https://gcc.gnu.org/bugzilla/show_bug.cgi?id=125417
pietro <pietro at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
URL| |https://inbox.sourceware.or
| |g/gcc-patches/2026052520225
| |4.13130-2-pietro@sociotechn
| |ical.xyz/T/#u
Status|UNCONFIRMED |WAITING
Ever confirmed|0 |1
Last reconfirmed| |2026-05-25
Assignee|unassigned at gcc dot gnu.org |pietro at gcc dot
gnu.org
--- Comment #1 from pietro <pietro at gcc dot gnu.org> ---
I was able to reproduce CFarm 215 (i386-solaris), on trunk:
=== libffi Summary for unix/-m64 ===
# of expected passes 742
# of unexpected failures 712
=== libffi Summary for unix/-m32 ===
# of expected passes 4068
Reverting the change to libffi.exp to use "-I${libffi_include}
-I${srcdir}/../include -I${libffi_include}/..", I get:
=== libffi Summary for unix/-m64 ===
# of expected passes 1662
=== libffi Summary for unix/-m32 ===
# of expected passes 4068
The following has equal results but is more readable IMHO "-I${libffi_include}
-I${blddirffi} -I${srcdir}/../include"
Patch sent to gcc-patches[1] and forwarded upstream[2].
[1]
https://inbox.sourceware.org/gcc-patches/[email protected]/T/#u
[2] https://github.com/libffi/libffi/pull/965
