rj-jesus wrote:

> official criteria is here: 
> https://llvm.org/docs/HowToReleaseLLVM.html#release-patch-rules
> 
> it's not obviously very safe to me and I can't speak as to whether this is a 
> bug fix, so I'll defer to @davemgreen for that.
> 
> But generally speaking from a release maintainer perspective the less time 
> patches have had on main the more risk they carry, so for backports it's 
> worth asking if it's really necessary for this release. We are still early in 
> the release cycle with rc2 next Friday, so there is opportunity to fix 
> issues, but just worth bearing in mind.

Yeah, I'm okay if this is considered too high risk as I appreciate that this 
change came a bit late and is not obviously very safe. The backport is 
addressing reports of missed attempts to use the RNG intrinsics due to the 
feature being available at compile time when it shouldn't be for the target 
specified, so it would be good if the issue was fixed, but I'll trust your and 
@davemgreen's judgment. :) In any case, we'd still appreciate it if #177139 
could be backported as it should be a smaller and more self-contained change.

https://github.com/llvm/llvm-project/pull/177171
_______________________________________________
llvm-branch-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

Reply via email to