On Sun, 20 Sep 2026, Matt Turner wrote: > Extend the movmisalign mode iterator to HImode, SImode and DImode and > derive the access size from the mode rather than hardcoding 8. The > memcpy is then folded at gimple time into a load or a store of a > type with reduced alignment, which expands to the unaligned sequence > directly, with no stack slot, no frame, and no by-pieces byte copy. > -msafe-partial keeps emitting byte stores for the store case.
Doesn't the new default violate the Linux kernel's requirement for memory writes not to violate the atomicity of their adjacent locations, the lack of fulfilment of which was the motivation to drop non-BWX CPU support? Maciej
