On 12/29/2017 07:01 PM, Paul Eggert wrote:
Thanks, but -Wunused-parameter issues so many false alarms that I think I'd
rather just ask people to compile with -Wno-unused-parameter. That's what
coreutils does, and Emacs, and so forth.
Okay, thanks.
Have a nice day,
Berny
Thanks, but -Wunused-parameter issues so many false alarms that I think I'd
rather just ask people to compile with -Wno-unused-parameter. That's what
coreutils does, and Emacs, and so forth.
GCC-7.2.1 complains:
../gl/lib/stat-time.h:215:47: error: unused parameter 'st' \
[-Werror=unused-parameter]
stat_time_normalize (int result, struct stat *st)
^~
* lib/stat-time.h (stat_time_normalize): Silence compiler regarding
the unused p
Bruno Haible, on ven. 29 déc. 2017 17:35:47 +0100, wrote:
> Lesson learned: Hurd is a cross-compilation target, not only Linux (where it
> is well-known, because of systems like busybox or buildroot).
Yes, in Debian we cross-build between Linux & Hurd :)
Samuel
Hello Samuel,
> While cross-compilling packages using gnulib, we are getting different
> results on GNU/Hurd, because the test results are not preseeded in the
> .m4 files. The attached patch updates them.
Thanks! Applied (with additional whilespace changes in calloc.m4, mbr*.m4,
printf.m4, unget
Hello,
While cross-compilling packages using gnulib, we are getting different
results on GNU/Hurd, because the test results are not preseeded in the
.m4 files. The attached patch updates them.
Samuel
m4/calloc.m4: Add GNU/Hurd guess.
m4/cbrtl.m4: Likewise.
m4/ceil.m4: Likewise.
m4/ceilf.m4: Like