search for: iov_to_buf

From: Amos Kong <akong at redhat.com> Currently virtio-net code relys on the layout of descriptor, this patchset removed the assumptions and introduced a control command to set mac address. Last patch is a trivial renaming. V2: check guest's iov_len V3: fix of migration compatibility make mac field in config space read-only when new feature is acked V4: add fix of descriptor layout

From: Amos Kong <akong at redhat.com> Currently virtio-net code relys on the layout of descriptor, this patchset removed the assumptions and introduced a control command to set mac address. Last patch is a trivial renaming. V2: check guest's iov_len V3: fix of migration compatibility make mac field in config space read-only when new feature is acked V4: add fix of descriptor layout

...uced a control command to set mac address. Last patch is a trivial renaming. V2: check guest's iov_len V3: fix of migration compatibility make mac field in config space read-only when new feature is acked V4: add fix of descriptor layout assumptions, trivial rename V5: fix endianness after iov_to_buf copy Amos Kong (2): virtio-net: introduce a new macaddr control virtio-net: rename ctrl rx commands Michael S. Tsirkin (1): virtio-net: remove layout assumptions for ctrl vq hw/pc_piix.c | 4 ++ hw/virtio-net.c | 142 +++++++++++++++++++++++++++++++++--------------------- hw/virtio...

...uced a control command to set mac address. Last patch is a trivial renaming. V2: check guest's iov_len V3: fix of migration compatibility make mac field in config space read-only when new feature is acked V4: add fix of descriptor layout assumptions, trivial rename V5: fix endianness after iov_to_buf copy Amos Kong (2): virtio-net: introduce a new macaddr control virtio-net: rename ctrl rx commands Michael S. Tsirkin (1): virtio-net: remove layout assumptions for ctrl vq hw/pc_piix.c | 4 ++ hw/virtio-net.c | 142 +++++++++++++++++++++++++++++++++--------------------- hw/virtio...

add modern header. This patch is for virtio 1.0 branch, doesn't apply to master. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- include/hw/virtio/virtio-balloon.h | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/include/hw/virtio/virtio-balloon.h b/include/hw/virtio/virtio-balloon.h index f863bfe..79eca67 100644 --- a/include/hw/virtio/virtio-balloon.h +++

add modern header. This patch is for virtio 1.0 branch, doesn't apply to master. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- include/hw/virtio/virtio-balloon.h | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/include/hw/virtio/virtio-balloon.h b/include/hw/virtio/virtio-balloon.h index f863bfe..79eca67 100644 --- a/include/hw/virtio/virtio-balloon.h +++

...BalloonStat legacy_stat; > + VirtIOBalloonStatModern modern_stat; > size_t offset = 0; > qemu_timeval tv; > > @@ -253,14 +254,28 @@ static void virtio_balloon_receive_stats(VirtIODevice *vdev, VirtQueue *vq) > */ > reset_stats(s); > > - while (iov_to_buf(elem->out_sg, elem->out_num, offset, &stat, sizeof(stat)) > - == sizeof(stat)) { > - uint16_t tag = virtio_tswap16(vdev, stat.tag); > - uint64_t val = virtio_tswap64(vdev, stat.val); > + if (virtio_has_feature(vdev, VIRTIO_F_VERSION_1)) { > +...

...BalloonStat legacy_stat; > + VirtIOBalloonStatModern modern_stat; > size_t offset = 0; > qemu_timeval tv; > > @@ -253,14 +254,28 @@ static void virtio_balloon_receive_stats(VirtIODevice *vdev, VirtQueue *vq) > */ > reset_stats(s); > > - while (iov_to_buf(elem->out_sg, elem->out_num, offset, &stat, sizeof(stat)) > - == sizeof(stat)) { > - uint16_t tag = virtio_tswap16(vdev, stat.tag); > - uint64_t val = virtio_tswap64(vdev, stat.val); > + if (virtio_has_feature(vdev, VIRTIO_F_VERSION_1)) { > +...

...m; - VirtIOBalloonStat stat; + VirtIOBalloonStat legacy_stat; + VirtIOBalloonStatModern modern_stat; size_t offset = 0; qemu_timeval tv; @@ -253,14 +254,28 @@ static void virtio_balloon_receive_stats(VirtIODevice *vdev, VirtQueue *vq) */ reset_stats(s); - while (iov_to_buf(elem->out_sg, elem->out_num, offset, &stat, sizeof(stat)) - == sizeof(stat)) { - uint16_t tag = virtio_tswap16(vdev, stat.tag); - uint64_t val = virtio_tswap64(vdev, stat.val); + if (virtio_has_feature(vdev, VIRTIO_F_VERSION_1)) { + while (iov_to_buf(elem...

...OBalloonStatModern modern_stat; > > size_t offset = 0; > > qemu_timeval tv; > > > > @@ -253,14 +254,28 @@ static void virtio_balloon_receive_stats(VirtIODevice *vdev, VirtQueue *vq) > > */ > > reset_stats(s); > > > > - while (iov_to_buf(elem->out_sg, elem->out_num, offset, &stat, sizeof(stat)) > > - == sizeof(stat)) { > > - uint16_t tag = virtio_tswap16(vdev, stat.tag); > > - uint64_t val = virtio_tswap64(vdev, stat.val); > > + if (virtio_has_feature(vdev, VIRTIO_F_VERSI...

...v_base; > > + if (elem->out_sg[0].iov_len != sizeof(*hdr)) { > > + error_report("invalid header size: %u", > > + (unsigned)elem->out_sg[0].iov_len); > > + exit(1); > > + } > > Please use iov_to_buf() instead of directly accessing out_sg[]. Virtio > devices are not supposed to assume a particular iovec layout. In other > words, virtio_pstore_hdr could be split across multiple out_sg[] iovecs. > > You must also copy in data (similar to Linux syscall implementations) to > preve...

...v_base; > > + if (elem->out_sg[0].iov_len != sizeof(*hdr)) { > > + error_report("invalid header size: %u", > > + (unsigned)elem->out_sg[0].iov_len); > > + exit(1); > > + } > > Please use iov_to_buf() instead of directly accessing out_sg[]. Virtio > devices are not supposed to assume a particular iovec layout. In other > words, virtio_pstore_hdr could be split across multiple out_sg[] iovecs. > > You must also copy in data (similar to Linux syscall implementations) to > preve...

...OK : + VIRTIO_RDMA_CTRL_ERR; +} + +int virtio_rdma_query_port(VirtIORdma *rdev, struct iovec *in, + struct iovec *out) +{ + struct ibv_port_attr attr = {}; + struct cmd_query_port cmd = {}; + int offs; + size_t s; + + s = iov_to_buf(in, 1, 0, &cmd, sizeof(cmd)); + if (s != sizeof(cmd)) { + return VIRTIO_RDMA_CTRL_ERR; + } + + if (cmd.port != 1) { + return VIRTIO_RDMA_CTRL_ERR; + } + + attr.state = IBV_PORT_ACTIVE; + attr.max_mtu = attr.active_mtu = IBV_MTU_1024; + attr.gid_tbl_len = 256;...

...tmap_bytes = 0, free_pages_count = 0; > + > + elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); > + if (!elem) { > + return; > + } > + s->free_pages_vq_elem = elem; > + > + if (!elem->out_num) { > + return; > + } > + > + iov_to_buf(elem->out_sg, elem->out_num, offset, > + &free_pages_count, sizeof(uint64_t)); > + > + offset += sizeof(uint64_t); > + iov_to_buf(elem->out_sg, elem->out_num, offset, > + &bitmap_bytes, sizeof(uint64_t)); > + > + offset +...

...QueueElement *elem; + size_t offset = 0; + uint64_t bitmap_bytes = 0, free_pages_count = 0; + + elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); + if (!elem) { + return; + } + s->free_pages_vq_elem = elem; + + if (!elem->out_num) { + return; + } + + iov_to_buf(elem->out_sg, elem->out_num, offset, + &free_pages_count, sizeof(uint64_t)); + + offset += sizeof(uint64_t); + iov_to_buf(elem->out_sg, elem->out_num, offset, + &bitmap_bytes, sizeof(uint64_t)); + + offset += sizeof(uint64_t); + iov_to_buf...

The current QEMU live migration implementation mark the all the guest's RAM pages as dirtied in the ram bulk stage, all these pages will be processed and that takes quit a lot of CPU cycles. >From guest's point of view, it doesn't care about the content in free pages. We can make use of this fact and skip processing the free pages in the ram bulk stage, it can save a lot CPU cycles

The current QEMU live migration implementation mark the all the guest's RAM pages as dirtied in the ram bulk stage, all these pages will be processed and that takes quit a lot of CPU cycles. >From guest's point of view, it doesn't care about the content in free pages. We can make use of this fact and skip processing the free pages in the ram bulk stage, it can save a lot CPU cycles

...(elem->out_sg[0].iov_len != sizeof(*hdr)) { > > > + error_report("invalid header size: %u", > > > + (unsigned)elem->out_sg[0].iov_len); > > > + exit(1); > > > + } > > > > Please use iov_to_buf() instead of directly accessing out_sg[]. Virtio > > devices are not supposed to assume a particular iovec layout. In other > > words, virtio_pstore_hdr could be split across multiple out_sg[] iovecs. > > > > You must also copy in data (similar to Linux syscall implementa...

...vdev, VirtQueue *vq) > pa = (hwaddr) p << VIRTIO_BALLOON_PFN_SHIFT; > offset += 4; > > - if (vq == s->icvq) { > + if (vq == s->icvq || vq == s->dcvq) { > uint32_t psize_ptr; > if (iov_to_buf(elem->out_sg, elem->out_num, offset, &psize_ptr, 4) != 4) { > break; > @@ -383,8 +385,9 @@ static void virtio_balloon_handle_output(VirtIODevice *vdev, VirtQueue *vq) > balloon_inflate_page(s, section.mr, >...

...s exactly the suggested optimization point: skip unmapped > pages. > > > > > > > You'd better check the live migration code. > > What's there to check in migration code? > Here's the extent of what balloon does on output: > > > while (iov_to_buf(elem->out_sg, elem->out_num, offset, &pfn, 4) == 4) > { > ram_addr_t pa; > ram_addr_t addr; > int p = virtio_ldl_p(vdev, &pfn); > > pa = (ram_addr_t) p << VIRTIO_BALLOON_PFN_SHIFT; > offset += 4; &...