[llvm-commits] [llvm] r80445 - in /llvm/trunk: lib/Target/X86/X86InstrInfo.td test/CodeGen/X86/cmov-i8-eflags.ll
Dan Gohman
gohman at apple.com
Sat Aug 29 15:19:15 PDT 2009
Author: djg
Date: Sat Aug 29 17:19:15 2009
New Revision: 80445
URL: http://llvm.org/viewvc/llvm-project?rev=80445&view=rev
Log:
CMOV_GR8 clobbers EFLAGS when its expansion involves an xor to set
a register to 0. This fixes PR4814.
Added:
llvm/trunk/test/CodeGen/X86/cmov-i8-eflags.ll
Modified:
llvm/trunk/lib/Target/X86/X86InstrInfo.td
Modified: llvm/trunk/lib/Target/X86/X86InstrInfo.td
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/X86/X86InstrInfo.td?rev=80445&r1=80444&r2=80445&view=diff
==============================================================================
--- llvm/trunk/lib/Target/X86/X86InstrInfo.td (original)
+++ llvm/trunk/lib/Target/X86/X86InstrInfo.td Sat Aug 29 17:19:15 2009
@@ -1038,8 +1038,10 @@
// X86 doesn't have 8-bit conditional moves. Use a customDAGSchedInserter to
// emit control flow. An alternative to this is to mark i8 SELECT as Promote,
// however that requires promoting the operands, and can induce additional
-// i8 register pressure.
-let usesCustomDAGSchedInserter = 1, isTwoAddress = 0 in
+// i8 register pressure. Note that CMOV_GR8 is conservatively considered to
+// clobber EFLAGS, because if one of the operands is zero, the expansion
+// could involve an xor.
+let usesCustomDAGSchedInserter = 1, isTwoAddress = 0, Defs = [EFLAGS] in
def CMOV_GR8 : I<0, Pseudo,
(outs GR8:$dst), (ins GR8:$src1, GR8:$src2, i8imm:$cond),
"#CMOV_GR8 PSEUDO!",
Added: llvm/trunk/test/CodeGen/X86/cmov-i8-eflags.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/cmov-i8-eflags.ll?rev=80445&view=auto
==============================================================================
--- llvm/trunk/test/CodeGen/X86/cmov-i8-eflags.ll (added)
+++ llvm/trunk/test/CodeGen/X86/cmov-i8-eflags.ll Sat Aug 29 17:19:15 2009
@@ -0,0 +1,55 @@
+; RUN: llvm-as < %s | llc -march=x86-64 | %prcontext {setne %al} 1 | grep test | count 2
+; PR4814
+
+; CodeGen shouldn't try to do a setne after an expanded 8-bit conditional
+; move without recomputing EFLAGS, because the expansion of the conditional
+; move with control flow may clobber EFLAGS (e.g., with xor, to set the
+; register to zero).
+
+; The prcontext usage above is a little awkward; the important part is that
+; there's a test before the setne.
+
+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"
+
+ at g_3 = external global i8 ; <i8*> [#uses=1]
+ at g_96 = external global i8 ; <i8*> [#uses=2]
+ at g_100 = external global i8 ; <i8*> [#uses=2]
+ at _2E_str = external constant [15 x i8], align 1 ; <[15 x i8]*> [#uses=1]
+
+define i32 @main() nounwind {
+entry:
+ %0 = load i8* @g_3, align 1 ; <i8> [#uses=2]
+ %1 = sext i8 %0 to i32 ; <i32> [#uses=1]
+ %.lobit.i = lshr i8 %0, 7 ; <i8> [#uses=1]
+ %tmp.i = zext i8 %.lobit.i to i32 ; <i32> [#uses=1]
+ %tmp.not.i = xor i32 %tmp.i, 1 ; <i32> [#uses=1]
+ %iftmp.17.0.i.i = ashr i32 %1, %tmp.not.i ; <i32> [#uses=1]
+ %retval56.i.i = trunc i32 %iftmp.17.0.i.i to i8 ; <i8> [#uses=1]
+ %2 = icmp eq i8 %retval56.i.i, 0 ; <i1> [#uses=2]
+ %g_96.promoted.i = load i8* @g_96 ; <i8> [#uses=3]
+ %3 = icmp eq i8 %g_96.promoted.i, 0 ; <i1> [#uses=2]
+ br i1 %3, label %func_4.exit.i, label %bb.i.i.i
+
+bb.i.i.i: ; preds = %entry
+ %4 = volatile load i8* @g_100, align 1 ; <i8> [#uses=0]
+ br label %func_4.exit.i
+
+func_4.exit.i: ; preds = %bb.i.i.i, %entry
+ %.not.i = xor i1 %2, true ; <i1> [#uses=1]
+ %brmerge.i = or i1 %3, %.not.i ; <i1> [#uses=1]
+ %.mux.i = select i1 %2, i8 %g_96.promoted.i, i8 0 ; <i8> [#uses=1]
+ br i1 %brmerge.i, label %func_1.exit, label %bb.i.i
+
+bb.i.i: ; preds = %func_4.exit.i
+ %5 = volatile load i8* @g_100, align 1 ; <i8> [#uses=0]
+ br label %func_1.exit
+
+func_1.exit: ; preds = %bb.i.i, %func_4.exit.i
+ %g_96.tmp.0.i = phi i8 [ %g_96.promoted.i, %bb.i.i ], [ %.mux.i, %func_4.exit.i ] ; <i8> [#uses=2]
+ store i8 %g_96.tmp.0.i, i8* @g_96
+ %6 = zext i8 %g_96.tmp.0.i to i32 ; <i32> [#uses=1]
+ %7 = tail call i32 (i8*, ...)* @printf(i8* noalias getelementptr ([15 x i8]* @_2E_str, i64 0, i64 0), i32 %6) nounwind ; <i32> [#uses=0]
+ ret i32 0
+}
+
+declare i32 @printf(i8* nocapture, ...) nounwind
More information about the llvm-commits
mailing list