))
-@device_function atomic_xchg!(p::LLVMPtr{Float64,$as}, val::Float64) =
- reinterpret(Float64, atomic_xchg!(reinterpret(LLVMPtr{UInt64,$as}, p),
- reinterpret(UInt64, val)))
+ reinterpret(
+ Float32, atomic_xchg!(
+ reinterpret(LLVMPtr{UInt32, $as}, p),
+ reinterpret(UInt32, val)
+ )
+ )
+ @device_function atomic_xchg!(p::LLVMPtr{Float64, $as}, val::Float64) =
+ reinterpret(
+ Float64, atomic_xchg!(
+ reinterpret(LLVMPtr{UInt64, $as}, p),
+ reinterpret(UInt64, val)
+ )
+ )
@device_function atomic_cmpxchg!(p::LLVMPtr{Float32,$as}, cmp::Float32, val::Float32) =
reinterpret(Float32, atomic_cmpxchg!(reinterpret(LLVMPtr{UInt32,$as}, p),
reinterpret(UInt32, cmp),
reinterpret(UInt32, val)))
-@device_function atomic_cmpxchg!(p::LLVMPtr{Float64,$as}, cmp::Float64, val::Float64) =
- reinterpret(Float64, atomic_cmpxchg!(reinterpret(LLVMPtr{UInt64,$as}, p),
- reinterpret(UInt64, cmp),
- reinterpret(UInt64, val)))
+ @device_function atomic_cmpxchg!(p::LLVMPtr{Float64, $as}, cmp::Float64, val::Float64) =
+ reinterpret(
+ Float64, atomic_cmpxchg!(
+ reinterpret(LLVMPtr{UInt64, $as}, p),
+ reinterpret(UInt64, cmp),
+ reinterpret(UInt64, val)
+ )
+ )
end
end
@@ -256,7 +268,8 @@ for (op,impl) in [(+) => atomic_add!,
Base.max => atomic_max!,
Base.min => atomic_min!]
@eval @inline atomic_arrayset(A::AbstractArray{T}, I::Integer, ::typeof($op),
- val::T) where {T <: Union{atomic_integer_types...}} =
+ val::T
+ ) where {T <: Union{atomic_integer_types...}} =
$impl(pointer(A, I), val)
end
diff --git a/test/atomics.jl b/test/atomics.jl
index 068e8cd..f1724bf 100644
--- a/test/atomics.jl
+++ b/test/atomics.jl
@@ -1,16 +1,16 @@
@testset "atomics" begin
-function atomic_count(counter)
- OpenCL.@atomic counter[] += 1
- return
-end
+ function atomic_count(counter)
+ OpenCL.@atomic counter[] += 1
+ return
+ end
-@testset "atomic_add! ($T)" for T in [Int32, UInt32, Int64, UInt64]
- if sizeof(T) == 4 || "cl_khr_int64_extended_atomics" in cl.device().extensions
- a = OpenCL.zeros(T)
- @opencl global_size=1000 atomic_count(a)
- @test OpenCL.@allowscalar a[] == 1000
+ @testset "atomic_add! ($T)" for T in [Int32, UInt32, Int64, UInt64]
+ if sizeof(T) == 4 || "cl_khr_int64_extended_atomics" in cl.device().extensions
+ a = OpenCL.zeros(T)
+ @opencl global_size = 1000 atomic_count(a)
+ @test OpenCL.@allowscalar a[] == 1000
+ end
end
-end
end |
|
Ok, it works now! I was just missing a dispatch so it previously was using the fallback CAS implementation for everything. Still, that shouldn't error, just be slow, so something else is going wrong here. |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #347 +/- ##
=======================================
Coverage 78.86% 78.86%
=======================================
Files 12 12
Lines 672 672
=======================================
Hits 530 530
Misses 142 142 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
|
Any ideas why this is now failing on Windows? (and Mac?) |
|
Windows is #317, macOS hangs seem related... |
|
Hm, the timeouts seem to happen on #348 too. So maybe unrelated? |
|
@maleadt it looks like you still need to manually approve the workflow run even if you're the one commiting 😅 |
Do you have that code around still? Would be good to have that as an issue; invalid SPIR-V codegen seems bad enough to investigate. |
|
It was simply using the fallback definition here with an Int64 input, so that code should still reproduce it. I can try reducing it to a standalone reproducer tomorrow |
2 participants
This doesn't work as-is, when running the tests, I get: