* [PATCH v2] vhost-vdpa: add support for vIOMMU
@ 2022-10-19 6:39 Cindy Lu
2022-10-20 2:14 ` Jason Wang
0 siblings, 1 reply; 5+ messages in thread
From: Cindy Lu @ 2022-10-19 6:39 UTC (permalink / raw)
To: lulu, mst, jasowang, eperezma; +Cc: qemu-devel, stefanha
Add support for vIOMMU. Register a memory listener to dma_as in
dev_start
- during region_add register a specific IOMMU notifier, and store all notifiers in a list.
- during region_del, compare and delete the IOMMU notifier from the list
Verified in vp_vdpa and vdpa_sim_net driver
Signed-off-by: Cindy Lu <lulu@redhat.com>
---
hw/virtio/vhost-vdpa.c | 218 ++++++++++++++++++++++++++++++++-
include/hw/virtio/vhost-vdpa.h | 11 ++
2 files changed, 226 insertions(+), 3 deletions(-)
diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
index 3ff9ce3501..8f36423c28 100644
--- a/hw/virtio/vhost-vdpa.c
+++ b/hw/virtio/vhost-vdpa.c
@@ -26,6 +26,7 @@
#include "cpu.h"
#include "trace.h"
#include "qapi/error.h"
+#include "hw/virtio/virtio-access.h"
/*
* Return one past the end of the end of section. Be careful with uint64_t
@@ -146,9 +147,8 @@ static void vhost_vdpa_iotlb_batch_begin_once(struct vhost_vdpa *v)
v->iotlb_batch_begin_sent = true;
}
-static void vhost_vdpa_listener_commit(MemoryListener *listener)
+static void vhost_vdpa_iotlb_batch_end(struct vhost_vdpa *v)
{
- struct vhost_vdpa *v = container_of(listener, struct vhost_vdpa, listener);
struct vhost_dev *dev = v->dev;
struct vhost_msg_v2 msg = {};
int fd = v->device_fd;
@@ -172,6 +172,18 @@ static void vhost_vdpa_listener_commit(MemoryListener *listener)
v->iotlb_batch_begin_sent = false;
}
+static void vhost_vdpa_listener_commit(MemoryListener *listener)
+{
+ struct vhost_vdpa *v = container_of(listener, struct vhost_vdpa, listener);
+ return vhost_vdpa_iotlb_batch_end(v);
+}
+
+static void vhost_vdpa_listener_iommu_commit(MemoryListener *listener)
+{
+ struct vhost_vdpa *v =
+ container_of(listener, struct vhost_vdpa, iommu_listener);
+ return vhost_vdpa_iotlb_batch_end(v);
+}
static void vhost_vdpa_listener_region_add(MemoryListener *listener,
MemoryRegionSection *section)
@@ -186,6 +198,9 @@ static void vhost_vdpa_listener_region_add(MemoryListener *listener,
v->iova_range.last)) {
return;
}
+ if (memory_region_is_iommu(section->mr)) {
+ return;
+ }
if (unlikely((section->offset_within_address_space & ~TARGET_PAGE_MASK) !=
(section->offset_within_region & ~TARGET_PAGE_MASK))) {
@@ -260,6 +275,9 @@ static void vhost_vdpa_listener_region_del(MemoryListener *listener,
v->iova_range.last)) {
return;
}
+ if (memory_region_is_iommu(section->mr)) {
+ return;
+ }
if (unlikely((section->offset_within_address_space & ~TARGET_PAGE_MASK) !=
(section->offset_within_region & ~TARGET_PAGE_MASK))) {
@@ -312,6 +330,180 @@ static const MemoryListener vhost_vdpa_memory_listener = {
.region_del = vhost_vdpa_listener_region_del,
};
+/* Called with rcu_read_lock held. */
+static bool vhost_vdpa_get_xlat_addr(IOMMUTLBEntry *iotlb, void **vaddr,
+ ram_addr_t *ram_addr, bool *read_only)
+{
+ MemoryRegion *mr;
+ hwaddr xlat;
+ hwaddr len = iotlb->addr_mask + 1;
+ bool writable = iotlb->perm & IOMMU_WO;
+
+ /*
+ * The IOMMU TLB entry we have just covers translation through
+ * this IOMMU to its immediate target. We need to translate
+ * it the rest of the way through to memory.
+ */
+ mr = address_space_translate(&address_space_memory, iotlb->translated_addr,
+ &xlat, &len, writable, MEMTXATTRS_UNSPECIFIED);
+ if (!memory_region_is_ram(mr)) {
+ error_report("iommu map to non memory area %" HWADDR_PRIx "", xlat);
+ return false;
+ } else if (memory_region_has_ram_discard_manager(mr)) {
+ RamDiscardManager *rdm = memory_region_get_ram_discard_manager(mr);
+ MemoryRegionSection tmp = {
+ .mr = mr,
+ .offset_within_region = xlat,
+ .size = int128_make64(len),
+ };
+
+ if (!ram_discard_manager_is_populated(rdm, &tmp)) {
+ error_report("iommu map to discarded memory (e.g., unplugged via"
+ " virtio-mem): %" HWADDR_PRIx "",
+ iotlb->translated_addr);
+ return false;
+ }
+ }
+ /*
+ * Translation truncates length to the IOMMU page size,
+ * check that it did not truncate too much.
+ */
+ if (len & iotlb->addr_mask) {
+ error_report("iommu has granularity incompatible with target AS");
+ return false;
+ }
+
+ if (vaddr) {
+ *vaddr = memory_region_get_ram_ptr(mr) + xlat;
+ }
+
+ if (ram_addr) {
+ *ram_addr = memory_region_get_ram_addr(mr) + xlat;
+ }
+
+ if (read_only) {
+ *read_only = !writable || mr->readonly;
+ }
+
+ return true;
+}
+
+static void vhost_vdpa_iommu_map_notify(IOMMUNotifier *n, IOMMUTLBEntry *iotlb)
+{
+ struct vdpa_iommu *iommu = container_of(n, struct vdpa_iommu, n);
+
+ hwaddr iova = iotlb->iova + iommu->iommu_offset;
+ struct vhost_vdpa *v = iommu->dev;
+ void *vaddr;
+ int ret;
+
+ if (iotlb->target_as != &address_space_memory) {
+ error_report("Wrong target AS \"%s\", only system memory is allowed",
+ iotlb->target_as->name ? iotlb->target_as->name : "none");
+ return;
+ }
+ RCU_READ_LOCK_GUARD();
+ vhost_vdpa_iotlb_batch_begin_once(v);
+
+ if ((iotlb->perm & IOMMU_RW) != IOMMU_NONE) {
+ bool read_only;
+
+ if (!vhost_vdpa_get_xlat_addr(iotlb, &vaddr, NULL, &read_only)) {
+ return;
+ }
+ ret =
+ vhost_vdpa_dma_map(v, iova, iotlb->addr_mask + 1, vaddr, read_only);
+ if (ret) {
+ error_report("vhost_vdpa_dma_map(%p, 0x%" HWADDR_PRIx ", "
+ "0x%" HWADDR_PRIx ", %p) = %d (%m)",
+ v, iova, iotlb->addr_mask + 1, vaddr, ret);
+ }
+ } else {
+ ret = vhost_vdpa_dma_unmap(v, iova, iotlb->addr_mask + 1);
+ if (ret) {
+ error_report("vhost_vdpa_dma_unmap(%p, 0x%" HWADDR_PRIx ", "
+ "0x%" HWADDR_PRIx ") = %d (%m)",
+ v, iova, iotlb->addr_mask + 1, ret);
+ }
+ }
+}
+
+static void vhost_vdpa_iommu_region_add(MemoryListener *listener,
+ MemoryRegionSection *section)
+{
+ struct vhost_vdpa *v =
+ container_of(listener, struct vhost_vdpa, iommu_listener);
+
+ struct vdpa_iommu *iommu;
+ Int128 end;
+ int iommu_idx;
+ IOMMUMemoryRegion *iommu_mr;
+ int ret;
+
+ if (!memory_region_is_iommu(section->mr)) {
+ return;
+ }
+
+ iommu_mr = IOMMU_MEMORY_REGION(section->mr);
+
+ iommu = g_malloc0(sizeof(*iommu));
+ end =
+ int128_add(int128_make64(section->offset_within_region), section->size);
+ end = int128_sub(end, int128_one());
+ iommu_idx =
+ memory_region_iommu_attrs_to_index(iommu_mr, MEMTXATTRS_UNSPECIFIED);
+ iommu->iommu_mr = iommu_mr;
+
+ iommu_notifier_init(
+ &iommu->n, vhost_vdpa_iommu_map_notify, IOMMU_NOTIFIER_IOTLB_EVENTS,
+ section->offset_within_region, int128_get64(end), iommu_idx);
+ iommu->iommu_offset =
+ section->offset_within_address_space - section->offset_within_region;
+ iommu->dev = v;
+
+ ret = memory_region_register_iommu_notifier(section->mr, &iommu->n, NULL);
+ if (ret) {
+ g_free(iommu);
+ return;
+ }
+
+ QLIST_INSERT_HEAD(&v->iommu_list, iommu, iommu_next);
+ memory_region_iommu_replay(iommu->iommu_mr, &iommu->n);
+
+ return;
+}
+
+static void vhost_vdpa_iommu_region_del(MemoryListener *listener,
+ MemoryRegionSection *section)
+{
+ struct vhost_vdpa *v =
+ container_of(listener, struct vhost_vdpa, iommu_listener);
+
+ struct vdpa_iommu *iommu;
+
+ if (!memory_region_is_iommu(section->mr)) {
+ return;
+ }
+
+ QLIST_FOREACH(iommu, &v->iommu_list, iommu_next)
+ {
+ if (MEMORY_REGION(iommu->iommu_mr) == section->mr &&
+ iommu->n.start == section->offset_within_region) {
+ memory_region_unregister_iommu_notifier(section->mr, &iommu->n);
+ QLIST_REMOVE(iommu, iommu_next);
+ g_free(iommu);
+ break;
+ }
+ }
+}
+
+static const MemoryListener vhost_vdpa_iommu_listener = {
+ .name = "vhost-vdpa-iommu",
+ .region_add = vhost_vdpa_iommu_region_add,
+ .region_del = vhost_vdpa_iommu_region_del,
+ .commit = vhost_vdpa_listener_iommu_commit,
+};
+
static int vhost_vdpa_call(struct vhost_dev *dev, unsigned long int request,
void *arg)
{
@@ -433,6 +625,18 @@ static int vhost_vdpa_init_svq(struct vhost_dev *hdev, struct vhost_vdpa *v,
return 0;
}
+static bool vdpa_dev_has_iommu(struct vhost_dev *dev)
+{
+ VirtIODevice *vdev = dev->vdev;
+
+ if (vdev) {
+ return virtio_bus_device_iommu_enabled(vdev) &&
+ virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM);
+ } else {
+ return false;
+ }
+}
+
static int vhost_vdpa_init(struct vhost_dev *dev, void *opaque, Error **errp)
{
struct vhost_vdpa *v;
@@ -454,6 +658,7 @@ static int vhost_vdpa_init(struct vhost_dev *dev, void *opaque, Error **errp)
v->dev = dev;
dev->opaque = opaque ;
v->listener = vhost_vdpa_memory_listener;
+ v->iommu_listener = vhost_vdpa_iommu_listener;
v->msg_type = VHOST_IOTLB_MSG_V2;
ret = vhost_vdpa_init_svq(dev, v, errp);
if (ret) {
@@ -587,7 +792,6 @@ static int vhost_vdpa_cleanup(struct vhost_dev *dev)
v = dev->opaque;
trace_vhost_vdpa_cleanup(dev, v);
vhost_vdpa_host_notifiers_uninit(dev, dev->nvqs);
- memory_listener_unregister(&v->listener);
vhost_vdpa_svq_cleanup(dev);
dev->opaque = NULL;
@@ -1128,11 +1332,19 @@ static int vhost_vdpa_dev_start(struct vhost_dev *dev, bool started)
if (started) {
memory_listener_register(&v->listener, &address_space_memory);
+ if (vdpa_dev_has_iommu(dev)) {
+ memory_listener_register(&v->iommu_listener, dev->vdev->dma_as);
+ }
+
return vhost_vdpa_add_status(dev, VIRTIO_CONFIG_S_DRIVER_OK);
} else {
vhost_vdpa_reset_device(dev);
vhost_vdpa_add_status(dev, VIRTIO_CONFIG_S_ACKNOWLEDGE |
VIRTIO_CONFIG_S_DRIVER);
+
+ if (vdpa_dev_has_iommu(dev)) {
+ memory_listener_unregister(&v->iommu_listener);
+ }
memory_listener_unregister(&v->listener);
return 0;
diff --git a/include/hw/virtio/vhost-vdpa.h b/include/hw/virtio/vhost-vdpa.h
index d10a89303e..22cbce7934 100644
--- a/include/hw/virtio/vhost-vdpa.h
+++ b/include/hw/virtio/vhost-vdpa.h
@@ -30,6 +30,7 @@ typedef struct vhost_vdpa {
uint32_t msg_type;
bool iotlb_batch_begin_sent;
MemoryListener listener;
+ MemoryListener iommu_listener;
struct vhost_vdpa_iova_range iova_range;
uint64_t acked_features;
bool shadow_vqs_enabled;
@@ -41,8 +42,18 @@ typedef struct vhost_vdpa {
void *shadow_vq_ops_opaque;
struct vhost_dev *dev;
VhostVDPAHostNotifier notifier[VIRTIO_QUEUE_MAX];
+ QLIST_HEAD(, vdpa_iommu) iommu_list;
+ IOMMUNotifier n;
} VhostVDPA;
+struct vdpa_iommu {
+ struct vhost_vdpa *dev;
+ IOMMUMemoryRegion *iommu_mr;
+ hwaddr iommu_offset;
+ IOMMUNotifier n;
+ QLIST_ENTRY(vdpa_iommu) iommu_next;
+};
+
int vhost_vdpa_dma_map(struct vhost_vdpa *v, hwaddr iova, hwaddr size,
void *vaddr, bool readonly);
int vhost_vdpa_dma_unmap(struct vhost_vdpa *v, hwaddr iova, hwaddr size);
--
2.34.3
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH v2] vhost-vdpa: add support for vIOMMU
2022-10-19 6:39 [PATCH v2] vhost-vdpa: add support for vIOMMU Cindy Lu
@ 2022-10-20 2:14 ` Jason Wang
2022-10-23 14:45 ` Cindy Lu
0 siblings, 1 reply; 5+ messages in thread
From: Jason Wang @ 2022-10-20 2:14 UTC (permalink / raw)
To: Cindy Lu; +Cc: mst, eperezma, qemu-devel, stefanha
On Wed, Oct 19, 2022 at 2:39 PM Cindy Lu <lulu@redhat.com> wrote:
>
> Add support for vIOMMU. Register a memory listener to dma_as in
> dev_start
> - during region_add register a specific IOMMU notifier, and store all notifiers in a list.
> - during region_del, compare and delete the IOMMU notifier from the list
>
> Verified in vp_vdpa and vdpa_sim_net driver
>
> Signed-off-by: Cindy Lu <lulu@redhat.com>
> ---
> hw/virtio/vhost-vdpa.c | 218 ++++++++++++++++++++++++++++++++-
> include/hw/virtio/vhost-vdpa.h | 11 ++
> 2 files changed, 226 insertions(+), 3 deletions(-)
>
> diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
> index 3ff9ce3501..8f36423c28 100644
> --- a/hw/virtio/vhost-vdpa.c
> +++ b/hw/virtio/vhost-vdpa.c
> @@ -26,6 +26,7 @@
> #include "cpu.h"
> #include "trace.h"
> #include "qapi/error.h"
> +#include "hw/virtio/virtio-access.h"
>
> /*
> * Return one past the end of the end of section. Be careful with uint64_t
> @@ -146,9 +147,8 @@ static void vhost_vdpa_iotlb_batch_begin_once(struct vhost_vdpa *v)
> v->iotlb_batch_begin_sent = true;
> }
>
> -static void vhost_vdpa_listener_commit(MemoryListener *listener)
> +static void vhost_vdpa_iotlb_batch_end(struct vhost_vdpa *v)
> {
> - struct vhost_vdpa *v = container_of(listener, struct vhost_vdpa, listener);
> struct vhost_dev *dev = v->dev;
> struct vhost_msg_v2 msg = {};
> int fd = v->device_fd;
> @@ -172,6 +172,18 @@ static void vhost_vdpa_listener_commit(MemoryListener *listener)
>
> v->iotlb_batch_begin_sent = false;
> }
> +static void vhost_vdpa_listener_commit(MemoryListener *listener)
> +{
> + struct vhost_vdpa *v = container_of(listener, struct vhost_vdpa, listener);
> + return vhost_vdpa_iotlb_batch_end(v);
> +}
> +
> +static void vhost_vdpa_listener_iommu_commit(MemoryListener *listener)
> +{
> + struct vhost_vdpa *v =
> + container_of(listener, struct vhost_vdpa, iommu_listener);
> + return vhost_vdpa_iotlb_batch_end(v);
> +}
>
> static void vhost_vdpa_listener_region_add(MemoryListener *listener,
> MemoryRegionSection *section)
> @@ -186,6 +198,9 @@ static void vhost_vdpa_listener_region_add(MemoryListener *listener,
> v->iova_range.last)) {
> return;
> }
> + if (memory_region_is_iommu(section->mr)) {
> + return;
> + }
>
> if (unlikely((section->offset_within_address_space & ~TARGET_PAGE_MASK) !=
> (section->offset_within_region & ~TARGET_PAGE_MASK))) {
> @@ -260,6 +275,9 @@ static void vhost_vdpa_listener_region_del(MemoryListener *listener,
> v->iova_range.last)) {
> return;
> }
> + if (memory_region_is_iommu(section->mr)) {
> + return;
> + }
>
> if (unlikely((section->offset_within_address_space & ~TARGET_PAGE_MASK) !=
> (section->offset_within_region & ~TARGET_PAGE_MASK))) {
> @@ -312,6 +330,180 @@ static const MemoryListener vhost_vdpa_memory_listener = {
> .region_del = vhost_vdpa_listener_region_del,
> };
>
> +/* Called with rcu_read_lock held. */
> +static bool vhost_vdpa_get_xlat_addr(IOMMUTLBEntry *iotlb, void **vaddr,
> + ram_addr_t *ram_addr, bool *read_only)
So this looks exact the same as what VFIO did, try to unify them?
> +{
> + MemoryRegion *mr;
> + hwaddr xlat;
> + hwaddr len = iotlb->addr_mask + 1;
> + bool writable = iotlb->perm & IOMMU_WO;
> +
> + /*
> + * The IOMMU TLB entry we have just covers translation through
> + * this IOMMU to its immediate target. We need to translate
> + * it the rest of the way through to memory.
> + */
> + mr = address_space_translate(&address_space_memory, iotlb->translated_addr,
> + &xlat, &len, writable, MEMTXATTRS_UNSPECIFIED);
> + if (!memory_region_is_ram(mr)) {
> + error_report("iommu map to non memory area %" HWADDR_PRIx "", xlat);
> + return false;
> + } else if (memory_region_has_ram_discard_manager(mr)) {
> + RamDiscardManager *rdm = memory_region_get_ram_discard_manager(mr);
> + MemoryRegionSection tmp = {
> + .mr = mr,
> + .offset_within_region = xlat,
> + .size = int128_make64(len),
> + };
> +
> + if (!ram_discard_manager_is_populated(rdm, &tmp)) {
> + error_report("iommu map to discarded memory (e.g., unplugged via"
> + " virtio-mem): %" HWADDR_PRIx "",
> + iotlb->translated_addr);
> + return false;
> + }
> + }
> + /*
> + * Translation truncates length to the IOMMU page size,
> + * check that it did not truncate too much.
> + */
> + if (len & iotlb->addr_mask) {
> + error_report("iommu has granularity incompatible with target AS");
> + return false;
> + }
> +
> + if (vaddr) {
> + *vaddr = memory_region_get_ram_ptr(mr) + xlat;
> + }
> +
> + if (ram_addr) {
> + *ram_addr = memory_region_get_ram_addr(mr) + xlat;
> + }
> +
> + if (read_only) {
> + *read_only = !writable || mr->readonly;
> + }
> +
> + return true;
> +}
> +
> +static void vhost_vdpa_iommu_map_notify(IOMMUNotifier *n, IOMMUTLBEntry *iotlb)
> +{
> + struct vdpa_iommu *iommu = container_of(n, struct vdpa_iommu, n);
> +
> + hwaddr iova = iotlb->iova + iommu->iommu_offset;
> + struct vhost_vdpa *v = iommu->dev;
> + void *vaddr;
> + int ret;
> +
> + if (iotlb->target_as != &address_space_memory) {
> + error_report("Wrong target AS \"%s\", only system memory is allowed",
> + iotlb->target_as->name ? iotlb->target_as->name : "none");
> + return;
> + }
> + RCU_READ_LOCK_GUARD();
> + vhost_vdpa_iotlb_batch_begin_once(v);
> +
> + if ((iotlb->perm & IOMMU_RW) != IOMMU_NONE) {
> + bool read_only;
> +
> + if (!vhost_vdpa_get_xlat_addr(iotlb, &vaddr, NULL, &read_only)) {
> + return;
> + }
> + ret =
> + vhost_vdpa_dma_map(v, iova, iotlb->addr_mask + 1, vaddr, read_only);
> + if (ret) {
> + error_report("vhost_vdpa_dma_map(%p, 0x%" HWADDR_PRIx ", "
> + "0x%" HWADDR_PRIx ", %p) = %d (%m)",
> + v, iova, iotlb->addr_mask + 1, vaddr, ret);
> + }
> + } else {
> + ret = vhost_vdpa_dma_unmap(v, iova, iotlb->addr_mask + 1);
> + if (ret) {
> + error_report("vhost_vdpa_dma_unmap(%p, 0x%" HWADDR_PRIx ", "
> + "0x%" HWADDR_PRIx ") = %d (%m)",
> + v, iova, iotlb->addr_mask + 1, ret);
> + }
> + }
> +}
> +
> +static void vhost_vdpa_iommu_region_add(MemoryListener *listener,
> + MemoryRegionSection *section)
> +{
> + struct vhost_vdpa *v =
> + container_of(listener, struct vhost_vdpa, iommu_listener);
> +
> + struct vdpa_iommu *iommu;
> + Int128 end;
> + int iommu_idx;
> + IOMMUMemoryRegion *iommu_mr;
> + int ret;
> +
> + if (!memory_region_is_iommu(section->mr)) {
> + return;
> + }
> +
> + iommu_mr = IOMMU_MEMORY_REGION(section->mr);
> +
> + iommu = g_malloc0(sizeof(*iommu));
> + end =
> + int128_add(int128_make64(section->offset_within_region), section->size);
I'd suggest to tweak the line breaking here and below, e.g you can
keep int128_add and other stuffs as long as it does not exceeds 70 or
80 characters at the line where "end =" sits.
> + end = int128_sub(end, int128_one());
> + iommu_idx =
> + memory_region_iommu_attrs_to_index(iommu_mr, MEMTXATTRS_UNSPECIFIED);
> + iommu->iommu_mr = iommu_mr;
> +
> + iommu_notifier_init(
> + &iommu->n, vhost_vdpa_iommu_map_notify, IOMMU_NOTIFIER_IOTLB_EVENTS,
> + section->offset_within_region, int128_get64(end), iommu_idx);
> + iommu->iommu_offset =
> + section->offset_within_address_space - section->offset_within_region;
> + iommu->dev = v;
> +
> + ret = memory_region_register_iommu_notifier(section->mr, &iommu->n, NULL);
> + if (ret) {
> + g_free(iommu);
> + return;
> + }
> +
> + QLIST_INSERT_HEAD(&v->iommu_list, iommu, iommu_next);
> + memory_region_iommu_replay(iommu->iommu_mr, &iommu->n);
> +
> + return;
> +}
> +
> +static void vhost_vdpa_iommu_region_del(MemoryListener *listener,
> + MemoryRegionSection *section)
> +{
> + struct vhost_vdpa *v =
> + container_of(listener, struct vhost_vdpa, iommu_listener);
> +
> + struct vdpa_iommu *iommu;
> +
> + if (!memory_region_is_iommu(section->mr)) {
> + return;
> + }
> +
> + QLIST_FOREACH(iommu, &v->iommu_list, iommu_next)
> + {
> + if (MEMORY_REGION(iommu->iommu_mr) == section->mr &&
> + iommu->n.start == section->offset_within_region) {
> + memory_region_unregister_iommu_notifier(section->mr, &iommu->n);
> + QLIST_REMOVE(iommu, iommu_next);
> + g_free(iommu);
> + break;
> + }
> + }
> +}
> +
> +static const MemoryListener vhost_vdpa_iommu_listener = {
> + .name = "vhost-vdpa-iommu",
> + .region_add = vhost_vdpa_iommu_region_add,
> + .region_del = vhost_vdpa_iommu_region_del,
> + .commit = vhost_vdpa_listener_iommu_commit,
Unless there's a special requirement. I'd suggest to reuse the
existing listener (since you've already had if (iommu_region()) check
there).
Thanks
> +};
> +
> static int vhost_vdpa_call(struct vhost_dev *dev, unsigned long int request,
> void *arg)
> {
> @@ -433,6 +625,18 @@ static int vhost_vdpa_init_svq(struct vhost_dev *hdev, struct vhost_vdpa *v,
> return 0;
> }
>
> +static bool vdpa_dev_has_iommu(struct vhost_dev *dev)
> +{
> + VirtIODevice *vdev = dev->vdev;
> +
> + if (vdev) {
> + return virtio_bus_device_iommu_enabled(vdev) &&
> + virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM);
> + } else {
> + return false;
> + }
> +}
> +
> static int vhost_vdpa_init(struct vhost_dev *dev, void *opaque, Error **errp)
> {
> struct vhost_vdpa *v;
> @@ -454,6 +658,7 @@ static int vhost_vdpa_init(struct vhost_dev *dev, void *opaque, Error **errp)
> v->dev = dev;
> dev->opaque = opaque ;
> v->listener = vhost_vdpa_memory_listener;
> + v->iommu_listener = vhost_vdpa_iommu_listener;
> v->msg_type = VHOST_IOTLB_MSG_V2;
> ret = vhost_vdpa_init_svq(dev, v, errp);
> if (ret) {
> @@ -587,7 +792,6 @@ static int vhost_vdpa_cleanup(struct vhost_dev *dev)
> v = dev->opaque;
> trace_vhost_vdpa_cleanup(dev, v);
> vhost_vdpa_host_notifiers_uninit(dev, dev->nvqs);
> - memory_listener_unregister(&v->listener);
> vhost_vdpa_svq_cleanup(dev);
>
> dev->opaque = NULL;
> @@ -1128,11 +1332,19 @@ static int vhost_vdpa_dev_start(struct vhost_dev *dev, bool started)
>
> if (started) {
> memory_listener_register(&v->listener, &address_space_memory);
> + if (vdpa_dev_has_iommu(dev)) {
> + memory_listener_register(&v->iommu_listener, dev->vdev->dma_as);
> + }
> +
> return vhost_vdpa_add_status(dev, VIRTIO_CONFIG_S_DRIVER_OK);
> } else {
> vhost_vdpa_reset_device(dev);
> vhost_vdpa_add_status(dev, VIRTIO_CONFIG_S_ACKNOWLEDGE |
> VIRTIO_CONFIG_S_DRIVER);
> +
> + if (vdpa_dev_has_iommu(dev)) {
> + memory_listener_unregister(&v->iommu_listener);
> + }
> memory_listener_unregister(&v->listener);
>
> return 0;
> diff --git a/include/hw/virtio/vhost-vdpa.h b/include/hw/virtio/vhost-vdpa.h
> index d10a89303e..22cbce7934 100644
> --- a/include/hw/virtio/vhost-vdpa.h
> +++ b/include/hw/virtio/vhost-vdpa.h
> @@ -30,6 +30,7 @@ typedef struct vhost_vdpa {
> uint32_t msg_type;
> bool iotlb_batch_begin_sent;
> MemoryListener listener;
> + MemoryListener iommu_listener;
> struct vhost_vdpa_iova_range iova_range;
> uint64_t acked_features;
> bool shadow_vqs_enabled;
> @@ -41,8 +42,18 @@ typedef struct vhost_vdpa {
> void *shadow_vq_ops_opaque;
> struct vhost_dev *dev;
> VhostVDPAHostNotifier notifier[VIRTIO_QUEUE_MAX];
> + QLIST_HEAD(, vdpa_iommu) iommu_list;
> + IOMMUNotifier n;
> } VhostVDPA;
>
> +struct vdpa_iommu {
> + struct vhost_vdpa *dev;
> + IOMMUMemoryRegion *iommu_mr;
> + hwaddr iommu_offset;
> + IOMMUNotifier n;
> + QLIST_ENTRY(vdpa_iommu) iommu_next;
> +};
> +
> int vhost_vdpa_dma_map(struct vhost_vdpa *v, hwaddr iova, hwaddr size,
> void *vaddr, bool readonly);
> int vhost_vdpa_dma_unmap(struct vhost_vdpa *v, hwaddr iova, hwaddr size);
> --
> 2.34.3
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2] vhost-vdpa: add support for vIOMMU
2022-10-20 2:14 ` Jason Wang
@ 2022-10-23 14:45 ` Cindy Lu
2022-10-24 1:51 ` Jason Wang
0 siblings, 1 reply; 5+ messages in thread
From: Cindy Lu @ 2022-10-23 14:45 UTC (permalink / raw)
To: Jason Wang; +Cc: mst, eperezma, qemu-devel, stefanha
On Thu, 20 Oct 2022 at 10:15, Jason Wang <jasowang@redhat.com> wrote:
>
> On Wed, Oct 19, 2022 at 2:39 PM Cindy Lu <lulu@redhat.com> wrote:
> >
> > Add support for vIOMMU. Register a memory listener to dma_as in
> > dev_start
> > - during region_add register a specific IOMMU notifier, and store all notifiers in a list.
> > - during region_del, compare and delete the IOMMU notifier from the list
> >
> > Verified in vp_vdpa and vdpa_sim_net driver
> >
> > Signed-off-by: Cindy Lu <lulu@redhat.com>
> > ---
> > hw/virtio/vhost-vdpa.c | 218 ++++++++++++++++++++++++++++++++-
> > include/hw/virtio/vhost-vdpa.h | 11 ++
> > 2 files changed, 226 insertions(+), 3 deletions(-)
> >
> > diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
> > index 3ff9ce3501..8f36423c28 100644
> > --- a/hw/virtio/vhost-vdpa.c
> > +++ b/hw/virtio/vhost-vdpa.c
> > @@ -26,6 +26,7 @@
> > #include "cpu.h"
> > #include "trace.h"
> > #include "qapi/error.h"
> > +#include "hw/virtio/virtio-access.h"
> >
> > /*
> > * Return one past the end of the end of section. Be careful with uint64_t
> > @@ -146,9 +147,8 @@ static void vhost_vdpa_iotlb_batch_begin_once(struct vhost_vdpa *v)
> > v->iotlb_batch_begin_sent = true;
> > }
> >
> > -static void vhost_vdpa_listener_commit(MemoryListener *listener)
> > +static void vhost_vdpa_iotlb_batch_end(struct vhost_vdpa *v)
> > {
> > - struct vhost_vdpa *v = container_of(listener, struct vhost_vdpa, listener);
> > struct vhost_dev *dev = v->dev;
> > struct vhost_msg_v2 msg = {};
> > int fd = v->device_fd;
> > @@ -172,6 +172,18 @@ static void vhost_vdpa_listener_commit(MemoryListener *listener)
> >
> > v->iotlb_batch_begin_sent = false;
> > }
> > +static void vhost_vdpa_listener_commit(MemoryListener *listener)
> > +{
> > + struct vhost_vdpa *v = container_of(listener, struct vhost_vdpa, listener);
> > + return vhost_vdpa_iotlb_batch_end(v);
> > +}
> > +
> > +static void vhost_vdpa_listener_iommu_commit(MemoryListener *listener)
> > +{
> > + struct vhost_vdpa *v =
> > + container_of(listener, struct vhost_vdpa, iommu_listener);
> > + return vhost_vdpa_iotlb_batch_end(v);
> > +}
> >
> > static void vhost_vdpa_listener_region_add(MemoryListener *listener,
> > MemoryRegionSection *section)
> > @@ -186,6 +198,9 @@ static void vhost_vdpa_listener_region_add(MemoryListener *listener,
> > v->iova_range.last)) {
> > return;
> > }
> > + if (memory_region_is_iommu(section->mr)) {
> > + return;
> > + }
> >
> > if (unlikely((section->offset_within_address_space & ~TARGET_PAGE_MASK) !=
> > (section->offset_within_region & ~TARGET_PAGE_MASK))) {
> > @@ -260,6 +275,9 @@ static void vhost_vdpa_listener_region_del(MemoryListener *listener,
> > v->iova_range.last)) {
> > return;
> > }
> > + if (memory_region_is_iommu(section->mr)) {
> > + return;
> > + }
> >
> > if (unlikely((section->offset_within_address_space & ~TARGET_PAGE_MASK) !=
> > (section->offset_within_region & ~TARGET_PAGE_MASK))) {
> > @@ -312,6 +330,180 @@ static const MemoryListener vhost_vdpa_memory_listener = {
> > .region_del = vhost_vdpa_listener_region_del,
> > };
> >
> > +/* Called with rcu_read_lock held. */
> > +static bool vhost_vdpa_get_xlat_addr(IOMMUTLBEntry *iotlb, void **vaddr,
> > + ram_addr_t *ram_addr, bool *read_only)
>
> So this looks exact the same as what VFIO did, try to unify them?
>
for this function, I have tried but seems the vfio does not complied
in some arch by default
such as m68k-softmmu, if we want to use this function, maybe we need
to enable the VFIO in
that arch, not sure if this is worth？
Thanks
cindy
> > +{
> > + MemoryRegion *mr;
> > + hwaddr xlat;
> > + hwaddr len = iotlb->addr_mask + 1;
> > + bool writable = iotlb->perm & IOMMU_WO;
> > +
> > + /*
> > + * The IOMMU TLB entry we have just covers translation through
> > + * this IOMMU to its immediate target. We need to translate
> > + * it the rest of the way through to memory.
> > + */
> > + mr = address_space_translate(&address_space_memory, iotlb->translated_addr,
> > + &xlat, &len, writable, MEMTXATTRS_UNSPECIFIED);
> > + if (!memory_region_is_ram(mr)) {
> > + error_report("iommu map to non memory area %" HWADDR_PRIx "", xlat);
> > + return false;
> > + } else if (memory_region_has_ram_discard_manager(mr)) {
> > + RamDiscardManager *rdm = memory_region_get_ram_discard_manager(mr);
> > + MemoryRegionSection tmp = {
> > + .mr = mr,
> > + .offset_within_region = xlat,
> > + .size = int128_make64(len),
> > + };
> > +
> > + if (!ram_discard_manager_is_populated(rdm, &tmp)) {
> > + error_report("iommu map to discarded memory (e.g., unplugged via"
> > + " virtio-mem): %" HWADDR_PRIx "",
> > + iotlb->translated_addr);
> > + return false;
> > + }
> > + }
> > + /*
> > + * Translation truncates length to the IOMMU page size,
> > + * check that it did not truncate too much.
> > + */
> > + if (len & iotlb->addr_mask) {
> > + error_report("iommu has granularity incompatible with target AS");
> > + return false;
> > + }
> > +
> > + if (vaddr) {
> > + *vaddr = memory_region_get_ram_ptr(mr) + xlat;
> > + }
> > +
> > + if (ram_addr) {
> > + *ram_addr = memory_region_get_ram_addr(mr) + xlat;
> > + }
> > +
> > + if (read_only) {
> > + *read_only = !writable || mr->readonly;
> > + }
> > +
> > + return true;
> > +}
> > +
> > +static void vhost_vdpa_iommu_map_notify(IOMMUNotifier *n, IOMMUTLBEntry *iotlb)
> > +{
> > + struct vdpa_iommu *iommu = container_of(n, struct vdpa_iommu, n);
> > +
> > + hwaddr iova = iotlb->iova + iommu->iommu_offset;
> > + struct vhost_vdpa *v = iommu->dev;
> > + void *vaddr;
> > + int ret;
> > +
> > + if (iotlb->target_as != &address_space_memory) {
> > + error_report("Wrong target AS \"%s\", only system memory is allowed",
> > + iotlb->target_as->name ? iotlb->target_as->name : "none");
> > + return;
> > + }
> > + RCU_READ_LOCK_GUARD();
> > + vhost_vdpa_iotlb_batch_begin_once(v);
> > +
> > + if ((iotlb->perm & IOMMU_RW) != IOMMU_NONE) {
> > + bool read_only;
> > +
> > + if (!vhost_vdpa_get_xlat_addr(iotlb, &vaddr, NULL, &read_only)) {
> > + return;
> > + }
> > + ret =
> > + vhost_vdpa_dma_map(v, iova, iotlb->addr_mask + 1, vaddr, read_only);
> > + if (ret) {
> > + error_report("vhost_vdpa_dma_map(%p, 0x%" HWADDR_PRIx ", "
> > + "0x%" HWADDR_PRIx ", %p) = %d (%m)",
> > + v, iova, iotlb->addr_mask + 1, vaddr, ret);
> > + }
> > + } else {
> > + ret = vhost_vdpa_dma_unmap(v, iova, iotlb->addr_mask + 1);
> > + if (ret) {
> > + error_report("vhost_vdpa_dma_unmap(%p, 0x%" HWADDR_PRIx ", "
> > + "0x%" HWADDR_PRIx ") = %d (%m)",
> > + v, iova, iotlb->addr_mask + 1, ret);
> > + }
> > + }
> > +}
> > +
> > +static void vhost_vdpa_iommu_region_add(MemoryListener *listener,
> > + MemoryRegionSection *section)
> > +{
> > + struct vhost_vdpa *v =
> > + container_of(listener, struct vhost_vdpa, iommu_listener);
> > +
> > + struct vdpa_iommu *iommu;
> > + Int128 end;
> > + int iommu_idx;
> > + IOMMUMemoryRegion *iommu_mr;
> > + int ret;
> > +
> > + if (!memory_region_is_iommu(section->mr)) {
> > + return;
> > + }
> > +
> > + iommu_mr = IOMMU_MEMORY_REGION(section->mr);
> > +
> > + iommu = g_malloc0(sizeof(*iommu));
> > + end =
> > + int128_add(int128_make64(section->offset_within_region), section->size);
>
> I'd suggest to tweak the line breaking here and below, e.g you can
> keep int128_add and other stuffs as long as it does not exceeds 70 or
> 80 characters at the line where "end =" sits.
>
sure will change this
> > + end = int128_sub(end, int128_one());
> > + iommu_idx =
> > + memory_region_iommu_attrs_to_index(iommu_mr, MEMTXATTRS_UNSPECIFIED);
> > + iommu->iommu_mr = iommu_mr;
> > +
> > + iommu_notifier_init(
> > + &iommu->n, vhost_vdpa_iommu_map_notify, IOMMU_NOTIFIER_IOTLB_EVENTS,
> > + section->offset_within_region, int128_get64(end), iommu_idx);
> > + iommu->iommu_offset =
> > + section->offset_within_address_space - section->offset_within_region;
> > + iommu->dev = v;
> > +
> > + ret = memory_region_register_iommu_notifier(section->mr, &iommu->n, NULL);
> > + if (ret) {
> > + g_free(iommu);
> > + return;
> > + }
> > +
> > + QLIST_INSERT_HEAD(&v->iommu_list, iommu, iommu_next);
> > + memory_region_iommu_replay(iommu->iommu_mr, &iommu->n);
> > +
> > + return;
> > +}
> > +
> > +static void vhost_vdpa_iommu_region_del(MemoryListener *listener,
> > + MemoryRegionSection *section)
> > +{
> > + struct vhost_vdpa *v =
> > + container_of(listener, struct vhost_vdpa, iommu_listener);
> > +
> > + struct vdpa_iommu *iommu;
> > +
> > + if (!memory_region_is_iommu(section->mr)) {
> > + return;
> > + }
> > +
> > + QLIST_FOREACH(iommu, &v->iommu_list, iommu_next)
> > + {
> > + if (MEMORY_REGION(iommu->iommu_mr) == section->mr &&
> > + iommu->n.start == section->offset_within_region) {
> > + memory_region_unregister_iommu_notifier(section->mr, &iommu->n);
> > + QLIST_REMOVE(iommu, iommu_next);
> > + g_free(iommu);
> > + break;
> > + }
> > + }
> > +}
> > +
> > +static const MemoryListener vhost_vdpa_iommu_listener = {
> > + .name = "vhost-vdpa-iommu",
> > + .region_add = vhost_vdpa_iommu_region_add,
> > + .region_del = vhost_vdpa_iommu_region_del,
> > + .commit = vhost_vdpa_listener_iommu_commit,
>
> Unless there's a special requirement. I'd suggest to reuse the
> existing listener (since you've already had if (iommu_region()) check
> there).
>
> Thanks
>
sure Thanks Jason , will provide a new version soon
Thanks
Cindy
> > +};
> > +
> > static int vhost_vdpa_call(struct vhost_dev *dev, unsigned long int request,
> > void *arg)
> > {
> > @@ -433,6 +625,18 @@ static int vhost_vdpa_init_svq(struct vhost_dev *hdev, struct vhost_vdpa *v,
> > return 0;
> > }
> >
> > +static bool vdpa_dev_has_iommu(struct vhost_dev *dev)
> > +{
> > + VirtIODevice *vdev = dev->vdev;
> > +
> > + if (vdev) {
> > + return virtio_bus_device_iommu_enabled(vdev) &&
> > + virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM);
> > + } else {
> > + return false;
> > + }
> > +}
> > +
> > static int vhost_vdpa_init(struct vhost_dev *dev, void *opaque, Error **errp)
> > {
> > struct vhost_vdpa *v;
> > @@ -454,6 +658,7 @@ static int vhost_vdpa_init(struct vhost_dev *dev, void *opaque, Error **errp)
> > v->dev = dev;
> > dev->opaque = opaque ;
> > v->listener = vhost_vdpa_memory_listener;
> > + v->iommu_listener = vhost_vdpa_iommu_listener;
> > v->msg_type = VHOST_IOTLB_MSG_V2;
> > ret = vhost_vdpa_init_svq(dev, v, errp);
> > if (ret) {
> > @@ -587,7 +792,6 @@ static int vhost_vdpa_cleanup(struct vhost_dev *dev)
> > v = dev->opaque;
> > trace_vhost_vdpa_cleanup(dev, v);
> > vhost_vdpa_host_notifiers_uninit(dev, dev->nvqs);
> > - memory_listener_unregister(&v->listener);
> > vhost_vdpa_svq_cleanup(dev);
> >
> > dev->opaque = NULL;
> > @@ -1128,11 +1332,19 @@ static int vhost_vdpa_dev_start(struct vhost_dev *dev, bool started)
> >
> > if (started) {
> > memory_listener_register(&v->listener, &address_space_memory);
> > + if (vdpa_dev_has_iommu(dev)) {
> > + memory_listener_register(&v->iommu_listener, dev->vdev->dma_as);
> > + }
> > +
> > return vhost_vdpa_add_status(dev, VIRTIO_CONFIG_S_DRIVER_OK);
> > } else {
> > vhost_vdpa_reset_device(dev);
> > vhost_vdpa_add_status(dev, VIRTIO_CONFIG_S_ACKNOWLEDGE |
> > VIRTIO_CONFIG_S_DRIVER);
> > +
> > + if (vdpa_dev_has_iommu(dev)) {
> > + memory_listener_unregister(&v->iommu_listener);
> > + }
> > memory_listener_unregister(&v->listener);
> >
> > return 0;
> > diff --git a/include/hw/virtio/vhost-vdpa.h b/include/hw/virtio/vhost-vdpa.h
> > index d10a89303e..22cbce7934 100644
> > --- a/include/hw/virtio/vhost-vdpa.h
> > +++ b/include/hw/virtio/vhost-vdpa.h
> > @@ -30,6 +30,7 @@ typedef struct vhost_vdpa {
> > uint32_t msg_type;
> > bool iotlb_batch_begin_sent;
> > MemoryListener listener;
> > + MemoryListener iommu_listener;
> > struct vhost_vdpa_iova_range iova_range;
> > uint64_t acked_features;
> > bool shadow_vqs_enabled;
> > @@ -41,8 +42,18 @@ typedef struct vhost_vdpa {
> > void *shadow_vq_ops_opaque;
> > struct vhost_dev *dev;
> > VhostVDPAHostNotifier notifier[VIRTIO_QUEUE_MAX];
> > + QLIST_HEAD(, vdpa_iommu) iommu_list;
> > + IOMMUNotifier n;
> > } VhostVDPA;
> >
> > +struct vdpa_iommu {
> > + struct vhost_vdpa *dev;
> > + IOMMUMemoryRegion *iommu_mr;
> > + hwaddr iommu_offset;
> > + IOMMUNotifier n;
> > + QLIST_ENTRY(vdpa_iommu) iommu_next;
> > +};
> > +
> > int vhost_vdpa_dma_map(struct vhost_vdpa *v, hwaddr iova, hwaddr size,
> > void *vaddr, bool readonly);
> > int vhost_vdpa_dma_unmap(struct vhost_vdpa *v, hwaddr iova, hwaddr size);
> > --
> > 2.34.3
> >
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2] vhost-vdpa: add support for vIOMMU
2022-10-23 14:45 ` Cindy Lu
@ 2022-10-24 1:51 ` Jason Wang
2022-10-24 5:21 ` Cindy Lu
0 siblings, 1 reply; 5+ messages in thread
From: Jason Wang @ 2022-10-24 1:51 UTC (permalink / raw)
To: Cindy Lu; +Cc: mst, eperezma, qemu-devel, stefanha
On Sun, Oct 23, 2022 at 10:46 PM Cindy Lu <lulu@redhat.com> wrote:
>
> On Thu, 20 Oct 2022 at 10:15, Jason Wang <jasowang@redhat.com> wrote:
> >
> > On Wed, Oct 19, 2022 at 2:39 PM Cindy Lu <lulu@redhat.com> wrote:
> > >
> > > Add support for vIOMMU. Register a memory listener to dma_as in
> > > dev_start
> > > - during region_add register a specific IOMMU notifier, and store all notifiers in a list.
> > > - during region_del, compare and delete the IOMMU notifier from the list
> > >
> > > Verified in vp_vdpa and vdpa_sim_net driver
> > >
> > > Signed-off-by: Cindy Lu <lulu@redhat.com>
> > > ---
> > > hw/virtio/vhost-vdpa.c | 218 ++++++++++++++++++++++++++++++++-
> > > include/hw/virtio/vhost-vdpa.h | 11 ++
> > > 2 files changed, 226 insertions(+), 3 deletions(-)
> > >
> > > diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
> > > index 3ff9ce3501..8f36423c28 100644
> > > --- a/hw/virtio/vhost-vdpa.c
> > > +++ b/hw/virtio/vhost-vdpa.c
> > > @@ -26,6 +26,7 @@
> > > #include "cpu.h"
> > > #include "trace.h"
> > > #include "qapi/error.h"
> > > +#include "hw/virtio/virtio-access.h"
> > >
> > > /*
> > > * Return one past the end of the end of section. Be careful with uint64_t
> > > @@ -146,9 +147,8 @@ static void vhost_vdpa_iotlb_batch_begin_once(struct vhost_vdpa *v)
> > > v->iotlb_batch_begin_sent = true;
> > > }
> > >
> > > -static void vhost_vdpa_listener_commit(MemoryListener *listener)
> > > +static void vhost_vdpa_iotlb_batch_end(struct vhost_vdpa *v)
> > > {
> > > - struct vhost_vdpa *v = container_of(listener, struct vhost_vdpa, listener);
> > > struct vhost_dev *dev = v->dev;
> > > struct vhost_msg_v2 msg = {};
> > > int fd = v->device_fd;
> > > @@ -172,6 +172,18 @@ static void vhost_vdpa_listener_commit(MemoryListener *listener)
> > >
> > > v->iotlb_batch_begin_sent = false;
> > > }
> > > +static void vhost_vdpa_listener_commit(MemoryListener *listener)
> > > +{
> > > + struct vhost_vdpa *v = container_of(listener, struct vhost_vdpa, listener);
> > > + return vhost_vdpa_iotlb_batch_end(v);
> > > +}
> > > +
> > > +static void vhost_vdpa_listener_iommu_commit(MemoryListener *listener)
> > > +{
> > > + struct vhost_vdpa *v =
> > > + container_of(listener, struct vhost_vdpa, iommu_listener);
> > > + return vhost_vdpa_iotlb_batch_end(v);
> > > +}
> > >
> > > static void vhost_vdpa_listener_region_add(MemoryListener *listener,
> > > MemoryRegionSection *section)
> > > @@ -186,6 +198,9 @@ static void vhost_vdpa_listener_region_add(MemoryListener *listener,
> > > v->iova_range.last)) {
> > > return;
> > > }
> > > + if (memory_region_is_iommu(section->mr)) {
> > > + return;
> > > + }
> > >
> > > if (unlikely((section->offset_within_address_space & ~TARGET_PAGE_MASK) !=
> > > (section->offset_within_region & ~TARGET_PAGE_MASK))) {
> > > @@ -260,6 +275,9 @@ static void vhost_vdpa_listener_region_del(MemoryListener *listener,
> > > v->iova_range.last)) {
> > > return;
> > > }
> > > + if (memory_region_is_iommu(section->mr)) {
> > > + return;
> > > + }
> > >
> > > if (unlikely((section->offset_within_address_space & ~TARGET_PAGE_MASK) !=
> > > (section->offset_within_region & ~TARGET_PAGE_MASK))) {
> > > @@ -312,6 +330,180 @@ static const MemoryListener vhost_vdpa_memory_listener = {
> > > .region_del = vhost_vdpa_listener_region_del,
> > > };
> > >
> > > +/* Called with rcu_read_lock held. */
> > > +static bool vhost_vdpa_get_xlat_addr(IOMMUTLBEntry *iotlb, void **vaddr,
> > > + ram_addr_t *ram_addr, bool *read_only)
> >
> > So this looks exact the same as what VFIO did, try to unify them?
> >
> for this function, I have tried but seems the vfio does not complied
> in some arch by default
> such as m68k-softmmu, if we want to use this function, maybe we need
> to enable the VFIO in
> that arch, not sure if this is worth？
> Thanks
> cindy
The function is nothing vfio specific, so let's move it somewhere e.g memory.c?
Thanks
>
> > > +{
> > > + MemoryRegion *mr;
> > > + hwaddr xlat;
> > > + hwaddr len = iotlb->addr_mask + 1;
> > > + bool writable = iotlb->perm & IOMMU_WO;
> > > +
> > > + /*
> > > + * The IOMMU TLB entry we have just covers translation through
> > > + * this IOMMU to its immediate target. We need to translate
> > > + * it the rest of the way through to memory.
> > > + */
> > > + mr = address_space_translate(&address_space_memory, iotlb->translated_addr,
> > > + &xlat, &len, writable, MEMTXATTRS_UNSPECIFIED);
> > > + if (!memory_region_is_ram(mr)) {
> > > + error_report("iommu map to non memory area %" HWADDR_PRIx "", xlat);
> > > + return false;
> > > + } else if (memory_region_has_ram_discard_manager(mr)) {
> > > + RamDiscardManager *rdm = memory_region_get_ram_discard_manager(mr);
> > > + MemoryRegionSection tmp = {
> > > + .mr = mr,
> > > + .offset_within_region = xlat,
> > > + .size = int128_make64(len),
> > > + };
> > > +
> > > + if (!ram_discard_manager_is_populated(rdm, &tmp)) {
> > > + error_report("iommu map to discarded memory (e.g., unplugged via"
> > > + " virtio-mem): %" HWADDR_PRIx "",
> > > + iotlb->translated_addr);
> > > + return false;
> > > + }
> > > + }
> > > + /*
> > > + * Translation truncates length to the IOMMU page size,
> > > + * check that it did not truncate too much.
> > > + */
> > > + if (len & iotlb->addr_mask) {
> > > + error_report("iommu has granularity incompatible with target AS");
> > > + return false;
> > > + }
> > > +
> > > + if (vaddr) {
> > > + *vaddr = memory_region_get_ram_ptr(mr) + xlat;
> > > + }
> > > +
> > > + if (ram_addr) {
> > > + *ram_addr = memory_region_get_ram_addr(mr) + xlat;
> > > + }
> > > +
> > > + if (read_only) {
> > > + *read_only = !writable || mr->readonly;
> > > + }
> > > +
> > > + return true;
> > > +}
> > > +
> > > +static void vhost_vdpa_iommu_map_notify(IOMMUNotifier *n, IOMMUTLBEntry *iotlb)
> > > +{
> > > + struct vdpa_iommu *iommu = container_of(n, struct vdpa_iommu, n);
> > > +
> > > + hwaddr iova = iotlb->iova + iommu->iommu_offset;
> > > + struct vhost_vdpa *v = iommu->dev;
> > > + void *vaddr;
> > > + int ret;
> > > +
> > > + if (iotlb->target_as != &address_space_memory) {
> > > + error_report("Wrong target AS \"%s\", only system memory is allowed",
> > > + iotlb->target_as->name ? iotlb->target_as->name : "none");
> > > + return;
> > > + }
> > > + RCU_READ_LOCK_GUARD();
> > > + vhost_vdpa_iotlb_batch_begin_once(v);
> > > +
> > > + if ((iotlb->perm & IOMMU_RW) != IOMMU_NONE) {
> > > + bool read_only;
> > > +
> > > + if (!vhost_vdpa_get_xlat_addr(iotlb, &vaddr, NULL, &read_only)) {
> > > + return;
> > > + }
> > > + ret =
> > > + vhost_vdpa_dma_map(v, iova, iotlb->addr_mask + 1, vaddr, read_only);
> > > + if (ret) {
> > > + error_report("vhost_vdpa_dma_map(%p, 0x%" HWADDR_PRIx ", "
> > > + "0x%" HWADDR_PRIx ", %p) = %d (%m)",
> > > + v, iova, iotlb->addr_mask + 1, vaddr, ret);
> > > + }
> > > + } else {
> > > + ret = vhost_vdpa_dma_unmap(v, iova, iotlb->addr_mask + 1);
> > > + if (ret) {
> > > + error_report("vhost_vdpa_dma_unmap(%p, 0x%" HWADDR_PRIx ", "
> > > + "0x%" HWADDR_PRIx ") = %d (%m)",
> > > + v, iova, iotlb->addr_mask + 1, ret);
> > > + }
> > > + }
> > > +}
> > > +
> > > +static void vhost_vdpa_iommu_region_add(MemoryListener *listener,
> > > + MemoryRegionSection *section)
> > > +{
> > > + struct vhost_vdpa *v =
> > > + container_of(listener, struct vhost_vdpa, iommu_listener);
> > > +
> > > + struct vdpa_iommu *iommu;
> > > + Int128 end;
> > > + int iommu_idx;
> > > + IOMMUMemoryRegion *iommu_mr;
> > > + int ret;
> > > +
> > > + if (!memory_region_is_iommu(section->mr)) {
> > > + return;
> > > + }
> > > +
> > > + iommu_mr = IOMMU_MEMORY_REGION(section->mr);
> > > +
> > > + iommu = g_malloc0(sizeof(*iommu));
> > > + end =
> > > + int128_add(int128_make64(section->offset_within_region), section->size);
> >
> > I'd suggest to tweak the line breaking here and below, e.g you can
> > keep int128_add and other stuffs as long as it does not exceeds 70 or
> > 80 characters at the line where "end =" sits.
> >
> sure will change this
> > > + end = int128_sub(end, int128_one());
> > > + iommu_idx =
> > > + memory_region_iommu_attrs_to_index(iommu_mr, MEMTXATTRS_UNSPECIFIED);
> > > + iommu->iommu_mr = iommu_mr;
> > > +
> > > + iommu_notifier_init(
> > > + &iommu->n, vhost_vdpa_iommu_map_notify, IOMMU_NOTIFIER_IOTLB_EVENTS,
> > > + section->offset_within_region, int128_get64(end), iommu_idx);
> > > + iommu->iommu_offset =
> > > + section->offset_within_address_space - section->offset_within_region;
> > > + iommu->dev = v;
> > > +
> > > + ret = memory_region_register_iommu_notifier(section->mr, &iommu->n, NULL);
> > > + if (ret) {
> > > + g_free(iommu);
> > > + return;
> > > + }
> > > +
> > > + QLIST_INSERT_HEAD(&v->iommu_list, iommu, iommu_next);
> > > + memory_region_iommu_replay(iommu->iommu_mr, &iommu->n);
> > > +
> > > + return;
> > > +}
> > > +
> > > +static void vhost_vdpa_iommu_region_del(MemoryListener *listener,
> > > + MemoryRegionSection *section)
> > > +{
> > > + struct vhost_vdpa *v =
> > > + container_of(listener, struct vhost_vdpa, iommu_listener);
> > > +
> > > + struct vdpa_iommu *iommu;
> > > +
> > > + if (!memory_region_is_iommu(section->mr)) {
> > > + return;
> > > + }
> > > +
> > > + QLIST_FOREACH(iommu, &v->iommu_list, iommu_next)
> > > + {
> > > + if (MEMORY_REGION(iommu->iommu_mr) == section->mr &&
> > > + iommu->n.start == section->offset_within_region) {
> > > + memory_region_unregister_iommu_notifier(section->mr, &iommu->n);
> > > + QLIST_REMOVE(iommu, iommu_next);
> > > + g_free(iommu);
> > > + break;
> > > + }
> > > + }
> > > +}
> > > +
> > > +static const MemoryListener vhost_vdpa_iommu_listener = {
> > > + .name = "vhost-vdpa-iommu",
> > > + .region_add = vhost_vdpa_iommu_region_add,
> > > + .region_del = vhost_vdpa_iommu_region_del,
> > > + .commit = vhost_vdpa_listener_iommu_commit,
> >
> > Unless there's a special requirement. I'd suggest to reuse the
> > existing listener (since you've already had if (iommu_region()) check
> > there).
> >
> > Thanks
> >
> sure Thanks Jason , will provide a new version soon
> Thanks
> Cindy
> > > +};
> > > +
> > > static int vhost_vdpa_call(struct vhost_dev *dev, unsigned long int request,
> > > void *arg)
> > > {
> > > @@ -433,6 +625,18 @@ static int vhost_vdpa_init_svq(struct vhost_dev *hdev, struct vhost_vdpa *v,
> > > return 0;
> > > }
> > >
> > > +static bool vdpa_dev_has_iommu(struct vhost_dev *dev)
> > > +{
> > > + VirtIODevice *vdev = dev->vdev;
> > > +
> > > + if (vdev) {
> > > + return virtio_bus_device_iommu_enabled(vdev) &&
> > > + virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM);
> > > + } else {
> > > + return false;
> > > + }
> > > +}
> > > +
> > > static int vhost_vdpa_init(struct vhost_dev *dev, void *opaque, Error **errp)
> > > {
> > > struct vhost_vdpa *v;
> > > @@ -454,6 +658,7 @@ static int vhost_vdpa_init(struct vhost_dev *dev, void *opaque, Error **errp)
> > > v->dev = dev;
> > > dev->opaque = opaque ;
> > > v->listener = vhost_vdpa_memory_listener;
> > > + v->iommu_listener = vhost_vdpa_iommu_listener;
> > > v->msg_type = VHOST_IOTLB_MSG_V2;
> > > ret = vhost_vdpa_init_svq(dev, v, errp);
> > > if (ret) {
> > > @@ -587,7 +792,6 @@ static int vhost_vdpa_cleanup(struct vhost_dev *dev)
> > > v = dev->opaque;
> > > trace_vhost_vdpa_cleanup(dev, v);
> > > vhost_vdpa_host_notifiers_uninit(dev, dev->nvqs);
> > > - memory_listener_unregister(&v->listener);
> > > vhost_vdpa_svq_cleanup(dev);
> > >
> > > dev->opaque = NULL;
> > > @@ -1128,11 +1332,19 @@ static int vhost_vdpa_dev_start(struct vhost_dev *dev, bool started)
> > >
> > > if (started) {
> > > memory_listener_register(&v->listener, &address_space_memory);
> > > + if (vdpa_dev_has_iommu(dev)) {
> > > + memory_listener_register(&v->iommu_listener, dev->vdev->dma_as);
> > > + }
> > > +
> > > return vhost_vdpa_add_status(dev, VIRTIO_CONFIG_S_DRIVER_OK);
> > > } else {
> > > vhost_vdpa_reset_device(dev);
> > > vhost_vdpa_add_status(dev, VIRTIO_CONFIG_S_ACKNOWLEDGE |
> > > VIRTIO_CONFIG_S_DRIVER);
> > > +
> > > + if (vdpa_dev_has_iommu(dev)) {
> > > + memory_listener_unregister(&v->iommu_listener);
> > > + }
> > > memory_listener_unregister(&v->listener);
> > >
> > > return 0;
> > > diff --git a/include/hw/virtio/vhost-vdpa.h b/include/hw/virtio/vhost-vdpa.h
> > > index d10a89303e..22cbce7934 100644
> > > --- a/include/hw/virtio/vhost-vdpa.h
> > > +++ b/include/hw/virtio/vhost-vdpa.h
> > > @@ -30,6 +30,7 @@ typedef struct vhost_vdpa {
> > > uint32_t msg_type;
> > > bool iotlb_batch_begin_sent;
> > > MemoryListener listener;
> > > + MemoryListener iommu_listener;
> > > struct vhost_vdpa_iova_range iova_range;
> > > uint64_t acked_features;
> > > bool shadow_vqs_enabled;
> > > @@ -41,8 +42,18 @@ typedef struct vhost_vdpa {
> > > void *shadow_vq_ops_opaque;
> > > struct vhost_dev *dev;
> > > VhostVDPAHostNotifier notifier[VIRTIO_QUEUE_MAX];
> > > + QLIST_HEAD(, vdpa_iommu) iommu_list;
> > > + IOMMUNotifier n;
> > > } VhostVDPA;
> > >
> > > +struct vdpa_iommu {
> > > + struct vhost_vdpa *dev;
> > > + IOMMUMemoryRegion *iommu_mr;
> > > + hwaddr iommu_offset;
> > > + IOMMUNotifier n;
> > > + QLIST_ENTRY(vdpa_iommu) iommu_next;
> > > +};
> > > +
> > > int vhost_vdpa_dma_map(struct vhost_vdpa *v, hwaddr iova, hwaddr size,
> > > void *vaddr, bool readonly);
> > > int vhost_vdpa_dma_unmap(struct vhost_vdpa *v, hwaddr iova, hwaddr size);
> > > --
> > > 2.34.3
> > >
> >
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2] vhost-vdpa: add support for vIOMMU
2022-10-24 1:51 ` Jason Wang
@ 2022-10-24 5:21 ` Cindy Lu
0 siblings, 0 replies; 5+ messages in thread
From: Cindy Lu @ 2022-10-24 5:21 UTC (permalink / raw)
To: Jason Wang; +Cc: mst, eperezma, qemu-devel, stefanha
On Mon, 24 Oct 2022 at 09:51, Jason Wang <jasowang@redhat.com> wrote:
>
> On Sun, Oct 23, 2022 at 10:46 PM Cindy Lu <lulu@redhat.com> wrote:
> >
> > On Thu, 20 Oct 2022 at 10:15, Jason Wang <jasowang@redhat.com> wrote:
> > >
> > > On Wed, Oct 19, 2022 at 2:39 PM Cindy Lu <lulu@redhat.com> wrote:
> > > >
> > > > Add support for vIOMMU. Register a memory listener to dma_as in
> > > > dev_start
> > > > - during region_add register a specific IOMMU notifier, and store all notifiers in a list.
> > > > - during region_del, compare and delete the IOMMU notifier from the list
> > > >
> > > > Verified in vp_vdpa and vdpa_sim_net driver
> > > >
> > > > Signed-off-by: Cindy Lu <lulu@redhat.com>
> > > > ---
> > > > hw/virtio/vhost-vdpa.c | 218 ++++++++++++++++++++++++++++++++-
> > > > include/hw/virtio/vhost-vdpa.h | 11 ++
> > > > 2 files changed, 226 insertions(+), 3 deletions(-)
> > > >
> > > > diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
> > > > index 3ff9ce3501..8f36423c28 100644
> > > > --- a/hw/virtio/vhost-vdpa.c
> > > > +++ b/hw/virtio/vhost-vdpa.c
> > > > @@ -26,6 +26,7 @@
> > > > #include "cpu.h"
> > > > #include "trace.h"
> > > > #include "qapi/error.h"
> > > > +#include "hw/virtio/virtio-access.h"
> > > >
> > > > /*
> > > > * Return one past the end of the end of section. Be careful with uint64_t
> > > > @@ -146,9 +147,8 @@ static void vhost_vdpa_iotlb_batch_begin_once(struct vhost_vdpa *v)
> > > > v->iotlb_batch_begin_sent = true;
> > > > }
> > > >
> > > > -static void vhost_vdpa_listener_commit(MemoryListener *listener)
> > > > +static void vhost_vdpa_iotlb_batch_end(struct vhost_vdpa *v)
> > > > {
> > > > - struct vhost_vdpa *v = container_of(listener, struct vhost_vdpa, listener);
> > > > struct vhost_dev *dev = v->dev;
> > > > struct vhost_msg_v2 msg = {};
> > > > int fd = v->device_fd;
> > > > @@ -172,6 +172,18 @@ static void vhost_vdpa_listener_commit(MemoryListener *listener)
> > > >
> > > > v->iotlb_batch_begin_sent = false;
> > > > }
> > > > +static void vhost_vdpa_listener_commit(MemoryListener *listener)
> > > > +{
> > > > + struct vhost_vdpa *v = container_of(listener, struct vhost_vdpa, listener);
> > > > + return vhost_vdpa_iotlb_batch_end(v);
> > > > +}
> > > > +
> > > > +static void vhost_vdpa_listener_iommu_commit(MemoryListener *listener)
> > > > +{
> > > > + struct vhost_vdpa *v =
> > > > + container_of(listener, struct vhost_vdpa, iommu_listener);
> > > > + return vhost_vdpa_iotlb_batch_end(v);
> > > > +}
> > > >
> > > > static void vhost_vdpa_listener_region_add(MemoryListener *listener,
> > > > MemoryRegionSection *section)
> > > > @@ -186,6 +198,9 @@ static void vhost_vdpa_listener_region_add(MemoryListener *listener,
> > > > v->iova_range.last)) {
> > > > return;
> > > > }
> > > > + if (memory_region_is_iommu(section->mr)) {
> > > > + return;
> > > > + }
> > > >
> > > > if (unlikely((section->offset_within_address_space & ~TARGET_PAGE_MASK) !=
> > > > (section->offset_within_region & ~TARGET_PAGE_MASK))) {
> > > > @@ -260,6 +275,9 @@ static void vhost_vdpa_listener_region_del(MemoryListener *listener,
> > > > v->iova_range.last)) {
> > > > return;
> > > > }
> > > > + if (memory_region_is_iommu(section->mr)) {
> > > > + return;
> > > > + }
> > > >
> > > > if (unlikely((section->offset_within_address_space & ~TARGET_PAGE_MASK) !=
> > > > (section->offset_within_region & ~TARGET_PAGE_MASK))) {
> > > > @@ -312,6 +330,180 @@ static const MemoryListener vhost_vdpa_memory_listener = {
> > > > .region_del = vhost_vdpa_listener_region_del,
> > > > };
> > > >
> > > > +/* Called with rcu_read_lock held. */
> > > > +static bool vhost_vdpa_get_xlat_addr(IOMMUTLBEntry *iotlb, void **vaddr,
> > > > + ram_addr_t *ram_addr, bool *read_only)
> > >
> > > So this looks exact the same as what VFIO did, try to unify them?
> > >
> > for this function, I have tried but seems the vfio does not complied
> > in some arch by default
> > such as m68k-softmmu, if we want to use this function, maybe we need
> > to enable the VFIO in
> > that arch, not sure if this is worth？
> > Thanks
> > cindy
>
> The function is nothing vfio specific, so let's move it somewhere e.g memory.c?
>
> Thanks
>
sure, I will try this
Thanks
Cindy
> >
> > > > +{
> > > > + MemoryRegion *mr;
> > > > + hwaddr xlat;
> > > > + hwaddr len = iotlb->addr_mask + 1;
> > > > + bool writable = iotlb->perm & IOMMU_WO;
> > > > +
> > > > + /*
> > > > + * The IOMMU TLB entry we have just covers translation through
> > > > + * this IOMMU to its immediate target. We need to translate
> > > > + * it the rest of the way through to memory.
> > > > + */
> > > > + mr = address_space_translate(&address_space_memory, iotlb->translated_addr,
> > > > + &xlat, &len, writable, MEMTXATTRS_UNSPECIFIED);
> > > > + if (!memory_region_is_ram(mr)) {
> > > > + error_report("iommu map to non memory area %" HWADDR_PRIx "", xlat);
> > > > + return false;
> > > > + } else if (memory_region_has_ram_discard_manager(mr)) {
> > > > + RamDiscardManager *rdm = memory_region_get_ram_discard_manager(mr);
> > > > + MemoryRegionSection tmp = {
> > > > + .mr = mr,
> > > > + .offset_within_region = xlat,
> > > > + .size = int128_make64(len),
> > > > + };
> > > > +
> > > > + if (!ram_discard_manager_is_populated(rdm, &tmp)) {
> > > > + error_report("iommu map to discarded memory (e.g., unplugged via"
> > > > + " virtio-mem): %" HWADDR_PRIx "",
> > > > + iotlb->translated_addr);
> > > > + return false;
> > > > + }
> > > > + }
> > > > + /*
> > > > + * Translation truncates length to the IOMMU page size,
> > > > + * check that it did not truncate too much.
> > > > + */
> > > > + if (len & iotlb->addr_mask) {
> > > > + error_report("iommu has granularity incompatible with target AS");
> > > > + return false;
> > > > + }
> > > > +
> > > > + if (vaddr) {
> > > > + *vaddr = memory_region_get_ram_ptr(mr) + xlat;
> > > > + }
> > > > +
> > > > + if (ram_addr) {
> > > > + *ram_addr = memory_region_get_ram_addr(mr) + xlat;
> > > > + }
> > > > +
> > > > + if (read_only) {
> > > > + *read_only = !writable || mr->readonly;
> > > > + }
> > > > +
> > > > + return true;
> > > > +}
> > > > +
> > > > +static void vhost_vdpa_iommu_map_notify(IOMMUNotifier *n, IOMMUTLBEntry *iotlb)
> > > > +{
> > > > + struct vdpa_iommu *iommu = container_of(n, struct vdpa_iommu, n);
> > > > +
> > > > + hwaddr iova = iotlb->iova + iommu->iommu_offset;
> > > > + struct vhost_vdpa *v = iommu->dev;
> > > > + void *vaddr;
> > > > + int ret;
> > > > +
> > > > + if (iotlb->target_as != &address_space_memory) {
> > > > + error_report("Wrong target AS \"%s\", only system memory is allowed",
> > > > + iotlb->target_as->name ? iotlb->target_as->name : "none");
> > > > + return;
> > > > + }
> > > > + RCU_READ_LOCK_GUARD();
> > > > + vhost_vdpa_iotlb_batch_begin_once(v);
> > > > +
> > > > + if ((iotlb->perm & IOMMU_RW) != IOMMU_NONE) {
> > > > + bool read_only;
> > > > +
> > > > + if (!vhost_vdpa_get_xlat_addr(iotlb, &vaddr, NULL, &read_only)) {
> > > > + return;
> > > > + }
> > > > + ret =
> > > > + vhost_vdpa_dma_map(v, iova, iotlb->addr_mask + 1, vaddr, read_only);
> > > > + if (ret) {
> > > > + error_report("vhost_vdpa_dma_map(%p, 0x%" HWADDR_PRIx ", "
> > > > + "0x%" HWADDR_PRIx ", %p) = %d (%m)",
> > > > + v, iova, iotlb->addr_mask + 1, vaddr, ret);
> > > > + }
> > > > + } else {
> > > > + ret = vhost_vdpa_dma_unmap(v, iova, iotlb->addr_mask + 1);
> > > > + if (ret) {
> > > > + error_report("vhost_vdpa_dma_unmap(%p, 0x%" HWADDR_PRIx ", "
> > > > + "0x%" HWADDR_PRIx ") = %d (%m)",
> > > > + v, iova, iotlb->addr_mask + 1, ret);
> > > > + }
> > > > + }
> > > > +}
> > > > +
> > > > +static void vhost_vdpa_iommu_region_add(MemoryListener *listener,
> > > > + MemoryRegionSection *section)
> > > > +{
> > > > + struct vhost_vdpa *v =
> > > > + container_of(listener, struct vhost_vdpa, iommu_listener);
> > > > +
> > > > + struct vdpa_iommu *iommu;
> > > > + Int128 end;
> > > > + int iommu_idx;
> > > > + IOMMUMemoryRegion *iommu_mr;
> > > > + int ret;
> > > > +
> > > > + if (!memory_region_is_iommu(section->mr)) {
> > > > + return;
> > > > + }
> > > > +
> > > > + iommu_mr = IOMMU_MEMORY_REGION(section->mr);
> > > > +
> > > > + iommu = g_malloc0(sizeof(*iommu));
> > > > + end =
> > > > + int128_add(int128_make64(section->offset_within_region), section->size);
> > >
> > > I'd suggest to tweak the line breaking here and below, e.g you can
> > > keep int128_add and other stuffs as long as it does not exceeds 70 or
> > > 80 characters at the line where "end =" sits.
> > >
> > sure will change this
> > > > + end = int128_sub(end, int128_one());
> > > > + iommu_idx =
> > > > + memory_region_iommu_attrs_to_index(iommu_mr, MEMTXATTRS_UNSPECIFIED);
> > > > + iommu->iommu_mr = iommu_mr;
> > > > +
> > > > + iommu_notifier_init(
> > > > + &iommu->n, vhost_vdpa_iommu_map_notify, IOMMU_NOTIFIER_IOTLB_EVENTS,
> > > > + section->offset_within_region, int128_get64(end), iommu_idx);
> > > > + iommu->iommu_offset =
> > > > + section->offset_within_address_space - section->offset_within_region;
> > > > + iommu->dev = v;
> > > > +
> > > > + ret = memory_region_register_iommu_notifier(section->mr, &iommu->n, NULL);
> > > > + if (ret) {
> > > > + g_free(iommu);
> > > > + return;
> > > > + }
> > > > +
> > > > + QLIST_INSERT_HEAD(&v->iommu_list, iommu, iommu_next);
> > > > + memory_region_iommu_replay(iommu->iommu_mr, &iommu->n);
> > > > +
> > > > + return;
> > > > +}
> > > > +
> > > > +static void vhost_vdpa_iommu_region_del(MemoryListener *listener,
> > > > + MemoryRegionSection *section)
> > > > +{
> > > > + struct vhost_vdpa *v =
> > > > + container_of(listener, struct vhost_vdpa, iommu_listener);
> > > > +
> > > > + struct vdpa_iommu *iommu;
> > > > +
> > > > + if (!memory_region_is_iommu(section->mr)) {
> > > > + return;
> > > > + }
> > > > +
> > > > + QLIST_FOREACH(iommu, &v->iommu_list, iommu_next)
> > > > + {
> > > > + if (MEMORY_REGION(iommu->iommu_mr) == section->mr &&
> > > > + iommu->n.start == section->offset_within_region) {
> > > > + memory_region_unregister_iommu_notifier(section->mr, &iommu->n);
> > > > + QLIST_REMOVE(iommu, iommu_next);
> > > > + g_free(iommu);
> > > > + break;
> > > > + }
> > > > + }
> > > > +}
> > > > +
> > > > +static const MemoryListener vhost_vdpa_iommu_listener = {
> > > > + .name = "vhost-vdpa-iommu",
> > > > + .region_add = vhost_vdpa_iommu_region_add,
> > > > + .region_del = vhost_vdpa_iommu_region_del,
> > > > + .commit = vhost_vdpa_listener_iommu_commit,
> > >
> > > Unless there's a special requirement. I'd suggest to reuse the
> > > existing listener (since you've already had if (iommu_region()) check
> > > there).
> > >
> > > Thanks
> > >
> > sure Thanks Jason , will provide a new version soon
> > Thanks
> > Cindy
> > > > +};
> > > > +
> > > > static int vhost_vdpa_call(struct vhost_dev *dev, unsigned long int request,
> > > > void *arg)
> > > > {
> > > > @@ -433,6 +625,18 @@ static int vhost_vdpa_init_svq(struct vhost_dev *hdev, struct vhost_vdpa *v,
> > > > return 0;
> > > > }
> > > >
> > > > +static bool vdpa_dev_has_iommu(struct vhost_dev *dev)
> > > > +{
> > > > + VirtIODevice *vdev = dev->vdev;
> > > > +
> > > > + if (vdev) {
> > > > + return virtio_bus_device_iommu_enabled(vdev) &&
> > > > + virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM);
> > > > + } else {
> > > > + return false;
> > > > + }
> > > > +}
> > > > +
> > > > static int vhost_vdpa_init(struct vhost_dev *dev, void *opaque, Error **errp)
> > > > {
> > > > struct vhost_vdpa *v;
> > > > @@ -454,6 +658,7 @@ static int vhost_vdpa_init(struct vhost_dev *dev, void *opaque, Error **errp)
> > > > v->dev = dev;
> > > > dev->opaque = opaque ;
> > > > v->listener = vhost_vdpa_memory_listener;
> > > > + v->iommu_listener = vhost_vdpa_iommu_listener;
> > > > v->msg_type = VHOST_IOTLB_MSG_V2;
> > > > ret = vhost_vdpa_init_svq(dev, v, errp);
> > > > if (ret) {
> > > > @@ -587,7 +792,6 @@ static int vhost_vdpa_cleanup(struct vhost_dev *dev)
> > > > v = dev->opaque;
> > > > trace_vhost_vdpa_cleanup(dev, v);
> > > > vhost_vdpa_host_notifiers_uninit(dev, dev->nvqs);
> > > > - memory_listener_unregister(&v->listener);
> > > > vhost_vdpa_svq_cleanup(dev);
> > > >
> > > > dev->opaque = NULL;
> > > > @@ -1128,11 +1332,19 @@ static int vhost_vdpa_dev_start(struct vhost_dev *dev, bool started)
> > > >
> > > > if (started) {
> > > > memory_listener_register(&v->listener, &address_space_memory);
> > > > + if (vdpa_dev_has_iommu(dev)) {
> > > > + memory_listener_register(&v->iommu_listener, dev->vdev->dma_as);
> > > > + }
> > > > +
> > > > return vhost_vdpa_add_status(dev, VIRTIO_CONFIG_S_DRIVER_OK);
> > > > } else {
> > > > vhost_vdpa_reset_device(dev);
> > > > vhost_vdpa_add_status(dev, VIRTIO_CONFIG_S_ACKNOWLEDGE |
> > > > VIRTIO_CONFIG_S_DRIVER);
> > > > +
> > > > + if (vdpa_dev_has_iommu(dev)) {
> > > > + memory_listener_unregister(&v->iommu_listener);
> > > > + }
> > > > memory_listener_unregister(&v->listener);
> > > >
> > > > return 0;
> > > > diff --git a/include/hw/virtio/vhost-vdpa.h b/include/hw/virtio/vhost-vdpa.h
> > > > index d10a89303e..22cbce7934 100644
> > > > --- a/include/hw/virtio/vhost-vdpa.h
> > > > +++ b/include/hw/virtio/vhost-vdpa.h
> > > > @@ -30,6 +30,7 @@ typedef struct vhost_vdpa {
> > > > uint32_t msg_type;
> > > > bool iotlb_batch_begin_sent;
> > > > MemoryListener listener;
> > > > + MemoryListener iommu_listener;
> > > > struct vhost_vdpa_iova_range iova_range;
> > > > uint64_t acked_features;
> > > > bool shadow_vqs_enabled;
> > > > @@ -41,8 +42,18 @@ typedef struct vhost_vdpa {
> > > > void *shadow_vq_ops_opaque;
> > > > struct vhost_dev *dev;
> > > > VhostVDPAHostNotifier notifier[VIRTIO_QUEUE_MAX];
> > > > + QLIST_HEAD(, vdpa_iommu) iommu_list;
> > > > + IOMMUNotifier n;
> > > > } VhostVDPA;
> > > >
> > > > +struct vdpa_iommu {
> > > > + struct vhost_vdpa *dev;
> > > > + IOMMUMemoryRegion *iommu_mr;
> > > > + hwaddr iommu_offset;
> > > > + IOMMUNotifier n;
> > > > + QLIST_ENTRY(vdpa_iommu) iommu_next;
> > > > +};
> > > > +
> > > > int vhost_vdpa_dma_map(struct vhost_vdpa *v, hwaddr iova, hwaddr size,
> > > > void *vaddr, bool readonly);
> > > > int vhost_vdpa_dma_unmap(struct vhost_vdpa *v, hwaddr iova, hwaddr size);
> > > > --
> > > > 2.34.3
> > > >
> > >
> >
>
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2022-10-24 6:21 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-10-19 6:39 [PATCH v2] vhost-vdpa: add support for vIOMMU Cindy Lu
2022-10-20 2:14 ` Jason Wang
2022-10-23 14:45 ` Cindy Lu
2022-10-24 1:51 ` Jason Wang
2022-10-24 5:21 ` Cindy Lu
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.