Jeff Law <jeffreya...@gmail.com> writes: > On 2/7/25 5:51 AM, Oleg Endo wrote: >>> Hi, >>> >>> Can the issue be resolved in a target independent manner as suggested below? >>> Or is it better to deal with this in the target code? > That seems like a pretty heavy hammer though. For that reason alone I > think this is going to need some discussion and I believe the folks most > needed for that discussion are focused on release related issues.
Yeah, agreed. Prefetches ought to be restricted to performance-critical code, which is also the kind of code that would suffer from having extra blockage instructions. We do have a prefetch rtl code, so it should be possible for the scheduler to recognise prefetches and handle them in a more sensible way. That would be more complex though... Thanks, Richard
