[llvm] 3fc7388 - [MIPS] Add a baseline test showing current inefficient hidden sret lowering

[Alex Richardson via llvm-commits]

Author: Alex Richardson
Date: 2020-05-04T14:44:39+01:00
New Revision: 3fc738846e156acc7b26607d9a7c78210478a75d

URL: https://github.com/llvm/llvm-project/commit/3fc738846e156acc7b26607d9a7c78210478a75d
DIFF: https://github.com/llvm/llvm-project/commit/3fc738846e156acc7b26607d9a7c78210478a75d.diff

LOG: [MIPS] Add a baseline test showing current inefficient hidden sret lowering

SelectionDAGBuilder currently doesn't propagate the known alignment of
the sret parameter. This is inefficient for MIPS and highly inefficient for
our out-of-tree CHERI-extended MIPS since we don't have lwl/lwr so fall back
to byte loads for align == 1.

Added: 
    llvm/test/CodeGen/Mips/implicit-sret.ll

Modified: 
    

Removed: 
    


################################################################################
diff  --git a/llvm/test/CodeGen/Mips/implicit-sret.ll b/llvm/test/CodeGen/Mips/implicit-sret.ll
new file mode 100644
index 000000000000..3cc0892ae809
--- /dev/null
+++ b/llvm/test/CodeGen/Mips/implicit-sret.ll
@@ -0,0 +1,128 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=mips64-unknown-freebsd -O0 -o - %s | FileCheck %s
+
+; Previously SelectionDAGBuilder would always set alignment to 1 for hidden sret
+; parameters so we were generating ldl+ldr/lwl+lwr to load those values even
+; though we know that they are aligned (since we allocated an aligned frame index)
+
+declare dso_local void @use_sret(i32, i128, i64) unnamed_addr
+declare dso_local { i32, i128, i64 } @implicit_sret_decl() unnamed_addr
+
+define internal void @test() unnamed_addr nounwind {
+; CHECK-LABEL: test:
+; CHECK:       # %bb.0: # %start
+; CHECK-NEXT:    daddiu $sp, $sp, -48
+; CHECK-NEXT:    sd $ra, 40($sp) # 8-byte Folded Spill
+; CHECK-NEXT:    daddiu $4, $sp, 8
+; CHECK-NEXT:    jal implicit_sret_decl
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    # implicit-def: $at_64
+; CHECK-NEXT:    ldl $1, 24($sp)
+; CHECK-NEXT:    ldr $1, 31($sp)
+; CHECK-NEXT:    # implicit-def: $v0_64
+; CHECK-NEXT:    ldl $2, 16($sp)
+; CHECK-NEXT:    ldr $2, 23($sp)
+; CHECK-NEXT:    # implicit-def: $v1_64
+; CHECK-NEXT:    ldl $3, 32($sp)
+; CHECK-NEXT:    ldr $3, 39($sp)
+; CHECK-NEXT:    # implicit-def: $a1
+; CHECK-NEXT:    lwl $5, 8($sp)
+; CHECK-NEXT:    lwr $5, 11($sp)
+; CHECK-NEXT:    # implicit-def: $a0_64
+; CHECK-NEXT:    move $4, $5
+; CHECK-NEXT:    move $5, $2
+; CHECK-NEXT:    move $6, $1
+; CHECK-NEXT:    move $7, $3
+; CHECK-NEXT:    jal use_sret
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    ld $ra, 40($sp) # 8-byte Folded Reload
+; CHECK-NEXT:    daddiu $sp, $sp, 48
+; CHECK-NEXT:    jr $ra
+; CHECK-NEXT:    nop
+start:
+  %0 = call { i32, i128, i64 } @implicit_sret_decl()
+  %1 = extractvalue { i32, i128, i64 } %0, 0
+  %2 = extractvalue { i32, i128, i64 } %0, 1
+  %3 = extractvalue { i32, i128, i64 } %0, 2
+  call void @use_sret(i32 %1, i128 %2, i64 %3)
+  ret void
+}
+
+define internal { i32, i128, i64 } @implicit_sret_impl() unnamed_addr nounwind {
+; CHECK-LABEL: implicit_sret_impl:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    move $1, $4
+; CHECK-NEXT:    daddiu $2, $zero, 20
+; CHECK-NEXT:    sd $2, 16($4)
+; CHECK-NEXT:    daddiu $2, $zero, 0
+; CHECK-NEXT:    sd $zero, 8($4)
+; CHECK-NEXT:    daddiu $3, $zero, 30
+; CHECK-NEXT:    sd $3, 24($4)
+; CHECK-NEXT:    addiu $5, $zero, 10
+; CHECK-NEXT:    sw $5, 0($4)
+; CHECK-NEXT:    jr $ra
+; CHECK-NEXT:    nop
+  ret { i32, i128, i64 } { i32 10, i128 20, i64 30 }
+}
+
+declare dso_local { i32, i32, i32, i32, i32, i32 } @implicit_sret_decl2() unnamed_addr
+declare dso_local void @use_sret2(i32, i32, i32) unnamed_addr
+define internal void @test2() unnamed_addr nounwind {
+; CHECK-LABEL: test2:
+; CHECK:       # %bb.0: # %start
+; CHECK-NEXT:    daddiu $sp, $sp, -32
+; CHECK-NEXT:    sd $ra, 24($sp) # 8-byte Folded Spill
+; CHECK-NEXT:    daddiu $4, $sp, 0
+; CHECK-NEXT:    jal implicit_sret_decl2
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    # implicit-def: $at
+; CHECK-NEXT:    lwl $1, 20($sp)
+; CHECK-NEXT:    lwr $1, 23($sp)
+; CHECK-NEXT:    # implicit-def: $v0
+; CHECK-NEXT:    lwl $2, 12($sp)
+; CHECK-NEXT:    lwr $2, 15($sp)
+; CHECK-NEXT:    # implicit-def: $v1
+; CHECK-NEXT:    lwl $3, 4($sp)
+; CHECK-NEXT:    lwr $3, 7($sp)
+; CHECK-NEXT:    # implicit-def: $a0_64
+; CHECK-NEXT:    move $4, $3
+; CHECK-NEXT:    # implicit-def: $a1_64
+; CHECK-NEXT:    move $5, $2
+; CHECK-NEXT:    # implicit-def: $a2_64
+; CHECK-NEXT:    move $6, $1
+; CHECK-NEXT:    jal use_sret2
+; CHECK-NEXT:    nop
+; CHECK-NEXT:    ld $ra, 24($sp) # 8-byte Folded Reload
+; CHECK-NEXT:    daddiu $sp, $sp, 32
+; CHECK-NEXT:    jr $ra
+; CHECK-NEXT:    nop
+start:
+  %0 = call { i32, i32, i32, i32, i32, i32 } @implicit_sret_decl2()
+  %1 = extractvalue { i32, i32, i32, i32, i32, i32 } %0, 1
+  %2 = extractvalue { i32, i32, i32, i32, i32, i32 } %0, 3
+  %3 = extractvalue { i32, i32, i32, i32, i32, i32 } %0, 5
+  call void @use_sret2(i32 %1, i32 %2, i32 %3)
+  ret void
+}
+
+
+define internal { i32, i32, i32, i32, i32, i32 } @implicit_sret_impl2() unnamed_addr nounwind {
+; CHECK-LABEL: implicit_sret_impl2:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    move $1, $4
+; CHECK-NEXT:    addiu $2, $zero, 6
+; CHECK-NEXT:    sw $2, 20($4)
+; CHECK-NEXT:    addiu $2, $zero, 5
+; CHECK-NEXT:    sw $2, 16($4)
+; CHECK-NEXT:    addiu $2, $zero, 4
+; CHECK-NEXT:    sw $2, 12($4)
+; CHECK-NEXT:    addiu $2, $zero, 3
+; CHECK-NEXT:    sw $2, 8($4)
+; CHECK-NEXT:    addiu $2, $zero, 2
+; CHECK-NEXT:    sw $2, 4($4)
+; CHECK-NEXT:    addiu $2, $zero, 1
+; CHECK-NEXT:    sw $2, 0($4)
+; CHECK-NEXT:    jr $ra
+; CHECK-NEXT:    nop
+  ret { i32, i32, i32, i32, i32, i32 } { i32 1, i32 2, i32 3, i32 4, i32 5, i32 6 }
+}