https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116662
Jeffrey A. Law <law at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |jason at gcc dot gnu.org,
| |redi at gcc dot gnu.org
--- Comment #11 from Jeffrey A. Law <law at gcc dot gnu.org> ---
So I was about to submit a patch to fix this, but after reviewing Jason's
comments & cfe-dev, I'm a bit concerned about making this change.
Essentially the concern is this value has ABI implications; so I'm quite
hesitant to change it.
I think Jason's position is that while it has ABI implications, that anyone
using it in a way that exposes those ABI implications is using the feature in
an unsafe way and that we shouldn't even try to present a stable & correct
value as it inherently has to change based on underlying u-architectures.
I can buy that argument. But before I make such a change for RISC-V I'd like
to hear from Jon and Jason on that point.
