https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107943
David Malcolm <dmalcolm at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Summary|[13/14/15/16 Regression] |[13/14/15 Regression] gcc
|gcc -fanalyzer hangs in |-fanalyzer hangs in openssl
|openssl curve25519.c since |curve25519.c since
|r11-3840-gaf66094d03779377 |r11-3840-gaf66094d03779377
--- Comment #12 from David Malcolm <dmalcolm at gcc dot gnu.org> ---
Trying again with trunk.
On a release build of r16-7650-g64545dee159b10, with -O3 -fanalyzer:
* attachment 53998:
* takes 8-9s without -fanalyzer
* takes 54s with -fanalyzer
* attachment 54000:
* takes 0.6s without -fanalyzer
* takes 3.6s with -fanalyzer
This is much better than the timings reported earlier, so I'm going to treat
this as fixed in at least GCC 16 (and perhaps earlier).
Note that I tried to optimize the binding_key stuff from comment #7 in
r16-7053-gc28dd8467b1e35, so perhaps this helped.
