Should help live migration downtime on source host. Below are the
coalesced dma_unmap time series on 2 queue pair config (no
dedicated descriptor group ASID for SVQ).
109531@1693367276.853503:vhost_vdpa_reset_device dev: 0x55c933926890
109531@1693367276.853513:vhost_vdpa_add_status dev: 0x55c933926890 status: 0x3
109531@1693367276.853520:vhost_vdpa_flush_map dev: 0x55c933926890 doit: 1 svq_flush: 0 persist: 1
109531@1693367276.853524:vhost_vdpa_set_config_call dev: 0x55c933926890 fd: -1
109531@1693367276.853579:vhost_vdpa_iotlb_begin_batch vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 type: 5
109531@1693367276.853586:vhost_vdpa_dma_unmap vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 asid: 0 iova: 0x1000 size: 0x2000 type: 3
109531@1693367276.853600:vhost_vdpa_dma_unmap vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 asid: 0 iova: 0x3000 size: 0x1000 type: 3
109531@1693367276.853618:vhost_vdpa_dma_unmap vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 asid: 0 iova: 0x4000 size: 0x2000 type: 3
109531@1693367276.853625:vhost_vdpa_dma_unmap vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 asid: 0 iova: 0x6000 size: 0x1000 type: 3
109531@1693367276.853630:vhost_vdpa_dma_unmap vdpa:0x7fa2aa84c190 fd: 16 msg_type: 2 asid: 0 iova: 0x7000 size: 0x2000 type: 3
109531@1693367276.853636:vhost_vdpa_dma_unmap vdpa:0x7fa2aa84c190 fd: 16 msg_type: 2 asid: 0 iova: 0x9000 size: 0x1000 type: 3
109531@1693367276.853642:vhost_vdpa_dma_unmap vdpa:0x7fa2aa84c190 fd: 16 msg_type: 2 asid: 0 iova: 0xa000 size: 0x2000 type: 3
109531@1693367276.853648:vhost_vdpa_dma_unmap vdpa:0x7fa2aa84c190 fd: 16 msg_type: 2 asid: 0 iova: 0xc000 size: 0x1000 type: 3
109531@1693367276.853654:vhost_vdpa_dma_unmap vdpa:0x7fa2aa6b6190 fd: 16 msg_type: 2 asid: 0 iova: 0xf000 size: 0x1000 type: 3
109531@1693367276.853660:vhost_vdpa_dma_unmap vdpa:0x7fa2aa6b6190 fd: 16 msg_type: 2 asid: 0 iova: 0x10000 size: 0x1000 type: 3
109531@1693367276.853666:vhost_vdpa_dma_unmap vdpa:0x7fa2aa6b6190 fd: 16 msg_type: 2 asid: 0 iova: 0xd000 size: 0x1000 type: 3
109531@1693367276.853670:vhost_vdpa_dma_unmap vdpa:0x7fa2aa6b6190 fd: 16 msg_type: 2 asid: 0 iova: 0xe000 size: 0x1000 type: 3
109531@1693367276.853675:vhost_vdpa_iotlb_end_batch vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 type: 6
109531@1693367277.014697:vhost_vdpa_get_vq_index dev: 0x55c933925de0 idx: 0 vq idx: 0
109531@1693367277.014747:vhost_vdpa_get_vq_index dev: 0x55c933925de0 idx: 1 vq idx: 1
109531@1693367277.014753:vhost_vdpa_get_vq_index dev: 0x55c9339262e0 idx: 2 vq idx: 2
109531@1693367277.014756:vhost_vdpa_get_vq_index dev: 0x55c9339262e0 idx: 3 vq idx: 3
Signed-off-by: Si-Wei Liu <si-wei.liu@oracle.com>
---
hw/virtio/vhost-vdpa.c | 7 +--
include/hw/virtio/vhost-vdpa.h | 3 ++
net/vhost-vdpa.c | 112 +++++++++++++++++++++++++++--------------
3 files changed, 80 insertions(+), 42 deletions(-)
diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
index d98704a..4010fd9 100644
--- a/hw/virtio/vhost-vdpa.c
+++ b/hw/virtio/vhost-vdpa.c
@@ -1162,8 +1162,8 @@ static void vhost_vdpa_svq_unmap_ring(struct vhost_vdpa *v, hwaddr addr)
vhost_iova_tree_remove(v->shared->iova_tree, *result);
}
-static void vhost_vdpa_svq_unmap_rings(struct vhost_dev *dev,
- const VhostShadowVirtqueue *svq)
+void vhost_vdpa_svq_unmap_rings(struct vhost_dev *dev,
+ const VhostShadowVirtqueue *svq)
{
struct vhost_vdpa *v = dev->opaque;
struct vhost_vring_addr svq_addr;
@@ -1346,17 +1346,14 @@ static void vhost_vdpa_svqs_stop(struct vhost_dev *dev)
return;
}
- vhost_vdpa_dma_batch_begin_once(v->shared, v->address_space_id);
for (unsigned i = 0; i < v->shadow_vqs->len; ++i) {
VhostShadowVirtqueue *svq = g_ptr_array_index(v->shadow_vqs, i);
vhost_svq_stop(svq);
- vhost_vdpa_svq_unmap_rings(dev, svq);
event_notifier_cleanup(&svq->hdev_kick);
event_notifier_cleanup(&svq->hdev_call);
}
- vhost_vdpa_dma_batch_end_once(v->shared, v->address_space_id);
}
static void vhost_vdpa_suspend(struct vhost_dev *dev)
diff --git a/include/hw/virtio/vhost-vdpa.h b/include/hw/virtio/vhost-vdpa.h
index aa13679..f426e2c 100644
--- a/include/hw/virtio/vhost-vdpa.h
+++ b/include/hw/virtio/vhost-vdpa.h
@@ -112,6 +112,9 @@ int vhost_vdpa_dma_batch_end_once(VhostVDPAShared *s, uint32_t asid);
int vhost_vdpa_load_setup(VhostVDPAShared *s, AddressSpace *dma_as);
int vhost_vdpa_load_cleanup(VhostVDPAShared *s, bool vhost_will_start);
+void vhost_vdpa_svq_unmap_rings(struct vhost_dev *dev,
+ const VhostShadowVirtqueue *svq);
+
typedef struct vdpa_iommu {
VhostVDPAShared *dev_shared;
IOMMUMemoryRegion *iommu_mr;
diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c
index 683619f..41714d1 100644
--- a/net/vhost-vdpa.c
+++ b/net/vhost-vdpa.c
@@ -29,6 +29,7 @@
#include "migration/migration.h"
#include "migration/misc.h"
#include "hw/virtio/vhost.h"
+#include "hw/virtio/vhost-vdpa.h"
/* Todo:need to add the multiqueue support here */
typedef struct VhostVDPAState {
@@ -467,15 +468,89 @@ static int vhost_vdpa_net_data_load(NetClientState *nc)
return 0;
}
+static void vhost_vdpa_cvq_unmap_buf(struct vhost_vdpa *v, void *addr)
+{
+ VhostIOVATree *tree = v->shared->iova_tree;
+ DMAMap needle = {
+ /*
+ * No need to specify size or to look for more translations since
+ * this contiguous chunk was allocated by us.
+ */
+ .translated_addr = (hwaddr)(uintptr_t)addr,
+ };
+ const DMAMap *map = vhost_iova_tree_find_iova(tree, &needle);
+ int r;
+
+ if (unlikely(!map)) {
+ error_report("Cannot locate expected map");
+ return;
+ }
+
+ r = vhost_vdpa_dma_unmap(v->shared, v->address_space_id, map->iova,
+ map->size + 1);
+ if (unlikely(r != 0)) {
+ error_report("Device cannot unmap: %s(%d)", g_strerror(r), r);
+ }
+
+ vhost_iova_tree_remove(tree, *map);
+}
+
static void vhost_vdpa_net_client_stop(NetClientState *nc)
{
VhostVDPAState *s = DO_UPCAST(VhostVDPAState, nc, nc);
+ struct vhost_vdpa *v = &s->vhost_vdpa;
+ struct vhost_vdpa *last_vi = NULL;
+ bool has_cvq = v->dev->vq_index_end % 2;
+ int nvqp;
assert(nc->info->type == NET_CLIENT_DRIVER_VHOST_VDPA);
if (s->vhost_vdpa.index == 0) {
migration_remove_notifier(&s->migration_state);
}
+
+ if (v->dev->vq_index + v->dev->nvqs != v->dev->vq_index_end) {
+ return;
+ }
+
+ nvqp = (v->dev->vq_index_end + 1) / 2;
+ for (int i = 0; i < nvqp; ++i) {
+ VhostVDPAState *s_i = vhost_vdpa_net_get_nc_vdpa(s, i);
+ struct vhost_vdpa *v_i = &s_i->vhost_vdpa;
+
+ if (!v_i->shadow_vqs_enabled) {
+ continue;
+ }
+ if (!last_vi) {
+ vhost_vdpa_dma_batch_begin_once(v_i->shared,
+ v_i->address_space_id);
+ last_vi = v_i;
+ } else if (last_vi->address_space_id != v_i->address_space_id) {
+ vhost_vdpa_dma_batch_end_once(last_vi->shared,
+ last_vi->address_space_id);
+ vhost_vdpa_dma_batch_begin_once(v_i->shared,
+ v_i->address_space_id);
+ last_vi = v_i;
+ }
+
+ for (unsigned j = 0; j < v_i->shadow_vqs->len; ++j) {
+ VhostShadowVirtqueue *svq = g_ptr_array_index(v_i->shadow_vqs, j);
+
+ vhost_vdpa_svq_unmap_rings(v_i->dev, svq);
+ }
+ }
+ if (has_cvq) {
+ if (last_vi) {
+ assert(last_vi->address_space_id == v->address_space_id);
+ }
+ vhost_vdpa_cvq_unmap_buf(&s->vhost_vdpa, s->cvq_cmd_out_buffer);
+ vhost_vdpa_cvq_unmap_buf(&s->vhost_vdpa, s->status);
+ }
+ if (last_vi) {
+ vhost_vdpa_dma_batch_end_once(last_vi->shared,
+ last_vi->address_space_id);
+ last_vi = NULL;
+ }
}
static int vhost_vdpa_net_load_setup(NetClientState *nc, NICState *nic)
@@ -585,33 +660,6 @@ static int64_t vhost_vdpa_get_vring_desc_group(int device_fd,
return state.num;
}
-static void vhost_vdpa_cvq_unmap_buf(struct vhost_vdpa *v, void *addr)
-{
- VhostIOVATree *tree = v->shared->iova_tree;
- DMAMap needle = {
- /*
- * No need to specify size or to look for more translations since
- * this contiguous chunk was allocated by us.
- */
- .translated_addr = (hwaddr)(uintptr_t)addr,
- };
- const DMAMap *map = vhost_iova_tree_find_iova(tree, &needle);
- int r;
-
- if (unlikely(!map)) {
- error_report("Cannot locate expected map");
- return;
- }
-
- r = vhost_vdpa_dma_unmap(v->shared, v->address_space_id, map->iova,
- map->size + 1);
- if (unlikely(r != 0)) {
- error_report("Device cannot unmap: %s(%d)", g_strerror(r), r);
- }
-
- vhost_iova_tree_remove(tree, *map);
-}
-
/** Map CVQ buffer. */
static int vhost_vdpa_cvq_map_buf(struct vhost_vdpa *v, void *buf, size_t size,
bool write)
@@ -740,18 +788,8 @@ err:
static void vhost_vdpa_net_cvq_stop(NetClientState *nc)
{
- VhostVDPAState *s = DO_UPCAST(VhostVDPAState, nc, nc);
- struct vhost_vdpa *v = &s->vhost_vdpa;
-
assert(nc->info->type == NET_CLIENT_DRIVER_VHOST_VDPA);
- if (s->vhost_vdpa.shadow_vqs_enabled) {
- vhost_vdpa_dma_batch_begin_once(v->shared, v->address_space_id);
- vhost_vdpa_cvq_unmap_buf(&s->vhost_vdpa, s->cvq_cmd_out_buffer);
- vhost_vdpa_cvq_unmap_buf(&s->vhost_vdpa, s->status);
- vhost_vdpa_dma_batch_end_once(v->shared, v->address_space_id);
- }
-
vhost_vdpa_net_client_stop(nc);
}
--
1.8.3.1On Fri, Dec 8, 2023 at 2:51 AM Si-Wei Liu <si-wei.liu@oracle.com> wrote:
>
> Should help live migration downtime on source host. Below are the
> coalesced dma_unmap time series on 2 queue pair config (no
> dedicated descriptor group ASID for SVQ).
It's better to explain how we can batch into a single call (e.g do we
batch at the level of a whole device?)
>
> 109531@1693367276.853503:vhost_vdpa_reset_device dev: 0x55c933926890
> 109531@1693367276.853513:vhost_vdpa_add_status dev: 0x55c933926890 status: 0x3
> 109531@1693367276.853520:vhost_vdpa_flush_map dev: 0x55c933926890 doit: 1 svq_flush: 0 persist: 1
> 109531@1693367276.853524:vhost_vdpa_set_config_call dev: 0x55c933926890 fd: -1
> 109531@1693367276.853579:vhost_vdpa_iotlb_begin_batch vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 type: 5
> 109531@1693367276.853586:vhost_vdpa_dma_unmap vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 asid: 0 iova: 0x1000 size: 0x2000 type: 3
> 109531@1693367276.853600:vhost_vdpa_dma_unmap vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 asid: 0 iova: 0x3000 size: 0x1000 type: 3
> 109531@1693367276.853618:vhost_vdpa_dma_unmap vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 asid: 0 iova: 0x4000 size: 0x2000 type: 3
> 109531@1693367276.853625:vhost_vdpa_dma_unmap vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 asid: 0 iova: 0x6000 size: 0x1000 type: 3
> 109531@1693367276.853630:vhost_vdpa_dma_unmap vdpa:0x7fa2aa84c190 fd: 16 msg_type: 2 asid: 0 iova: 0x7000 size: 0x2000 type: 3
> 109531@1693367276.853636:vhost_vdpa_dma_unmap vdpa:0x7fa2aa84c190 fd: 16 msg_type: 2 asid: 0 iova: 0x9000 size: 0x1000 type: 3
> 109531@1693367276.853642:vhost_vdpa_dma_unmap vdpa:0x7fa2aa84c190 fd: 16 msg_type: 2 asid: 0 iova: 0xa000 size: 0x2000 type: 3
> 109531@1693367276.853648:vhost_vdpa_dma_unmap vdpa:0x7fa2aa84c190 fd: 16 msg_type: 2 asid: 0 iova: 0xc000 size: 0x1000 type: 3
> 109531@1693367276.853654:vhost_vdpa_dma_unmap vdpa:0x7fa2aa6b6190 fd: 16 msg_type: 2 asid: 0 iova: 0xf000 size: 0x1000 type: 3
> 109531@1693367276.853660:vhost_vdpa_dma_unmap vdpa:0x7fa2aa6b6190 fd: 16 msg_type: 2 asid: 0 iova: 0x10000 size: 0x1000 type: 3
> 109531@1693367276.853666:vhost_vdpa_dma_unmap vdpa:0x7fa2aa6b6190 fd: 16 msg_type: 2 asid: 0 iova: 0xd000 size: 0x1000 type: 3
> 109531@1693367276.853670:vhost_vdpa_dma_unmap vdpa:0x7fa2aa6b6190 fd: 16 msg_type: 2 asid: 0 iova: 0xe000 size: 0x1000 type: 3
> 109531@1693367276.853675:vhost_vdpa_iotlb_end_batch vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 type: 6
> 109531@1693367277.014697:vhost_vdpa_get_vq_index dev: 0x55c933925de0 idx: 0 vq idx: 0
> 109531@1693367277.014747:vhost_vdpa_get_vq_index dev: 0x55c933925de0 idx: 1 vq idx: 1
> 109531@1693367277.014753:vhost_vdpa_get_vq_index dev: 0x55c9339262e0 idx: 2 vq idx: 2
> 109531@1693367277.014756:vhost_vdpa_get_vq_index dev: 0x55c9339262e0 idx: 3 vq idx: 3
>
> Signed-off-by: Si-Wei Liu <si-wei.liu@oracle.com>
> ---
> hw/virtio/vhost-vdpa.c | 7 +--
> include/hw/virtio/vhost-vdpa.h | 3 ++
> net/vhost-vdpa.c | 112 +++++++++++++++++++++++++++--------------
> 3 files changed, 80 insertions(+), 42 deletions(-)
>
> diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
> index d98704a..4010fd9 100644
> --- a/hw/virtio/vhost-vdpa.c
> +++ b/hw/virtio/vhost-vdpa.c
> @@ -1162,8 +1162,8 @@ static void vhost_vdpa_svq_unmap_ring(struct vhost_vdpa *v, hwaddr addr)
> vhost_iova_tree_remove(v->shared->iova_tree, *result);
> }
>
> -static void vhost_vdpa_svq_unmap_rings(struct vhost_dev *dev,
> - const VhostShadowVirtqueue *svq)
> +void vhost_vdpa_svq_unmap_rings(struct vhost_dev *dev,
> + const VhostShadowVirtqueue *svq)
> {
> struct vhost_vdpa *v = dev->opaque;
> struct vhost_vring_addr svq_addr;
> @@ -1346,17 +1346,14 @@ static void vhost_vdpa_svqs_stop(struct vhost_dev *dev)
> return;
> }
>
> - vhost_vdpa_dma_batch_begin_once(v->shared, v->address_space_id);
> for (unsigned i = 0; i < v->shadow_vqs->len; ++i) {
> VhostShadowVirtqueue *svq = g_ptr_array_index(v->shadow_vqs, i);
>
> vhost_svq_stop(svq);
> - vhost_vdpa_svq_unmap_rings(dev, svq);
>
> event_notifier_cleanup(&svq->hdev_kick);
> event_notifier_cleanup(&svq->hdev_call);
> }
> - vhost_vdpa_dma_batch_end_once(v->shared, v->address_space_id);
> }
>
> static void vhost_vdpa_suspend(struct vhost_dev *dev)
> diff --git a/include/hw/virtio/vhost-vdpa.h b/include/hw/virtio/vhost-vdpa.h
> index aa13679..f426e2c 100644
> --- a/include/hw/virtio/vhost-vdpa.h
> +++ b/include/hw/virtio/vhost-vdpa.h
> @@ -112,6 +112,9 @@ int vhost_vdpa_dma_batch_end_once(VhostVDPAShared *s, uint32_t asid);
> int vhost_vdpa_load_setup(VhostVDPAShared *s, AddressSpace *dma_as);
> int vhost_vdpa_load_cleanup(VhostVDPAShared *s, bool vhost_will_start);
>
> +void vhost_vdpa_svq_unmap_rings(struct vhost_dev *dev,
> + const VhostShadowVirtqueue *svq);
> +
> typedef struct vdpa_iommu {
> VhostVDPAShared *dev_shared;
> IOMMUMemoryRegion *iommu_mr;
> diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c
> index 683619f..41714d1 100644
> --- a/net/vhost-vdpa.c
> +++ b/net/vhost-vdpa.c
> @@ -29,6 +29,7 @@
> #include "migration/migration.h"
> #include "migration/misc.h"
> #include "hw/virtio/vhost.h"
> +#include "hw/virtio/vhost-vdpa.h"
>
> /* Todo:need to add the multiqueue support here */
> typedef struct VhostVDPAState {
> @@ -467,15 +468,89 @@ static int vhost_vdpa_net_data_load(NetClientState *nc)
> return 0;
> }
>
> +static void vhost_vdpa_cvq_unmap_buf(struct vhost_vdpa *v, void *addr)
> +{
> + VhostIOVATree *tree = v->shared->iova_tree;
> + DMAMap needle = {
> + /*
> + * No need to specify size or to look for more translations since
> + * this contiguous chunk was allocated by us.
> + */
> + .translated_addr = (hwaddr)(uintptr_t)addr,
> + };
> + const DMAMap *map = vhost_iova_tree_find_iova(tree, &needle);
> + int r;
> +
> + if (unlikely(!map)) {
> + error_report("Cannot locate expected map");
> + return;
> + }
> +
> + r = vhost_vdpa_dma_unmap(v->shared, v->address_space_id, map->iova,
> + map->size + 1);
> + if (unlikely(r != 0)) {
> + error_report("Device cannot unmap: %s(%d)", g_strerror(r), r);
> + }
> +
> + vhost_iova_tree_remove(tree, *map);
> +}
> +
> static void vhost_vdpa_net_client_stop(NetClientState *nc)
> {
> VhostVDPAState *s = DO_UPCAST(VhostVDPAState, nc, nc);
> + struct vhost_vdpa *v = &s->vhost_vdpa;
> + struct vhost_vdpa *last_vi = NULL;
Nit: just curious what did "vi" mean here?
> + bool has_cvq = v->dev->vq_index_end % 2;
> + int nvqp;
>
> assert(nc->info->type == NET_CLIENT_DRIVER_VHOST_VDPA);
>
> if (s->vhost_vdpa.index == 0) {
> migration_remove_notifier(&s->migration_state);
> }
> +
> + if (v->dev->vq_index + v->dev->nvqs != v->dev->vq_index_end) {
> + return;
> + }
> +
> + nvqp = (v->dev->vq_index_end + 1) / 2;
> + for (int i = 0; i < nvqp; ++i) {
> + VhostVDPAState *s_i = vhost_vdpa_net_get_nc_vdpa(s, i);
> + struct vhost_vdpa *v_i = &s_i->vhost_vdpa;
> +
> + if (!v_i->shadow_vqs_enabled) {
> + continue;
> + }
> + if (!last_vi) {
> + vhost_vdpa_dma_batch_begin_once(v_i->shared,
> + v_i->address_space_id);
> + last_vi = v_i;
> + } else if (last_vi->address_space_id != v_i->address_space_id) {
> + vhost_vdpa_dma_batch_end_once(last_vi->shared,
> + last_vi->address_space_id);
> + vhost_vdpa_dma_batch_begin_once(v_i->shared,
> + v_i->address_space_id);
> + last_vi = v_i;
> + }
> +
> + for (unsigned j = 0; j < v_i->shadow_vqs->len; ++j) {
> + VhostShadowVirtqueue *svq = g_ptr_array_index(v_i->shadow_vqs, j);
> +
> + vhost_vdpa_svq_unmap_rings(v_i->dev, svq);
> + }
> + }
> + if (has_cvq) {
> + if (last_vi) {
> + assert(last_vi->address_space_id == v->address_space_id);
> + }
> + vhost_vdpa_cvq_unmap_buf(&s->vhost_vdpa, s->cvq_cmd_out_buffer);
> + vhost_vdpa_cvq_unmap_buf(&s->vhost_vdpa, s->status);
> + }
> + if (last_vi) {
> + vhost_vdpa_dma_batch_end_once(last_vi->shared,
> + last_vi->address_space_id);
> + last_vi = NULL;
> + }
The logic looks rather complicated, can we simplify it by:
batch_begin_once()
unmap()
batch_end_once()
?
Thanks
> }
>
> static int vhost_vdpa_net_load_setup(NetClientState *nc, NICState *nic)
> @@ -585,33 +660,6 @@ static int64_t vhost_vdpa_get_vring_desc_group(int device_fd,
> return state.num;
> }
>
> -static void vhost_vdpa_cvq_unmap_buf(struct vhost_vdpa *v, void *addr)
> -{
> - VhostIOVATree *tree = v->shared->iova_tree;
> - DMAMap needle = {
> - /*
> - * No need to specify size or to look for more translations since
> - * this contiguous chunk was allocated by us.
> - */
> - .translated_addr = (hwaddr)(uintptr_t)addr,
> - };
> - const DMAMap *map = vhost_iova_tree_find_iova(tree, &needle);
> - int r;
> -
> - if (unlikely(!map)) {
> - error_report("Cannot locate expected map");
> - return;
> - }
> -
> - r = vhost_vdpa_dma_unmap(v->shared, v->address_space_id, map->iova,
> - map->size + 1);
> - if (unlikely(r != 0)) {
> - error_report("Device cannot unmap: %s(%d)", g_strerror(r), r);
> - }
> -
> - vhost_iova_tree_remove(tree, *map);
> -}
> -
> /** Map CVQ buffer. */
> static int vhost_vdpa_cvq_map_buf(struct vhost_vdpa *v, void *buf, size_t size,
> bool write)
> @@ -740,18 +788,8 @@ err:
>
> static void vhost_vdpa_net_cvq_stop(NetClientState *nc)
> {
> - VhostVDPAState *s = DO_UPCAST(VhostVDPAState, nc, nc);
> - struct vhost_vdpa *v = &s->vhost_vdpa;
> -
> assert(nc->info->type == NET_CLIENT_DRIVER_VHOST_VDPA);
>
> - if (s->vhost_vdpa.shadow_vqs_enabled) {
> - vhost_vdpa_dma_batch_begin_once(v->shared, v->address_space_id);
> - vhost_vdpa_cvq_unmap_buf(&s->vhost_vdpa, s->cvq_cmd_out_buffer);
> - vhost_vdpa_cvq_unmap_buf(&s->vhost_vdpa, s->status);
> - vhost_vdpa_dma_batch_end_once(v->shared, v->address_space_id);
> - }
> -
> vhost_vdpa_net_client_stop(nc);
> }
>
> --
> 1.8.3.1
>
On Thu, Dec 7, 2023 at 7:51 PM Si-Wei Liu <si-wei.liu@oracle.com> wrote:
>
> Should help live migration downtime on source host. Below are the
> coalesced dma_unmap time series on 2 queue pair config (no
> dedicated descriptor group ASID for SVQ).
>
> 109531@1693367276.853503:vhost_vdpa_reset_device dev: 0x55c933926890
> 109531@1693367276.853513:vhost_vdpa_add_status dev: 0x55c933926890 status: 0x3
> 109531@1693367276.853520:vhost_vdpa_flush_map dev: 0x55c933926890 doit: 1 svq_flush: 0 persist: 1
> 109531@1693367276.853524:vhost_vdpa_set_config_call dev: 0x55c933926890 fd: -1
> 109531@1693367276.853579:vhost_vdpa_iotlb_begin_batch vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 type: 5
> 109531@1693367276.853586:vhost_vdpa_dma_unmap vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 asid: 0 iova: 0x1000 size: 0x2000 type: 3
> 109531@1693367276.853600:vhost_vdpa_dma_unmap vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 asid: 0 iova: 0x3000 size: 0x1000 type: 3
> 109531@1693367276.853618:vhost_vdpa_dma_unmap vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 asid: 0 iova: 0x4000 size: 0x2000 type: 3
> 109531@1693367276.853625:vhost_vdpa_dma_unmap vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 asid: 0 iova: 0x6000 size: 0x1000 type: 3
> 109531@1693367276.853630:vhost_vdpa_dma_unmap vdpa:0x7fa2aa84c190 fd: 16 msg_type: 2 asid: 0 iova: 0x7000 size: 0x2000 type: 3
> 109531@1693367276.853636:vhost_vdpa_dma_unmap vdpa:0x7fa2aa84c190 fd: 16 msg_type: 2 asid: 0 iova: 0x9000 size: 0x1000 type: 3
> 109531@1693367276.853642:vhost_vdpa_dma_unmap vdpa:0x7fa2aa84c190 fd: 16 msg_type: 2 asid: 0 iova: 0xa000 size: 0x2000 type: 3
> 109531@1693367276.853648:vhost_vdpa_dma_unmap vdpa:0x7fa2aa84c190 fd: 16 msg_type: 2 asid: 0 iova: 0xc000 size: 0x1000 type: 3
> 109531@1693367276.853654:vhost_vdpa_dma_unmap vdpa:0x7fa2aa6b6190 fd: 16 msg_type: 2 asid: 0 iova: 0xf000 size: 0x1000 type: 3
> 109531@1693367276.853660:vhost_vdpa_dma_unmap vdpa:0x7fa2aa6b6190 fd: 16 msg_type: 2 asid: 0 iova: 0x10000 size: 0x1000 type: 3
> 109531@1693367276.853666:vhost_vdpa_dma_unmap vdpa:0x7fa2aa6b6190 fd: 16 msg_type: 2 asid: 0 iova: 0xd000 size: 0x1000 type: 3
> 109531@1693367276.853670:vhost_vdpa_dma_unmap vdpa:0x7fa2aa6b6190 fd: 16 msg_type: 2 asid: 0 iova: 0xe000 size: 0x1000 type: 3
> 109531@1693367276.853675:vhost_vdpa_iotlb_end_batch vdpa:0x7fa2aa895190 fd: 16 msg_type: 2 type: 6
> 109531@1693367277.014697:vhost_vdpa_get_vq_index dev: 0x55c933925de0 idx: 0 vq idx: 0
> 109531@1693367277.014747:vhost_vdpa_get_vq_index dev: 0x55c933925de0 idx: 1 vq idx: 1
> 109531@1693367277.014753:vhost_vdpa_get_vq_index dev: 0x55c9339262e0 idx: 2 vq idx: 2
> 109531@1693367277.014756:vhost_vdpa_get_vq_index dev: 0x55c9339262e0 idx: 3 vq idx: 3
>
> Signed-off-by: Si-Wei Liu <si-wei.liu@oracle.com>
> ---
> hw/virtio/vhost-vdpa.c | 7 +--
> include/hw/virtio/vhost-vdpa.h | 3 ++
> net/vhost-vdpa.c | 112 +++++++++++++++++++++++++++--------------
> 3 files changed, 80 insertions(+), 42 deletions(-)
>
> diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
> index d98704a..4010fd9 100644
> --- a/hw/virtio/vhost-vdpa.c
> +++ b/hw/virtio/vhost-vdpa.c
> @@ -1162,8 +1162,8 @@ static void vhost_vdpa_svq_unmap_ring(struct vhost_vdpa *v, hwaddr addr)
> vhost_iova_tree_remove(v->shared->iova_tree, *result);
> }
>
> -static void vhost_vdpa_svq_unmap_rings(struct vhost_dev *dev,
> - const VhostShadowVirtqueue *svq)
> +void vhost_vdpa_svq_unmap_rings(struct vhost_dev *dev,
> + const VhostShadowVirtqueue *svq)
> {
> struct vhost_vdpa *v = dev->opaque;
> struct vhost_vring_addr svq_addr;
> @@ -1346,17 +1346,14 @@ static void vhost_vdpa_svqs_stop(struct vhost_dev *dev)
> return;
> }
>
> - vhost_vdpa_dma_batch_begin_once(v->shared, v->address_space_id);
> for (unsigned i = 0; i < v->shadow_vqs->len; ++i) {
> VhostShadowVirtqueue *svq = g_ptr_array_index(v->shadow_vqs, i);
>
> vhost_svq_stop(svq);
> - vhost_vdpa_svq_unmap_rings(dev, svq);
>
> event_notifier_cleanup(&svq->hdev_kick);
> event_notifier_cleanup(&svq->hdev_call);
> }
> - vhost_vdpa_dma_batch_end_once(v->shared, v->address_space_id);
> }
>
> static void vhost_vdpa_suspend(struct vhost_dev *dev)
> diff --git a/include/hw/virtio/vhost-vdpa.h b/include/hw/virtio/vhost-vdpa.h
> index aa13679..f426e2c 100644
> --- a/include/hw/virtio/vhost-vdpa.h
> +++ b/include/hw/virtio/vhost-vdpa.h
> @@ -112,6 +112,9 @@ int vhost_vdpa_dma_batch_end_once(VhostVDPAShared *s, uint32_t asid);
> int vhost_vdpa_load_setup(VhostVDPAShared *s, AddressSpace *dma_as);
> int vhost_vdpa_load_cleanup(VhostVDPAShared *s, bool vhost_will_start);
>
> +void vhost_vdpa_svq_unmap_rings(struct vhost_dev *dev,
> + const VhostShadowVirtqueue *svq);
> +
> typedef struct vdpa_iommu {
> VhostVDPAShared *dev_shared;
> IOMMUMemoryRegion *iommu_mr;
> diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c
> index 683619f..41714d1 100644
> --- a/net/vhost-vdpa.c
> +++ b/net/vhost-vdpa.c
> @@ -29,6 +29,7 @@
> #include "migration/migration.h"
> #include "migration/misc.h"
> #include "hw/virtio/vhost.h"
> +#include "hw/virtio/vhost-vdpa.h"
>
> /* Todo:need to add the multiqueue support here */
> typedef struct VhostVDPAState {
> @@ -467,15 +468,89 @@ static int vhost_vdpa_net_data_load(NetClientState *nc)
> return 0;
> }
>
> +static void vhost_vdpa_cvq_unmap_buf(struct vhost_vdpa *v, void *addr)
> +{
> + VhostIOVATree *tree = v->shared->iova_tree;
> + DMAMap needle = {
> + /*
> + * No need to specify size or to look for more translations since
> + * this contiguous chunk was allocated by us.
> + */
> + .translated_addr = (hwaddr)(uintptr_t)addr,
> + };
> + const DMAMap *map = vhost_iova_tree_find_iova(tree, &needle);
> + int r;
> +
> + if (unlikely(!map)) {
> + error_report("Cannot locate expected map");
> + return;
> + }
> +
> + r = vhost_vdpa_dma_unmap(v->shared, v->address_space_id, map->iova,
> + map->size + 1);
> + if (unlikely(r != 0)) {
> + error_report("Device cannot unmap: %s(%d)", g_strerror(r), r);
> + }
> +
> + vhost_iova_tree_remove(tree, *map);
> +}
> +
> static void vhost_vdpa_net_client_stop(NetClientState *nc)
> {
> VhostVDPAState *s = DO_UPCAST(VhostVDPAState, nc, nc);
> + struct vhost_vdpa *v = &s->vhost_vdpa;
> + struct vhost_vdpa *last_vi = NULL;
> + bool has_cvq = v->dev->vq_index_end % 2;
> + int nvqp;
>
> assert(nc->info->type == NET_CLIENT_DRIVER_VHOST_VDPA);
>
> if (s->vhost_vdpa.index == 0) {
> migration_remove_notifier(&s->migration_state);
> }
> +
> + if (v->dev->vq_index + v->dev->nvqs != v->dev->vq_index_end) {
> + return;
> + }
> +
> + nvqp = (v->dev->vq_index_end + 1) / 2;
> + for (int i = 0; i < nvqp; ++i) {
> + VhostVDPAState *s_i = vhost_vdpa_net_get_nc_vdpa(s, i);
> + struct vhost_vdpa *v_i = &s_i->vhost_vdpa;
> +
> + if (!v_i->shadow_vqs_enabled) {
> + continue;
> + }
> + if (!last_vi) {
> + vhost_vdpa_dma_batch_begin_once(v_i->shared,
> + v_i->address_space_id);
> + last_vi = v_i;
> + } else if (last_vi->address_space_id != v_i->address_space_id) {
> + vhost_vdpa_dma_batch_end_once(last_vi->shared,
> + last_vi->address_space_id);
> + vhost_vdpa_dma_batch_begin_once(v_i->shared,
> + v_i->address_space_id);
> + last_vi = v_i;
> + }
> +
> + for (unsigned j = 0; j < v_i->shadow_vqs->len; ++j) {
> + VhostShadowVirtqueue *svq = g_ptr_array_index(v_i->shadow_vqs, j);
> +
> + vhost_vdpa_svq_unmap_rings(v_i->dev, svq);
> + }
> + }
> + if (has_cvq) {
> + if (last_vi) {
> + assert(last_vi->address_space_id == v->address_space_id);
> + }
> + vhost_vdpa_cvq_unmap_buf(&s->vhost_vdpa, s->cvq_cmd_out_buffer);
> + vhost_vdpa_cvq_unmap_buf(&s->vhost_vdpa, s->status);
> + }
> + if (last_vi) {
> + vhost_vdpa_dma_batch_end_once(last_vi->shared,
> + last_vi->address_space_id);
> + last_vi = NULL;
> + }
Since we've delayed the guest unmap memory to _cleanup, why not delay
these unmaps to cleanup too?
> }
>
> static int vhost_vdpa_net_load_setup(NetClientState *nc, NICState *nic)
> @@ -585,33 +660,6 @@ static int64_t vhost_vdpa_get_vring_desc_group(int device_fd,
> return state.num;
> }
>
> -static void vhost_vdpa_cvq_unmap_buf(struct vhost_vdpa *v, void *addr)
> -{
> - VhostIOVATree *tree = v->shared->iova_tree;
> - DMAMap needle = {
> - /*
> - * No need to specify size or to look for more translations since
> - * this contiguous chunk was allocated by us.
> - */
> - .translated_addr = (hwaddr)(uintptr_t)addr,
> - };
> - const DMAMap *map = vhost_iova_tree_find_iova(tree, &needle);
> - int r;
> -
> - if (unlikely(!map)) {
> - error_report("Cannot locate expected map");
> - return;
> - }
> -
> - r = vhost_vdpa_dma_unmap(v->shared, v->address_space_id, map->iova,
> - map->size + 1);
> - if (unlikely(r != 0)) {
> - error_report("Device cannot unmap: %s(%d)", g_strerror(r), r);
> - }
> -
> - vhost_iova_tree_remove(tree, *map);
> -}
> -
> /** Map CVQ buffer. */
> static int vhost_vdpa_cvq_map_buf(struct vhost_vdpa *v, void *buf, size_t size,
> bool write)
> @@ -740,18 +788,8 @@ err:
>
> static void vhost_vdpa_net_cvq_stop(NetClientState *nc)
> {
> - VhostVDPAState *s = DO_UPCAST(VhostVDPAState, nc, nc);
> - struct vhost_vdpa *v = &s->vhost_vdpa;
> -
> assert(nc->info->type == NET_CLIENT_DRIVER_VHOST_VDPA);
>
> - if (s->vhost_vdpa.shadow_vqs_enabled) {
> - vhost_vdpa_dma_batch_begin_once(v->shared, v->address_space_id);
> - vhost_vdpa_cvq_unmap_buf(&s->vhost_vdpa, s->cvq_cmd_out_buffer);
> - vhost_vdpa_cvq_unmap_buf(&s->vhost_vdpa, s->status);
> - vhost_vdpa_dma_batch_end_once(v->shared, v->address_space_id);
> - }
> -
> vhost_vdpa_net_client_stop(nc);
> }
>
> --
> 1.8.3.1
>
© 2016 - 2026 Red Hat, Inc.