I found these while writing a test case for function summaries.
The coverage.paths field was updated with the wrong count, which would
for function summaries lead obviously wrong output like:
'Prime paths covered:0.00% of -1 (2 of 1 suppressed)'
The numbers in the gcov report itself were fine.
The missing else cause 'No conditions' to be output when it shouldn't:
Calls executed:50.00% of 2
Condition outcomes covered:50.00% of 4
No conditions
I apparently never tried these exact combinations with function
summaries, which are occasionally useful, but I mostly used the
detailed reports.
Bootstrapped and tested on x86_64-linux-gnu, pushed as obvious.
gcc/ChangeLog:
* gcov.cc (add_path_counts): Read the right field paths.size.
(function_summary): Add missing else.
---
gcc/gcov.cc | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/gcc/gcov.cc b/gcc/gcov.cc
index ceed689c1c9..bab0f893f03 100644
--- a/gcc/gcov.cc
+++ b/gcc/gcov.cc
@@ -3315,7 +3315,7 @@ add_condition_counts (coverage_info *coverage, const
block_info *block)
static void
add_path_counts (coverage_info &coverage, const function_info &fn)
{
- coverage.paths += fn.paths.path_count ();
+ coverage.paths += fn.paths.paths.size ();
coverage.paths_covered += fn.paths.covered_paths ();
coverage.paths_suppressed += fn.paths.suppressed_count ();
}
@@ -3441,7 +3441,7 @@ function_summary (const coverage_info *coverage)
format_gcov (coverage->conditions_covered,
coverage->conditions, 2),
coverage->conditions);
- if (coverage->conditions && coverage->conditions_suppressed > 0)
+ else if (coverage->conditions && coverage->conditions_suppressed > 0)
fnotice (stdout, "Condition outcomes covered:%s of %d"
" (%d of %d suppressed)\n",
format_gcov (coverage->conditions_covered,
--
2.47.3
