re-implement some of the FCVTZS/FCVTMS/FCVTPS/FCVTNS

[mr-c]

Recent Qemu confirms the different behavior that was seen on mac M1 and a Pixel 4a 5g phone versus our previous implementation.

To fix this:

1. Revert 339ffe4
2. Adjust the test vectors to match the hardware behavior (or that of a recent Qemu version)
3. Adjust the software implementations of those intrinsics to match the new test vectors

Note that Arm defines Round towards Zero (RZ) mode as

Means that results are rounded to the nearest representable number that is no greater in magnitude than the unrounded result.

[mr-c]

@yyctw Is this something you have time to do?

[vt-alt]

Is 0.9.0 planned anytime soon?

[mr-c]

Is 0.9.0 planned anytime soon?

0.8.4 is under way: #1228

Are you able to help fix the FCVTZS/FCVTMS/FCVTPS/FCVTNS implementations? Are there specific intrinsics that you need?

[vt-alt]

Thanks for clarification! Ah, excuse me, I'm no intended to fix the implementations. Reason for the question here, is we need some gcc14 fixes and I thought if we should backport them or just wait for the next release. And 0.9.0 is the only release I found in Milestones tab (and this is the only issue looked requiring coding the other one is for adding documentation).

[mr-c]

Thanks for clarification! Ah, excuse me, I'm no intended to fix the implementations. Reason for the question here, is we need some gcc14 fixes and I thought if we should backport them or just wait for the next release. And 0.9.0 is the only release I found in Milestones tab (and this is the only issue looked requiring coding the other one is for adding documentation).

:-) What gcc-14 fixes do you need? Who is 'we'?

[vt-alt]

ALT Linux distribution. Commits fixing these FTBFS issues on i586: https://git.altlinux.org/beehive/logs/Sisyphus-i586/latest/error/simde-0.8.2-alt1 I see a fellow maintainer is picking this commit #1251