On 11/5/25 23:38, Jørgen Kvalsvik wrote:
Hi,These patches are a mix of features, bugfixes, and quality of life improvements. * fixes a wrong line count when multiple blocks touch the same line * -fpath-coverage and -fcondition-coverage implies -ftest-coverage (.gcno, a prerequisite for running gcov) * gcov-dump prints paths and the correct offset of the conditions tag. I noticed a regression in the gcov test suite which I bisected to this commit. The breakage is from discriminators being added to the locations, which caused unintended misses in a hash set in profile.cc. commit 385d9937f0e23cbf9c62f0b2553a33ff70e56ecf Author: Jan Hubicka <[email protected]> Date: Fri Jul 11 13:01:13 2025 +0200 Rewrite assign_discriminators The prime paths test change is quite harmless, and the occasional update to tests like this should be expected. The pr-* tests do catch the real problem. Bootstrapped and regtested on x86_64-linux-gnu Jørgen Kvalsvik (5): gcov: Only consider pure locations in line mapping Update expected number of prime paths in test case gcov-dump: Print start-position for conditions gcc: -f*-coverage implies -ftest-coverage gcov-dump: Print PATHS tag gcc/gcc.cc | 1 + gcc/gcov-dump.cc | 14 +++++++++++++- gcc/profile.cc | 6 +++--- gcc/testsuite/g++.dg/gcov/gcov-18.C | 2 +- gcc/testsuite/g++.dg/gcov/gcov-22.C | 2 +- gcc/testsuite/gcc.misc-tests/gcov-19.c | 2 +- gcc/testsuite/gcc.misc-tests/gcov-29.c | 4 ++-- 7 files changed, 22 insertions(+), 9 deletions(-)
Ping.
