On Sat, 19 Jun 2021 at 18:28, Richard Henderson <[email protected]> wrote: > > Short story is that the first two patches resolve the observed > problem, by completely bypassing quite a lot of code in memory.c. > > Longer story is that we should either use that code in memory.c, > or we should bypass it to an even lower level, so that we don't > have multiple locations doing the partial-read assembly thing.
I haven't read the patchset yet, but my initial reaction is that we ought to be handling this stuff in memory.c, because that code is shared across all accelerators -- we would want the same behaviour for accesses to a device that doesn't itself handle a misalignment or a small access, whether we are using TCG or KVM or HVF or whatever... thanks -- PMM
