similar to: [LLVMdev] Regarding codegenprepare transformations

On Mar 15, 2010, at 7:11 AM, Sebastian Schlunke wrote: > Hi, > > i ran into a problem when using blockaddress() with a label in another function. It seems to me that LLVM tries to remove the label used in blockaddress because it seems like it is not used, but in fact it may be used somewhere with a indirectbr. > > I attached a small test-case that produces this error. (The

I see. But the block does not necessarily contain dead code. My original problem is more like this: define i32 @main() { entry: %target = bitcast i8* blockaddress(@test_fun, %test_label) to i8* call i32 @test_fun(i8* %target) ret i32 0 } define i32 @test_fun(i8* %target) { entry: indirectbr i8* %target, [label %test_label] test_label: ; assume some code here... br label %ret ret: ret

Thanks again. I executed the following command line llc -O3 comb.ll.bc -debug-pass=Arguments and got Pass Arguments: -targetdata -targetpassconfig -no-aa -tbaa -targetlibinfo -basicaa -collector-metadata -machinemoduleinfo -machine-branch-prob -preverify -domtree -verify -loops -loop-simplify -scalar-evolution -loop-simplify -iv-users -loop-reduce -gc-lowering -unreachableblockelim

On Mar 15, 2010, at 10:01 AM, Sebastian Schlunke wrote: > I see. But the block does not necessarily contain dead code. This case is now fixed in r98566, I will fix the 'dead block' case in a bit. -Chris > > My original problem is more like this: > > define i32 @main() { > entry: > %target = bitcast i8* blockaddress(@test_fun, %test_label) to i8* > > call

Hi, > I executed the following command line > > llc -O3 comb.ll.bc -debug-pass=Arguments > > and got > > Pass Arguments: -targetdata -targetpassconfig -no-aa -tbaa > -targetlibinfo -basicaa -collector-metadata -machinemoduleinfo > -machine-branch-prob -preverify -domtree -verify -loops -loop-simplify > -scalar-evolution -loop-simplify -iv-users -loop-reduce

Works like a charm! Thanks for the fast help. :) - Sebastian On Monday 15 March 2010 20:10:54 you wrote: > > On Mar 15, 2010, at 10:01 AM, Sebastian Schlunke wrote: > > > I see. But the block does not necessarily contain dead code. > > This case is now fixed in r98566, I will fix the 'dead block' case in a bit. > > -Chris > > > > > My

On Mar 15, 2010, at 9:41 AM, Chris Lattner wrote: > > On Mar 15, 2010, at 7:11 AM, Sebastian Schlunke wrote: > >> Hi, >> >> i ran into a problem when using blockaddress() with a label in another function. It seems to me that LLVM tries to remove the label used in blockaddress because it seems like it is not used, but in fact it may be used somewhere with a

Does LLVM have loop multiversioning ? it seems it does not with clang++ -O3 -mllvm -debug-pass=Arguments program.c -c bash-4.1$ clang++ -O3 -mllvm -debug-pass=Arguments fast_algorithms.c -c clang-3.6: warning: treating 'c' input as 'c++' when in C++ mode, this behavior is deprecated Pass Arguments: -datalayout -notti -basictti -x86tti -targetlibinfo -no-aa -tbaa -scoped-noalias

Dear LLVM, I'm trying to understand why opt tool does not know some of the registered passes, for example, -targetpassconfig, -gc-lowering, -unreachableblockelim, -unreachableblockelim, -stack-protector, -machinemoduleinfo. This prevents me from using bugpoint to reduce the crash in backend. As I understand, they should be registered and available through the command line in the same way as

Hi Duncan, Is it correct that opt does not understand codegen passes, but llc understands both analysis/transformation and codegen passes? I would like to modify bugpoint to work on this end. Thanks, - D. 2012/6/29 Duncan Sands <baldrick at free.fr>: > Hi Dima, > >> I'm trying to understand why opt tool does not know some of the >> registered passes, for example,

Hi Dima, > I'm trying to understand why opt tool does not know some of the > registered passes, for example, -targetpassconfig, -gc-lowering, > -unreachableblockelim, -unreachableblockelim, -stack-protector, > -machinemoduleinfo. This prevents me from using bugpoint to reduce the > crash in backend. As I understand, they should be registered and > available through the

Hi Dmitry, > Is it correct that opt does not understand codegen passes, but llc > understands both analysis/transformation and codegen passes? no. Ciao, Duncan. I would > like to modify bugpoint to work on this end. > > Thanks, > - D. > > 2012/6/29 Duncan Sands <baldrick at free.fr>: >> Hi Dima, >> >>> I'm trying to understand why opt

Hi, I am working on a patch for CodeGenPrepare, which introduces a use of TargetLoweringBase::InstructionOpcodeToISD (see [1] for the details). This is usual for CodeGenPrepare to use the TargetLowering class when it is available, however, using this particular function creates linking problems. ** Context ** The TargetLowering class is part of libLLVMCodeGen, which means that in theory every

---------- Forwarded message ---------- From: Qingan Li <ww345ww at gmail.com> Date: 2011/5/30 Subject: To: llvmdev at cs.uiuc.edu Hi, I wrote an analysis pass, myPass, inherited from both ModulePass and ProfileInfo, and this pass requires the CallGraph, i.e., * class myPass : public ModulePass, public ProfileInfo { ...};* * void myPass::getAnalysisUsage(AnalysisUsage &AU) const

Hi, I have come across a couple of cases where the code generated after CodeGenPrepare pass has "br i1 false .." with both true and false conditions preserved and this propagates further and remains the same in the final assembly code/executable. In CodeGenPrepare::runOnFunction, ConstantFoldTerminator (which handles the br i1 false condition) is called only once and if after the

I've developed a working LLVM back-end (based on LLVM 2.6) for a custom architecture with its own tool chain. This tool chain creates stand-alone programs from a single assembly. We used to use GCC, which supported producing a single machine assembly from multiple source files. I modified Clang to accept the architecture, but discovered that clang-cc (or the Clang Tool subclass inside Clang)

Sorry, then I don't understand. If I invoke backend with target->addPassesToEmitFile, it adds both opt and llc passes. Thus, PassManager contains both opt and llc passes after that. In case of crash, bugpoint reloads all passes from the failing pass manager to the separate instance of opt. If opt and llc both do not support such mixing, then I'm guessing I need to replace

> we lower llvm.objectsize later than we should Is there a well-accepted best (or even just better) place to lower objectsize? I ask because I sorta fear that these kinds of problems will become more pronounced as llvm.is.constant, which is also lowered in CGP, gains popularity. (To be clear, I think it totally makes sense to lower is.constant and objectsize in the same place. I'm just

On Nov 12, 2013, at 11:24 AM, Junbum Lim <junbums at gmail.com> wrote: > > I wonder why CodeGenPrepare breaks GEP into integer calculations (ptrtoin/add/inttopt) instead of directly sinking the address calculation using GEP into user's block. I believe it's primary for address mode matching where only part of the GEP can be folded (depending on the instruction set). Evan

On Nov 20, 2013, at 5:38 PM, Hal Finkel <hfinkel at anl.gov> wrote: > ----- Original Message ----- >> From: "Evan Cheng" <evan.cheng at apple.com> >> To: "Junbum Lim" <junbums at gmail.com> >> Cc: llvmdev at cs.uiuc.edu >> Sent: Wednesday, November 20, 2013 7:01:49 PM >> Subject: Re: [LLVMdev] sinking address computing in