similar to: [LLVMdev] using bugpoint to find miscompilation/code generation problems?

Hi David, > It then iterates until a simplified test case is created whose output is > still different from that of the native code. isn't that exactly what you want: a small test-case that shows the problem? Ciao, Duncan.

Yes, its what I want, but I can't figure out how to get bugpoint to accomplish this. -David ________________________________ From: Duncan Sands [mailto:baldrick at free.fr] Sent: Thu 2/18/2010 6:06 AM To: David Dunkle Cc: llvmdev at cs.uiuc.edu Subject: Re: [LLVMdev] using bugpoint to find miscompilation/code generation problems? Hi David, > It then iterates until a simplified test

Hi, I am trying to reduce what I believe to be a miscompilation bug. Running lli on my bitcode file causes a segmentation fault. However, running bugpoint as bugpoint file.bc gives me the following errors, /tmp/ccAdmNqH.o: In function `_ZL17bus_error_handleriP7siginfoPv': bugpoint-test-program.bc-Vega5s.cbe.c:(.text+0x1b4e1): undefined reference to `std::cerr' /tmp/ccAdmNqH.o: In

Hi all, I've been running into what appears to be an optimization bug in llvm, while compiling the latest ruby 1.9.3 with the latest llvm/clang 2.9 version. From hands on experimentation I do know that the gvn pass for the "vm.c" compilation is the culprit of the miscompilation (clang -O2/-O3 options produce the miscompilation, however -O1 is fine!). I've also tried to

To get x86 to work, you need to compile /arch/x86/boot and /arch/x86/boot/compressed with gcc. The rest can be compiled with llvm-gcc. The only adjustment you need to make is to /arch/x86/kernel/signal_32.c to fix the stack offset in sys_sigreturn. Also you must compile with -O0. instcombine is causing the crashes you are seeing. I know mem2reg, sccp, simplifycfg, and dce result in a working

Thanks, that helps a lot. > All chips (to date) with NEON have VFP3, so it's safe to assume that a -mfpu=neon will have VFP3, so all the decisions > about code generated for VFP3 can safely be assumed by targets with NEON. Just to confirm my understanding, can I correctly say in general that the llc code generator might blur distinctions between NEON and VFP3 when it can do so

On 27 May 2011 02:04, David Dunkle <ddunkle at arxan.com> wrote: > In all cases, I get code that looks pretty very the same; its like what > is below. However, I am expecting to see instruction level differences > between the vfp3 and neon versions. When I do the same with gcc 4.2 I do > see differences in the generated code. Hi David, You could see different instructions (as

I have a code generation question for ARM with VFP and NEON. I am generating code for the following function as a test: void FloatingPointTest(float f1, float f2, float f3) { float f4 = f1 * f2; if (f4 > f3) printf("%f\n",f2); else printf("%f\n",f3); } I have tried compiling with: 1. -mfloat-abi=softfp and -mfpu=neon 2.

On Fri, 22 Apr 2005, Markus F.X.J. Oberhumer wrote: > Chris Lattner wrote: >> On Fri, 22 Apr 2005, Markus F.X.J. Oberhumer wrote: >>> many thanks once more for your quick fix for this issue. Actually the >>> current CVS version is the first version since llvm-1.3 which does not >>> miscompile the UCL library, so this is a definitive improvement :-). And

> was wrong. At the time (~ 1 year ago) it was almost always > instcombine and almost always volatile related. Now bugpoint has some > ways to run the code by some external agent which I haven't > investigated yet. People have fixed a ton of bugs (volatile and otherwise) in instcombine since then. I will be very interested to hear what pass ends up being the problem here, and

On Mon, Nov 24, 2008 at 10:51 AM, John Regehr <regehr at cs.utah.edu> wrote: >> was wrong. At the time (~ 1 year ago) it was almost always >> instcombine and almost always volatile related. Now bugpoint has some >> ways to run the code by some external agent which I haven't >> investigated yet. > > People have fixed a ton of bugs (volatile and otherwise)

For people who want to try hacking on the linux kernel with llvm, here is a head start: http://llvm.org/~alenhar2/k.tbz This is a 2.6.27.5 kernel with a .config file for qemu/kvm with virtio devices (I also think it will work with the default devices). The build process uses llvm-gcc -> opt -> llc so you can add your own (per-file) passes to the build process or debug specific passes.

Hey that is great to hear that the volatile stuff is helping someone. Just broadly speaking do you know if the instcombine bug involves pointer code vs. scalar? I ask because intensive random testing has not found the bug that you are seeing. That says that either (1) the bug lies in a part of the program space we don't explore or (2) it does, but we haven't run the tests for long

I've made quite a bit of progress getting bugpoint to work with our (non-gcc) tools. In fact I caught the alignment bug I recently posted about using it. But it's not there yet. In particular, I am seeing this scenario a lot (comments in brackets): *** Found miscompiling pass: -instcombine Emitted bitcode to 'bugpoint-passinput.bc' [ Good! This is progress! ] *** You can

Hey everyone. It sounds like 426 people are in for a fun weekend. It's possible that some of you may even have bugs in your code (I know, I know, not YOU... :) If this is you, read on... You might be interested in trying out the 'bugpoint' tool. It can help you when your pass crashes on a testcase. Merely tell it the input testcase and the name of your pass, and it will try to

"Finkel, Hal J. via llvm-dev" <llvm-dev at lists.llvm.org> writes: > One concern that I have is that, from personal experience, the ability > for bugpoint to reduce the set of optimization passes applied in order > to reproduce a bug is extremely helpful. I understand your desire to > decouple the logic somewhat, and maybe there's some way to generalize > that

One other note I forgot... sig_sysreturn (arch/x86/kernel/signal_32.c) has a magic fudge factor to fix up stack layout differences between llvm and gcc. If when you hit userspace you get signal handling errors, look at the frame addresses printed out and adjust the fudge factor to make the frame addresses match. (the code automatically enables the debug output after the first signal frame

On Fri, 22 Apr 2005, Markus F.X.J. Oberhumer wrote: > I've finally got it working! The key point was to pass all bytecode > objects individually to bugpoint, and not to use the pre-linked bytecode from > gccld. nice! > After running for some time bugpoints exits saying: > > *** The following functions are being miscompiled: ucl_alloc main >

On Jul 8, 2009, at 12:42 AM, Viktor Kutuzov wrote: > Hello Evan, > > Thanks for looking at the patch. > >> This should use std::cerr and make sure it is wrapped inside the >> DEBUG >> macro. > > Will do. > >> Also, we don't want RemoteRunSafely.sh to be under utils/bugpoint. >> Can you move it to test-suite? Are you planning to change the

Hi, I am trying to build the Linux kernel with LLVM. 'ARCH=um' appears to work, now I am trying to get 'ARCH=x86' to work. So far it seems there is something wrong with the boot vga code (it finds no video modes), the acpi code, and the serial console code. I am now trying to compile drivers/ with llvm-gcc and the rest with gcc-4.2 (I have a wrapper script), I am compiling to