8ef460fc51
This patch adds parallel processing of chunks. When reducing very large inputs, e.g. functions with 500k basic blocks, processing chunks in parallel can significantly speed up the reduction. To allow modifying clones of the original module in parallel, each clone needs their own LLVMContext object. To achieve this, each job parses the input module with their own LLVMContext. In case a job successfully reduced the input, it serializes the result module as bitcode into a result array. To ensure parallel reduction produces the same results as serial reduction, only the first successfully reduced result is used, and results of other successful jobs are dropped. Processing resumes after the chunk that was successfully reduced. The number of threads to use can be configured using the -j option. It defaults to 1, which means serial processing. Reviewed By: Meinersbur Differential Revision: https://reviews.llvm.org/D113857
67 lines
2.8 KiB
LLVM
67 lines
2.8 KiB
LLVM
; RUN: llvm-reduce %s -o %t --delta-passes=operands-skip --test FileCheck --test-arg %s --test-arg --match-full-lines --test-arg --check-prefix=INTERESTING --test-arg --input-file
|
|
; RUN: FileCheck %s --input-file %t --check-prefixes=REDUCED
|
|
|
|
; RUN: llvm-reduce -j 2 %s -o %t.1 --delta-passes=operands-skip --test FileCheck --test-arg %s --test-arg --match-full-lines --test-arg --check-prefix=INTERESTING --test-arg --input-file
|
|
; RUN: FileCheck %s --input-file %t.1 --check-prefixes=REDUCED
|
|
|
|
; RUN: llvm-reduce -j 4 %s -o %t.2 --delta-passes=operands-skip --test FileCheck --test-arg %s --test-arg --match-full-lines --test-arg --check-prefix=INTERESTING --test-arg --input-file
|
|
; RUN: FileCheck %s --input-file %t.2 --check-prefixes=REDUCED
|
|
|
|
|
|
; INTERESTING: store i32 43, i32* {{(%imm|%indirect)}}, align 4
|
|
; REDUCED: store i32 43, i32* %imm, align 4
|
|
|
|
; INTERESTING: store i32 44, i32* {{(%imm|%indirect|%phi)}}, align 4
|
|
; REDUCED: store i32 44, i32* %phi, align 4
|
|
|
|
; INTERESTING: store i32 45, i32* {{(%imm|%indirect|%phi|%val)}}, align 4
|
|
; REDUCED: store i32 45, i32* %val, align 4
|
|
|
|
; INTERESTING: store i32 46, i32* {{(%imm|%indirect|%phi|%val|@Global)}}, align 4
|
|
; REDUCED: store i32 46, i32* @Global, align 4
|
|
|
|
; INTERESTING: store i32 47, i32* {{(%imm|%indirect|%phi|%val|@Global|%arg2)}}, align 4
|
|
; REDUCED: store i32 47, i32* %arg2, align 4
|
|
|
|
; INTERESTING: store i32 48, i32* {{(%imm|%indirect|%phi|%val|@Global|%arg2|%arg1)}}, align 4
|
|
; REDUCED: store i32 48, i32* %arg1, align 4
|
|
|
|
; INTERESTING: store i32 49, i32* {{(%imm|%indirect|%phi|%val|@Global|%arg2|%arg1|null)}}, align 4
|
|
; REDUCED: store i32 49, i32* null, align 4
|
|
|
|
; REDUCED: store i32 50, i32* %arg1, align 4
|
|
; REDUCED: store i32 51, i32* %arg1, align 4
|
|
|
|
@Global = global i32 42
|
|
|
|
define void @func(i32* %arg1, i32* %arg2) {
|
|
entry:
|
|
%val = getelementptr i32, i32* getelementptr (i32, i32* @Global, i32 1), i32 2
|
|
br i1 undef, label %branch, label %loop
|
|
|
|
branch:
|
|
%nondominating1 = getelementptr i32, i32* %val, i32 3
|
|
br label %loop
|
|
|
|
loop:
|
|
%phi = phi i32* [ null, %entry ], [ %nondominating1, %branch ], [ %nondominating2, %loop ]
|
|
%imm = getelementptr i32, i32* %phi, i32 4
|
|
%indirect = getelementptr i32, i32* %imm, i32 5
|
|
|
|
store i32 43, i32* %imm, align 4 ; Don't reduce to %indirect (not "more reduced" than %imm)
|
|
store i32 44, i32* %imm, align 4 ; Reduce to %phi
|
|
store i32 45, i32* %imm, align 4 ; Reduce to %val
|
|
store i32 46, i32* %imm, align 4 ; Reduce to @Global
|
|
store i32 47, i32* %imm, align 4 ; Reduce to %arg1
|
|
store i32 48, i32* %imm, align 4 ; Reduce to %arg2
|
|
store i32 49, i32* %imm, align 4 ; Reduce to null
|
|
|
|
%nondominating2 = getelementptr i32, i32* %indirect, i32 6
|
|
br i1 undef, label %loop, label %exit
|
|
|
|
exit:
|
|
store i32 50, i32* %arg2, align 4 ; Reduce to %arg1 (compactify function arguments)
|
|
store i32 51, i32* %arg1, align 4 ; Don't reduce
|
|
ret void
|
|
}
|