search for: setcond

Hi LLVMDev list I am building LLVM from the SVN trunk at 213638 on a W7/X86_64/Cygwin system and running make check leads to a series of failed assertions like ******************** FAIL: LLVM :: ExecutionEngine/MCJIT/test-setcond-fp.ll (6185 of 11245) ******************** TEST 'LLVM :: ExecutionEngine/MCJIT/test-setcond-fp.ll' FAILED ******************** Script: -- /cygdrive/z/dev/llvm/x64/static/Release+Asserts/bin/lli -use-mcjit -mtriple=x86_64-unknown-cygwin-elf /cygdrive/z/dev/llvm/x64/llvm/test/ExecutionEngi...

...; wrote: > > Hi LLVMDev list > > I am building LLVM from the SVN trunk at 213638 on a > W7/X86_64/Cygwin system and running make check leads to a series > of failed assertions like > > ******************** > FAIL: LLVM :: ExecutionEngine/MCJIT/test-setcond-fp.ll (6185 of 11245) > ******************** TEST 'LLVM :: > ExecutionEngine/MCJIT/test-setcond-fp.ll' FAILED ******************** > Script: > -- > /cygdrive/z/dev/llvm/x64/static/Release+Asserts/bin/lli -use-mcjit > -mtriple=x86_64-unknown-cygwin-e...

Slowly we are getting to the point, that we miss enough optimization opportunities as the result of our own passes. For this we need to fix AlgebraicOpt to be able to handle mods on sources without creating new issues. The last patch enables looping opts. v2: update commit author Karol Herbst (3): nv50/ir: fix AlgebraicOpt for slcts with mods nv50/ir: handle logops with NOT in AlgebraicOpt

On Sun, 9 May 2004, Reid Spencer wrote: > Before I file a bug, I thought I'd check to make sure something hasn't > changed. Stacker used to work in 1.2, now it fails on six tests. All six > tests pertain to use of the SetLT, SetGT, SetLE, and SetGE SetCondInst. > In my test programs they all fail, regardless of the values. > > The *only* thing I've done to Stacker is to change the base type > involved in the comparison from int to long. Is there something > "different" about the comparison of longs? > > The code gen...

On Sun, 2004-05-09 at 15:25, Chris Lattner wrote: > That sounds like a bug. I'll look into it. I assume you are seeing this > problem with the X86 backend? > Yes. > > Perhaps the cast from bool to long doesn't work the same as for bool > > to int? Should it? > > Yes, you should get a long 0 or 1. That's what I thought. > Please do, I'll also

Signed-off-by: Ilia Mirkin <imirkin at alum.mit.edu> --- Pretty sure there's nothing wrong with it, but it looks odd in the code. src/gallium/drivers/nouveau/codegen/nv50_ir_emit_gk110.cpp | 2 ++ src/gallium/drivers/nouveau/codegen/nv50_ir_emit_gm107.cpp | 7 +++++-- src/gallium/drivers/nouveau/codegen/nv50_ir_emit_nvc0.cpp | 2 ++ 3 files changed, 9 insertions(+), 2 deletions(-)

Slowly we are getting to the point, that we miss enough optimization opportunities as the result of our own passes. For this we need to fix AlgebraicOpt to be able to handle mods on sources without creating new issues. The last patch enables looping opts. Karol Herbst (3): nv50/ir: fix AlgebraicOpt for slcts with mods nv50/ir: handle logops with NOT in AlgebraicOpt nv50/ir: run some

...e != TYPE_U32) > + return; > + > + i->getSrc(t)->getInsn()->dType = TYPE_F32; > + if (i->src(t).mod != Modifier(0)) { > + assert(i->src(t).mod == Modifier(NV50_IR_MOD_NOT)); > + i->src(t).mod = Modifier(0); > + cmp->setCond = reverseCondCode(cmp->setCond); > + } > + i->op = OP_MOV; > + i->setSrc(s, NULL); > + if (t) { > + i->setSrc(0, i->getSrc(t)); > + i->setSrc(t, NULL); > + } > + } > + break; > + > case OP_SHL: &...

.../gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp @@ -1856,6 +1856,12 @@ AlgebraicOpt::handleLOGOP(Instruction *logop) set0 = cloneForward(func, set0); set1 = cloneShallow(func, set1); + + if (logop->src(0).mod == Modifier(NV50_IR_MOD_NOT)) + set0->asCmp()->setCond = inverseCondCode(set0->asCmp()->setCond); + if (logop->src(1).mod == Modifier(NV50_IR_MOD_NOT)) + set1->asCmp()->setCond = inverseCondCode(set1->asCmp()->setCond); + logop->bb->insertAfter(logop, set1); logop->bb->insertAfter(logop, set0)...

.../nv50_ir_peephole.cpp @@ -187,7 +187,8 @@ LoadPropagation::checkSwapSrc01(Instruction *insn) return; } - if (insn->op == OP_SET) + if (insn->op == OP_SET || insn->op == OP_SET_AND || + insn->op == OP_SET_OR || insn->op == OP_SET_XOR) insn->asCmp()->setCond = reverseCondCode(insn->asCmp()->setCond); else if (insn->op == OP_SLCT) -- 1.8.5.5

...ero.ll LLVM :: ExecutionEngine/test-global.ll LLVM :: ExecutionEngine/test-interp-vec-arithm_float.ll LLVM :: ExecutionEngine/test-interp-vec-arithm_int.ll LLVM :: ExecutionEngine/test-interp-vec-loadstore.ll LLVM :: ExecutionEngine/test-interp-vec-logical.ll LLVM :: ExecutionEngine/test-interp-vec-setcond-fp.ll LLVM :: ExecutionEngine/test-interp-vec-setcond-int.ll LLVM :: ExecutionEngine/test-loadstore.ll LLVM :: ExecutionEngine/test-local.ll LLVM :: ExecutionEngine/test-logical.ll LLVM :: ExecutionEngine/test-loop.ll LLVM :: ExecutionEngine/test-phi.ll LLVM :: ExecutionEngine/test-ret.ll LLVM :: E...

...IT/test-local.ll LLVM :: ExecutionEngine/MCJIT/test-logical.ll LLVM :: ExecutionEngine/MCJIT/test-loop.ll LLVM :: ExecutionEngine/MCJIT/test-phi.ll LLVM :: ExecutionEngine/MCJIT/test-ret.ll LLVM :: ExecutionEngine/MCJIT/test-return.ll LLVM :: ExecutionEngine/MCJIT/test-setcond-fp.ll LLVM :: ExecutionEngine/MCJIT/test-setcond-int.ll LLVM :: ExecutionEngine/MCJIT/test-shift.ll Expected Passes : 10774 Expected Failures : 82 Unsupported Tests : 14 Unexpected Failures: 47

Before I file a bug, I thought I'd check to make sure something hasn't changed. Stacker used to work in 1.2, now it fails on six tests. All six tests pertain to use of the SetLT, SetGT, SetLE, and SetGE SetCondInst. In my test programs they all fail, regardless of the values. The *only* thing I've done to Stacker is to change the base type involved in the comparison from int to long. Is there something "different" about the comparison of longs? The code generated by Stacker Compiler is lik...

On Mon, Jun 18, 2012 at 04:56:53PM -0700, reed kotler wrote: > I don't see any tests in either test or test-suite for -use-mcjit. For ARM, we need to manually switch to use mcjit, say $ make check-all LIT_ARGS=--param=jit_impl=mcjit Regards, chenwj -- Wei-Ren Chen (陳韋任) Computer Systems Lab, Institute of Information Science, Academia Sinica, Taiwan (R.O.C.) Tel:886-2-2788-3799

...turn; + if (cmp->dType != TYPE_U32) + return; - i->getSrc(t)->getInsn()->dType = TYPE_F32; - if (i->src(t).mod != Modifier(0)) { - assert(i->src(t).mod == Modifier(NV50_IR_MOD_NOT)); - i->src(t).mod = Modifier(0); - cmp->setCond = inverseCondCode(cmp->setCond); - } - i->op = OP_MOV; - i->setSrc(s, NULL); - if (t) { - i->setSrc(0, i->getSrc(t)); - i->setSrc(t, NULL); + cmp->dType = TYPE_F32; + if (i->src(t).mod != Modifier(0)) { + assert(...

...ExecutionEngine/MCJIT/test-logical.ll > LLVM :: ExecutionEngine/MCJIT/test-loop.ll > LLVM :: ExecutionEngine/MCJIT/test-phi.ll > LLVM :: ExecutionEngine/MCJIT/test-ret.ll > LLVM :: ExecutionEngine/MCJIT/test-return.ll > LLVM :: ExecutionEngine/MCJIT/test-setcond-fp.ll > LLVM :: ExecutionEngine/MCJIT/test-setcond-int.ll > LLVM :: ExecutionEngine/MCJIT/test-shift.ll > > Expected Passes : 10774 > Expected Failures : 82 > Unsupported Tests : 14 > Unexpected Failures: 47 > > ___________________________...

I don't see any tests in either test or test-suite for -use-mcjit. Are we not testing this yet? There are lots of other llc options. What is our plan for testing these?

...n... > > assert(FibF->abegin() != FibF->aend()); // Make sure there's an arg > > > > Argument &ArgX = FibF->afront(); // Get the arg > > ArgX.setName("AnArg"); // Give it a nice symbolic name for fun. > > > > SetCondInst* CondInst > > = new SetCondInst( Instruction::SetLE, > > &ArgX, Two ); > > > > BB->getInstList().push_back(CondInst); > > > > // Create the true_block > > BasicBlock* true_bb = new BasicBlock("arg<=2"); > &gt...

...4c92a1efb5..bd60a84998 100644 --- a/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp +++ b/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp @@ -1797,10 +1797,10 @@ AlgebraicOpt::handleSLCT(Instruction *slct) if (slct->getSrc(2)->asImm()->compare(slct->asCmp()->setCond, 0.0f)) slct->setSrc(0, slct->getSrc(1)); } else - if (slct->getSrc(0) != slct->getSrc(1)) { + if (slct->getSrc(0) != slct->getSrc(1) || slct->src(0).mod != slct->src(1).mod) return; - } - slct->op = OP_MOV; + slct->op = slct->src(0).m...

...argument of the add1 function... > assert(FibF->abegin() != FibF->aend()); // Make sure there's an arg > > Argument &ArgX = FibF->afront(); // Get the arg > ArgX.setName("AnArg"); // Give it a nice symbolic name for fun. > > SetCondInst* CondInst > = new SetCondInst( Instruction::SetLE, > &ArgX, Two ); > > BB->getInstList().push_back(CondInst); > > // Create the true_block > BasicBlock* true_bb = new BasicBlock("arg<=2"); > > > // Create the retu...