On Thu, 2022-09-29 at 19:35 +0200, Paolo Bonzini wrote: > On 9/29/22 18:39, Christoph Hellwig wrote: > > On Thu, Sep 29, 2022 at 10:37:22AM -0600, Keith Busch wrote: > > > > I am aware, and I've submitted the fix to qemu here: > > > > > > > > https://lists.nongnu.org/archive/html/qemu-block/2022-09/msg00398.html > > > > > > I don't think so. Memory alignment and length granularity are two completely > > > different concepts. If anything, the kernel's ABI had been that the length > > > requirement was also required for the memory alignment, not the other way > > > around. That usage will continue working with this kernel patch. Yes, this is how I also understand it - for example for O_DIRECT on a file which resides on 4K block device, you have to use page aligned buffers. But here after the patch, 512 aligned buffer starts working as well - If I understand you correctly the ABI didn't guarantee that such usage would fail, but rather that it might fail. > > > > Well, Linus does treat anything that breaks significant userspace > > as a regression. Qemu certainly is significant, but that might depend > > on bit how common configurations hitting this issue are. > > Seeing the QEMU patch, I agree that it's a QEMU bug though. I'm > surprised it has ever worked. > > It requires 4K sectors in the host but not in the guest, and can be > worked around (if not migrating) by disabling O_DIRECT. I think it's > not that awful, but we probably should do some extra releases of QEMU > stable branches. > > Paolo > I must admit I am out of the loop on the exact requirements of the O_DIRECT. If I understand that correctly, after the patch in question, qemu is able to use just 512 bytes aligned buffer to read a single 4K block from the disk, which supposed to fail but wasn't guarnteed to fail. Later qemu it submits iovec which also reads a 4K block but in two parts, and if I understand that correctly, each part (iov) is considered to be a separate IO operation, and thus each has to be in my case 4K in size, and its memory buffer *should* also be 4K aligned. (but it can work with smaller alignement as well). Assuming that I understand all of this correctly, I agree with Paolo that this is qemu bug, but I do fear that it can cause quite some problems for users, especially for users that use outdated qemu version. It might be too much to ask, but maybe add a Kconfig option to keep legacy behavier for those that need it? Best regards, Maxim Levitsky
