On Mon, 16 Nov 2026 at 17:05, Philippe Mathieu-Daudé <[email protected]> wrote: > On 16/11/26 17:53, Peter Maydell wrote: > > For aarch64 in particular I would prefer to deprecate-and-drop > > big-endian support (probably starting with the deprecation for > > 12.0, rather than pushing it into 11.2), since the kernel just > > did that in 7.4.
> Tangent to this, it makes sense to deprecate quickly enough on > QEMU side when a host-related project we use deprecates / removes > it; when this is guest related I don't see the urge, in particular > when the maintenance is almost zero cost. For the particular > aarch64_be I agree what we lacks is coverage testing; but for > other features I'd prefer we keep them as long as there is interest > within QEMU users, independent of external projects change like > the Linux kernel. I'm not in a huge hurry about deprecating aarch64_be (as you say it's not a huge cost to us to carry it, though it is a small extra amount of compilation time in CI). It's more that I would favour dropping it over doing any future work or refactoring on it. In terms of QEMU user interest, I really don't think we actually have much of it (though of course it's difficult to tell since users don't tell us about stuff that works for them). About the only thing I can remember is https://gitlab.com/qemu-project/qemu/-/work_items/3958 which is a report from one user that it doesn't work for them at all! (My suspicion is that this is because their userspace is built with some custom local kernel / libc changes which aren't upstream.) As Richard says, BE AArch64 on Linux has pretty much never been something with much serious interest, and the config has been marked BROKEN in Linux for some time prior to removal. I don't think there is a big stock of BE userspace binaries out there that people want to try to run. thanks -- PMM
