1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2025-01-31 20:51:52 +01:00
llvm-mirror/test/CodeGen/X86/bswap-rotate.ll
Simon Pilgrim d654e7d40c [X86] Handle COPYs of physregs better (regalloc hints)
Enable enableMultipleCopyHints() on X86.

Original Patch by @jonpa:

While enabling the mischeduler for SystemZ, it was discovered that for some reason a test needed one extra seemingly needless COPY (test/CodeGen/SystemZ/call-03.ll). The handling for that is resulted in this patch, which improves the register coalescing by providing not just one copy hint, but a sorted list of copy hints. On SystemZ, this gives ~12500 less register moves on SPEC, as well as marginally less spilling.

Instead of improving just the SystemZ backend, the improvement has been implemented in common-code (calculateSpillWeightAndHint(). This gives a lot of test failures, but since this should be a general improvement I hope that the involved targets will help and review the test updates.

Differential Revision: https://reviews.llvm.org/D38128

llvm-svn: 342578
2018-09-19 18:59:08 +00:00

29 lines
933 B
LLVM

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=i686-unknown-unknown -mcpu=i686 | FileCheck %s --check-prefix=X86
; RUN: llc < %s -mtriple=x86_64-unknown-unknown | FileCheck %s --check-prefix=X64
; Combine BSWAP (lowered to rolw 8) with a second rotate.
; This test checks for combining rotates with inconsistent constant value types.
define i16 @combine_bswap_rotate(i16 %a0) {
; X86-LABEL: combine_bswap_rotate:
; X86: # %bb.0:
; X86-NEXT: movzwl {{[0-9]+}}(%esp), %eax
; X86-NEXT: rolw $9, %ax
; X86-NEXT: retl
;
; X64-LABEL: combine_bswap_rotate:
; X64: # %bb.0:
; X64-NEXT: movl %edi, %eax
; X64-NEXT: rolw $9, %ax
; X64-NEXT: # kill: def $ax killed $ax killed $eax
; X64-NEXT: retq
%1 = call i16 @llvm.bswap.i16(i16 %a0)
%2 = shl i16 %1, 1
%3 = lshr i16 %1, 15
%4 = or i16 %2, %3
ret i16 %4
}
declare i16 @llvm.bswap.i16(i16)