mirror of
https://github.com/RPCS3/llvm.git
synced 2024-12-13 14:47:00 +00:00
50b8835451
Summary: As a side-quest for D6629 jvoung pointed out that I should use -verify-machineinstrs and this found a bug in x86-32's handling of EFLAGS for PUSHF/POPF. This patch fixes the use/def, and adds -verify-machineinstrs to all x86 tests which contain 'EFLAGS'. One exception: this patch leaves inline-asm-fpstack.ll as-is because it fails -verify-machineinstrs in a way unrelated to EFLAGS. This patch also modifies cmpxchg-clobber-flags.ll along the lines of what D6629 already does by also testing i386. Test Plan: ninja check Reviewers: t.p.northover, jvoung Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D6687 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@224359 91177308-0d34-0410-b5e6-96231b3b80d8
81 lines
2.6 KiB
LLVM
81 lines
2.6 KiB
LLVM
; RUN: llc < %s -verify-machineinstrs -disable-fp-elim -disable-machine-dce -verify-coalescing
|
|
target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
|
|
target triple = "x86_64-apple-macosx10.7.0"
|
|
|
|
; This test case has a sub-register join followed by a remat:
|
|
;
|
|
; 256L %vreg2<def> = COPY %vreg7:sub_32bit<kill>; GR32:%vreg2 GR64:%vreg7
|
|
; Considering merging %vreg2 with %vreg7:sub_32bit
|
|
; Cross-class to GR64.
|
|
; RHS = %vreg2 = [256d,272d:0) 0@256d
|
|
; LHS = %vreg7 = [208d,256d:0)[304L,480L:0) 0@208d
|
|
; updated: 272L %vreg0<def> = COPY %vreg7:sub_32bit<kill>; GR32:%vreg0 GR64:%vreg7
|
|
; Joined. Result = %vreg7 = [208d,272d:0)[304L,480L:0) 0@208d
|
|
;
|
|
; 272L %vreg10:sub_32bit<def> = COPY %vreg7:sub_32bit<kill>, %vreg10<imp-def>; GR64:%vreg10,%vreg7
|
|
; Considering merging %vreg7 with %vreg10
|
|
; RHS = %vreg7 = [208d,272d:0)[304L,480L:0) 0@208d
|
|
; LHS = %vreg10 = [16d,64L:2)[64L,160L:1)[192L,240L:1)[272d,304L:3)[304L,352d:1)[352d,400d:0)[400d,400S:4) 0@352d 1@64L-phidef 2@16d-phikill 3@272d-phikill 4@400d
|
|
; Remat: %vreg10<def> = MOV64r0 %vreg10<imp-def>, %EFLAGS<imp-def,dead>, %vreg10<imp-def>; GR64:%vreg10
|
|
; Shrink: %vreg7 = [208d,272d:0)[304L,480L:0) 0@208d
|
|
; live-in at 240L
|
|
; live-in at 416L
|
|
; live-in at 320L
|
|
; live-in at 304L
|
|
; Shrunk: %vreg7 = [208d,256d:0)[304L,480L:0) 0@208d
|
|
;
|
|
; The COPY at 256L is rewritten as a partial def, and that would artificially
|
|
; extend the live range of %vreg7 to end at 256d. When the joined copy is
|
|
; removed, -verify-coalescing complains about the dangling kill.
|
|
;
|
|
; <rdar://problem/9967101>
|
|
|
|
define void @f1() nounwind uwtable ssp {
|
|
bb:
|
|
br label %bb1
|
|
|
|
bb1:
|
|
%tmp = phi i32 [ 0, %bb ], [ %tmp21, %bb20 ]
|
|
br label %bb2
|
|
|
|
bb2:
|
|
br i1 undef, label %bb5, label %bb8
|
|
|
|
bb4:
|
|
br i1 undef, label %bb2, label %bb20
|
|
|
|
bb5:
|
|
br i1 undef, label %bb4, label %bb20
|
|
|
|
bb8:
|
|
%tmp9 = phi i32 [ %tmp24, %bb23 ], [ 0, %bb2 ]
|
|
br i1 false, label %bb41, label %bb10
|
|
|
|
bb10:
|
|
%tmp11 = sub nsw i32 %tmp9, %tmp
|
|
br i1 false, label %bb2, label %bb26
|
|
|
|
bb20:
|
|
%tmp21 = phi i32 [ undef, %bb4 ], [ undef, %bb5 ], [ %tmp9, %bb27 ], [ undef, %bb32 ]
|
|
%tmp22 = phi i32 [ undef, %bb4 ], [ undef, %bb5 ], [ %tmp11, %bb27 ], [ undef, %bb32 ]
|
|
br label %bb1
|
|
|
|
bb23:
|
|
%tmp24 = add nsw i32 %tmp9, 1
|
|
br label %bb8
|
|
|
|
bb26:
|
|
br i1 undef, label %bb27, label %bb32
|
|
|
|
bb27:
|
|
%tmp28 = zext i32 %tmp11 to i64
|
|
%tmp30 = icmp eq i64 undef, %tmp28
|
|
br i1 %tmp30, label %bb20, label %bb27
|
|
|
|
bb32:
|
|
br i1 undef, label %bb20, label %bb23
|
|
|
|
bb41:
|
|
ret void
|
|
}
|