https://gcc.gnu.org/bugzilla/show_bug.cgi?id=124175
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Keywords| |alias, missed-optimization
Status|UNCONFIRMED |NEW
Blocks| |49774
Ever confirmed|0 |1
Last reconfirmed| |2026-02-20
--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
The __asm__("nop") inhibits PRE. The restrict qualification does not help
disambiguating against subroutine1() since that's not implemented in GCC.
Otherwise FSM threading would perform the optimization.
So, IMO, a dup of the restrict vs. function call bug which I think exists
somewhere ('p' is not "directly" passed to subroutine1()).
Referenced Bugs:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=49774
[Bug 49774] [meta-bug] restrict qualification aliasing issues