similar to: [LLVMdev] instprof tests down in ARM build

Just curious, what's the reason we don't make the github mirrors the official mirrors? ...besides not knowing how to revive them when they go down. :) -Greg On Tue, Mar 4, 2014 at 5:35 PM, Steven Noonan <steven at uplinklabs.net> wrote: > Someone should enable the smart HTTP protocol: > > https://www.kernel.org/pub/software/scm/git/docs/git-http-backend.html > > On

> LLVM has its own mirror at http://llvm.org/git/llvm.git, if that suits your needs. The llvm.org mirror seems to be quite a bit slower (~3.5x). I think it's because github lets me use the git protocol whereas llvm.org suggests http. When I try to clone "git at llvm.org:git/llvm.git", I'm asked for a password for user 'git'. -Greg On Tue, Mar 4, 2014 at 3:54 PM,

As a temporary fix, you can replace this line in sanitizer_linux_libcdep.c: const uptr kThreadDescriptorSize = FIRST_32_SECOND_64(1216, 2304); with const uptr kThreadDescriptorSize = FIRST_32_SECOND_64(1168, 2304); The tests should pass after that. I need to figure out which ifdefs to put this under, so I might not be able to land the fix until Monday. On Fri, May 31, 2013 at 8:48 PM, Greg

Those changes shouldn't affect ARM at all, since everything is under #if defined(__i386__) || defined(__x86_64__). What version of glibc are you building with on x86? On Fri, May 31, 2013 at 7:16 PM, Greg Fitzgerald <garious at gmail.com> wrote: > The failures happen on x86 Linux, Ubuntu Lucid. On ARM Android, my > example code segfaults, whereas before it worked. I

> What version of glibc are you building with on x86? 2.11.1 for 64-bit x86 linux $ ldd --version ldd (Ubuntu EGLIBC 2.11.1-0ubuntu7.8) 2.11.1 On Fri, May 31, 2013 at 8:24 AM, Sergey Matveev <earthdok at google.com> wrote: > Those changes shouldn't affect ARM at all, since everything is under #if > defined(__i386__) || defined(__x86_64__). > > What version of glibc are

It's been 4 days since Github's llvm mirror has been updated. Anybody know who maintains this? https://github.com/llvm-mirror Thanks, Greg

> const uptr kThreadDescriptorSize = FIRST_32_SECOND_64(1168, 2304); Yes, that change causes all tests to pass. > I need to figure out which ifdefs to put this under, so I might not be able to land the fix until Monday. Okay, no worries, thanks for doing this. I've moved over to release_33 for the short-term. With the one change mentioned earlier (#include <stdint.h>), asan

> Not that I know of, and I always build compiler-rt natively, together with Clang and LLVM. :( Do you mean "natively on ARM" or "natively on X86 with the ARM target enabled"? If the former, do you have a script to merge the target libs into the host's install directory? If the latter, that'll only build the X86 versions of the compiler-rt libraries. > Would

> Cool, can you use clang 3.3 then? :) I can, but digging deeper I see that the compiler-rt sanitizer tests depend on just-built-clang for its object instrumentation. The next time the instrumentation changes, I'd expect those tests to break. If the lit tests that require -fsanitize were moved to the clang repo, then I think it'd be safe to build compiler-rt with clang 3.3 or gcc

Okay, dropping gcc 4.4.3 makes sense. How do you feel about using clang 3.2 (and the upcoming 3.3) instead of tip-of-the-trunk clang? It looks like everything works great, but that you just need to make those UB tests 'unsupported' since they fail with "libclang_rt.ubsan was built without __int128 support". Thanks, Greg On Mon, May 27, 2013 at 12:36 AM, Alexey Samsonov

> What is the exact line you use to configure build tree, and the output you see? cmake ../.. \ -G Ninja \ -DCMAKE_INSTALL_PREFIX=ship \ -DCMAKE_BUILD_TYPE=Release \ -DLLVM_ENABLE_ASSERTIONS=ON \ -DLLVM_TARGETS_TO_BUILD=ARM \ -DLLVM_DEFAULT_TARGET_TRIPLE=arm-none-linux-gnueabi \ -DLLVM_TARGET_ARCH=arm-none-linux-gnueabi \ -DLLVM_LIT_ARGS=-v ninja check-all

> it assumes that compiler-rt is checked out to > llvm/projects/compiler-rt. Apparently, this is a problem. I have a patch for this ready. I'll send it to you and llvm-commits. Most of the tests pass with "make check-all" but the recently-added lsan tests are all failing. Do those fail for you as well? If so, can we XFAIL them for now and try to keep the "make

You can email the list, what url says that? -Chris > On Jan 20, 2015, at 9:31 AM, Greg Fitzgerald <garious at gmail.com> wrote: > > Ping > >> On Mon, Jan 19, 2015 at 9:52 AM, Greg Fitzgerald <garious at gmail.com> wrote: >> I am not a member of the llvm-admin email list and on the description >> of that page it says, "DO NOT MAIL THIS LIST!"

For me, UBsan fails with clang 3.2 and passes with clang 3.3. Using a fixed version allows you to build all clang/llvm/compiler-rt with one compiler. It simplifies the build process quite a bit. Also better for isolating regressions in compiler-rt, especially if you use git-bisect. Greg On May 29, 2013, at 12:30 AM, Alexey Samsonov <samsonov at google.com> wrote: > UBsan tests work

On 26 November 2013 18:46, Greg Fitzgerald <garious at gmail.com> wrote: > Here's where we left off: > http://lists.cs.uiuc.edu/pipermail/llvmdev/2013-February/059099.html Back then you mentioned a Chromium build, are you still running that routinely with -integrated-as? That would give a reasonable reassurance about general-usage, if not weird features. Cheers. Tim.

I blame this line in lsan/lit_tests/lit.cfg: # Setup attributes common for all compiler-rt projects. compiler_rt_lit_cfg = os.path.join(llvm_src_root, "projects", "compiler-rt", "lib", "lit.common.cfg") On Fri, May 24, 2013 at 2:53 PM, Alexey Samsonov <samsonov at google.com>wrote: > > On Fri, May 24,

> Android runtime is special, we build it in a separate build tree configured with > -DCMAKE_TOOLCHAIN_FILE=$LLVM_CHECKOUT/cmake/platforms/Android.cmake This worked great, thanks! Would you mind tweaking Android.cmake so that I can override the location of the C compiler? The current version forces me to use the just-built-clang and that the new build directory be in a sibling directory.

I'm hoping to revive the LLD standalone CMake build. I'm new to this build but it looks like it borrowed code from an old version of compiler-rt, which I did some work on last year. Like compiler-rt, I'd like to get the LLD build up running with only CMAKE_PREFIX_PATH instead of defining custom variables like LLD_PATH_TO_LLVM_BUILD and LLD_PATH_TO_LLVM_SOURCE. Any objection to that?

> Why does fork-exec for llvm-symbolizer work, but simple exec(self) does not? Because the llvm-symbolizer the runtime finds is built for the host architecture. This is weird, yes, but once we integrate the symbolizer, it goes away. > Could we write a ulimit-like utility that would do setrlimit and then > exec the specified binary > %run %ulimit -s 8192 %t? I like that idea. How

On Sat, May 25, 2013 at 4:12 AM, Greg Fitzgerald <garious at gmail.com> wrote: > When I build compiler-rt with clang 3.2, all lsan tests pass. The only > failing tests I see are in ubsan: > > Failing Tests (6): > UndefinedBehaviorSanitizer :: Float/cast-overflow.cpp > UndefinedBehaviorSanitizer :: Integer/add-overflow.cpp > UndefinedBehaviorSanitizer ::