[llvm] [RISCV] Fix double counting scalar CSRs with Zcmp when emitting cfi_offset for RVV CSRs. (PR #117408)

via llvm-commits llvm-commits at lists.llvm.org
Fri Nov 22 16:35:49 PST 2024


llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-backend-risc-v

Author: Craig Topper (topperc)

<details>
<summary>Changes</summary>

getCalleeSavedStackSize() already contains RVPushStackSize. Don't subtract it again.

---
Full diff: https://github.com/llvm/llvm-project/pull/117408.diff


2 Files Affected:

- (modified) llvm/lib/Target/RISCV/RISCVFrameLowering.cpp (+1-2) 
- (modified) llvm/test/CodeGen/RISCV/rvv/rvv-cfi-info.ll (+189-42) 


``````````diff
diff --git a/llvm/lib/Target/RISCV/RISCVFrameLowering.cpp b/llvm/lib/Target/RISCV/RISCVFrameLowering.cpp
index f0bc74e331db46..546bfe6a90f7a7 100644
--- a/llvm/lib/Target/RISCV/RISCVFrameLowering.cpp
+++ b/llvm/lib/Target/RISCV/RISCVFrameLowering.cpp
@@ -1752,8 +1752,7 @@ void RISCVFrameLowering::emitCalleeSavedRVVPrologCFI(
   if (!HasFP) {
     uint64_t ScalarLocalVarSize =
         MFI.getStackSize() - RVFI->getCalleeSavedStackSize() -
-        RVFI->getRVPushStackSize() - RVFI->getVarArgsSaveSize() +
-        RVFI->getRVVPadding();
+        RVFI->getVarArgsSaveSize() + RVFI->getRVVPadding();
     FixedSize -= ScalarLocalVarSize;
   }
 
diff --git a/llvm/test/CodeGen/RISCV/rvv/rvv-cfi-info.ll b/llvm/test/CodeGen/RISCV/rvv/rvv-cfi-info.ll
index e78bb323f4d3c5..0ae2c2ef9c9d33 100644
--- a/llvm/test/CodeGen/RISCV/rvv/rvv-cfi-info.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/rvv-cfi-info.ll
@@ -3,57 +3,63 @@
 ; RUN:   | FileCheck -check-prefix=OMIT-FP %s
 ; RUN: llc -mtriple=riscv64 -mattr=+v,+m -verify-machineinstrs -frame-pointer=all < %s \
 ; RUN:   | FileCheck -check-prefix=NO-OMIT-FP %s
+; RUN: llc -mtriple=riscv64 -mattr=+v,+m,+zcmp -verify-machineinstrs < %s \
+; RUN:   | FileCheck -check-prefix=OMIT-FP-ZCMP %s
+; RUN: llc -mtriple=riscv64 -mattr=+v,+m,+zcmp -verify-machineinstrs -frame-pointer=all < %s \
+; RUN:   | FileCheck -check-prefix=NO-OMIT-FP-ZCMP %s
 
 define riscv_vector_cc <vscale x 1 x i32> @test_vector_callee_cfi(<vscale x 1 x i32> %va) {
 ; OMIT-FP-LABEL: test_vector_callee_cfi:
 ; OMIT-FP:       # %bb.0: # %entry
-; OMIT-FP-NEXT:    addi sp, sp, -16
-; OMIT-FP-NEXT:    .cfi_def_cfa_offset 16
+; OMIT-FP-NEXT:    addi sp, sp, -48
+; OMIT-FP-NEXT:    .cfi_def_cfa_offset 48
+; OMIT-FP-NEXT:    sd s1, 40(sp) # 8-byte Folded Spill
+; OMIT-FP-NEXT:    .cfi_offset s1, -8
 ; OMIT-FP-NEXT:    csrr a0, vlenb
 ; OMIT-FP-NEXT:    slli a1, a0, 3
 ; OMIT-FP-NEXT:    sub a0, a1, a0
 ; OMIT-FP-NEXT:    sub sp, sp, a0
-; OMIT-FP-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x07, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 7 * vlenb
+; OMIT-FP-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x07, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 7 * vlenb
 ; OMIT-FP-NEXT:    csrr a0, vlenb
 ; OMIT-FP-NEXT:    li a1, 6
 ; OMIT-FP-NEXT:    mul a0, a0, a1
 ; OMIT-FP-NEXT:    add a0, sp, a0
-; OMIT-FP-NEXT:    addi a0, a0, 16
+; OMIT-FP-NEXT:    addi a0, a0, 32
 ; OMIT-FP-NEXT:    vs1r.v v1, (a0) # Unknown-size Folded Spill
 ; OMIT-FP-NEXT:    csrr a0, vlenb
 ; OMIT-FP-NEXT:    slli a0, a0, 2
 ; OMIT-FP-NEXT:    add a0, sp, a0
-; OMIT-FP-NEXT:    addi a0, a0, 16
+; OMIT-FP-NEXT:    addi a0, a0, 32
 ; OMIT-FP-NEXT:    vs2r.v v2, (a0) # Unknown-size Folded Spill
-; OMIT-FP-NEXT:    addi a0, sp, 16
+; OMIT-FP-NEXT:    addi a0, sp, 32
 ; OMIT-FP-NEXT:    vs4r.v v4, (a0) # Unknown-size Folded Spill
-; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x61, 0x08, 0x11, 0x7f, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v1 @ cfa - 1 * vlenb
-; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x62, 0x08, 0x11, 0x7d, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v2 @ cfa - 3 * vlenb
-; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x63, 0x08, 0x11, 0x7e, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v3 @ cfa - 2 * vlenb
-; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x64, 0x08, 0x11, 0x79, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v4 @ cfa - 7 * vlenb
-; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x65, 0x08, 0x11, 0x7a, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v5 @ cfa - 6 * vlenb
-; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x66, 0x08, 0x11, 0x7b, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v6 @ cfa - 5 * vlenb
-; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x67, 0x08, 0x11, 0x7c, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v7 @ cfa - 4 * vlenb
+; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x61, 0x0b, 0x11, 0x70, 0x22, 0x11, 0x7f, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v1 @ cfa - 16 - 1 * vlenb
+; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x62, 0x0b, 0x11, 0x70, 0x22, 0x11, 0x7d, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v2 @ cfa - 16 - 3 * vlenb
+; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x63, 0x0b, 0x11, 0x70, 0x22, 0x11, 0x7e, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v3 @ cfa - 16 - 2 * vlenb
+; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x64, 0x0b, 0x11, 0x70, 0x22, 0x11, 0x79, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v4 @ cfa - 16 - 7 * vlenb
+; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x65, 0x0b, 0x11, 0x70, 0x22, 0x11, 0x7a, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v5 @ cfa - 16 - 6 * vlenb
+; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x66, 0x0b, 0x11, 0x70, 0x22, 0x11, 0x7b, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v6 @ cfa - 16 - 5 * vlenb
+; OMIT-FP-NEXT:    .cfi_escape 0x10, 0x67, 0x0b, 0x11, 0x70, 0x22, 0x11, 0x7c, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v7 @ cfa - 16 - 4 * vlenb
 ; OMIT-FP-NEXT:    #APP
 ; OMIT-FP-NEXT:    #NO_APP
 ; OMIT-FP-NEXT:    csrr a0, vlenb
 ; OMIT-FP-NEXT:    li a1, 6
 ; OMIT-FP-NEXT:    mul a0, a0, a1
 ; OMIT-FP-NEXT:    add a0, sp, a0
-; OMIT-FP-NEXT:    addi a0, a0, 16
+; OMIT-FP-NEXT:    addi a0, a0, 32
 ; OMIT-FP-NEXT:    vl1r.v v1, (a0) # Unknown-size Folded Reload
 ; OMIT-FP-NEXT:    csrr a0, vlenb
 ; OMIT-FP-NEXT:    slli a0, a0, 2
 ; OMIT-FP-NEXT:    add a0, sp, a0
-; OMIT-FP-NEXT:    addi a0, a0, 16
+; OMIT-FP-NEXT:    addi a0, a0, 32
 ; OMIT-FP-NEXT:    vl2r.v v2, (a0) # Unknown-size Folded Reload
-; OMIT-FP-NEXT:    addi a0, sp, 16
+; OMIT-FP-NEXT:    addi a0, sp, 32
 ; OMIT-FP-NEXT:    vl4r.v v4, (a0) # Unknown-size Folded Reload
 ; OMIT-FP-NEXT:    csrr a0, vlenb
 ; OMIT-FP-NEXT:    slli a1, a0, 3
 ; OMIT-FP-NEXT:    sub a0, a1, a0
 ; OMIT-FP-NEXT:    add sp, sp, a0
-; OMIT-FP-NEXT:    .cfi_def_cfa sp, 16
+; OMIT-FP-NEXT:    .cfi_def_cfa sp, 48
 ; OMIT-FP-NEXT:    .cfi_restore v1
 ; OMIT-FP-NEXT:    .cfi_restore v2
 ; OMIT-FP-NEXT:    .cfi_restore v3
@@ -61,19 +67,23 @@ define riscv_vector_cc <vscale x 1 x i32> @test_vector_callee_cfi(<vscale x 1 x
 ; OMIT-FP-NEXT:    .cfi_restore v5
 ; OMIT-FP-NEXT:    .cfi_restore v6
 ; OMIT-FP-NEXT:    .cfi_restore v7
-; OMIT-FP-NEXT:    addi sp, sp, 16
+; OMIT-FP-NEXT:    ld s1, 40(sp) # 8-byte Folded Reload
+; OMIT-FP-NEXT:    .cfi_restore s1
+; OMIT-FP-NEXT:    addi sp, sp, 48
 ; OMIT-FP-NEXT:    .cfi_def_cfa_offset 0
 ; OMIT-FP-NEXT:    ret
 ;
 ; NO-OMIT-FP-LABEL: test_vector_callee_cfi:
 ; NO-OMIT-FP:       # %bb.0: # %entry
-; NO-OMIT-FP-NEXT:    addi sp, sp, -32
-; NO-OMIT-FP-NEXT:    .cfi_def_cfa_offset 32
-; NO-OMIT-FP-NEXT:    sd ra, 24(sp) # 8-byte Folded Spill
-; NO-OMIT-FP-NEXT:    sd s0, 16(sp) # 8-byte Folded Spill
+; NO-OMIT-FP-NEXT:    addi sp, sp, -48
+; NO-OMIT-FP-NEXT:    .cfi_def_cfa_offset 48
+; NO-OMIT-FP-NEXT:    sd ra, 40(sp) # 8-byte Folded Spill
+; NO-OMIT-FP-NEXT:    sd s0, 32(sp) # 8-byte Folded Spill
+; NO-OMIT-FP-NEXT:    sd s1, 24(sp) # 8-byte Folded Spill
 ; NO-OMIT-FP-NEXT:    .cfi_offset ra, -8
 ; NO-OMIT-FP-NEXT:    .cfi_offset s0, -16
-; NO-OMIT-FP-NEXT:    addi s0, sp, 32
+; NO-OMIT-FP-NEXT:    .cfi_offset s1, -24
+; NO-OMIT-FP-NEXT:    addi s0, sp, 48
 ; NO-OMIT-FP-NEXT:    .cfi_def_cfa s0, 0
 ; NO-OMIT-FP-NEXT:    csrr a0, vlenb
 ; NO-OMIT-FP-NEXT:    slli a1, a0, 3
@@ -81,44 +91,44 @@ define riscv_vector_cc <vscale x 1 x i32> @test_vector_callee_cfi(<vscale x 1 x
 ; NO-OMIT-FP-NEXT:    sub sp, sp, a0
 ; NO-OMIT-FP-NEXT:    csrr a0, vlenb
 ; NO-OMIT-FP-NEXT:    sub a0, s0, a0
-; NO-OMIT-FP-NEXT:    addi a0, a0, -32
+; NO-OMIT-FP-NEXT:    addi a0, a0, -48
 ; NO-OMIT-FP-NEXT:    vs1r.v v1, (a0) # Unknown-size Folded Spill
 ; NO-OMIT-FP-NEXT:    csrr a0, vlenb
 ; NO-OMIT-FP-NEXT:    slli a1, a0, 1
 ; NO-OMIT-FP-NEXT:    add a0, a1, a0
 ; NO-OMIT-FP-NEXT:    sub a0, s0, a0
-; NO-OMIT-FP-NEXT:    addi a0, a0, -32
+; NO-OMIT-FP-NEXT:    addi a0, a0, -48
 ; NO-OMIT-FP-NEXT:    vs2r.v v2, (a0) # Unknown-size Folded Spill
 ; NO-OMIT-FP-NEXT:    csrr a0, vlenb
 ; NO-OMIT-FP-NEXT:    slli a1, a0, 3
 ; NO-OMIT-FP-NEXT:    sub a0, a1, a0
 ; NO-OMIT-FP-NEXT:    sub a0, s0, a0
-; NO-OMIT-FP-NEXT:    addi a0, a0, -32
+; NO-OMIT-FP-NEXT:    addi a0, a0, -48
 ; NO-OMIT-FP-NEXT:    vs4r.v v4, (a0) # Unknown-size Folded Spill
-; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x61, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x7f, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v1 @ cfa - 32 - 1 * vlenb
-; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x62, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x7d, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v2 @ cfa - 32 - 3 * vlenb
-; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x63, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x7e, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v3 @ cfa - 32 - 2 * vlenb
-; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x64, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x79, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v4 @ cfa - 32 - 7 * vlenb
-; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x65, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x7a, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v5 @ cfa - 32 - 6 * vlenb
-; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x66, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x7b, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v6 @ cfa - 32 - 5 * vlenb
-; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x67, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x7c, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v7 @ cfa - 32 - 4 * vlenb
+; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x61, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x7f, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v1 @ cfa - 48 - 1 * vlenb
+; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x62, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x7d, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v2 @ cfa - 48 - 3 * vlenb
+; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x63, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x7e, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v3 @ cfa - 48 - 2 * vlenb
+; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x64, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x79, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v4 @ cfa - 48 - 7 * vlenb
+; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x65, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x7a, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v5 @ cfa - 48 - 6 * vlenb
+; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x66, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x7b, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v6 @ cfa - 48 - 5 * vlenb
+; NO-OMIT-FP-NEXT:    .cfi_escape 0x10, 0x67, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x7c, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v7 @ cfa - 48 - 4 * vlenb
 ; NO-OMIT-FP-NEXT:    #APP
 ; NO-OMIT-FP-NEXT:    #NO_APP
 ; NO-OMIT-FP-NEXT:    csrr a0, vlenb
 ; NO-OMIT-FP-NEXT:    sub a0, s0, a0
-; NO-OMIT-FP-NEXT:    addi a0, a0, -32
+; NO-OMIT-FP-NEXT:    addi a0, a0, -48
 ; NO-OMIT-FP-NEXT:    vl1r.v v1, (a0) # Unknown-size Folded Reload
 ; NO-OMIT-FP-NEXT:    csrr a0, vlenb
 ; NO-OMIT-FP-NEXT:    slli a1, a0, 1
 ; NO-OMIT-FP-NEXT:    add a0, a1, a0
 ; NO-OMIT-FP-NEXT:    sub a0, s0, a0
-; NO-OMIT-FP-NEXT:    addi a0, a0, -32
+; NO-OMIT-FP-NEXT:    addi a0, a0, -48
 ; NO-OMIT-FP-NEXT:    vl2r.v v2, (a0) # Unknown-size Folded Reload
 ; NO-OMIT-FP-NEXT:    csrr a0, vlenb
 ; NO-OMIT-FP-NEXT:    slli a1, a0, 3
 ; NO-OMIT-FP-NEXT:    sub a0, a1, a0
 ; NO-OMIT-FP-NEXT:    sub a0, s0, a0
-; NO-OMIT-FP-NEXT:    addi a0, a0, -32
+; NO-OMIT-FP-NEXT:    addi a0, a0, -48
 ; NO-OMIT-FP-NEXT:    vl4r.v v4, (a0) # Unknown-size Folded Reload
 ; NO-OMIT-FP-NEXT:    .cfi_restore v1
 ; NO-OMIT-FP-NEXT:    .cfi_restore v2
@@ -127,18 +137,155 @@ define riscv_vector_cc <vscale x 1 x i32> @test_vector_callee_cfi(<vscale x 1 x
 ; NO-OMIT-FP-NEXT:    .cfi_restore v5
 ; NO-OMIT-FP-NEXT:    .cfi_restore v6
 ; NO-OMIT-FP-NEXT:    .cfi_restore v7
-; NO-OMIT-FP-NEXT:    addi sp, s0, -32
-; NO-OMIT-FP-NEXT:    .cfi_def_cfa sp, 32
-; NO-OMIT-FP-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
-; NO-OMIT-FP-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
+; NO-OMIT-FP-NEXT:    addi sp, s0, -48
+; NO-OMIT-FP-NEXT:    .cfi_def_cfa sp, 48
+; NO-OMIT-FP-NEXT:    ld ra, 40(sp) # 8-byte Folded Reload
+; NO-OMIT-FP-NEXT:    ld s0, 32(sp) # 8-byte Folded Reload
+; NO-OMIT-FP-NEXT:    ld s1, 24(sp) # 8-byte Folded Reload
 ; NO-OMIT-FP-NEXT:    .cfi_restore ra
 ; NO-OMIT-FP-NEXT:    .cfi_restore s0
-; NO-OMIT-FP-NEXT:    addi sp, sp, 32
+; NO-OMIT-FP-NEXT:    .cfi_restore s1
+; NO-OMIT-FP-NEXT:    addi sp, sp, 48
 ; NO-OMIT-FP-NEXT:    .cfi_def_cfa_offset 0
 ; NO-OMIT-FP-NEXT:    ret
+;
+; OMIT-FP-ZCMP-LABEL: test_vector_callee_cfi:
+; OMIT-FP-ZCMP:       # %bb.0: # %entry
+; OMIT-FP-ZCMP-NEXT:    cm.push {ra, s0-s1}, -48
+; OMIT-FP-ZCMP-NEXT:    .cfi_def_cfa_offset 48
+; OMIT-FP-ZCMP-NEXT:    .cfi_offset s1, -8
+; OMIT-FP-ZCMP-NEXT:    csrr a0, vlenb
+; OMIT-FP-ZCMP-NEXT:    slli a1, a0, 3
+; OMIT-FP-ZCMP-NEXT:    sub a0, a1, a0
+; OMIT-FP-ZCMP-NEXT:    sub sp, sp, a0
+; OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x07, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 7 * vlenb
+; OMIT-FP-ZCMP-NEXT:    csrr a0, vlenb
+; OMIT-FP-ZCMP-NEXT:    li a1, 6
+; OMIT-FP-ZCMP-NEXT:    mul a0, a0, a1
+; OMIT-FP-ZCMP-NEXT:    add a0, a0, sp
+; OMIT-FP-ZCMP-NEXT:    addi a0, a0, 16
+; OMIT-FP-ZCMP-NEXT:    vs1r.v v1, (a0) # Unknown-size Folded Spill
+; OMIT-FP-ZCMP-NEXT:    csrr a0, vlenb
+; OMIT-FP-ZCMP-NEXT:    slli a0, a0, 2
+; OMIT-FP-ZCMP-NEXT:    add a0, a0, sp
+; OMIT-FP-ZCMP-NEXT:    addi a0, a0, 16
+; OMIT-FP-ZCMP-NEXT:    vs2r.v v2, (a0) # Unknown-size Folded Spill
+; OMIT-FP-ZCMP-NEXT:    addi a0, sp, 16
+; OMIT-FP-ZCMP-NEXT:    vs4r.v v4, (a0) # Unknown-size Folded Spill
+; OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x61, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x7f, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v1 @ cfa - 32 - 1 * vlenb
+; OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x62, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x7d, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v2 @ cfa - 32 - 3 * vlenb
+; OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x63, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x7e, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v3 @ cfa - 32 - 2 * vlenb
+; OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x64, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x79, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v4 @ cfa - 32 - 7 * vlenb
+; OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x65, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x7a, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v5 @ cfa - 32 - 6 * vlenb
+; OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x66, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x7b, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v6 @ cfa - 32 - 5 * vlenb
+; OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x67, 0x0b, 0x11, 0x60, 0x22, 0x11, 0x7c, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v7 @ cfa - 32 - 4 * vlenb
+; OMIT-FP-ZCMP-NEXT:    #APP
+; OMIT-FP-ZCMP-NEXT:    #NO_APP
+; OMIT-FP-ZCMP-NEXT:    csrr a0, vlenb
+; OMIT-FP-ZCMP-NEXT:    li a1, 6
+; OMIT-FP-ZCMP-NEXT:    mul a0, a0, a1
+; OMIT-FP-ZCMP-NEXT:    add a0, a0, sp
+; OMIT-FP-ZCMP-NEXT:    addi a0, a0, 16
+; OMIT-FP-ZCMP-NEXT:    vl1r.v v1, (a0) # Unknown-size Folded Reload
+; OMIT-FP-ZCMP-NEXT:    csrr a0, vlenb
+; OMIT-FP-ZCMP-NEXT:    slli a0, a0, 2
+; OMIT-FP-ZCMP-NEXT:    add a0, a0, sp
+; OMIT-FP-ZCMP-NEXT:    addi a0, a0, 16
+; OMIT-FP-ZCMP-NEXT:    vl2r.v v2, (a0) # Unknown-size Folded Reload
+; OMIT-FP-ZCMP-NEXT:    addi a0, sp, 16
+; OMIT-FP-ZCMP-NEXT:    vl4r.v v4, (a0) # Unknown-size Folded Reload
+; OMIT-FP-ZCMP-NEXT:    csrr a0, vlenb
+; OMIT-FP-ZCMP-NEXT:    slli a1, a0, 3
+; OMIT-FP-ZCMP-NEXT:    sub a0, a1, a0
+; OMIT-FP-ZCMP-NEXT:    add sp, sp, a0
+; OMIT-FP-ZCMP-NEXT:    .cfi_def_cfa sp, 48
+; OMIT-FP-ZCMP-NEXT:    .cfi_restore v1
+; OMIT-FP-ZCMP-NEXT:    .cfi_restore v2
+; OMIT-FP-ZCMP-NEXT:    .cfi_restore v3
+; OMIT-FP-ZCMP-NEXT:    .cfi_restore v4
+; OMIT-FP-ZCMP-NEXT:    .cfi_restore v5
+; OMIT-FP-ZCMP-NEXT:    .cfi_restore v6
+; OMIT-FP-ZCMP-NEXT:    .cfi_restore v7
+; OMIT-FP-ZCMP-NEXT:    cm.popret {ra, s0-s1}, 48
+;
+; NO-OMIT-FP-ZCMP-LABEL: test_vector_callee_cfi:
+; NO-OMIT-FP-ZCMP:       # %bb.0: # %entry
+; NO-OMIT-FP-ZCMP-NEXT:    addi sp, sp, -48
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_def_cfa_offset 48
+; NO-OMIT-FP-ZCMP-NEXT:    sd ra, 40(sp) # 8-byte Folded Spill
+; NO-OMIT-FP-ZCMP-NEXT:    sd s0, 32(sp) # 8-byte Folded Spill
+; NO-OMIT-FP-ZCMP-NEXT:    sd s1, 24(sp) # 8-byte Folded Spill
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_offset ra, -8
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_offset s0, -16
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_offset s1, -24
+; NO-OMIT-FP-ZCMP-NEXT:    addi s0, sp, 48
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_def_cfa s0, 0
+; NO-OMIT-FP-ZCMP-NEXT:    csrr a0, vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    slli a1, a0, 3
+; NO-OMIT-FP-ZCMP-NEXT:    sub a0, a1, a0
+; NO-OMIT-FP-ZCMP-NEXT:    sub sp, sp, a0
+; NO-OMIT-FP-ZCMP-NEXT:    csrr a0, vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    sub a0, s0, a0
+; NO-OMIT-FP-ZCMP-NEXT:    addi a0, a0, -48
+; NO-OMIT-FP-ZCMP-NEXT:    vs1r.v v1, (a0) # Unknown-size Folded Spill
+; NO-OMIT-FP-ZCMP-NEXT:    csrr a0, vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    slli a1, a0, 1
+; NO-OMIT-FP-ZCMP-NEXT:    add a0, a0, a1
+; NO-OMIT-FP-ZCMP-NEXT:    sub a0, s0, a0
+; NO-OMIT-FP-ZCMP-NEXT:    addi a0, a0, -48
+; NO-OMIT-FP-ZCMP-NEXT:    vs2r.v v2, (a0) # Unknown-size Folded Spill
+; NO-OMIT-FP-ZCMP-NEXT:    csrr a0, vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    slli a1, a0, 3
+; NO-OMIT-FP-ZCMP-NEXT:    sub a0, a1, a0
+; NO-OMIT-FP-ZCMP-NEXT:    sub a0, s0, a0
+; NO-OMIT-FP-ZCMP-NEXT:    addi a0, a0, -48
+; NO-OMIT-FP-ZCMP-NEXT:    vs4r.v v4, (a0) # Unknown-size Folded Spill
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x61, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x7f, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v1 @ cfa - 48 - 1 * vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x62, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x7d, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v2 @ cfa - 48 - 3 * vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x63, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x7e, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v3 @ cfa - 48 - 2 * vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x64, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x79, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v4 @ cfa - 48 - 7 * vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x65, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x7a, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v5 @ cfa - 48 - 6 * vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x66, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x7b, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v6 @ cfa - 48 - 5 * vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_escape 0x10, 0x67, 0x0b, 0x11, 0x50, 0x22, 0x11, 0x7c, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # $v7 @ cfa - 48 - 4 * vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    #APP
+; NO-OMIT-FP-ZCMP-NEXT:    #NO_APP
+; NO-OMIT-FP-ZCMP-NEXT:    csrr a0, vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    sub a0, s0, a0
+; NO-OMIT-FP-ZCMP-NEXT:    addi a0, a0, -48
+; NO-OMIT-FP-ZCMP-NEXT:    vl1r.v v1, (a0) # Unknown-size Folded Reload
+; NO-OMIT-FP-ZCMP-NEXT:    csrr a0, vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    slli a1, a0, 1
+; NO-OMIT-FP-ZCMP-NEXT:    add a0, a0, a1
+; NO-OMIT-FP-ZCMP-NEXT:    sub a0, s0, a0
+; NO-OMIT-FP-ZCMP-NEXT:    addi a0, a0, -48
+; NO-OMIT-FP-ZCMP-NEXT:    vl2r.v v2, (a0) # Unknown-size Folded Reload
+; NO-OMIT-FP-ZCMP-NEXT:    csrr a0, vlenb
+; NO-OMIT-FP-ZCMP-NEXT:    slli a1, a0, 3
+; NO-OMIT-FP-ZCMP-NEXT:    sub a0, a1, a0
+; NO-OMIT-FP-ZCMP-NEXT:    sub a0, s0, a0
+; NO-OMIT-FP-ZCMP-NEXT:    addi a0, a0, -48
+; NO-OMIT-FP-ZCMP-NEXT:    vl4r.v v4, (a0) # Unknown-size Folded Reload
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_restore v1
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_restore v2
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_restore v3
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_restore v4
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_restore v5
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_restore v6
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_restore v7
+; NO-OMIT-FP-ZCMP-NEXT:    addi sp, s0, -48
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_def_cfa sp, 48
+; NO-OMIT-FP-ZCMP-NEXT:    ld ra, 40(sp) # 8-byte Folded Reload
+; NO-OMIT-FP-ZCMP-NEXT:    ld s0, 32(sp) # 8-byte Folded Reload
+; NO-OMIT-FP-ZCMP-NEXT:    ld s1, 24(sp) # 8-byte Folded Reload
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_restore ra
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_restore s0
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_restore s1
+; NO-OMIT-FP-ZCMP-NEXT:    addi sp, sp, 48
+; NO-OMIT-FP-ZCMP-NEXT:    .cfi_def_cfa_offset 0
+; NO-OMIT-FP-ZCMP-NEXT:    ret
 entry:
   call void asm sideeffect "",
-  "~{v1},~{v2},~{v3},~{v4},~{v5},~{v6},~{v7}"()
+  "~{v1},~{v2},~{v3},~{v4},~{v5},~{v6},~{v7},~{s1}"()
 
   ret <vscale x 1 x i32> %va
 }

``````````

</details>


https://github.com/llvm/llvm-project/pull/117408


More information about the llvm-commits mailing list