Date: Mon, 8 Jul 2024 17:24:36 +0000 (UTC)
On Tue, 2 Jul 2024, Richard Smith via Liaison wrote:
> GCC and Clang folks asked the psABI maintainers
> <https://groups.google.com/g/ia32-abi/c/Tlu6Hs-ohPY> to do something about
> it, and it seems like they didn't. So we're kind of stuck. No-one wants to
> break their compiler's ABI without an official decision from the owners of
> the ABI, and we don't have an official decision. Maybe someone could try to
> resurrect that thread.
I don't think that 32-bit ABI was maintained very much even before HJ Lu
retired, and it may be maintained even less since then. (HJ said he'd
document whatever was implemented for the _BitInt ABI for 32-bit x86 in
GCC, but that hasn't gone into the document either.)
> GCC and Clang folks asked the psABI maintainers
> <https://groups.google.com/g/ia32-abi/c/Tlu6Hs-ohPY> to do something about
> it, and it seems like they didn't. So we're kind of stuck. No-one wants to
> break their compiler's ABI without an official decision from the owners of
> the ABI, and we don't have an official decision. Maybe someone could try to
> resurrect that thread.
I don't think that 32-bit ABI was maintained very much even before HJ Lu
retired, and it may be maintained even less since then. (HJ said he'd
document whatever was implemented for the _BitInt ABI for 32-bit x86 in
GCC, but that hasn't gone into the document either.)
-- Joseph S. Myers josmyers_at_[hidden]
Received on 2024-07-08 17:24:42