similar to: [LLVMdev] lnt parameters

The issues seems to be this line in the tools Makefile timeit-target: timeit.c $(LD_ENV_OVERRIDES) $(LCC) -o $@ $< $(LDFLAGS) $(CFLAGS) $(TARGET_FLAGS) -O3 It should not add target flags if we are simulating the target on the host. On 06/06/2013 06:59 PM, reed kotler wrote: > I want to get lnt to use qemu for the execution. > > In that case, RHOST= is not set. > > But I

I want to get lnt to use qemu for the execution. In that case, RHOST= is not set. But I change the Arch because I am going to run in cross mode. Then I'm setting RUNUNDER to be a script which runs qemu. In this case it builds timeit-target as a Mips which fails because this is running on x86. ~/mysandbox/bin/lnt runtest nt --sandbox ~/mysandbox --cc /local/llvmpb_a/install/bin/clang

is test-suite being run still by itself (independent of llvm)? if so, are there public build bots for this? is lnt going to replace test-suite as the normal mechanism for running test-suite? Tia. Reed

On 4 June 2013 22:08, Reed Kotler <rkotler at mips.com> wrote: > is test-suite being run still by itself (independent of llvm)? > AFAIK, yes. http://llvm.org/docs/lnt/quickstart.html if so, are there public build bots for this? > We have one: http://lab.llvm.org:8011/builders/clang-native-arm-lnt Setting it up was just copy&paste on Zorg's config (and fixing the bugs

Okay to push this change? On 07/23/2013 05:17 PM, reed kotler wrote: > Hi Daniel, > > In this case we are not using lnt under Qemu user mode for benchmarking; > just as a way to run test-suite to test whether the code is correct. > > Qemu user mode emulates target instructions, but when it gets a Unix > Kernel trap, it uses the host to emulate those. > > For example,

On Tue, Jun 4, 2013 at 5:52 PM, Renato Golin <renato.golin at linaro.org>wrote: > On 4 June 2013 22:08, Reed Kotler <rkotler at mips.com> wrote: > >> is test-suite being run still by itself (independent of llvm)? >> > > AFAIK, yes. http://llvm.org/docs/lnt/quickstart.html > > > if so, are there public build bots for this? >> > > We have

One could argue that mclinker is doing something good or not by how it's using this class but I don't see the need for parser<bool> to be final. That is a subjective opinion that mclinker needs to be changed. I think that "final" was added to some of these command line classes to avoid some kind of clang warning. That seems wrong to me that the tools are dictating

On 03/19/2015 08:55 AM, David Blaikie wrote: > > > On Thu, Mar 19, 2015 at 4:30 AM, Reed Kotler <Reed.Kotler at imgtec.com > <mailto:Reed.Kotler at imgtec.com>> wrote: > > One could argue that mclinker is doing something good or not by > how it's using this class > but I don't see the need for parser<bool> to be final. That is a >

//===----------------------------------------------------------------------===// // FalseParser //===----------------------------------------------------------------------===// class FalseParser : public parser<bool> { public: explicit FalseParser(Option &O) : parser<bool>(O) { } // parse - Return true on error. bool parse(cl::Option& O, StringRef ArgName, StringRef

Well, you are an mclinker contributor and Google uses mclinker and now it's broken as the result of your change. I still don't see any justification to making a change in a public interface that is used by other non LLVM projects to fix some issue with clang warnings. People should be able to derive from those classes. I can't understand your reasoning as to why these classes must

On 03/19/2015 09:24 AM, David Blaikie wrote: > > > On Thu, Mar 19, 2015 at 9:18 AM, Reed Kotler <reed.kotler at imgtec.com > <mailto:reed.kotler at imgtec.com>> wrote: > > Well, you are an mclinker contributor > > > Me personally? Not that I know of. Sorry. I thought i had seen your name in an mclinker commit. > > and Google uses mclinker >

On 03/10/2014 03:37 PM, Eric Christopher wrote: > cc'ing Bill as Release Manager > > On Mon, Mar 10, 2014 at 2:53 PM, reed kotler <rkotler at mips.com> wrote: >> I'm trying to understand how Mips will fit into the official release >> procedure. >> >> There are two Mips compilers: >> 1) The X86 resident llvm/clang compiler that can generate

On 03/19/2015 09:38 AM, David Blaikie wrote: > > > On Thu, Mar 19, 2015 at 9:34 AM, Reed Kotler <reed.kotler at imgtec.com > <mailto:reed.kotler at imgtec.com>> wrote: > > On 03/19/2015 09:24 AM, David Blaikie wrote: >> >> >> On Thu, Mar 19, 2015 at 9:18 AM, Reed Kotler >> <reed.kotler at imgtec.com <mailto:reed.kotler at

On 03/19/2015 09:57 AM, David Blaikie wrote: > > > On Thu, Mar 19, 2015 at 9:52 AM, Reed Kotler <reed.kotler at imgtec.com > <mailto:reed.kotler at imgtec.com>> wrote: > > On 03/19/2015 09:38 AM, David Blaikie wrote: >> >> >> On Thu, Mar 19, 2015 at 9:34 AM, Reed Kotler >> <reed.kotler at imgtec.com <mailto:reed.kotler at

While porting my backends to llvm-3.2, I found a few places where the optimizers could have performed better. I believe the mainstream targets can also benefits from my tweaks. But before upstreaming my changes, I would like to quantify their merits on other applications --- not just my domain specific codes. In a word, it seemed the right time for me to start using LNT :) I followed the

Hi Since the buildbot host was updated to Fedora F20 levels, the clang buildbot has been failing with the following: $ /home/clangbuild/clang-ppc64-1/lnt.venv/bin/lnt runtest --verbose --submit http://llvm.org/perf/submitRun --commit=1 nt --sandbox nt --no-timestamp --cc /home/clangbuild/clang-ppc64-1/llvm.install.1/bin/clang --cxx /home/clangbuild/clang-ppc64-1/llvm.install.1/bin/clang++

On 21 Apr 2016, at 17:44, Sergey Yakoushkin <sergey.yakoushkin at gmail.com<mailto:sergey.yakoushkin at gmail.com>> wrote: Hi Kristof, The way we use LNT, we would run different configuration (e.g. -O3 vs -Os) as different "machines" in LNT's model. O2/O3 is indeed bad example. We're also using different machines for Os/O3 - such parameters apply to all

Hello, When I was running the LNT testing for LLVM test-suite, it failed to submit the results to database. The command is: lnt runtest --submit db nt --sandbox sandbox --cc clang --cxx clang++ --test-suite ./test-suite-3.2/ --llvm-src ~/projects/llvm/llvm --llvm-obj ~/projects/llvm/llvm_build/ --test-style=nightly --only-test MultiSource/Benchmarks/mediabench/ When submitting the results

I followed the lnt quickstart <http://llvm.org/docs/lnt/quickstart.html> directions but got this diagnostic when doing the setup: bash-3.2$ ~/mysandbox/bin/python ~/lnt/setup.py develop /Users/dcallahan/mysandbox/lib/python2.7/site-packages/setuptools/dist.py:284: UserWarning: The version spec\ ified requires normalization, consider using '0.4.1.dev0' instead of

> 2. Running lnt several times on the same unmodified clang+llvm binaries gives > different results (execution time can vary wildly : ~200%). I am running lnt > on linux/x86_64. I tried to deactivate the cpufreq thing, the machine was not > loaded, but this did not change. Is there any way to run the tests multiple > times and use statistics to get reproducible numbers (within a