On 10/8/21 03:44, Siddhesh Poyarekar wrote:
(from about 4% to 70% in bash), but that could well be due to the _chk
I should also clarify that this is for memcpy. For all fortifiable functions, the coverage percentage went from 30.81% to 84.5% for bash. Below is the full table. Please note that this is only based on symbols emitted in the end as I didn't want to rebuild the _FORTIFIED_SOURCE=2 binaries, so it does not take into account the fact that _chk could get folded to regular calls if we know at compile time that it's safe to do so.
No more posting patches at 4am; it only leads to more clarification follow-ups :/
Func F(2) U(2) Cov(2) F(3) U(3) Cov(3) ---------------------------------------------------------------- asprintf 1 0 100.00% 1 0 100.00% *confstr 0 2 0.00% 1 1 50.00% fdelt 10 0 100.00% 10 0 100.00% *fgets 0 1 0.00% 1 0 100.00% fprintf 82 0 100.00% 82 0 100.00% getcwd 0 3 0.00% 0 3 0.00% *getgroups 0 1 0.00% 1 0 100.00% gethostname 0 1 0.00% 0 1 0.00% longjmp 23 0 100.00% 23 0 100.00% mbsnrtowcs 0 2 0.00% 0 2 0.00% *mbsrtowcs 0 1 0.00% 1 0 100.00% *mbstowcs 0 10 0.00% 5 5 50.00% *memcpy 7 170 3.95% 116 40 74.36% *memmove 4 21 16.00% 12 17 41.38% *memset 0 24 0.00% 3 21 12.50% printf 150 0 100.00% 150 0 100.00% *read 0 19 0.00% 8 11 42.11% *readlink 0 3 0.00% 0 3 0.00% snprintf 11 0 100.00% 11 0 100.00% sprintf 28 0 100.00% 28 0 100.00% *strcat 0 5 0.00% 8 0 100.00% *strcpy 6 440 1.35% 413 36 91.98% *strncpy 2 52 3.70% 33 19 63.46% vfprintf 14 0 100.00% 14 0 100.00% vsnprintf 4 0 100.00% 4 0 100.00% wcrtomb 0 7 0.00% 0 7 0.00% *wcsrtombs 0 3 0.00% 2 1 66.67% wctomb 0 3 0.00% 0 3 0.00% ----------------------------------------------------------------- Total 342 768 30.81% 927 170 84.50%
