[RISCV] Use iXLen for ptr<->int casts in vararg.ll (#74426)

Also use ABI alignment for ptr sized objects.

This makes the code more sane and avoids only loading part of what was
stored by vastart on RV64.

Author: topperc

llvm/test/CodeGen/RISCV/vararg.ll

@@ -1,25 +1,25 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=riscv32 -verify-machineinstrs < %s \
+; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -verify-machineinstrs \
 ; RUN:   | FileCheck -check-prefix=ILP32-ILP32F-FPELIM %s
-; RUN: llc -mtriple=riscv32 -verify-machineinstrs -frame-pointer=all < %s \
+; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -verify-machineinstrs -frame-pointer=all \
 ; RUN:   | FileCheck -check-prefix=ILP32-ILP32F-WITHFP %s
-; RUN: llc -mtriple=riscv32 -mattr=+d -verify-machineinstrs < %s \
+; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -mattr=+d -verify-machineinstrs \
 ; RUN:   | FileCheck -check-prefix=RV32D-ILP32-ILP32F-ILP32D-FPELIM %s
-; RUN: llc -mtriple=riscv32 -mattr=+d -target-abi ilp32f \
-; RUN:     -verify-machineinstrs < %s \
+; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -mattr=+d -target-abi ilp32f \
+; RUN:     -verify-machineinstrs \
 ; RUN:   | FileCheck -check-prefix=RV32D-ILP32-ILP32F-ILP32D-FPELIM %s
-; RUN: llc -mtriple=riscv32 -mattr=+d -target-abi ilp32d \
-; RUN:     -verify-machineinstrs < %s \
+; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -mattr=+d -target-abi ilp32d \
+; RUN:     -verify-machineinstrs \
 ; RUN:   | FileCheck -check-prefix=RV32D-ILP32-ILP32F-ILP32D-FPELIM %s
-; RUN: llc -mtriple=riscv64 -verify-machineinstrs < %s \
+; RUN: sed 's/iXLen/i64/g' %s | llc -mtriple=riscv64 -verify-machineinstrs \
 ; RUN:   | FileCheck -check-prefix=LP64-LP64F-LP64D-FPELIM %s
-; RUN: llc -mtriple=riscv64 -mattr=+d -target-abi lp64f \
-; RUN:     -verify-machineinstrs < %s \
+; RUN: sed 's/iXLen/i64/g' %s | llc -mtriple=riscv64 -mattr=+d -target-abi lp64f \
+; RUN:     -verify-machineinstrs \
 ; RUN:   | FileCheck -check-prefix=LP64-LP64F-LP64D-FPELIM %s
-; RUN: llc -mtriple=riscv64 -mattr=+d -target-abi lp64d \
-; RUN:     -verify-machineinstrs < %s \
+; RUN: sed 's/iXLen/i64/g' %s | llc -mtriple=riscv64 -mattr=+d -target-abi lp64d \
+; RUN:     -verify-machineinstrs \
 ; RUN:   | FileCheck -check-prefix=LP64-LP64F-LP64D-FPELIM %s
-; RUN: llc -mtriple=riscv64 -verify-machineinstrs -frame-pointer=all < %s \
+; RUN: sed 's/iXLen/i64/g' %s | llc -mtriple=riscv64 -verify-machineinstrs -frame-pointer=all \
 ; RUN:   | FileCheck -check-prefix=LP64-LP64F-LP64D-WITHFP %s
 
 ; The same vararg calling convention is used for ilp32/ilp32f/ilp32d and for
@@ -556,24 +556,16 @@ define i64 @va2(ptr %fmt, ...) nounwind {
 ; LP64-LP64F-LP64D-FPELIM-LABEL: va2:
 ; LP64-LP64F-LP64D-FPELIM:       # %bb.0:
 ; LP64-LP64F-LP64D-FPELIM-NEXT:    addi sp, sp, -80
+; LP64-LP64F-LP64D-FPELIM-NEXT:    mv a0, a1
 ; LP64-LP64F-LP64D-FPELIM-NEXT:    sd a7, 72(sp)
 ; LP64-LP64F-LP64D-FPELIM-NEXT:    sd a6, 64(sp)
 ; LP64-LP64F-LP64D-FPELIM-NEXT:    sd a5, 56(sp)
 ; LP64-LP64F-LP64D-FPELIM-NEXT:    sd a4, 48(sp)
 ; LP64-LP64F-LP64D-FPELIM-NEXT:    sd a3, 40(sp)
 ; LP64-LP64F-LP64D-FPELIM-NEXT:    sd a2, 32(sp)
 ; LP64-LP64F-LP64D-FPELIM-NEXT:    sd a1, 24(sp)
-; LP64-LP64F-LP64D-FPELIM-NEXT:    addi a0, sp, 24
-; LP64-LP64F-LP64D-FPELIM-NEXT:    sd a0, 8(sp)
-; LP64-LP64F-LP64D-FPELIM-NEXT:    lw a0, 8(sp)
-; LP64-LP64F-LP64D-FPELIM-NEXT:    addi a0, a0, 7
-; LP64-LP64F-LP64D-FPELIM-NEXT:    slli a1, a0, 32
-; LP64-LP64F-LP64D-FPELIM-NEXT:    srli a1, a1, 32
-; LP64-LP64F-LP64D-FPELIM-NEXT:    addi a1, a1, 8
+; LP64-LP64F-LP64D-FPELIM-NEXT:    addi a1, sp, 39
 ; LP64-LP64F-LP64D-FPELIM-NEXT:    sd a1, 8(sp)
-; LP64-LP64F-LP64D-FPELIM-NEXT:    srliw a0, a0, 3
-; LP64-LP64F-LP64D-FPELIM-NEXT:    slli a0, a0, 3
-; LP64-LP64F-LP64D-FPELIM-NEXT:    ld a0, 0(a0)
 ; LP64-LP64F-LP64D-FPELIM-NEXT:    addi sp, sp, 80
 ; LP64-LP64F-LP64D-FPELIM-NEXT:    ret
 ;
@@ -583,37 +575,30 @@ define i64 @va2(ptr %fmt, ...) nounwind {
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    sd ra, 24(sp) # 8-byte Folded Spill
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    sd s0, 16(sp) # 8-byte Folded Spill
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    addi s0, sp, 32
+; LP64-LP64F-LP64D-WITHFP-NEXT:    mv a0, a1
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    sd a7, 56(s0)
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    sd a6, 48(s0)
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    sd a5, 40(s0)
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    sd a4, 32(s0)
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    sd a3, 24(s0)
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    sd a2, 16(s0)
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    sd a1, 8(s0)
-; LP64-LP64F-LP64D-WITHFP-NEXT:    addi a0, s0, 8
-; LP64-LP64F-LP64D-WITHFP-NEXT:    sd a0, -24(s0)
-; LP64-LP64F-LP64D-WITHFP-NEXT:    lw a0, -24(s0)
-; LP64-LP64F-LP64D-WITHFP-NEXT:    addi a0, a0, 7
-; LP64-LP64F-LP64D-WITHFP-NEXT:    slli a1, a0, 32
-; LP64-LP64F-LP64D-WITHFP-NEXT:    srli a1, a1, 32
-; LP64-LP64F-LP64D-WITHFP-NEXT:    addi a1, a1, 8
+; LP64-LP64F-LP64D-WITHFP-NEXT:    addi a1, s0, 23
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    sd a1, -24(s0)
-; LP64-LP64F-LP64D-WITHFP-NEXT:    srliw a0, a0, 3
-; LP64-LP64F-LP64D-WITHFP-NEXT:    slli a0, a0, 3
-; LP64-LP64F-LP64D-WITHFP-NEXT:    ld a0, 0(a0)
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    addi sp, sp, 96
 ; LP64-LP64F-LP64D-WITHFP-NEXT:    ret
   %va = alloca ptr
   call void @llvm.va_start(ptr %va)
-  %argp.cur = load i32, ptr %va, align 4
-  %1 = add i32 %argp.cur, 7
-  %2 = and i32 %1, -8
-  %argp.cur.aligned = inttoptr i32 %1 to ptr
+  %argp.cur = load ptr, ptr %va
+  %ptrint = ptrtoint ptr %argp.cur to iXLen
+  %1 = add iXLen %ptrint, 7
+  %2 = and iXLen %1, -8
+  %argp.cur.aligned = inttoptr iXLen %1 to ptr
   %argp.next = getelementptr inbounds i8, ptr %argp.cur.aligned, i32 8
-  store ptr %argp.next, ptr %va, align 4
-  %3 = inttoptr i32 %2 to ptr
+  store ptr %argp.next, ptr %va
+  %3 = inttoptr iXLen %2 to ptr
   %4 = load double, ptr %3, align 8
   %5 = bitcast double %4 to i64
   call void @llvm.va_end(ptr %va)