- hw/arm/fsl-imx: Do not ignore Error argument - hw/net/cadence_gem.c: spelling fixes: Octects - tests/qtest: check the return value - libvhost-user: Fix VHOST_USER_GET_MAX_MEM_SLOTS reply mainline inclusion commit 69a5daec06f423843ce1bb9be5fb049314996f78 category: bugfix - io_uring: fix short read slow path mainline inclusion commit c06fc7ce147e57ab493bad9263f1601b8298484b category: bugfix - libvhost-user: Fix VHOST_USER_ADD_MEM_REG reply mainline inclusion commit 7f27d20ded2f480f3e66d03f90ea71507b834276 category: bugfix - qsd: Unlink absolute PID file path mainline inclusion commit 9d8f8233b9fa525a7e37350fbc18877051128c5d category: bugfix - net: Fix a misleading error message - vdpa: stop all svq on device deletion - vhost: release virtqueue objects in error path - vhost: fix the fd leak - virtio: i2c: Check notifier helpers for VIRTIO_CONFIG_IRQ_IDX - hw/virtio: fix typo in VIRTIO_CONFIG_IRQ_IDX comments - virtio-net: clear guest_announce feature if no cvq backend - vdpa: fix VHOST_BACKEND_F_IOTLB_ASID flag check - vdpa: do not block migration if device has cvq and x-svq=on - vdpa net: block migration if the device has CVQ - vdpa: Return -EIO if device ack is VIRTIO_NET_ERR in _load_mq() - vdpa: Return -EIO if device ack is VIRTIO_NET_ERR in _load_mac() - vdpa: fix not using CVQ buffer in case of error - vdpa: Fix possible use-after-free for VirtQueueElement - hw/virtio: fix vhost_user_read tracepoint - vhost: Fix false positive out-of-bounds - vhost: fix possible wrap in SVQ descriptor ring - vhost: move iova_tree set to vhost_svq_start - vhost: Always store new kick fd on vhost_svq_set_svq_kick_fd - virtio-crypto: verify src&dst buffer length for sym request - vdpa: commit all host notifier MRs in a single MR transaction - vdpa: harden the error path if get_iova_range failed - vdpa-dev: get iova range explicitly - virtio-pci: add support for configure interrupt - virtio-mmio: add support for configure interrupt - virtio-net: add support for configure interrupt - vhost: add support for configure interrupt - virtio: add support for configure interrupt - vhost-vdpa: add support for config interrupt - vhost: introduce new VhostOps vhost_set_config_call - virtio-pci: decouple the single vector from the interrupt process - virtio-pci: decouple notifier from interrupt process - virtio: introduce macro VIRTIO_CONFIG_IRQ_IDX - vdpa: do not handle VIRTIO_NET_F_GUEST_ANNOUNCE in vhost-vdpa - vdpa: handle VIRTIO_NET_CTRL_ANNOUNCE in vhost_vdpa_net_handle_ctrl_avail - vhost: fix vq dirty bitmap syncing when vIOMMU is enabled - hw/virtio: gracefully handle unset vhost_dev vdev - hw/virtio/vhost: Fix typo in comment. - vdpa: always start CVQ in SVQ mode if possible - vdpa: add shadow_data to vhost_vdpa - vdpa: store x-svq parameter in VhostVDPAState - vdpa: add asid parameter to vhost_vdpa_dma_map/unmap - vdpa: allocate SVQ array unconditionally - vdpa: move SVQ vring features check to net/ - vdpa: request iova_range only once - vdpa: add vhost_vdpa_net_valid_svq_features - vhost: allocate SVQ device file descriptors at device start - vhost: set SVQ device call handler at SVQ start - vdpa: use v->shadow_vqs_enabled in vhost_vdpa_svqs_start & stop - vhost: enable vrings in vhost_dev_start() for vhost-user devices - vhost-vdpa: fix assert !virtio_net_get_subqueue(nc)->async_tx.elem in virtio_net_reset - net/vhost-vdpa.c: Fix clang compilation failure - vhost-vdpa: allow passing opened vhostfd to vhost-vdpa - vdpa: Remove shadow CVQ command check - vdpa: Delete duplicated vdpa_feature_bits entry - hw/virtio: add some vhost-user trace events - vdpa: Allow MQ feature in SVQ - virtio-net: Update virtio-net curr_queue_pairs in vdpa backends - vdpa: validate MQ CVQ commands - vdpa: Add vhost_vdpa_net_load_mq - vdpa: extract vhost_vdpa_net_load_mac from vhost_vdpa_net_load - vdpa: Make VhostVDPAState cvq_cmd_in_buffer control ack type - vdpa: Delete CVQ migration blocker - vdpa: Add virtio-net mac address via CVQ at start - vhost_net: add NetClientState->load() callback - vdpa: extract vhost_vdpa_net_cvq_add from vhost_vdpa_net_handle_ctrl_avail - vdpa: Move command buffers map to start of net device - vdpa: add net_vhost_vdpa_cvq_info NetClientInfo - vhost_net: Add NetClientInfo stop callback - vhost_net: Add NetClientInfo start callback - vdpa: Use ring hwaddr at vhost_vdpa_svq_unmap_ring - vdpa: Make SVQ vring unmapping return void - vdpa: Remove SVQ vring from iova_tree at shutdown - util: accept iova_tree_remove_parameter by value - vdpa: do not save failed dma maps in SVQ iova tree - vdpa: Skip the maps not in the iova tree - vdpa: Fix file descriptor leak on get features error - vdpa: Fix memory listener deletions of iova tree - vhost: Get vring base from vq, not svq - vdpa: Add x-svq to NetdevVhostVDPAOptions - vdpa: Add device migration blocker - vdpa: Extract get features part from vhost_vdpa_get_max_queue_pairs - vdpa: Buffer CVQ support on shadow virtqueue - vdpa: manual forward CVQ buffers - vdpa: Export vhost_vdpa_dma_map and unmap calls - vhost: Add svq avail_handler callback - vhost: add vhost_svq_poll - vhost: Expose vhost_svq_add - vhost: add vhost_svq_push_elem - vhost: Track number of descs in SVQDescState - vhost: Add SVQDescState - vhost: Decouple vhost_svq_add from VirtQueueElement - vhost: Check for queue full at vhost_svq_add - vhost: Move vhost_svq_kick call to vhost_svq_add - vhost: Reorder vhost_svq_kick - vdpa: Avoid compiler to squash reads to used idx - virtio-net: Expose ctrl virtqueue logic - virtio-net: Expose MAC_TABLE_ENTRIES - vhost: move descriptor translation to vhost_svq_vring_write_descs - util: Return void on iova_tree_remove - virtio-net: don't handle mq request in userspace handler for vhost-vdpa - vhost-vdpa: change name and polarity for vhost_vdpa_one_time_request() - vhost-vdpa: backend feature should set only once - vhost-vdpa: fix improper cleanup in net_init_vhost_vdpa - virtio-net: align ctrl_vq index for non-mq guest for vhost_vdpa - virtio: add vhost support for virtio devices - include/hw: start documenting the vhost API - hw/virtio: add vhost_user_[read|write] trace points - vhost: Fix element in vhost_svq_add failure - vdpa: Fix index calculus at vhost_vdpa_svqs_start - vdpa: Fix bad index calculus at vhost_vdpa_get_vring_base - vhost: Fix device's used descriptor dequeue - vhost: Track descriptor chain in private at SVQ - vdpa: Add missing tracing to batch mapping functions - vhost-vdpa: fix typo in a comment - virtio: fix --enable-vhost-user build on non-Linux - vdpa: Expose VHOST_F_LOG_ALL on SVQ - vdpa: Never set log_base addr if SVQ is enabled - vdpa: Adapt vhost_vdpa_get_vring_base to SVQ - vdpa: Add custom IOTLB translations to SVQ - vhost: Add VhostIOVATree - util: add iova_tree_find_iova - util: Add iova_tree_alloc_map - vhost: Shadow virtqueue buffers forwarding - vdpa: adapt vhost_ops callbacks to svq - virtio: Add vhost_svq_get_vring_addr - vhost: Add vhost_svq_valid_features to shadow vq - vhost: Add Shadow VirtQueue call forwarding capabilities - vhost: Add Shadow VirtQueue kick forwarding capabilities - vhost: Add VhostShadowVirtqueue - vdpa: Make ncs autofree - Revert "virtio: introduce macro IRTIO_CONFIG_IRQ_IDX" - Revert "virtio-pci: decouple notifier from interrupt process" - Revert "virtio-pci: decouple the single vector from the interrupt process" - Revert "vhost-vdpa: add support for config interrupt" - Revert "virtio: add support for configure interrupt" - Revert "vhost: add support for configure interrupt" - Revert "virtio-net: add support for configure interrupt" - Revert "virtio-mmio: add support for configure interrupt" - Revert "virtio-pci: add support for configure interrupt" - Revert "vhost: introduce new VhostOps vhost_set_config_call" - virtio: signal after wrapping packed used_idx - target/i386: Adjust feature level according to FEAT_7_1_EDX - target/i386: Add new CPU model GraniteRapids - target/i386: Add support for PREFETCHIT0/1 in CPUID enumeration - target/i386: Add support for AVX-NE-CONVERT in CPUID enumeration - target/i386: Add support for AVX-VNNI-INT8 in CPUID enumeration - target/i386: Add support for AVX-IFMA in CPUID enumeration - target/i386: Add support for AMX-FP16 in CPUID enumeration - target/i386: Add support for CMPCCXADD in CPUID enumeration - tracetool: avoid invalid escape in Python string - hw/pvrdma: Protect against buggy or malicious guest driver - vga: avoid crash if no default vga card mainline inclusion commit 6985d8ede92494f3b791de01e8ee9306eb6d5e4a category: bugfix - qom/object: Remove circular include dependency mainline inclusion commit 5bba9bcfbb42e7c016626420e148a1bf1b080835 category: bugfix - artist: set memory region owners for buffers to the artist device mainline inclusion commit 39fbaeca096a9bf6cbe2af88572c1cb2aa62aa8c category: bugfix - virtio-iommu: Fix the partial copy of probe request mainline inclusion commit 45461aace83d961e933b27519b81d17b4c690514 category: bugfix - e1000: set RX descriptor status in a separate operation mainline inclusion commit 034d00d4858161e1d4cff82d8d230bce874a04d3 category: bugfix - vhost: introduce new VhostOps vhost_set_config_call - vhost: stick to -errno error return convention - vhost-user: stick to -errno error return convention - vhost-vdpa: stick to -errno error return convention - virtio-pci: add support for configure interrupt - virtio-mmio: add support for configure interrupt - virtio-net: add support for configure interrupt - vhost: add support for configure interrupt - virtio: add support for configure interrupt - vhost-vdpa: add support for config interrupt - virtio-pci: decouple the single vector from the interrupt process - virtio-pci: decouple notifier from interrupt process - virtio: introduce macro IRTIO_CONFIG_IRQ_IDX - pci: Fix the update of interrupt disable bit in PCI_COMMAND register - hw/timer/npcm7xx_timer: Prevent timer from counting down past zero - tpm_crb: mark command buffer as dirty on request completion mainline inclusion commit e37a0ef4605e5d2041785ff3fc89ca6021faf7a0 category: bugfix - pci: fix overflow in snprintf string formatting mainline inclusion commit 36f18c6989a3d1ff1d7a0e50b0868ef3958299b4 category: bugfix - hw/usb/hcd-ehci: fix writeback order mainline inclusion commit f471e8b060798f26a7fc339c6152f82f22a7b33d category: bugfix - qemu-timer: Skip empty timer lists before locking in qemu_clock_deadline_ns_all mainline inclusion commit 3f42906c9ab2c777a895b48b87b8107167e4a275 category: bugfix - semihosting/config: Merge --semihosting-config option groups mainline inclusion commit 90c072e063737e9e8f431489bbd334452f89056e category: bugfix - semihosting: fix memleak at semihosting_arg_fallback - target/i386: Export GDS_NO bit to guests Signed-off-by: Jiabo Feng <fengjiabo1@huawei.com>
109 lines
4.5 KiB
Diff
109 lines
4.5 KiB
Diff
From fdc4e7386f7fa5f0d47d6d7b582a01854073206e Mon Sep 17 00:00:00 2001
|
|
From: Hawkins Jiawei <yin31149@gmail.com>
|
|
Date: Tue, 9 May 2023 16:48:17 +0800
|
|
Subject: [PATCH] vhost: fix possible wrap in SVQ descriptor ring
|
|
MIME-Version: 1.0
|
|
Content-Type: text/plain; charset=UTF-8
|
|
Content-Transfer-Encoding: 8bit
|
|
|
|
QEMU invokes vhost_svq_add() when adding a guest's element
|
|
into SVQ. In vhost_svq_add(), it uses vhost_svq_available_slots()
|
|
to check whether QEMU can add the element into SVQ. If there is
|
|
enough space, then QEMU combines some out descriptors and some
|
|
in descriptors into one descriptor chain, and adds it into
|
|
`svq->vring.desc` by vhost_svq_vring_write_descs().
|
|
|
|
Yet the problem is that, `svq->shadow_avail_idx - svq->shadow_used_idx`
|
|
in vhost_svq_available_slots() returns the number of occupied elements,
|
|
or the number of descriptor chains, instead of the number of occupied
|
|
descriptors, which may cause wrapping in SVQ descriptor ring.
|
|
|
|
Here is an example. In vhost_handle_guest_kick(), QEMU forwards
|
|
as many available buffers to device by virtqueue_pop() and
|
|
vhost_svq_add_element(). virtqueue_pop() returns a guest's element,
|
|
and then this element is added into SVQ by vhost_svq_add_element(),
|
|
a wrapper to vhost_svq_add(). If QEMU invokes virtqueue_pop() and
|
|
vhost_svq_add_element() `svq->vring.num` times,
|
|
vhost_svq_available_slots() thinks QEMU just ran out of slots and
|
|
everything should work fine. But in fact, virtqueue_pop() returns
|
|
`svq->vring.num` elements or descriptor chains, more than
|
|
`svq->vring.num` descriptors due to guest memory fragmentation,
|
|
and this causes wrapping in SVQ descriptor ring.
|
|
|
|
This bug is valid even before marking the descriptors used.
|
|
If the guest memory is fragmented, SVQ must add chains
|
|
so it can try to add more descriptors than possible.
|
|
|
|
This patch solves it by adding `num_free` field in
|
|
VhostShadowVirtqueue structure and updating this field
|
|
in vhost_svq_add() and vhost_svq_get_buf(), to record
|
|
the number of free descriptors.
|
|
|
|
Fixes: 100890f7ca ("vhost: Shadow virtqueue buffers forwarding")
|
|
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
|
|
Acked-by: Eugenio Pérez <eperezma@redhat.com>
|
|
Message-Id: <20230509084817.3973-1-yin31149@gmail.com>
|
|
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
|
|
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
|
|
Tested-by: Lei Yang <leiyang@redhat.com>
|
|
Signed-off-by: fangyi <eric.fangyi@huawei.com>
|
|
---
|
|
hw/virtio/vhost-shadow-virtqueue.c | 5 ++++-
|
|
hw/virtio/vhost-shadow-virtqueue.h | 3 +++
|
|
2 files changed, 7 insertions(+), 1 deletion(-)
|
|
|
|
diff --git a/hw/virtio/vhost-shadow-virtqueue.c b/hw/virtio/vhost-shadow-virtqueue.c
|
|
index 7d6afcb528..8d99edb196 100644
|
|
--- a/hw/virtio/vhost-shadow-virtqueue.c
|
|
+++ b/hw/virtio/vhost-shadow-virtqueue.c
|
|
@@ -66,7 +66,7 @@ bool vhost_svq_valid_features(uint64_t features, Error **errp)
|
|
*/
|
|
static uint16_t vhost_svq_available_slots(const VhostShadowVirtqueue *svq)
|
|
{
|
|
- return svq->vring.num - (svq->shadow_avail_idx - svq->shadow_used_idx);
|
|
+ return svq->num_free;
|
|
}
|
|
|
|
/**
|
|
@@ -255,6 +255,7 @@ int vhost_svq_add(VhostShadowVirtqueue *svq, const struct iovec *out_sg,
|
|
return -EINVAL;
|
|
}
|
|
|
|
+ svq->num_free -= ndescs;
|
|
svq->desc_state[qemu_head].elem = elem;
|
|
svq->desc_state[qemu_head].ndescs = ndescs;
|
|
vhost_svq_kick(svq);
|
|
@@ -426,6 +427,7 @@ static VirtQueueElement *vhost_svq_get_buf(VhostShadowVirtqueue *svq,
|
|
last_used_chain = vhost_svq_last_desc_of_chain(svq, num, used_elem.id);
|
|
svq->desc_next[last_used_chain] = svq->free_head;
|
|
svq->free_head = used_elem.id;
|
|
+ svq->num_free += num;
|
|
|
|
*len = used_elem.len;
|
|
return g_steal_pointer(&svq->desc_state[used_elem.id].elem);
|
|
@@ -636,6 +638,7 @@ void vhost_svq_start(VhostShadowVirtqueue *svq, VirtIODevice *vdev,
|
|
svq->iova_tree = iova_tree;
|
|
|
|
svq->vring.num = virtio_queue_get_num(vdev, virtio_get_queue_index(vq));
|
|
+ svq->num_free = svq->vring.num;
|
|
driver_size = vhost_svq_driver_area_size(svq);
|
|
device_size = vhost_svq_device_area_size(svq);
|
|
svq->vring.desc = qemu_memalign(qemu_real_host_page_size, driver_size);
|
|
diff --git a/hw/virtio/vhost-shadow-virtqueue.h b/hw/virtio/vhost-shadow-virtqueue.h
|
|
index 926a4897b1..6efe051a70 100644
|
|
--- a/hw/virtio/vhost-shadow-virtqueue.h
|
|
+++ b/hw/virtio/vhost-shadow-virtqueue.h
|
|
@@ -107,6 +107,9 @@ typedef struct VhostShadowVirtqueue {
|
|
|
|
/* Next head to consume from the device */
|
|
uint16_t last_used_idx;
|
|
+
|
|
+ /* Size of SVQ vring free descriptors */
|
|
+ uint16_t num_free;
|
|
} VhostShadowVirtqueue;
|
|
|
|
bool vhost_svq_valid_features(uint64_t features, Error **errp);
|
|
--
|
|
2.27.0
|
|
|