[RFC PATCH] x86: Ignore rounding for vcvt[, u]si2sd under r32 and vcvt[,u]dq2pd instead of reporting bad

[Jiang, Haochen]

They will not be #UD after my confirmation with HW.

So it is safe to go ahead.

Thx,
Haochen

From: Christian Ludloff <ludloff@gmail.com>
Sent: Tuesday, January 14, 2025 5:06 PM

The toolchains are priority #2 here.

The various chips with AVX10.2 are priority #1, i.e. did they get the {er.ignored} right for the two CVTs I listed, or will they unexpectedly #UD upon U=0...

--
C.

On Tue, Jan 14, 2025, 09:35 Haochen Jiang <haochen.jiang@intel.com<mailto:haochen.jiang@intel.com>> wrote:
Hi all,

I just did a quick try on the rounding ignore in disassembler for
vcvt[,u]si2sd under r32 and vcvt[,u]dq2pd. It is doable in a small
patch following, with some change in OP_Rounding. So I called it
an RFC patch to see if we should do so.

For assembler part, it is much easier to accept that inline asm. I
will do that in parallel with disassembler RFC patch review.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://sourceware.org/pipermail/binutils/attachments/20250116/98c19723/attachment-0001.htm>