[clang] [llvm] [LLVM][SROA] Teach SROA how to "bitcast" between fixed and scalable vectors. (PR #130973)

[Nikita Popov via cfe-commits]

================
@@ -26,11 +26,15 @@ typedef vbool64_t fixed_bool64_t __attribute__((riscv_rvv_vector_bits(__riscv_v_
 //
 // CHECK-128-LABEL: @call_bool32_ff(
 // CHECK-128-NEXT:  entry:
+// CHECK-128-NEXT:    [[SAVED_VALUE:%.*]] = alloca <1 x i8>, align 1
----------------
nikic wrote:

I'd expect `<vscale x 4 x i1>` to also be broken. We need the fixed size to be a multiple of 8 to compute correct sizes.

https://github.com/llvm/llvm-project/pull/130973