peakflops: make eltype a static parameter of the method
#1253
Conversation
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
Member
Author
|
The change eliminates all invalidations in the sysimage from running this code as a user: struct I <: Integer end
function Base.Int8(::I) end
function Base.UInt8(::I) endReproducer: ./julia -E 'using SnoopCompileCore; i = (@snoop_invalidations (struct I <: Integer end; function Base.Int8(::I) end; function Base.UInt8(::I) end)); using SnoopCompile; length(uinvalidated(i))'The eliminated invalidations: {
"invalidation_count": 8,
"trees": [
{
"method": "Int8(::Main.RawInvalidations.I) @ Main.RawInvalidations ~/invalidations/script/reproducer.jl:11",
"reason": "inserting",
"mt_backedges": [
{
"type": "Tuple{Type{Int8}, Integer}",
"tree": {
"method_instance": {
"method": "convert(::Type{T}, x::Number) where T<:Number @ Base number.jl:7",
"method_instance": "MethodInstance for convert(::Type{Int8}, ::Integer)"
},
"children": [
{
"method_instance": {
"method": "fill!(a::Union{Array{Int8}, Array{UInt8}}, x::Integer) @ Base array.jl:430",
"method_instance": "MethodInstance for fill!(::Union{Matrix{Int8}, Matrix{UInt8}}, ::Integer)"
},
"children": [
{
"method_instance": {
"method": "ones(::Type{T}, dims::NTuple{N, Integer}) where {T, N} @ Base array.jl:594",
"method_instance": "MethodInstance for ones(::DataType, ::Tuple{Int64, Int64})"
},
"children": [
{
"method_instance": {
"method": "ones(::Type{T}, dims::Union{Integer, AbstractUnitRange}...) where T @ Base array.jl:591",
"method_instance": "MethodInstance for ones(::DataType, ::Int64, ::Int64)"
},
"children": [
{
"method_instance": {
"method": "var\"#peakflops#332\"(eltype::DataType, ntrials::Integer, parallel::Bool, ::typeof(LinearAlgebra.peakflops), n::Integer) @ LinearAlgebra ~/tmp/jl/jl/master-69a22cf4bc/share/julia/stdlib/v1.13/LinearAlgebra/src/LinearAlgebra.jl:753",
"method_instance": "MethodInstance for LinearAlgebra.var\"#peakflops#332\"(::DataType, ::Int64, ::Bool, ::typeof(LinearAlgebra.peakflops), ::Int64)"
},
"children": [
]
},
{
"method_instance": {
"method": "var\"#peakflops#332\"(eltype::DataType, ntrials::Integer, parallel::Bool, ::typeof(LinearAlgebra.peakflops), n::Integer) @ LinearAlgebra ~/tmp/jl/jl/master-69a22cf4bc/share/julia/stdlib/v1.13/LinearAlgebra/src/LinearAlgebra.jl:753",
"method_instance": "MethodInstance for LinearAlgebra.var\"#peakflops#332\"(::DataType, ::Int64, ::Bool, ::typeof(LinearAlgebra.peakflops), ::Int64)"
},
"children": [
]
},
{
"method_instance": {
"method": "var\"#peakflops#332\"(eltype::DataType, ntrials::Integer, parallel::Bool, ::typeof(LinearAlgebra.peakflops), n::Integer) @ LinearAlgebra ~/tmp/jl/jl/master-69a22cf4bc/share/julia/stdlib/v1.13/LinearAlgebra/src/LinearAlgebra.jl:753",
"method_instance": "MethodInstance for LinearAlgebra.var\"#peakflops#332\"(::DataType, ::Int64, ::Bool, ::typeof(LinearAlgebra.peakflops), ::Int64)"
},
"children": [
]
},
{
"method_instance": {
"method": "var\"#peakflops#332\"(eltype::DataType, ntrials::Integer, parallel::Bool, ::typeof(LinearAlgebra.peakflops), n::Integer) @ LinearAlgebra ~/tmp/jl/jl/master-69a22cf4bc/share/julia/stdlib/v1.13/LinearAlgebra/src/LinearAlgebra.jl:753",
"method_instance": "MethodInstance for LinearAlgebra.var\"#peakflops#332\"(::DataType, ::Int64, ::Bool, ::typeof(LinearAlgebra.peakflops), ::Int64)"
},
"children": [
{
"method_instance": {
"method": "peakflops(n::Integer; eltype, ntrials, parallel) @ LinearAlgebra ~/tmp/jl/jl/master-69a22cf4bc/share/julia/stdlib/v1.13/LinearAlgebra/src/LinearAlgebra.jl:753",
"method_instance": "MethodInstance for LinearAlgebra.peakflops(::Int64)"
},
"children": [
]
},
{
"method_instance": {
"method": "peakflops(n::Integer; eltype, ntrials, parallel) @ LinearAlgebra ~/tmp/jl/jl/master-69a22cf4bc/share/julia/stdlib/v1.13/LinearAlgebra/src/LinearAlgebra.jl:753",
"method_instance": "MethodInstance for LinearAlgebra.peakflops(::Int64)"
},
"children": [
{
"method_instance": {
"method": "peakflops(; ...) @ LinearAlgebra ~/tmp/jl/jl/master-69a22cf4bc/share/julia/stdlib/v1.13/LinearAlgebra/src/LinearAlgebra.jl:753",
"method_instance": "MethodInstance for LinearAlgebra.peakflops()"
},
"children": [
]
}
]
}
]
}
]
}
]
}
]
}
]
}
}
],
"backedges": [
],
"mt_cache": [
],
"mt_disable": [
]
},
{
"method": "UInt8(::Main.RawInvalidations.I) @ Main.RawInvalidations ~/invalidations/script/reproducer.jl:12",
"reason": "inserting",
"mt_backedges": [
{
"type": "Tuple{Type{UInt8}, Integer}",
"tree": {
"method_instance": {
"method": "convert(::Type{T}, x::Number) where T<:Number @ Base number.jl:7",
"method_instance": "MethodInstance for convert(::Type{UInt8}, ::Integer)"
},
"children": [
{
"method_instance": {
"method": "fill!(a::Union{Array{Int8}, Array{UInt8}}, x::Integer) @ Base array.jl:430",
"method_instance": "MethodInstance for fill!(::Union{Matrix{Int8}, Matrix{UInt8}}, ::Integer)"
},
"children": [
]
}
]
}
}
],
"backedges": [
],
"mt_cache": [
],
"mt_disable": [
]
}
]
} |
Member
|
Is this called when the sysimage is created? I don't see where (and that is required AFAIU to cause invalidations). |
Member
Author
|
No idea why I suspect that any method callable without arguments gets precompiled. No idea where this is implemented, if so. |
Otherwise `eltype` is inferred as merely `DataType`. Too bad that constprop doesn't fix this on its own. This makes the sysimage more resistant to invalidation.
Member
Author
|
Fixed the merge conflict. |
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## master #1253 +/- ##
==========================================
- Coverage 92.03% 91.97% -0.06%
==========================================
Files 34 34
Lines 15459 15471 +12
==========================================
+ Hits 14227 14230 +3
- Misses 1232 1241 +9 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Member
Author
|
bump |
Member
|
The change looks ok to me |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
Otherwise
eltypeis inferred as merelyDataType. Too bad that constprop doesn't fix this on its own.This will make the sysimage more resistant to invalidation, once JuliaLang/julia updates LinearAlgebra.