6502/llvm-6502
1
0
Fork 0
mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2025-11-02 22:23:10 +00:00
Code Issues Projects Releases Wiki Activity
Files
11947113de24f14f0bb7c7895b5087f001839e5d
llvm-6502/test/Transforms/InstCombine/load-cmp.ll
Hal Finkel 5ecf528fc2 [InstCombine] Remove unnecessary variable indexing into single-element arrays 
This change addresses a deficiency pointed out in PR22629. To copy from the bug
report:

[from the bug report]

Consider this code:

int f(int x) {
  int a[] = {12};
  return a[x];
}

GCC knows to optimize this to

movl     $12, %eax
ret

The code generated by recent Clang at -O3 is:

movslq   %edi, %rax
movl     .L_ZZ1fiE1a(,%rax,4), %eax
retq

.L_ZZ1fiE1a:
  .long    12                      # 0xc

[end from the bug report]

This definitely seems worth fixing. I've also seen this kind of code before (as
the base case of generic vector wrapper templates with one element).

The general idea is to look at the GEP feeding a load or a store, which has
some variable as its first non-zero index, and determine if that index must be
zero (or else an out-of-bounds access would occur). We can do this for allocas
and globals with constant initializers where we know the maximum size of the
underlying object. When we find such a GEP, we create a new one for the memory
access with that first variable index replaced with a constant zero.

Even if we can't eliminate the memory access (and sometimes we can't), it is
still useful because it removes unnecessary indexing calculations.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@229959 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-20 03:05:53 +00:00

373 lines
9.9 KiB
LLVM
Raw Blame History

; RUN: opt -instcombine -S < %s | FileCheck -check-prefix=NODL %s
; RUN: opt -instcombine -S -default-data-layout="p:32:32:32-p1:16:16:16-n8:16:32:64" < %s | FileCheck -check-prefix=P32 %s
@G16 = internal constant [10 x i16] [i16 35, i16 82, i16 69, i16 81, i16 85,
i16 73, i16 82, i16 69, i16 68, i16 0]
@G16_as1 = internal addrspace(1) constant [10 x i16] [i16 35, i16 82, i16 69, i16 81, i16 85,
i16 73, i16 82, i16 69, i16 68, i16 0]
@GD = internal constant [6 x double]
[double -10.0, double 1.0, double 4.0, double 2.0, double -20.0, double -40.0]
%Foo = type { i32, i32, i32, i32 }
@GS = internal constant %Foo { i32 1, i32 4, i32 9, i32 14 }
@GStructArr = internal constant [4 x %Foo] [ %Foo { i32 1, i32 4, i32 9, i32 14 },
%Foo { i32 5, i32 4, i32 6, i32 11 },
%Foo { i32 6, i32 5, i32 9, i32 20 },
%Foo { i32 12, i32 3, i32 9, i32 8 } ]
define i1 @test1(i32 %X) {
%P = getelementptr inbounds [10 x i16]* @G16, i32 0, i32 %X
%Q = load i16* %P
%R = icmp eq i16 %Q, 0
ret i1 %R
; NODL-LABEL: @test1(
; NODL-NEXT: %R = icmp eq i32 %X, 9
; NODL-NEXT: ret i1 %R
; P32-LABEL: @test1(
; P32-NEXT: %R = icmp eq i32 %X, 9
; P32-NEXT: ret i1 %R
}
define i1 @test1_noinbounds(i32 %X) {
%P = getelementptr [10 x i16]* @G16, i32 0, i32 %X
%Q = load i16* %P
%R = icmp eq i16 %Q, 0
ret i1 %R
; NODL-LABEL: @test1_noinbounds(
; NODL-NEXT: %P = getelementptr [10 x i16]* @G16, i32 0, i32 %X
; P32-LABEL: @test1_noinbounds(
; P32-NEXT: %R = icmp eq i32 %X, 9
; P32-NEXT: ret i1 %R
}
define i1 @test1_noinbounds_i64(i64 %X) {
%P = getelementptr [10 x i16]* @G16, i64 0, i64 %X
%Q = load i16* %P
%R = icmp eq i16 %Q, 0
ret i1 %R
; NODL-LABEL: @test1_noinbounds_i64(
; NODL-NEXT: %P = getelementptr [10 x i16]* @G16, i64 0, i64 %X
; P32-LABEL: @test1_noinbounds_i64(
; P32: %R = icmp eq i32 %1, 9
; P32-NEXT: ret i1 %R
}
define i1 @test1_noinbounds_as1(i32 %x) {
%p = getelementptr [10 x i16] addrspace(1)* @G16_as1, i16 0, i32 %x
%q = load i16 addrspace(1)* %p
%r = icmp eq i16 %q, 0
ret i1 %r
; P32-LABEL: @test1_noinbounds_as1(
; P32-NEXT: trunc i32 %x to i16
; P32-NEXT: %r = icmp eq i16 %1, 9
; P32-NEXT: ret i1 %r
}
define i1 @test2(i32 %X) {
%P = getelementptr inbounds [10 x i16]* @G16, i32 0, i32 %X
%Q = load i16* %P
%R = icmp slt i16 %Q, 85
ret i1 %R
; NODL-LABEL: @test2(
; NODL-NEXT: %R = icmp ne i32 %X, 4
; NODL-NEXT: ret i1 %R
}
define i1 @test3(i32 %X) {
%P = getelementptr inbounds [6 x double]* @GD, i32 0, i32 %X
%Q = load double* %P
%R = fcmp oeq double %Q, 1.0
ret i1 %R
; NODL-LABEL: @test3(
; NODL-NEXT: %R = icmp eq i32 %X, 1
; NODL-NEXT: ret i1 %R
; P32-LABEL: @test3(
; P32-NEXT: %R = icmp eq i32 %X, 1
; P32-NEXT: ret i1 %R
}
define i1 @test4(i32 %X) {
%P = getelementptr inbounds [10 x i16]* @G16, i32 0, i32 %X
%Q = load i16* %P
%R = icmp sle i16 %Q, 73
ret i1 %R
; NODL-LABEL: @test4(
; NODL-NEXT: lshr i32 933, %X
; NODL-NEXT: and i32 {{.*}}, 1
; NODL-NEXT: %R = icmp ne i32 {{.*}}, 0
; NODL-NEXT: ret i1 %R
; P32-LABEL: @test4(
; P32-NEXT: lshr i32 933, %X
; P32-NEXT: and i32 {{.*}}, 1
; P32-NEXT: %R = icmp ne i32 {{.*}}, 0
; P32-NEXT: ret i1 %R
}
define i1 @test4_i16(i16 %X) {
%P = getelementptr inbounds [10 x i16]* @G16, i32 0, i16 %X
%Q = load i16* %P
%R = icmp sle i16 %Q, 73
ret i1 %R
; NODL-LABEL: @test4_i16(
; NODL-NEXT: lshr i16 933, %X
; NODL-NEXT: and i16 {{.*}}, 1
; NODL-NEXT: %R = icmp ne i16 {{.*}}, 0
; NODL-NEXT: ret i1 %R
; P32-LABEL: @test4_i16(
; P32-NEXT: sext i16 %X to i32
; P32-NEXT: lshr i32 933, %1
; P32-NEXT: and i32 {{.*}}, 1
; P32-NEXT: %R = icmp ne i32 {{.*}}, 0
; P32-NEXT: ret i1 %R
}
define i1 @test5(i32 %X) {
%P = getelementptr inbounds [10 x i16]* @G16, i32 0, i32 %X
%Q = load i16* %P
%R = icmp eq i16 %Q, 69
ret i1 %R
; NODL-LABEL: @test5(
; NODL-NEXT: icmp eq i32 %X, 2
; NODL-NEXT: icmp eq i32 %X, 7
; NODL-NEXT: %R = or i1
; NODL-NEXT: ret i1 %R
; P32-LABEL: @test5(
; P32-NEXT: icmp eq i32 %X, 2
; P32-NEXT: icmp eq i32 %X, 7
; P32-NEXT: %R = or i1
; P32-NEXT: ret i1 %R
}
define i1 @test6(i32 %X) {
%P = getelementptr inbounds [6 x double]* @GD, i32 0, i32 %X
%Q = load double* %P
%R = fcmp ogt double %Q, 0.0
ret i1 %R
; NODL-LABEL: @test6(
; NODL-NEXT: add i32 %X, -1
; NODL-NEXT: %R = icmp ult i32 {{.*}}, 3
; NODL-NEXT: ret i1 %R
; P32-LABEL: @test6(
; P32-NEXT: add i32 %X, -1
; P32-NEXT: %R = icmp ult i32 {{.*}}, 3
; P32-NEXT: ret i1 %R
}
define i1 @test7(i32 %X) {
%P = getelementptr inbounds [6 x double]* @GD, i32 0, i32 %X
%Q = load double* %P
%R = fcmp olt double %Q, 0.0
ret i1 %R
; NODL-LABEL: @test7(
; NODL-NEXT: add i32 %X, -1
; NODL-NEXT: %R = icmp ugt i32 {{.*}}, 2
; NODL-NEXT: ret i1 %R
; P32-LABEL: @test7(
; P32-NEXT: add i32 %X, -1
; P32-NEXT: %R = icmp ugt i32 {{.*}}, 2
; P32-NEXT: ret i1 %R
}
define i1 @test8(i32 %X) {
%P = getelementptr inbounds [10 x i16]* @G16, i32 0, i32 %X
%Q = load i16* %P
%R = and i16 %Q, 3
%S = icmp eq i16 %R, 0
ret i1 %S
; NODL-LABEL: @test8(
; NODL-NEXT: and i32 %X, -2
; NODL-NEXT: icmp eq i32 {{.*}}, 8
; NODL-NEXT: ret i1
; P32-LABEL: @test8(
; P32-NEXT: and i32 %X, -2
; P32-NEXT: icmp eq i32 {{.*}}, 8
; P32-NEXT: ret i1
}
@GA = internal constant [4 x { i32, i32 } ] [
{ i32, i32 } { i32 1, i32 0 },
{ i32, i32 } { i32 2, i32 1 },
{ i32, i32 } { i32 3, i32 1 },
{ i32, i32 } { i32 4, i32 0 }
]
define i1 @test9(i32 %X) {
%P = getelementptr inbounds [4 x { i32, i32 } ]* @GA, i32 0, i32 %X, i32 1
%Q = load i32* %P
%R = icmp eq i32 %Q, 1
ret i1 %R
; NODL-LABEL: @test9(
; NODL-NEXT: add i32 %X, -1
; NODL-NEXT: icmp ult i32 {{.*}}, 2
; NODL-NEXT: ret i1
; P32-LABEL: @test9(
; P32-NEXT: add i32 %X, -1
; P32-NEXT: icmp ult i32 {{.*}}, 2
; P32-NEXT: ret i1
}
define i1 @test10_struct(i32 %x) {
; NODL-LABEL: @test10_struct(
; NODL: getelementptr inbounds %Foo* @GS, i32 %x, i32 0
; P32-LABEL: @test10_struct(
; P32: ret i1 false
%p = getelementptr inbounds %Foo* @GS, i32 %x, i32 0
%q = load i32* %p
%r = icmp eq i32 %q, 9
ret i1 %r
}
define i1 @test10_struct_noinbounds(i32 %x) {
; NODL-LABEL: @test10_struct_noinbounds(
; NODL: getelementptr %Foo* @GS, i32 %x, i32 0
; P32-LABEL: @test10_struct_noinbounds(
; P32: getelementptr %Foo* @GS, i32 %x, i32 0
%p = getelementptr %Foo* @GS, i32 %x, i32 0
%q = load i32* %p
%r = icmp eq i32 %q, 9
ret i1 %r
}
; Test that the GEP indices are converted before we ever get here
; Index < ptr size
define i1 @test10_struct_i16(i16 %x){
; NODL-LABEL: @test10_struct_i16(
; NODL: getelementptr inbounds %Foo* @GS, i16 %x, i32 0
; P32-LABEL: @test10_struct_i16(
; P32: ret i1 false
%p = getelementptr inbounds %Foo* @GS, i16 %x, i32 0
%q = load i32* %p
%r = icmp eq i32 %q, 0
ret i1 %r
}
; Test that the GEP indices are converted before we ever get here
; Index > ptr size
define i1 @test10_struct_i64(i64 %x){
; NODL-LABEL: @test10_struct_i64(
; NODL: getelementptr inbounds %Foo* @GS, i64 %x, i32 0
; P32-LABEL: @test10_struct_i64(
; P32: ret i1 false
%p = getelementptr inbounds %Foo* @GS, i64 %x, i32 0
%q = load i32* %p
%r = icmp eq i32 %q, 0
ret i1 %r
}
define i1 @test10_struct_noinbounds_i16(i16 %x) {
; NODL-LABEL: @test10_struct_noinbounds_i16(
; NODL: getelementptr %Foo* @GS, i16 %x, i32 0
; P32-LABEL: @test10_struct_noinbounds_i16(
; P32: %1 = sext i16 %x to i32
; P32: getelementptr %Foo* @GS, i32 %1, i32 0
%p = getelementptr %Foo* @GS, i16 %x, i32 0
%q = load i32* %p
%r = icmp eq i32 %q, 0
ret i1 %r
}
define i1 @test10_struct_arr(i32 %x) {
; NODL-LABEL: @test10_struct_arr(
; NODL-NEXT: %r = icmp ne i32 %x, 1
; NODL-NEXT: ret i1 %r
; P32-LABEL: @test10_struct_arr(
; P32-NEXT: %r = icmp ne i32 %x, 1
; P32-NEXT: ret i1 %r
%p = getelementptr inbounds [4 x %Foo]* @GStructArr, i32 0, i32 %x, i32 2
%q = load i32* %p
%r = icmp eq i32 %q, 9
ret i1 %r
}
define i1 @test10_struct_arr_noinbounds(i32 %x) {
; NODL-LABEL: @test10_struct_arr_noinbounds(
; NODL-NEXT %p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i32 %x, i32 2
; P32-LABEL: @test10_struct_arr_noinbounds(
; P32-NEXT %p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i32 %x, i32 2
%p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i32 %x, i32 2
%q = load i32* %p
%r = icmp eq i32 %q, 9
ret i1 %r
}
define i1 @test10_struct_arr_i16(i16 %x) {
; NODL-LABEL: @test10_struct_arr_i16(
; NODL-NEXT: %r = icmp ne i16 %x, 1
; NODL-NEXT: ret i1 %r
; P32-LABEL: @test10_struct_arr_i16(
; P32-NEXT: %r = icmp ne i16 %x, 1
; P32-NEXT: ret i1 %r
%p = getelementptr inbounds [4 x %Foo]* @GStructArr, i16 0, i16 %x, i32 2
%q = load i32* %p
%r = icmp eq i32 %q, 9
ret i1 %r
}
define i1 @test10_struct_arr_i64(i64 %x) {
; NODL-LABEL: @test10_struct_arr_i64(
; NODL-NEXT: %r = icmp ne i64 %x, 1
; NODL-NEXT: ret i1 %r
; P32-LABEL: @test10_struct_arr_i64(
; P32-NEXT: trunc i64 %x to i32
; P32-NEXT: %r = icmp ne i32 %1, 1
; P32-NEXT: ret i1 %r
%p = getelementptr inbounds [4 x %Foo]* @GStructArr, i64 0, i64 %x, i32 2
%q = load i32* %p
%r = icmp eq i32 %q, 9
ret i1 %r
}
define i1 @test10_struct_arr_noinbounds_i16(i16 %x) {
; NODL-LABEL: @test10_struct_arr_noinbounds_i16(
; NODL-NEXT: %p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i16 %x, i32 2
; P32-LABEL: @test10_struct_arr_noinbounds_i16(
; P32-NEXT: %r = icmp ne i16 %x, 1
%p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i16 %x, i32 2
%q = load i32* %p
%r = icmp eq i32 %q, 9
ret i1 %r
}
define i1 @test10_struct_arr_noinbounds_i64(i64 %x) {
; FIXME: Should be no trunc?
; NODL-LABEL: @test10_struct_arr_noinbounds_i64(
; NODL-NEXT: %p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i64 %x, i32 2
; P32-LABEL: @test10_struct_arr_noinbounds_i64(
; P32: %r = icmp ne i32 %1, 1
; P32-NEXT: ret i1 %r
%p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i64 %x, i32 2
%q = load i32* %p
%r = icmp eq i32 %q, 9
ret i1 %r
}
Reference in New Issue View Git Blame Copy Permalink