similar to: [PATCH v2 2/2] vhost: relax used address alignment

commit 5d9a07b0de512b77bf28d2401e5fe3351f00a240 ("vhost: relax used address alignment") fixed the alignment for the used virtual address, but not for the physical address used for logging. That's a mistake: alignment should clearly be the same for virtual and physical addresses, Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- drivers/vhost/vhost.c | 2 +- 1 file

commit 5d9a07b0de512b77bf28d2401e5fe3351f00a240 ("vhost: relax used address alignment") fixed the alignment for the used virtual address, but not for the physical address used for logging. That's a mistake: alignment should clearly be the same for virtual and physical addresses, Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- drivers/vhost/vhost.c | 2 +- 1 file

virtio 1.0 only requires used address to be 4 byte aligned, vhost required 8 bytes (size of vring_used_elem). Fix up vhost to match that. Additionally, while vhost correctly requires 8 byte alignment for log, it's unconnected to used ring: it's a consequence that log has u64 entries. Tweak code to make that clearer. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> ---

virtio 1.0 only requires used address to be 4 byte aligned, vhost required 8 bytes (size of vring_used_elem). Fix up vhost to match that. Additionally, while vhost correctly requires 8 byte alignment for log, it's unconnected to used ring: it's a consequence that log has u64 entries. Tweak code to make that clearer. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> ---

On 2020/4/22 ??10:58, Michael S. Tsirkin wrote: > The ring element addresses are passed between components with different > alignments assumptions. Thus, if guest/userspace selects a pointer and > host then gets and dereferences it, we might need to decrease the > compiler-selected alignment to prevent compiler on the host from > assuming pointer is aligned. > > This actually

On 2020/4/21 ??4:46, Michael S. Tsirkin wrote: > The ring element addresses are passed between components with different > alignments assumptions. Thus, if guest/userspace selects a pointer and > host then gets and dereferences it, we might need to decrease the > compiler-selected alignment to prevent compiler on the host from > assuming pointer is aligned. > > This actually

With PUBLISH_USED_IDX, guest tells us which used entries it has consumed. This can be used to reduce the number of interrupts: after we write a used entry, if the guest has not yet consumed the previous entry, or if the guest has already consumed the new entry, we do not need to interrupt. This imporves bandwidth by 30% under some workflows. Signed-off-by: Michael S. Tsirkin <mst at

With PUBLISH_USED_IDX, guest tells us which used entries it has consumed. This can be used to reduce the number of interrupts: after we write a used entry, if the guest has not yet consumed the previous entry, or if the guest has already consumed the new entry, we do not need to interrupt. This imporves bandwidth by 30% under some workflows. Signed-off-by: Michael S. Tsirkin <mst at

The ring element addresses are passed between components with different alignments assumptions. Thus, if guest/userspace selects a pointer and host then gets and dereferences it, we might need to decrease the compiler-selected alignment to prevent compiler on the host from assuming pointer is aligned. This actually triggers on ARM with -mabi=apcs-gnu - which is a deprecated configuration, but it

The ring element addresses are passed between components with different alignments assumptions. Thus, if guest/userspace selects a pointer and host then gets and dereferences it, we might need to decrease the compiler-selected alignment to prevent compiler on the host from assuming pointer is aligned. This actually triggers on ARM with -mabi=apcs-gnu - which is a deprecated configuration, but it

The ring element addresses are passed between components with different alignments assumptions. Thus, if guest/userspace selects a pointer and host then gets and dereferences it, we might need to decrease the compiler-selected alignment to prevent compiler on the host from assuming pointer is aligned. This actually triggers on ARM with -mabi=apcs-gnu - which is a deprecated configuration, but it

The ring element addresses are passed between components with different alignments assumptions. Thus, if guest/userspace selects a pointer and host then gets and dereferences it, we might need to decrease the compiler-selected alignment to prevent compiler on the host from assuming pointer is aligned. This actually triggers on ARM with -mabi=apcs-gnu - which is a deprecated configuration, but it

With PUBLISH_USED_IDX, guest tells us which used entries it has consumed. This can be used to reduce the number of interrupts: after we write a used entry, if the guest has not yet consumed the previous entry, or if the guest has already consumed the new entry, we do not need to interrupt. This imporves bandwidth by 30% under some workflows. Signed-off-by: Michael S. Tsirkin <mst at

With PUBLISH_USED_IDX, guest tells us which used entries it has consumed. This can be used to reduce the number of interrupts: after we write a used entry, if the guest has not yet consumed the previous entry, or if the guest has already consumed the new entry, we do not need to interrupt. This imporves bandwidth by 30% under some workflows. Signed-off-by: Michael S. Tsirkin <mst at

This series tries to access virtqueue metadata through kernel virtual address instead of copy_user() friends since they had too much overheads like checks, spec barriers or even hardware feature toggling. This is done through setup kernel address through direct mapping and co-opreate VM management with MMU notifiers. Test shows about 23% improvement on TX PPS. TCP_STREAM doesn't see obvious

tree: https://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost.git vhost head: 90dc7026cdc9a9cbdc815cf1d8d29aa7b776ebf1 commit: ce4e61f610058ff9311f21d7584746fcfde9577e [27/32] virtio: force spec specified alignment on types config: i386-allyesconfig (attached as .config) compiler: gcc-7 (Ubuntu 7.5.0-6ubuntu2) 7.5.0 reproduce: git checkout ce4e61f610058ff9311f21d7584746fcfde9577e

The VHOST_VRING_F_LEGACY_BIG_ENDIAN flag informs the kernel that the associated device is big endian. Of course, this only makes sense for legacy virtio devices since modern virtio devices are always little endian. It will be used by the vhost memory accessors to byteswap vring data when we have a legacy device, in case host and guest endianness differ. Signed-off-by: Greg Kurz <gkurz at

The VHOST_VRING_F_LEGACY_BIG_ENDIAN flag informs the kernel that the associated device is big endian. Of course, this only makes sense for legacy virtio devices since modern virtio devices are always little endian. It will be used by the vhost memory accessors to byteswap vring data when we have a legacy device, in case host and guest endianness differ. Signed-off-by: Greg Kurz <gkurz at

Hi: This series tries to access virtqueue metadata through kernel virtual address instead of copy_user() friends since they had too much overheads like checks, spec barriers or even hardware feature toggling like SMAP. This is done through setup kernel address through direct mapping and co-opreate VM management with MMU notifiers. Test shows about 23% improvement on TX PPS. TCP_STREAM

Hi: This series tries to access virtqueue metadata through kernel virtual address instead of copy_user() friends since they had too much overheads like checks, spec barriers or even hardware feature toggling like SMAP. This is done through setup kernel address through direct mapping and co-opreate VM management with MMU notifiers. Test shows about 23% improvement on TX PPS. TCP_STREAM