Rewrite early and late optimize pipeline with NewPM#2711
Merged
Conversation
Contributor
|
Your PR requires formatting changes to meet the project's style guidelines. Click here to view the suggested changes.diff --git a/src/compiler/optimize.jl b/src/compiler/optimize.jl
index 027f1b22..98afb1bc 100644
--- a/src/compiler/optimize.jl
+++ b/src/compiler/optimize.jl
@@ -73,89 +73,89 @@ function optimize!(mod::LLVM.Module, tm::LLVM.TargetMachine)
run!(pb, mod, tm)
end
- function middle_optimize!(second_stage=false)
- @dispose pb = NewPMPassBuilder() begin
- registerEnzymeAndPassPipeline!(pb)
- register!(pb, RewriteGenericMemoryPass())
- add!(pb, NewPMAAManager()) do aam
- add!(aam, ScopedNoAliasAA())
- add!(aam, TypeBasedAA())
- add!(aam, BasicAA())
+ function middle_optimize!(second_stage = false)
+ return @dispose pb = NewPMPassBuilder() begin
+ registerEnzymeAndPassPipeline!(pb)
+ register!(pb, RewriteGenericMemoryPass())
+ add!(pb, NewPMAAManager()) do aam
+ add!(aam, ScopedNoAliasAA())
+ add!(aam, TypeBasedAA())
+ add!(aam, BasicAA())
end
- add!(pb, NewPMModulePassManager()) do mpm
- add!(mpm, RewriteGenericMemoryPass())
- add!(mpm, CPUFeaturesPass()) # why is this duplicated?
-
- add!(mpm, NewPMFunctionPassManager()) do fpm
- add!(fpm, InstCombinePass())
- add!(fpm, JLInstSimplifyPass())
- add!(fpm, SimplifyCFGPass())
- add!(fpm, SROAPass())
- add!(fpm, InstCombinePass())
- add!(fpm, JLInstSimplifyPass())
- add!(fpm, JumpThreadingPass())
- add!(fpm, CorrelatedValuePropagationPass())
- add!(fpm, InstCombinePass())
- add!(fpm, JLInstSimplifyPass())
- add!(fpm, ReassociatePass())
- add!(fpm, EarlyCSEPass())
- add!(fpm, AllocOptPass())
-
- add!(fpm, NewPMLoopPassManager(use_memory_ssa=true)) do lpm
- add!(lpm, LoopIdiomRecognizePass())
- add!(lpm, LoopRotatePass())
- add!(lpm, LowerSIMDLoopPass())
- add!(lpm, LICMPass())
- add!(lpm, JuliaLICMPass())
- add!(lpm, SimpleLoopUnswitchPass())
+ add!(pb, NewPMModulePassManager()) do mpm
+ add!(mpm, RewriteGenericMemoryPass())
+ add!(mpm, CPUFeaturesPass()) # why is this duplicated?
+
+ add!(mpm, NewPMFunctionPassManager()) do fpm
+ add!(fpm, InstCombinePass())
+ add!(fpm, JLInstSimplifyPass())
+ add!(fpm, SimplifyCFGPass())
+ add!(fpm, SROAPass())
+ add!(fpm, InstCombinePass())
+ add!(fpm, JLInstSimplifyPass())
+ add!(fpm, JumpThreadingPass())
+ add!(fpm, CorrelatedValuePropagationPass())
+ add!(fpm, InstCombinePass())
+ add!(fpm, JLInstSimplifyPass())
+ add!(fpm, ReassociatePass())
+ add!(fpm, EarlyCSEPass())
+ add!(fpm, AllocOptPass())
+
+ add!(fpm, NewPMLoopPassManager(use_memory_ssa = true)) do lpm
+ add!(lpm, LoopIdiomRecognizePass())
+ add!(lpm, LoopRotatePass())
+ add!(lpm, LowerSIMDLoopPass())
+ add!(lpm, LICMPass())
+ add!(lpm, JuliaLICMPass())
+ add!(lpm, SimpleLoopUnswitchPass())
+ end
+
+ add!(fpm, InstCombinePass())
+ add!(fpm, JLInstSimplifyPass())
+ add!(fpm, NewPMLoopPassManager()) do lpm
+ add!(lpm, IndVarSimplifyPass())
+ add!(lpm, LoopDeletionPass())
end
-
- add!(fpm, InstCombinePass())
- add!(fpm, JLInstSimplifyPass())
- add!(fpm, NewPMLoopPassManager()) do lpm
- add!(lpm, IndVarSimplifyPass())
- add!(lpm, LoopDeletionPass())
+ add!(fpm, LoopUnrollPass(opt_level = 2)) # what opt level?
+ add!(fpm, AllocOptPass())
+ add!(fpm, SROAPass())
+ add!(fpm, GVNPass())
+
+ # This InstCombine needs to be after GVN
+ # Otherwise it will generate load chains in GPU code...
+ add!(fpm, InstCombinePass())
+ add!(fpm, JLInstSimplifyPass())
+ add!(fpm, MemCpyOptPass())
+ add!(fpm, SCCPPass())
+ add!(fpm, InstCombinePass())
+ add!(fpm, JLInstSimplifyPass())
+ add!(fpm, JumpThreadingPass())
+ add!(fpm, DSEPass())
+ add!(fpm, AllocOptPass())
+ add!(fpm, SimplifyCFGPass())
+
+
+ add!(fpm, NewPMLoopPassManager()) do lpm
+ add!(lpm, LoopIdiomRecognizePass())
+ add!(lpm, LoopDeletionPass())
end
- add!(fpm, LoopUnrollPass(opt_level=2)) # what opt level?
- add!(fpm, AllocOptPass())
- add!(fpm, SROAPass())
- add!(fpm, GVNPass())
-
- # This InstCombine needs to be after GVN
- # Otherwise it will generate load chains in GPU code...
- add!(fpm, InstCombinePass())
- add!(fpm, JLInstSimplifyPass())
- add!(fpm, MemCpyOptPass())
- add!(fpm, SCCPPass())
- add!(fpm, InstCombinePass())
- add!(fpm, JLInstSimplifyPass())
- add!(fpm, JumpThreadingPass())
- add!(fpm, DSEPass())
- add!(fpm, AllocOptPass())
- add!(fpm, SimplifyCFGPass())
-
-
- add!(fpm, NewPMLoopPassManager()) do lpm
- add!(lpm, LoopIdiomRecognizePass())
- add!(lpm, LoopDeletionPass())
+ add!(fpm, JumpThreadingPass())
+ add!(fpm, CorrelatedValuePropagationPass())
+ if second_stage
+
+ add!(fpm, ADCEPass())
+ add!(fpm, InstCombinePass())
+ add!(fpm, JLInstSimplifyPass())
+
+ # GC passes
+ add!(fpm, GCInvariantVerifierPass(strong = false))
+ add!(fpm, SimplifyCFGPass())
+ add!(fpm, InstCombinePass())
+ add!(fpm, JLInstSimplifyPass())
+ end # second_stage
end
- add!(fpm, JumpThreadingPass())
- add!(fpm, CorrelatedValuePropagationPass())
- if second_stage
-
- add!(fpm, ADCEPass())
- add!(fpm, InstCombinePass())
- add!(fpm, JLInstSimplifyPass())
-
- # GC passes
- add!(fpm, GCInvariantVerifierPass(strong=false))
- add!(fpm, SimplifyCFGPass())
- add!(fpm, InstCombinePass())
- add!(fpm, JLInstSimplifyPass())
- end # second_stage
- end
end
- run!(pb, mod, tm)
+ run!(pb, mod, tm)
end
end # middle_optimize!
@@ -180,7 +180,7 @@ function optimize!(mod::LLVM.Module, tm::LLVM.TargetMachine)
end
removeDeadArgs!(mod, tm)
detect_writeonly!(mod)
- nodecayed_phis!(mod)
+ return nodecayed_phis!(mod)
end
function addOptimizationPasses!(mpm::LLVM.NewPMPassManager)
@@ -200,7 +200,7 @@ function addOptimizationPasses!(mpm::LLVM.NewPMPassManager)
add!(mpm, AlwaysInlinerPass())
- add!(mpm, NewPMFunctionPassManager()) do fpm
+ return add!(mpm, NewPMFunctionPassManager()) do fpm
# Running `memcpyopt` between this and `sroa` seems to give `sroa` a hard time
# merging the `alloca` for the unboxed data and the `alloca` created by the `alloc_opt`
# pass.
@@ -224,7 +224,7 @@ function addOptimizationPasses!(mpm::LLVM.NewPMPassManager)
# remove those before optimizing loops.
add!(fpm, AllocOptPass())
- add!(fpm, NewPMLoopPassManager(use_memory_ssa=true)) do lpm
+ add!(fpm, NewPMLoopPassManager(use_memory_ssa = true)) do lpm
add!(lpm, LoopRotatePass())
# moving IndVarSimplify here prevented removing the loop in perf_sumcartesian(10:-1:1)
add!(lpm, LoopIdiomRecognizePass())
@@ -240,7 +240,7 @@ function addOptimizationPasses!(mpm::LLVM.NewPMPassManager)
add!(lpm, IndVarSimplifyPass())
add!(lpm, LoopDeletionPass())
end
- add!(fpm, LoopUnrollPass(opt_level=2))
+ add!(fpm, LoopUnrollPass(opt_level = 2))
# Run our own SROA on heap objects before LLVM's
add!(fpm, AllocOptPass())
@@ -284,13 +284,13 @@ function addOptimizationPasses!(mpm::LLVM.NewPMPassManager)
end
function addMachinePasses!(mpm::LLVM.NewPMPassManager)
- add!(mpm, NewPMFunctionPassManager()) do fpm
+ return add!(mpm, NewPMFunctionPassManager()) do fpm
if VERSION < v"1.12.0-DEV.1390"
add!(fpm, CombineMulAddPass())
end
add!(fpm, DivRemPairsPass())
add!(fpm, DemoteFloat16Pass())
- add!(fpm, GVNPass())
+ add!(fpm, GVNPass())
end
end
@@ -317,7 +317,7 @@ function addJuliaLegalizationPasses!(mpm::LLVM.NewPMPassManager, lower_intrinsic
end
# We need these two passes and the instcombine below
# after GC lowering to let LLVM do some constant propagation on the tags.
- # and remove some unnecessary write barrier checks.
+ # and remove some unnecessary write barrier checks.
add!(mpm, NewPMFunctionPassManager()) do fpm
add!(fpm, GVNPass())
add!(fpm, SCCPPass())
@@ -330,10 +330,12 @@ function addJuliaLegalizationPasses!(mpm::LLVM.NewPMPassManager, lower_intrinsic
add!(fpm, InstCombinePass())
add!(fpm, JLInstSimplifyPass())
aggressiveSimplifyCFGOptions =
- (forward_switch_cond=true,
- switch_range_to_icmp=true,
- switch_to_lookup=true,
- hoist_common_insts=true)
+ (
+ forward_switch_cond = true,
+ switch_range_to_icmp = true,
+ switch_to_lookup = true,
+ hoist_common_insts = true,
+ )
add!(fpm, SimplifyCFGPass(; aggressiveSimplifyCFGOptions...))
end
else
diff --git a/src/llvm/transforms.jl b/src/llvm/transforms.jl
index 5462fd43..67255c3a 100644
--- a/src/llvm/transforms.jl
+++ b/src/llvm/transforms.jl
@@ -2372,7 +2372,7 @@ end
function rewrite_generic_memory!(mod::LLVM.Module)
@static if VERSION < v"1.11-"
return false
- else
+ else
for f in functions(mod), bb in blocks(f)
iter = LLVM.API.LLVMGetFirstInstruction(bb)
while iter != C_NULL
@@ -2381,7 +2381,7 @@ function rewrite_generic_memory!(mod::LLVM.Module)
if !isa(inst, LLVM.LoadInst)
continue
end
-
+
if isa(operands(inst)[1], LLVM.ConstantExpr)
legal2, obj = absint(inst)
if legal2 && obj isa Memory && obj == typeof(obj).instance |
vchuravy
force-pushed
the
vc/split_newpm2
branch
from
46d3e06 to
6cd4a76
Compare
vchuravy
force-pushed
the
vc/split_newpm2
branch
from
6cd4a76 to
e653a48
Compare
vchuravy
commented
Oct 28, 2025
vchuravy
force-pushed
the
vc/split_newpm2
branch
from
26f7d6d to
1a1a74a
Compare
vchuravy
changed the title
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #2711 +/- ##
==========================================
- Coverage 72.61% 72.48% -0.14%
==========================================
Files 58 58
Lines 18746 18635 -111
==========================================
- Hits 13613 13508 -105
+ Misses 5133 5127 -6 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Contributor
Benchmark Results
Benchmark PlotsA plot of the benchmark results has been uploaded as an artifact at https://github.com/EnzymeAD/Enzyme.jl/actions/runs/18924163508/artifacts/4411690206. |
This was referenced Oct 28, 2025
Closed
vchuravy
force-pushed
the
vc/split_newpm2
branch
from
dcbfa45 to
00c7c44
Compare
vchuravy
force-pushed
the
vc/split_newpm2
branch
from
00c7c44 to
06460fb
Compare
vchuravy
changed the title
Member
Author
|
Sigh the |
Member
Assuming you're referring to the |
Member
|
that test has been blocked by a julia patch release for a while =/. not in our control -- specifically JuliaLang/julia@8230e8b |
Member
Author
|
Well that's good then xD |
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
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.
No description provided.