Re: [PATCH] Remove ia64-linux-gnu

[Sam James <sam@gentoo.org>]

Frank Scheiner <frank.scheiner@web.de> writes:

> Dear Adhemerval, all,
>
> On 04.11.23 20:02, Adhemerval Zanella wrote:
>> Linux 6.7 removed ia64 from the official tree [1], following the general
>> principle that a glibc port needs upstream support for the architecture
>> in all the components it depends on (binutils, GCC, and the Linux
>> kernel).
>
>> Apart from the removal of sysdeps/ia64 and
>> sysdeps/unix/sysv/linux/ia64, there are updates to various comments
>> referencing ia64 for which removal of those references seemed
>> appropriate. The configuration is removed from README and
>> build-many-glibcs.py.
>
>> The CONTRIBUTED-BY, elf/elf.h, manual/contrib.texi
>> (the porting mention), *.po files, config.guess, and longlong.h are not
>> changed.
>
>> For Linux it allows cleanup some clone2 support on multiple
>> files.
>> The following bug can be closed as WONTFIX: BZ 22634 [2], BZ
>> 14250 [3], BZ 21634 [4], BZ 10163 [5], BZ 16401 [6], and BZ 11585 [7].
>
>> [1] >
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=43ff221426d33db909f7159fdf620c3b052e2d1c
>> [2] https://sourceware.org/bugzilla/show_bug.cgi?id=22634
>> [3] https://sourceware.org/bugzilla/show_bug.cgi?id=14250
>> [4] https://sourceware.org/bugzilla/show_bug.cgi?id=21634
>> [5] https://sourceware.org/bugzilla/show_bug.cgi?id=10163
>> [6] https://sourceware.org/bugzilla/show_bug.cgi?id=16401
>> [7] https://sourceware.org/bugzilla/show_bug.cgi?id=11585
>
> The speed this happens really surprises me and I hope there is no need
> to rush with this removal, I would like to comment on your points in [8]
> but need some more time for further research into this matter to get a
> better picture of how much work ia64 is in the glibc.
>

The compromise that glaubitz reached with the kernel folks was to remove
it after 6.6 so it was in the next LTS, so I feel like removing it just
now is indeed a bit premature.

Also, I worry that this might make things tricky to backport later for
older glibc branches. Regardless of what happens with ia64 being
maintained out of tree or not, we don't intend on removing it from Gentoo
immediately - possibly (although I don't make a promise here) as late as
when the last LTS kernel supporting it ages out. I'd prefer to not have
to pin ia64 to an older glibc and keep testing that branch just for one
arch.

> [8]: https://sourceware.org/pipermail/libc-alpha/2023-November/152497.html
>
> Only so much now, we would like to maintain Linux/ia64 (there is the
> possibility that this returns to the kernel in a year's time, see [9])
> and we also - as you already found out - started to work on the math
> issues for ia64 in the glibc. We started this roughly just a month ago
> and together with Aurelien's patch could make about 80 tests pass (30
> FAIL => PASS, 50 UNSUPPORTED => PASS). That is not a very big thing, but
> compared to what happened in this regard in the last ten years, it's not
> bad I'd say, for a month's time.
>
> [9]:
> https://lore.kernel.org/linux-ia64/CAHk-=whFLZ67ffzt1juryCYcYz6eL_XjQF8WucDzwUR5H65+rA@mail.gmail.com/
>

thanks,
sam