[PATCH] D74953: [profile] Don't dump counters when forking and don't reset when calling exec** functions

marco-c accepted this revision. marco-c added a comment. This revision is now accepted and ready to land. Could you test the last iteration of the patch on Mozilla's CI (with the workaround for the mismatch in LLVM version used by Rust)? Comment at: llvm/lib/Transforms/Instrum

[PATCH] D74953: [profile] Don't dump counters when forking and don't reset when calling exec** functions

marco-c added a comment. Also, as we discussed, https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93623#c5 regarding exec. Comment at: compiler-rt/lib/profile/GCDAProfiling.c:665 + gcov_lock(); + // Avoid a concurrent modification of the lists during the fork + pid = fork(); ---

[PATCH] D54195: Fix linker option for -fprofile-arcs -ftest-coverage

marco-c accepted this revision. marco-c added a comment. This revision is now accepted and ready to land. Thanks! Comment at: test/Driver/clang_f_opts.c:89 +// RUN: %clang -### -fprofile-arcs -ftest-coverage %s 2>&1 | FileCheck -check-prefix=CHECK-u %s +// CHECK-u-NOT: "-u{{.*

[PATCH] D52034: [Clang] Add options -Xclang -coverage-filter and -Xclang -coverage-exclude to filter the files to instrument with gcov

marco-c added inline comments. Comment at: include/clang/Driver/CC1Options.td:236 +def coverage_exclude_EQ : Joined<["-"], "coverage-exclude=">, + Alias; def coverage_exit_block_before_body : Flag<["-"], "coverage-exit-block-before-body">, calixte wrote: > vsk

[PATCH] D45454: Add llvm_gcov_flush to be called outside a shared library

marco-c added a comment. OK! Sounds good to me to keep it hidden until https://reviews.llvm.org/D48538 is done (I'm going to try to finish it soon). Il 29/06/2018 19:34, David Li via Phabricator ha scritto: > davidxl added a comment. > > With the current gcov_flush implementation in LLVM, makin

[PATCH] D45454: Add llvm_gcov_flush to be called outside a shared library

marco-c added a comment. Why keeping a __gcov_flush which is incompatible with GCC and previous versions of LLVM and adding a __llvm_gcov_flush which is compatible? I feel the other way around (or even just having an unhidden "__gcov_flush" and not introducing "__llvm_gcov_flush") would be mor

[PATCH] D45454: Make __gcov_flush visible outside a shared library

marco-c added a comment. Yes, the behavior changed very recently, I would be surprised if somebody came to depend on it. It's more likely that some clients are depending on the old behavior. - Marco. Il 26/06/2018 22:43, Stephen Hines via Phabricator ha scritto: > srhines added a comment. > >

[PATCH] D45454: Make __gcov_flush visible outside a shared library

marco-c added a comment. In https://reviews.llvm.org/D45454#1070884, @belleyb wrote: > @chh I had a chance to try out your proposed changes. It's not causing us any > trouble. In fact, `__gcov_flush()` is not even used at all (at least in LLVM > 5.0.1).. I can recompile llvm, compiler_rt and cl

[PATCH] D38124: Hide some symbols to avoid a crash on shutdown when using code coverage

marco-c added a comment. The change was already positively reviewed with the request for a test, which has been written since then. Repository: rCRT Compiler Runtime https://reviews.llvm.org/D38124 ___ cfe-commits mailing list cfe-commits@lists.

[PATCH] D38221: Add CoreOption flag to "-coverage" option to make it available for clang-cl

marco-c created this revision. The -coverage option is not a CoreOption, so it is not available to clang-cl. This patch adds the CoreOption flag to "-coverage" to allow it to be used with clang-cl. https://reviews.llvm.org/D38221 Files: include/clang/Driver/Options.td test/Driver/coverage.