Skip to contentSkip to conversation

VIRTIO-IOMMU probe request support and MSI bypass on ARM

PATCH Superseded Eric Auger 4 participants 15 messages May 8, 2020 – May 22, 2020

v2 of 7 Superseded Latest is v7 →

lore
Share & tools
lore

Eric Auger

author
#0 lore raw

v2 cover VIRTIO-IOMMU probe request support and MSI bypass on ARM

By default the virtio-iommu translates MSI transactions. This behavior is inherited from ARM SMMU. However the virt machine code knows where the MSI doorbells are, so we can easily declare those regions as VIRTIO_IOMMU_RESV_MEM_T_MSI. With that setting the guest iommu subsystem will not need to map MSIs. This setup will simplify the VFIO integration.

In this series, the ITS or GICV2M doorbells are declared as HW MSI regions to be bypassed by the VIRTIO-IOMMU.

This also paves the way to the x86 integration where the MSI region, [0xFEE00000,0xFEEFFFFF], will be exposed by the q35 machine. However this will be handled in a separate series when not-DT support gets resolved.

Best Regards

Eric

This series can be found at:
https://github.com/eauger/qemu/tree/v5.0.0-virtio-iommu-msi-bypass-v2

History:

v1 -> v2:
- check which MSI controller is in use and advertise the corresponding MSI doorbell
- managed for both ITS and GICv2M
- various fixes spotted by Peter and Jean-Philippe, see individual logs

v1: Most of those patches were respinned from
[PATCH for-5.0 v11 00/20] VIRTIO-IOMMU device
except the last one which is new

Eric Auger (5):
qdev: Introduce DEFINE_PROP_RESERVED_REGION
virtio-iommu: Implement RESV_MEM probe request
virtio-iommu: Handle reserved regions in the translation process
virtio-iommu-pci: Add array of Interval properties hw/arm/virt: Let the virtio-iommu bypass MSIs

 include/exec/memory.h            |   6 ++
 include/hw/arm/virt.h            |   6 ++
 include/hw/qdev-properties.h     |   3 +
 include/hw/virtio/virtio-iommu.h |   2 +
 include/qemu/typedefs.h          |   1 +
 hw/arm/virt.c                    |  18 +++++
 hw/core/qdev-properties.c        |  89 ++++++++++++++++++++++++
 hw/virtio/virtio-iommu-pci.c     |   3 +
 hw/virtio/virtio-iommu.c         | 114 +++++++++++++++++++++++++++++--
 hw/virtio/trace-events           |   1 +
 10 files changed, 239 insertions(+), 4 deletions(-)
Signature · 1 line
2.20.1
author
#1 lore raw

PATCH v2 1/5 qdev: Introduce DEFINE_PROP_RESERVED_REGION

Introduce a new property defining a reserved region:
<low address>, <high address>, <type>.

This will be used to encode reserved IOVA regions.

For instance, in virtio-iommu use case, reserved IOVA regions will be passed by the machine code to the virtio-iommu-pci device (an array of those). The type of the reserved region will match the virtio_iommu_probe_resv_mem subtype value:
- VIRTIO_IOMMU_RESV_MEM_T_RESERVED (0)
- VIRTIO_IOMMU_RESV_MEM_T_MSI (1)

on PC/Q35 machine, this will be used to inform the virtio-iommu-pci device it should bypass the MSI region. The reserved region will be: 0xfee00000, 0xfeefffff, 1.

On ARM, we can declare the ITS MSI doorbell as an MSI region to prevent MSIs from being mapped on guest side.

---

v11 -> v12:
- rename into DEFINE_PROP_RESERVED_REGION
- do not use g_strsplit anymore, use endptr instead
- remove 0x references

---
 include/exec/memory.h        |  6 +++
 include/hw/qdev-properties.h |  3 ++
 include/qemu/typedefs.h      |  1 +
 hw/core/qdev-properties.c    | 89 ++++++++++++++++++++++++++++++++++++
 4 files changed, 99 insertions(+)
include/exec/memory.h +6 -0
index e000bd2f97..7e47afabe8 100644
@@ -57,6 +57,12 @@ struct MemoryRegionMmio {
     CPUWriteMemoryFunc *write[3];
 };
 
+struct ReservedRegion {
+    hwaddr low;
+    hwaddr high;
+    unsigned int type;
+};
+
 typedef struct IOMMUTLBEntry IOMMUTLBEntry;
 
 /* See address_space_translate: bit 0 is read, bit 1 is write.  */
include/hw/qdev-properties.h +3 -0
index f161604fb6..03bf850a7e 100644
@@ -19,6 +19,7 @@ extern const PropertyInfo qdev_prop_string;
 extern const PropertyInfo qdev_prop_chr;
 extern const PropertyInfo qdev_prop_tpm;
 extern const PropertyInfo qdev_prop_macaddr;
+extern const PropertyInfo qdev_prop_reserved_region;
 extern const PropertyInfo qdev_prop_on_off_auto;
 extern const PropertyInfo qdev_prop_multifd_compression;
 extern const PropertyInfo qdev_prop_losttickpolicy;
@@ -183,6 +184,8 @@ extern const PropertyInfo qdev_prop_pcie_link_width;
     DEFINE_PROP(_n, _s, _f, qdev_prop_drive_iothread, BlockBackend *)
 #define DEFINE_PROP_MACADDR(_n, _s, _f)         \
     DEFINE_PROP(_n, _s, _f, qdev_prop_macaddr, MACAddr)
+#define DEFINE_PROP_RESERVED_REGION(_n, _s, _f)         \
+    DEFINE_PROP(_n, _s, _f, qdev_prop_reserved_region, ReservedRegion)
 #define DEFINE_PROP_ON_OFF_AUTO(_n, _s, _f, _d) \
     DEFINE_PROP_SIGNED(_n, _s, _f, _d, qdev_prop_on_off_auto, OnOffAuto)
 #define DEFINE_PROP_MULTIFD_COMPRESSION(_n, _s, _f, _d) \
include/qemu/typedefs.h +1 -0
index ecf3cde26c..85c4f891f4 100644
@@ -59,6 +59,7 @@ typedef struct ISABus ISABus;
 typedef struct ISADevice ISADevice;
 typedef struct IsaDma IsaDma;
 typedef struct MACAddr MACAddr;
+typedef struct ReservedRegion ReservedRegion;
 typedef struct MachineClass MachineClass;
 typedef struct MachineState MachineState;
 typedef struct MemoryListener MemoryListener;
hw/core/qdev-properties.c +89 -0

101 lines not on this page · raw message

Signature · 1 line
2.20.1
author
#2 lore raw

PATCH v2 2/5 virtio-iommu: Implement RESV_MEM probe request

In reply to Eric Auger

This patch implements the PROBE request. At the moment, only THE RESV_MEM property is handled. The first goal is to report iommu wide reserved regions such as the MSI regions set by the machine code. On x86 this will be the IOAPIC MSI region, [0xFEE00000 - 0xFEEFFFFF], on ARM this may be the ITS doorbell.

In the future we may introduce per device reserved regions. This will be useful when protecting host assigned devices which may expose their own reserved regions

---

v1 -> v2:
- move the unlock back to the same place
- remove the push label and factorize the code after the out label
- fix a bunch of cpu_to_leX according to the latest spec revision
- do not remove sizeof(last) from free space
- check the ep exists

---
 include/hw/virtio/virtio-iommu.h |  2 +
 hw/virtio/virtio-iommu.c         | 94 ++++++++++++++++++++++++++++++--
 hw/virtio/trace-events           |  1 +
 3 files changed, 93 insertions(+), 4 deletions(-)
include/hw/virtio/virtio-iommu.h +2 -0
index e653004d7c..49eb105cd8 100644
@@ -53,6 +53,8 @@ typedef struct VirtIOIOMMU {
     GHashTable *as_by_busptr;
     IOMMUPciBus *iommu_pcibus_by_bus_num[PCI_BUS_MAX];
     PCIBus *primary_bus;
+    ReservedRegion *reserved_regions;
+    uint32_t nb_reserved_regions;
     GTree *domains;
     QemuMutex mutex;
     GTree *endpoints;
hw/virtio/virtio-iommu.c +90 -4

147 lines not on this page · raw message

hw/virtio/trace-events +1 -0
index e83500bee9..5550475691 100644
@@ -73,3 +73,4 @@ virtio_iommu_get_domain(uint32_t domain_id) "Alloc domain=%d"
 virtio_iommu_put_domain(uint32_t domain_id) "Free domain=%d"
 virtio_iommu_translate_out(uint64_t virt_addr, uint64_t phys_addr, uint32_t sid) "0x%"PRIx64" -> 0x%"PRIx64 " for sid=%d"
 virtio_iommu_report_fault(uint8_t reason, uint32_t flags, uint32_t endpoint, uint64_t addr) "FAULT reason=%d flags=%d endpoint=%d address =0x%"PRIx64
+virtio_iommu_fill_resv_property(uint32_t devid, uint8_t subtype, uint64_t start, uint64_t end) "dev= %d, type=%d start=0x%"PRIx64" end=0x%"PRIx64
Signature · 1 line
2.20.1
#3 lore raw

RE: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Eric,

Quoted 110 lines

-----Original Message-----
From: Eric Auger <[email protected]>
Sent: Friday, May 8, 2020 11:01 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; jean- [email protected]; Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]
Subject: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

External Email

----------------------------------------------------------------------
This patch implements the PROBE request. At the moment, only THE RESV_MEM property is handled. The first goal is to report iommu wide reserved regions such as the MSI regions set by the machine code. On x86 this will be the IOAPIC MSI region, [0xFEE00000 - 0xFEEFFFFF], on ARM this may be the ITS doorbell.

In the future we may introduce per device reserved regions.
This will be useful when protecting host assigned devices which may expose their own reserved regions

---

v1 -> v2:
- move the unlock back to the same place
- remove the push label and factorize the code after the out label
- fix a bunch of cpu_to_leX according to the latest spec revision
- do not remove sizeof(last) from free space
- check the ep exists

---
 include/hw/virtio/virtio-iommu.h |  2 +
 hw/virtio/virtio-iommu.c         | 94 ++++++++++++++++++++++++++++++--
 hw/virtio/trace-events           |  1 +
 3 files changed, 93 insertions(+), 4 deletions(-)
include/hw/virtio/virtio-iommu.h +2 -0
index e653004d7c..49eb105cd8 100644
@@ -53,6 +53,8 @@ typedef struct VirtIOIOMMU {
     GHashTable *as_by_busptr;
     IOMMUPciBus *iommu_pcibus_by_bus_num[PCI_BUS_MAX];
     PCIBus *primary_bus;
+    ReservedRegion *reserved_regions;
+    uint32_t nb_reserved_regions;
     GTree *domains;
     QemuMutex mutex;
     GTree *endpoints;
hw/virtio/virtio-iommu.c index renamed from hw/virtio/virtio-iommu.c +0 -0

22ba8848c2..35d772e021 100644

hw/virtio/virtio-iommu.c +45 -0
@@ -38,6 +38,7 @@
 
 /* Max size */
 #define VIOMMU_DEFAULT_QUEUE_SIZE 256
+#define VIOMMU_PROBE_SIZE 512
 
 typedef struct VirtIOIOMMUDomain {
     uint32_t id;
@@ -378,6 +379,65 @@ static int virtio_iommu_unmap(VirtIOIOMMU *s,
     return ret;
 }
 
+static ssize_t virtio_iommu_fill_resv_mem_prop(VirtIOIOMMU *s, uint32_t ep,
+                                               uint8_t *buf, size_t
+free) {
+    struct virtio_iommu_probe_resv_mem prop = {};
+    size_t size = sizeof(prop), length = size - sizeof(prop.head), total;
+    int i;
+
+    total = size * s->nb_reserved_regions;
+
+    if (total > free) {
+        return -ENOSPC;
+    }
+
+    for (i = 0; i < s->nb_reserved_regions; i++) {
+        prop.head.type = cpu_to_le16(VIRTIO_IOMMU_PROBE_T_RESV_MEM);
+        prop.head.length = cpu_to_le16(length);
+        prop.subtype = s->reserved_regions[i].type;
+        prop.start = cpu_to_le64(s->reserved_regions[i].low);
+        prop.end = cpu_to_le64(s->reserved_regions[i].high);
+
+        memcpy(buf, &prop, size);
+
+        trace_virtio_iommu_fill_resv_property(ep, prop.subtype,
+                                              prop.start, prop.end);
+        buf += size;
+    }
+    return total;
+}
+
+/**
+ * virtio_iommu_probe - Fill the probe request buffer with
+ * the properties the device is able to return and add a NONE
+ * property at the end.
+ */
+static int virtio_iommu_probe(VirtIOIOMMU *s,
+                              struct virtio_iommu_req_probe *req,
+                              uint8_t *buf) {
+    uint32_t ep_id = le32_to_cpu(req->endpoint);
+    size_t free = VIOMMU_PROBE_SIZE;
+    ssize_t count;
+
+    if (!virtio_iommu_mr(s, ep_id)) {
+        return VIRTIO_IOMMU_S_NOENT;
+    }

hw/virtio/virtio-iommu.c static int virtio_iommu_unmap(VirtIOIOMMU *s, · line 435

Quoted 10 lines
hw/virtio/virtio-iommu.c +10 -0
+
+    count = virtio_iommu_fill_resv_mem_prop(s, ep_id, buf, free);
+    if (count < 0) {
+        return VIRTIO_IOMMU_S_INVAL;
+    }
+    buf += count;
+    free -= count;
+
+    /* Fill the rest with zeroes */
+    memset(buf, 0, free);

No need to fill with zero here as "buf" is set to zero on allocation, no?

Thanks
-Bharat

Quoted 103 lines
hw/virtio/virtio-iommu.c +4 -0
+
+    return VIRTIO_IOMMU_S_OK;
+}
+
 static int virtio_iommu_iov_to_req(struct iovec *iov,
                                    unsigned int iov_cnt,
                                    void *req, size_t req_sz) @@ -407,15 +467,27 @@

virtio_iommu_handle_req(detach)
virtio_iommu_handle_req(map)
virtio_iommu_handle_req(unmap)

hw/virtio/virtio-iommu.c +11 -1
+static int virtio_iommu_handle_probe(VirtIOIOMMU *s,
+                                     struct iovec *iov,
+                                     unsigned int iov_cnt,
+                                     uint8_t *buf) {
+    struct virtio_iommu_req_probe req;
+    int ret = virtio_iommu_iov_to_req(iov, iov_cnt, &req, sizeof(req));
+
+    return ret ? ret : virtio_iommu_probe(s, &req, buf); }
+
 static void virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue *vq)  {
     VirtIOIOMMU *s = VIRTIO_IOMMU(vdev);
     struct virtio_iommu_req_head head;
     struct virtio_iommu_req_tail tail = {};
+    size_t output_size = sizeof(tail), sz;
     VirtQueueElement *elem;
     unsigned int iov_cnt;
     struct iovec *iov;
-    size_t sz;
+    void *buf = NULL;
 
     for (;;) {
         elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); @@ -452,6 +524,17 @@

static void virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue *vq)

         case VIRTIO_IOMMU_T_UNMAP:
             tail.status = virtio_iommu_handle_unmap(s, iov, iov_cnt);
             break;
hw/virtio/virtio-iommu.c +16 -3
+        case VIRTIO_IOMMU_T_PROBE:
+        {
+            struct virtio_iommu_req_tail *ptail;
+
+            output_size = s->config.probe_size + sizeof(tail);
+            buf = g_malloc0(output_size);
+
+            ptail = (struct virtio_iommu_req_tail *)
+                        (buf + s->config.probe_size);
+            ptail->status = virtio_iommu_handle_probe(s, iov, iov_cnt, buf);
+        }
         default:
             tail.status = VIRTIO_IOMMU_S_UNSUPP;
         }
@@ -459,12 +542,13 @@ static void
 virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue *vq)
 
 out:
         sz = iov_from_buf(elem->in_sg, elem->in_num, 0,
-                          &tail, sizeof(tail));
-        assert(sz == sizeof(tail));
+                          buf ? buf : &tail, output_size);
+        assert(sz == output_size);
 
-        virtqueue_push(vq, elem, sizeof(tail));
+        virtqueue_push(vq, elem, sz);
         virtio_notify(vdev, vq);
         g_free(elem);
+        g_free(buf);
     }
 }
@@ -667,6 +751,7 @@ static void virtio_iommu_device_realize(DeviceState *dev,
 Error **errp)
     s->config.page_size_mask = TARGET_PAGE_MASK;
     s->config.input_range.end = -1UL;
     s->config.domain_range.end = 32;
+    s->config.probe_size = VIOMMU_PROBE_SIZE;
 
     virtio_add_feature(&s->features, VIRTIO_RING_F_EVENT_IDX);
     virtio_add_feature(&s->features, VIRTIO_RING_F_INDIRECT_DESC); @@ -676,6
+761,7 @@ static void virtio_iommu_device_realize(DeviceState *dev, Error
**errp)
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MAP_UNMAP);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_BYPASS);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MMIO);
hw/virtio/virtio-iommu.c +1 -0
+    virtio_add_feature(&s->features, VIRTIO_IOMMU_F_PROBE);
 
     qemu_mutex_init(&s->mutex);
hw/virtio/trace-events index renamed from hw/virtio/trace-events +0 -0

e83500bee9..5550475691 100644

hw/virtio/trace-events +0 -0
@@ -73,3 +73,4 @@ virtio_iommu_get_domain(uint32_t domain_id) "Alloc

domain=%d"
virtio_iommu_put_domain(uint32_t domain_id) "Free domain=%d" virtio_iommu_translate_out(uint64_t virt_addr, uint64_t phys_addr, uint32_t sid)
"0x%"PRIx64" -> 0x%"PRIx64 " for sid=%d"
virtio_iommu_report_fault(uint8_t reason, uint32_t flags, uint32_t endpoint,
uint64_t addr) "FAULT reason=%d flags=%d endpoint=%d address =0x%"PRIx64

hw/virtio/trace-events +3 -1
+virtio_iommu_fill_resv_property(uint32_t devid, uint8_t subtype,
+uint64_t start, uint64_t end) "dev= %d, type=%d start=0x%"PRIx64"
+end=0x%"PRIx64
--

2.20.1

author
#4 lore raw

Re: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Bharat,

Quoted 116 lines · Bharat Bhushan

Hi Eric,

Quoted 114 lines

-----Original Message-----
From: Eric Auger <[email protected]>
Sent: Friday, May 8, 2020 11:01 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; jean- [email protected]; Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]
Subject: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

External Email

----------------------------------------------------------------------
This patch implements the PROBE request. At the moment, only THE RESV_MEM property is handled. The first goal is to report iommu wide reserved regions such as the MSI regions set by the machine code. On x86 this will be the IOAPIC MSI region, [0xFEE00000 - 0xFEEFFFFF], on ARM this may be the ITS doorbell.

In the future we may introduce per device reserved regions.
This will be useful when protecting host assigned devices which may expose their own reserved regions

---

v1 -> v2:
- move the unlock back to the same place
- remove the push label and factorize the code after the out label
- fix a bunch of cpu_to_leX according to the latest spec revision
- do not remove sizeof(last) from free space
- check the ep exists

---
 include/hw/virtio/virtio-iommu.h |  2 +
 hw/virtio/virtio-iommu.c         | 94 ++++++++++++++++++++++++++++++--
 hw/virtio/trace-events           |  1 +
 3 files changed, 93 insertions(+), 4 deletions(-)
include/hw/virtio/virtio-iommu.h +2 -0
index e653004d7c..49eb105cd8 100644
@@ -53,6 +53,8 @@ typedef struct VirtIOIOMMU {
     GHashTable *as_by_busptr;
     IOMMUPciBus *iommu_pcibus_by_bus_num[PCI_BUS_MAX];
     PCIBus *primary_bus;
+    ReservedRegion *reserved_regions;
+    uint32_t nb_reserved_regions;
     GTree *domains;
     QemuMutex mutex;
     GTree *endpoints;
hw/virtio/virtio-iommu.c index renamed from hw/virtio/virtio-iommu.c +0 -0

22ba8848c2..35d772e021 100644

hw/virtio/virtio-iommu.c +49 -0
@@ -38,6 +38,7 @@
 
 /* Max size */
 #define VIOMMU_DEFAULT_QUEUE_SIZE 256
+#define VIOMMU_PROBE_SIZE 512
 
 typedef struct VirtIOIOMMUDomain {
     uint32_t id;
@@ -378,6 +379,65 @@ static int virtio_iommu_unmap(VirtIOIOMMU *s,
     return ret;
 }
 
+static ssize_t virtio_iommu_fill_resv_mem_prop(VirtIOIOMMU *s, uint32_t ep,
+                                               uint8_t *buf, size_t
+free) {
+    struct virtio_iommu_probe_resv_mem prop = {};
+    size_t size = sizeof(prop), length = size - sizeof(prop.head), total;
+    int i;
+
+    total = size * s->nb_reserved_regions;
+
+    if (total > free) {
+        return -ENOSPC;
+    }
+
+    for (i = 0; i < s->nb_reserved_regions; i++) {
+        prop.head.type = cpu_to_le16(VIRTIO_IOMMU_PROBE_T_RESV_MEM);
+        prop.head.length = cpu_to_le16(length);
+        prop.subtype = s->reserved_regions[i].type;
+        prop.start = cpu_to_le64(s->reserved_regions[i].low);
+        prop.end = cpu_to_le64(s->reserved_regions[i].high);
+
+        memcpy(buf, &prop, size);
+
+        trace_virtio_iommu_fill_resv_property(ep, prop.subtype,
+                                              prop.start, prop.end);
+        buf += size;
+    }
+    return total;
+}
+
+/**
+ * virtio_iommu_probe - Fill the probe request buffer with
+ * the properties the device is able to return and add a NONE
+ * property at the end.
+ */
+static int virtio_iommu_probe(VirtIOIOMMU *s,
+                              struct virtio_iommu_req_probe *req,
+                              uint8_t *buf) {
+    uint32_t ep_id = le32_to_cpu(req->endpoint);
+    size_t free = VIOMMU_PROBE_SIZE;
+    ssize_t count;
+
+    if (!virtio_iommu_mr(s, ep_id)) {
+        return VIRTIO_IOMMU_S_NOENT;
+    }
+
+    count = virtio_iommu_fill_resv_mem_prop(s, ep_id, buf, free);
+    if (count < 0) {
+        return VIRTIO_IOMMU_S_INVAL;
Quoted 8 lines

hw/virtio/virtio-iommu.c static int virtio_iommu_unmap(VirtIOIOMMU *s, · line 436

Quoted 6 lines
hw/virtio/virtio-iommu.c +6 -0
+    }
+    buf += count;
+    free -= count;
+
+    /* Fill the rest with zeroes */
+    memset(buf, 0, free);

No need to fill with zero here as "buf" is set to zero on allocation, no?

You're right. I will remove this in the next version.

Thanks

Eric

Quoted 109 lines

Thanks
-Bharat

Quoted 103 lines
hw/virtio/virtio-iommu.c +4 -0
+
+    return VIRTIO_IOMMU_S_OK;
+}
+
 static int virtio_iommu_iov_to_req(struct iovec *iov,
                                    unsigned int iov_cnt,
                                    void *req, size_t req_sz) @@ -407,15 +467,27 @@

virtio_iommu_handle_req(detach)
virtio_iommu_handle_req(map)
virtio_iommu_handle_req(unmap)

hw/virtio/virtio-iommu.c +11 -1
+static int virtio_iommu_handle_probe(VirtIOIOMMU *s,
+                                     struct iovec *iov,
+                                     unsigned int iov_cnt,
+                                     uint8_t *buf) {
+    struct virtio_iommu_req_probe req;
+    int ret = virtio_iommu_iov_to_req(iov, iov_cnt, &req, sizeof(req));
+
+    return ret ? ret : virtio_iommu_probe(s, &req, buf); }
+
 static void virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue *vq)  {
     VirtIOIOMMU *s = VIRTIO_IOMMU(vdev);
     struct virtio_iommu_req_head head;
     struct virtio_iommu_req_tail tail = {};
+    size_t output_size = sizeof(tail), sz;
     VirtQueueElement *elem;
     unsigned int iov_cnt;
     struct iovec *iov;
-    size_t sz;
+    void *buf = NULL;
 
     for (;;) {
         elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); @@ -452,6 +524,17 @@

static void virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue *vq)

         case VIRTIO_IOMMU_T_UNMAP:
             tail.status = virtio_iommu_handle_unmap(s, iov, iov_cnt);
             break;
hw/virtio/virtio-iommu.c +16 -3
+        case VIRTIO_IOMMU_T_PROBE:
+        {
+            struct virtio_iommu_req_tail *ptail;
+
+            output_size = s->config.probe_size + sizeof(tail);
+            buf = g_malloc0(output_size);
+
+            ptail = (struct virtio_iommu_req_tail *)
+                        (buf + s->config.probe_size);
+            ptail->status = virtio_iommu_handle_probe(s, iov, iov_cnt, buf);
+        }
         default:
             tail.status = VIRTIO_IOMMU_S_UNSUPP;
         }
@@ -459,12 +542,13 @@ static void
 virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue *vq)
 
 out:
         sz = iov_from_buf(elem->in_sg, elem->in_num, 0,
-                          &tail, sizeof(tail));
-        assert(sz == sizeof(tail));
+                          buf ? buf : &tail, output_size);
+        assert(sz == output_size);
 
-        virtqueue_push(vq, elem, sizeof(tail));
+        virtqueue_push(vq, elem, sz);
         virtio_notify(vdev, vq);
         g_free(elem);
+        g_free(buf);
     }
 }
@@ -667,6 +751,7 @@ static void virtio_iommu_device_realize(DeviceState *dev,
 Error **errp)
     s->config.page_size_mask = TARGET_PAGE_MASK;
     s->config.input_range.end = -1UL;
     s->config.domain_range.end = 32;
+    s->config.probe_size = VIOMMU_PROBE_SIZE;
 
     virtio_add_feature(&s->features, VIRTIO_RING_F_EVENT_IDX);
     virtio_add_feature(&s->features, VIRTIO_RING_F_INDIRECT_DESC); @@ -676,6
+761,7 @@ static void virtio_iommu_device_realize(DeviceState *dev, Error
**errp)
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MAP_UNMAP);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_BYPASS);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MMIO);
hw/virtio/virtio-iommu.c +1 -0
+    virtio_add_feature(&s->features, VIRTIO_IOMMU_F_PROBE);
 
     qemu_mutex_init(&s->mutex);
hw/virtio/trace-events index renamed from hw/virtio/trace-events +0 -0

e83500bee9..5550475691 100644

hw/virtio/trace-events +0 -0
@@ -73,3 +73,4 @@ virtio_iommu_get_domain(uint32_t domain_id) "Alloc

domain=%d"
virtio_iommu_put_domain(uint32_t domain_id) "Free domain=%d" virtio_iommu_translate_out(uint64_t virt_addr, uint64_t phys_addr, uint32_t sid)
"0x%"PRIx64" -> 0x%"PRIx64 " for sid=%d"
virtio_iommu_report_fault(uint8_t reason, uint32_t flags, uint32_t endpoint,
uint64_t addr) "FAULT reason=%d flags=%d endpoint=%d address =0x%"PRIx64

hw/virtio/trace-events +3 -1
+virtio_iommu_fill_resv_property(uint32_t devid, uint8_t subtype,
+uint64_t start, uint64_t end) "dev= %d, type=%d start=0x%"PRIx64"
+end=0x%"PRIx64
--

2.20.1

#5 lore raw

RE: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Eric,

Quoted 185 lines

-----Original Message-----
From: Auger Eric <[email protected]>
Sent: Monday, May 11, 2020 12:26 PM
To: Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Bharat,

Quoted 121 lines · Bharat Bhushan

Hi Eric,

Quoted 119 lines

-----Original Message-----
From: Eric Auger <[email protected]>
Sent: Friday, May 8, 2020 11:01 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; jean- [email protected]; Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]
Subject: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

External Email

---------------------------------------------------------------------
- This patch implements the PROBE request. At the moment, only THE RESV_MEM property is handled. The first goal is to report iommu wide reserved regions such as the MSI regions set by the machine code. On x86 this will be the IOAPIC MSI region,
[0xFEE00000 - 0xFEEFFFFF], on ARM this may be the ITS doorbell.

In the future we may introduce per device reserved regions. This will be useful when protecting host assigned devices which may expose their own reserved regions

---

v1 -> v2:
- move the unlock back to the same place
- remove the push label and factorize the code after the out label
- fix a bunch of cpu_to_leX according to the latest spec revision
- do not remove sizeof(last) from free space
- check the ep exists

---
 include/hw/virtio/virtio-iommu.h |  2 +
 hw/virtio/virtio-iommu.c         | 94 ++++++++++++++++++++++++++++++--
 hw/virtio/trace-events           |  1 +
 3 files changed, 93 insertions(+), 4 deletions(-)
include/hw/virtio/virtio-iommu.h +2 -0
index e653004d7c..49eb105cd8 100644
@@ -53,6 +53,8 @@ typedef struct VirtIOIOMMU {
     GHashTable *as_by_busptr;
     IOMMUPciBus *iommu_pcibus_by_bus_num[PCI_BUS_MAX];
     PCIBus *primary_bus;
+    ReservedRegion *reserved_regions;
+    uint32_t nb_reserved_regions;
     GTree *domains;
     QemuMutex mutex;
     GTree *endpoints;
hw/virtio/virtio-iommu.c +0 -0

index
22ba8848c2..35d772e021 100644

hw/virtio/virtio-iommu.c +49 -0
@@ -38,6 +38,7 @@
 
 /* Max size */
 #define VIOMMU_DEFAULT_QUEUE_SIZE 256
+#define VIOMMU_PROBE_SIZE 512
 
 typedef struct VirtIOIOMMUDomain {
     uint32_t id;
@@ -378,6 +379,65 @@ static int virtio_iommu_unmap(VirtIOIOMMU *s,
     return ret;
 }
 
+static ssize_t virtio_iommu_fill_resv_mem_prop(VirtIOIOMMU *s, uint32_t ep,
+                                               uint8_t *buf, size_t
+free) {
+    struct virtio_iommu_probe_resv_mem prop = {};
+    size_t size = sizeof(prop), length = size - sizeof(prop.head), total;
+    int i;
+
+    total = size * s->nb_reserved_regions;
+
+    if (total > free) {
+        return -ENOSPC;
+    }
+
+    for (i = 0; i < s->nb_reserved_regions; i++) {
+        prop.head.type = cpu_to_le16(VIRTIO_IOMMU_PROBE_T_RESV_MEM);
+        prop.head.length = cpu_to_le16(length);
+        prop.subtype = s->reserved_regions[i].type;
+        prop.start = cpu_to_le64(s->reserved_regions[i].low);
+        prop.end = cpu_to_le64(s->reserved_regions[i].high);
+
+        memcpy(buf, &prop, size);
+
+        trace_virtio_iommu_fill_resv_property(ep, prop.subtype,
+                                              prop.start, prop.end);
+        buf += size;
+    }
+    return total;
+}
+
+/**
+ * virtio_iommu_probe - Fill the probe request buffer with
+ * the properties the device is able to return and add a NONE
+ * property at the end.
+ */
+static int virtio_iommu_probe(VirtIOIOMMU *s,
+                              struct virtio_iommu_req_probe *req,
+                              uint8_t *buf) {
+    uint32_t ep_id = le32_to_cpu(req->endpoint);
+    size_t free = VIOMMU_PROBE_SIZE;
+    ssize_t count;
+
+    if (!virtio_iommu_mr(s, ep_id)) {
+        return VIRTIO_IOMMU_S_NOENT;
+    }
+
+    count = virtio_iommu_fill_resv_mem_prop(s, ep_id, buf, free);
+    if (count < 0) {
+        return VIRTIO_IOMMU_S_INVAL;
Quoted 8 lines

hw/virtio/virtio-iommu.c static int virtio_iommu_unmap(VirtIOIOMMU *s, · line 436

Quoted 6 lines
hw/virtio/virtio-iommu.c +6 -0
+    }
+    buf += count;
+    free -= count;
+
+    /* Fill the rest with zeroes */
+    memset(buf, 0, free);

No need to fill with zero here as "buf" is set to zero on allocation, no?

You're right. I will remove this in the next version.

Thanks

Eric

Quoted 16 lines

Thanks
-Bharat

Quoted 12 lines
hw/virtio/virtio-iommu.c +4 -1
+
+    return VIRTIO_IOMMU_S_OK;
+}
+
 static int virtio_iommu_iov_to_req(struct iovec *iov,
                                    unsigned int iov_cnt,
                                    void *req, size_t req_sz) @@
-407,15 +467,27 @@

virtio_iommu_handle_req(detach)
virtio_iommu_handle_req(map)
virtio_iommu_handle_req(unmap)

Quoted 11 lines

hw/virtio/virtio-iommu.c virtio_iommu_handle_req(detach)

Quoted 11 lines
hw/virtio/virtio-iommu.c +10 -0
+static int virtio_iommu_handle_probe(VirtIOIOMMU *s,
+                                     struct iovec *iov,
+                                     unsigned int iov_cnt,
+                                     uint8_t *buf) {
+    struct virtio_iommu_req_probe req;
+    int ret = virtio_iommu_iov_to_req(iov, iov_cnt, &req,
+sizeof(req));
+
+    return ret ? ret : virtio_iommu_probe(s, &req, buf); }
+
 static void virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue

*vq) {

Quoted 10 lines
Quoted 10 lines
     VirtIOIOMMU *s = VIRTIO_IOMMU(vdev);
     struct virtio_iommu_req_head head;
     struct virtio_iommu_req_tail tail = {};
hw/virtio/virtio-iommu.c +2 -1
+    size_t output_size = sizeof(tail), sz;
     VirtQueueElement *elem;
     unsigned int iov_cnt;
     struct iovec *iov;
-    size_t sz;
+    void *buf = NULL;
Quoted 8 lines
Quoted 3 lines
Quoted 3 lines

for (;;) {
elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); @@
-452,6 +524,17 @@ static void virtio_iommu_handle_command(VirtIODevice

*vdev, VirtQueue *vq)

Quoted 4 lines
Quoted 4 lines
         case VIRTIO_IOMMU_T_UNMAP:
             tail.status = virtio_iommu_handle_unmap(s, iov, iov_cnt);
             break;

+ case VIRTIO_IOMMU_T_PROBE:

As per spec
"
If the device does not offer the VIRTIO_IOMMU_F_PROBE feature, and if the driver sends a VIRTIO_- IOMMU_T_PROBE request, then the device SHOULD NOT write the buffer and SHOULD set the used length to zero.
"
So we should check if device supports "VIRTIO_IOMMU_F_PROBE" before proceed?

Thanks
-Bharat

Quoted 70 lines
Quoted 70 lines
Quoted 68 lines
hw/virtio/virtio-iommu.c +15 -3
+        {
+            struct virtio_iommu_req_tail *ptail;
+
+            output_size = s->config.probe_size + sizeof(tail);
+            buf = g_malloc0(output_size);
+
+            ptail = (struct virtio_iommu_req_tail *)
+                        (buf + s->config.probe_size);
+            ptail->status = virtio_iommu_handle_probe(s, iov, iov_cnt, buf);
+        }
         default:
             tail.status = VIRTIO_IOMMU_S_UNSUPP;
         }
@@ -459,12 +542,13 @@ static void
 virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue *vq)
 
 out:
         sz = iov_from_buf(elem->in_sg, elem->in_num, 0,
-                          &tail, sizeof(tail));
-        assert(sz == sizeof(tail));
+                          buf ? buf : &tail, output_size);
+        assert(sz == output_size);
 
-        virtqueue_push(vq, elem, sizeof(tail));
+        virtqueue_push(vq, elem, sz);
         virtio_notify(vdev, vq);
         g_free(elem);
+        g_free(buf);
     }
 }
@@ -667,6 +751,7 @@ static void
 virtio_iommu_device_realize(DeviceState *dev, Error **errp)
     s->config.page_size_mask = TARGET_PAGE_MASK;
     s->config.input_range.end = -1UL;
     s->config.domain_range.end = 32;
+    s->config.probe_size = VIOMMU_PROBE_SIZE;
 
     virtio_add_feature(&s->features, VIRTIO_RING_F_EVENT_IDX);
     virtio_add_feature(&s->features, VIRTIO_RING_F_INDIRECT_DESC);

@@ -676,6
+761,7 @@ static void virtio_iommu_device_realize(DeviceState *dev, +Error
**errp)

     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MAP_UNMAP);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_BYPASS);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MMIO);
hw/virtio/virtio-iommu.c +1 -0
+    virtio_add_feature(&s->features, VIRTIO_IOMMU_F_PROBE);
 
     qemu_mutex_init(&s->mutex);
hw/virtio/trace-events index renamed from hw/virtio/trace-events +0 -0

e83500bee9..5550475691 100644

hw/virtio/trace-events +0 -0
@@ -73,3 +73,4 @@ virtio_iommu_get_domain(uint32_t domain_id) "Alloc

domain=%d"
virtio_iommu_put_domain(uint32_t domain_id) "Free domain=%d" virtio_iommu_translate_out(uint64_t virt_addr, uint64_t phys_addr,
uint32_t sid) "0x%"PRIx64" -> 0x%"PRIx64 " for sid=%d"
virtio_iommu_report_fault(uint8_t reason, uint32_t flags, uint32_t
endpoint, uint64_t addr) "FAULT reason=%d flags=%d endpoint=%d address =0x%"PRIx64

hw/virtio/trace-events +3 -1
+virtio_iommu_fill_resv_property(uint32_t devid, uint8_t subtype,
+uint64_t start, uint64_t end) "dev= %d, type=%d start=0x%"PRIx64"
+end=0x%"PRIx64
--

2.20.1

author
#6 lore raw

Re: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Bharat,

Quoted 196 lines · Bharat Bhushan

Hi Eric,

Quoted 185 lines

-----Original Message-----
From: Auger Eric <[email protected]>
Sent: Monday, May 11, 2020 12:26 PM
To: Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Bharat,

Quoted 121 lines · Bharat Bhushan

Hi Eric,

Quoted 119 lines

-----Original Message-----
From: Eric Auger <[email protected]>
Sent: Friday, May 8, 2020 11:01 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; jean- [email protected]; Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]
Subject: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

External Email

---------------------------------------------------------------------
- This patch implements the PROBE request. At the moment, only THE RESV_MEM property is handled. The first goal is to report iommu wide reserved regions such as the MSI regions set by the machine code. On x86 this will be the IOAPIC MSI region,
[0xFEE00000 - 0xFEEFFFFF], on ARM this may be the ITS doorbell.

In the future we may introduce per device reserved regions. This will be useful when protecting host assigned devices which may expose their own reserved regions

---

v1 -> v2:
- move the unlock back to the same place
- remove the push label and factorize the code after the out label
- fix a bunch of cpu_to_leX according to the latest spec revision
- do not remove sizeof(last) from free space
- check the ep exists

---
 include/hw/virtio/virtio-iommu.h |  2 +
 hw/virtio/virtio-iommu.c         | 94 ++++++++++++++++++++++++++++++--
 hw/virtio/trace-events           |  1 +
 3 files changed, 93 insertions(+), 4 deletions(-)
include/hw/virtio/virtio-iommu.h +2 -0
index e653004d7c..49eb105cd8 100644
@@ -53,6 +53,8 @@ typedef struct VirtIOIOMMU {
     GHashTable *as_by_busptr;
     IOMMUPciBus *iommu_pcibus_by_bus_num[PCI_BUS_MAX];
     PCIBus *primary_bus;
+    ReservedRegion *reserved_regions;
+    uint32_t nb_reserved_regions;
     GTree *domains;
     QemuMutex mutex;
     GTree *endpoints;
hw/virtio/virtio-iommu.c +0 -0

index
22ba8848c2..35d772e021 100644

hw/virtio/virtio-iommu.c +49 -0
@@ -38,6 +38,7 @@
 
 /* Max size */
 #define VIOMMU_DEFAULT_QUEUE_SIZE 256
+#define VIOMMU_PROBE_SIZE 512
 
 typedef struct VirtIOIOMMUDomain {
     uint32_t id;
@@ -378,6 +379,65 @@ static int virtio_iommu_unmap(VirtIOIOMMU *s,
     return ret;
 }
 
+static ssize_t virtio_iommu_fill_resv_mem_prop(VirtIOIOMMU *s, uint32_t ep,
+                                               uint8_t *buf, size_t
+free) {
+    struct virtio_iommu_probe_resv_mem prop = {};
+    size_t size = sizeof(prop), length = size - sizeof(prop.head), total;
+    int i;
+
+    total = size * s->nb_reserved_regions;
+
+    if (total > free) {
+        return -ENOSPC;
+    }
+
+    for (i = 0; i < s->nb_reserved_regions; i++) {
+        prop.head.type = cpu_to_le16(VIRTIO_IOMMU_PROBE_T_RESV_MEM);
+        prop.head.length = cpu_to_le16(length);
+        prop.subtype = s->reserved_regions[i].type;
+        prop.start = cpu_to_le64(s->reserved_regions[i].low);
+        prop.end = cpu_to_le64(s->reserved_regions[i].high);
+
+        memcpy(buf, &prop, size);
+
+        trace_virtio_iommu_fill_resv_property(ep, prop.subtype,
+                                              prop.start, prop.end);
+        buf += size;
+    }
+    return total;
+}
+
+/**
+ * virtio_iommu_probe - Fill the probe request buffer with
+ * the properties the device is able to return and add a NONE
+ * property at the end.
+ */
+static int virtio_iommu_probe(VirtIOIOMMU *s,
+                              struct virtio_iommu_req_probe *req,
+                              uint8_t *buf) {
+    uint32_t ep_id = le32_to_cpu(req->endpoint);
+    size_t free = VIOMMU_PROBE_SIZE;
+    ssize_t count;
+
+    if (!virtio_iommu_mr(s, ep_id)) {
+        return VIRTIO_IOMMU_S_NOENT;
+    }
+
+    count = virtio_iommu_fill_resv_mem_prop(s, ep_id, buf, free);
+    if (count < 0) {
+        return VIRTIO_IOMMU_S_INVAL;
Quoted 8 lines

hw/virtio/virtio-iommu.c static int virtio_iommu_unmap(VirtIOIOMMU *s, · line 436

Quoted 6 lines
hw/virtio/virtio-iommu.c +6 -0
+    }
+    buf += count;
+    free -= count;
+
+    /* Fill the rest with zeroes */
+    memset(buf, 0, free);

No need to fill with zero here as "buf" is set to zero on allocation, no?

You're right. I will remove this in the next version.

Thanks

Eric

Quoted 16 lines

Thanks
-Bharat

Quoted 12 lines
hw/virtio/virtio-iommu.c +4 -1
+
+    return VIRTIO_IOMMU_S_OK;
+}
+
 static int virtio_iommu_iov_to_req(struct iovec *iov,
                                    unsigned int iov_cnt,
                                    void *req, size_t req_sz) @@
-407,15 +467,27 @@

virtio_iommu_handle_req(detach)
virtio_iommu_handle_req(map)
virtio_iommu_handle_req(unmap)

Quoted 11 lines

hw/virtio/virtio-iommu.c virtio_iommu_handle_req(detach)

Quoted 11 lines
hw/virtio/virtio-iommu.c +10 -0
+static int virtio_iommu_handle_probe(VirtIOIOMMU *s,
+                                     struct iovec *iov,
+                                     unsigned int iov_cnt,
+                                     uint8_t *buf) {
+    struct virtio_iommu_req_probe req;
+    int ret = virtio_iommu_iov_to_req(iov, iov_cnt, &req,
+sizeof(req));
+
+    return ret ? ret : virtio_iommu_probe(s, &req, buf); }
+
 static void virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue

*vq) {

Quoted 10 lines
Quoted 10 lines
     VirtIOIOMMU *s = VIRTIO_IOMMU(vdev);
     struct virtio_iommu_req_head head;
     struct virtio_iommu_req_tail tail = {};
hw/virtio/virtio-iommu.c +2 -1
+    size_t output_size = sizeof(tail), sz;
     VirtQueueElement *elem;
     unsigned int iov_cnt;
     struct iovec *iov;
-    size_t sz;
+    void *buf = NULL;
Quoted 8 lines
Quoted 3 lines
Quoted 3 lines

for (;;) {
elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); @@
-452,6 +524,17 @@ static void virtio_iommu_handle_command(VirtIODevice

*vdev, VirtQueue *vq)

Quoted 4 lines
Quoted 4 lines
         case VIRTIO_IOMMU_T_UNMAP:
             tail.status = virtio_iommu_handle_unmap(s, iov, iov_cnt);
             break;

+ case VIRTIO_IOMMU_T_PROBE:

Quoted 7 lines

As per spec
"
If the device does not offer the VIRTIO_IOMMU_F_PROBE feature, and if the driver sends a VIRTIO_- IOMMU_T_PROBE request, then the device SHOULD NOT write the buffer and SHOULD set the used length to zero.
"
So we should check if device supports "VIRTIO_IOMMU_F_PROBE" before proceed?

But are the device and from that patch onwards we do support the VIRTIO_IOMMU_F_PROBE feature, right?

Thanks

Eric

Quoted 75 lines

Thanks
-Bharat

Quoted 70 lines
Quoted 70 lines
Quoted 68 lines
hw/virtio/virtio-iommu.c +15 -3
+        {
+            struct virtio_iommu_req_tail *ptail;
+
+            output_size = s->config.probe_size + sizeof(tail);
+            buf = g_malloc0(output_size);
+
+            ptail = (struct virtio_iommu_req_tail *)
+                        (buf + s->config.probe_size);
+            ptail->status = virtio_iommu_handle_probe(s, iov, iov_cnt, buf);
+        }
         default:
             tail.status = VIRTIO_IOMMU_S_UNSUPP;
         }
@@ -459,12 +542,13 @@ static void
 virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue *vq)
 
 out:
         sz = iov_from_buf(elem->in_sg, elem->in_num, 0,
-                          &tail, sizeof(tail));
-        assert(sz == sizeof(tail));
+                          buf ? buf : &tail, output_size);
+        assert(sz == output_size);
 
-        virtqueue_push(vq, elem, sizeof(tail));
+        virtqueue_push(vq, elem, sz);
         virtio_notify(vdev, vq);
         g_free(elem);
+        g_free(buf);
     }
 }
@@ -667,6 +751,7 @@ static void
 virtio_iommu_device_realize(DeviceState *dev, Error **errp)
     s->config.page_size_mask = TARGET_PAGE_MASK;
     s->config.input_range.end = -1UL;
     s->config.domain_range.end = 32;
+    s->config.probe_size = VIOMMU_PROBE_SIZE;
 
     virtio_add_feature(&s->features, VIRTIO_RING_F_EVENT_IDX);
     virtio_add_feature(&s->features, VIRTIO_RING_F_INDIRECT_DESC);

@@ -676,6
+761,7 @@ static void virtio_iommu_device_realize(DeviceState *dev, +Error
**errp)

     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MAP_UNMAP);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_BYPASS);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MMIO);
hw/virtio/virtio-iommu.c +1 -0
+    virtio_add_feature(&s->features, VIRTIO_IOMMU_F_PROBE);
 
     qemu_mutex_init(&s->mutex);
hw/virtio/trace-events index renamed from hw/virtio/trace-events +0 -0

e83500bee9..5550475691 100644

hw/virtio/trace-events +0 -0
@@ -73,3 +73,4 @@ virtio_iommu_get_domain(uint32_t domain_id) "Alloc

domain=%d"
virtio_iommu_put_domain(uint32_t domain_id) "Free domain=%d" virtio_iommu_translate_out(uint64_t virt_addr, uint64_t phys_addr,
uint32_t sid) "0x%"PRIx64" -> 0x%"PRIx64 " for sid=%d"
virtio_iommu_report_fault(uint8_t reason, uint32_t flags, uint32_t
endpoint, uint64_t addr) "FAULT reason=%d flags=%d endpoint=%d address =0x%"PRIx64

hw/virtio/trace-events +3 -1
+virtio_iommu_fill_resv_property(uint32_t devid, uint8_t subtype,
+uint64_t start, uint64_t end) "dev= %d, type=%d start=0x%"PRIx64"
+end=0x%"PRIx64
--

2.20.1

#7 lore raw

RE: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Eric,

Quoted 219 lines

-----Original Message-----
From: Auger Eric <[email protected]>
Sent: Monday, May 11, 2020 2:19 PM
To: Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Bharat,

Quoted 83 lines · Bharat Bhushan

Hi Eric,

Quoted 81 lines

-----Original Message-----
From: Auger Eric <[email protected]>
Sent: Monday, May 11, 2020 12:26 PM
To: Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Bharat,

Quoted 69 lines · Bharat Bhushan

Hi Eric,

Quoted 67 lines

-----Original Message-----
From: Eric Auger <[email protected]>
Sent: Friday, May 8, 2020 11:01 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; jean- [email protected]; Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]
Subject: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

External Email

-------------------------------------------------------------------
--
- This patch implements the PROBE request. At the moment, only THE RESV_MEM property is handled. The first goal is to report iommu wide reserved regions such as the MSI regions set by the machine code. On
x86 this will be the IOAPIC MSI region,
[0xFEE00000 - 0xFEEFFFFF], on ARM this may be the ITS doorbell.

In the future we may introduce per device reserved regions. This will be useful when protecting host assigned devices which may expose their own reserved regions

---

v1 -> v2:
- move the unlock back to the same place
- remove the push label and factorize the code after the out label
- fix a bunch of cpu_to_leX according to the latest spec revision
- do not remove sizeof(last) from free space
- check the ep exists

---
 include/hw/virtio/virtio-iommu.h |  2 +
 hw/virtio/virtio-iommu.c         | 94 ++++++++++++++++++++++++++++++--
 hw/virtio/trace-events           |  1 +
 3 files changed, 93 insertions(+), 4 deletions(-)
include/hw/virtio/virtio-iommu.h +2 -0
index e653004d7c..49eb105cd8 100644
@@ -53,6 +53,8 @@ typedef struct VirtIOIOMMU {
     GHashTable *as_by_busptr;
     IOMMUPciBus *iommu_pcibus_by_bus_num[PCI_BUS_MAX];
     PCIBus *primary_bus;
+    ReservedRegion *reserved_regions;
+    uint32_t nb_reserved_regions;
     GTree *domains;
     QemuMutex mutex;
     GTree *endpoints;
hw/virtio/virtio-iommu.c +0 -0

index
22ba8848c2..35d772e021 100644

hw/virtio/virtio-iommu.c +1 -0
@@ -38,6 +38,7 @@
 
 /* Max size */
 #define VIOMMU_DEFAULT_QUEUE_SIZE 256
+#define VIOMMU_PROBE_SIZE 512
 
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines

hw/virtio/virtio-iommu.c static int virtio_iommu_unmap(VirtIOIOMMU *s, · line 382

Quoted 7 lines
hw/virtio/virtio-iommu.c +1 -0
 typedef struct VirtIOIOMMUDomain {
     uint32_t id;
@@ -378,6 +379,65 @@ static int virtio_iommu_unmap(VirtIOIOMMU *s,
     return ret;
 }
 
+static ssize_t virtio_iommu_fill_resv_mem_prop(VirtIOIOMMU *s, uint32_t

ep,

Quoted 106 lines
Quoted 106 lines
Quoted 48 lines
Quoted 48 lines
hw/virtio/virtio-iommu.c +48 -0
+                                               uint8_t *buf,
+size_t
+free) {
+    struct virtio_iommu_probe_resv_mem prop = {};
+    size_t size = sizeof(prop), length = size - sizeof(prop.head), total;
+    int i;
+
+    total = size * s->nb_reserved_regions;
+
+    if (total > free) {
+        return -ENOSPC;
+    }
+
+    for (i = 0; i < s->nb_reserved_regions; i++) {
+        prop.head.type = cpu_to_le16(VIRTIO_IOMMU_PROBE_T_RESV_MEM);
+        prop.head.length = cpu_to_le16(length);
+        prop.subtype = s->reserved_regions[i].type;
+        prop.start = cpu_to_le64(s->reserved_regions[i].low);
+        prop.end = cpu_to_le64(s->reserved_regions[i].high);
+
+        memcpy(buf, &prop, size);
+
+        trace_virtio_iommu_fill_resv_property(ep, prop.subtype,
+                                              prop.start, prop.end);
+        buf += size;
+    }
+    return total;
+}
+
+/**
+ * virtio_iommu_probe - Fill the probe request buffer with
+ * the properties the device is able to return and add a NONE
+ * property at the end.
+ */
+static int virtio_iommu_probe(VirtIOIOMMU *s,
+                              struct virtio_iommu_req_probe *req,
+                              uint8_t *buf) {
+    uint32_t ep_id = le32_to_cpu(req->endpoint);
+    size_t free = VIOMMU_PROBE_SIZE;
+    ssize_t count;
+
+    if (!virtio_iommu_mr(s, ep_id)) {
+        return VIRTIO_IOMMU_S_NOENT;
+    }
+
+    count = virtio_iommu_fill_resv_mem_prop(s, ep_id, buf, free);
+    if (count < 0) {
+        return VIRTIO_IOMMU_S_INVAL;
Quoted 8 lines

hw/virtio/virtio-iommu.c static int virtio_iommu_unmap(VirtIOIOMMU *s, · line 436

Quoted 6 lines
hw/virtio/virtio-iommu.c +6 -0
+    }
+    buf += count;
+    free -= count;
+
+    /* Fill the rest with zeroes */
+    memset(buf, 0, free);

No need to fill with zero here as "buf" is set to zero on allocation, no?

You're right. I will remove this in the next version.

Thanks

Eric

Quoted 16 lines

Thanks
-Bharat

Quoted 12 lines
hw/virtio/virtio-iommu.c +4 -1
+
+    return VIRTIO_IOMMU_S_OK;
+}
+
 static int virtio_iommu_iov_to_req(struct iovec *iov,
                                    unsigned int iov_cnt,
                                    void *req, size_t req_sz) @@
-407,15 +467,27 @@

virtio_iommu_handle_req(detach)
virtio_iommu_handle_req(map)
virtio_iommu_handle_req(unmap)

Quoted 12 lines
Quoted 11 lines
hw/virtio/virtio-iommu.c +10 -0
+static int virtio_iommu_handle_probe(VirtIOIOMMU *s,
+                                     struct iovec *iov,
+                                     unsigned int iov_cnt,
+                                     uint8_t *buf) {
+    struct virtio_iommu_req_probe req;
+    int ret = virtio_iommu_iov_to_req(iov, iov_cnt, &req,
+sizeof(req));
+
+    return ret ? ret : virtio_iommu_probe(s, &req, buf); }
+
 static void virtio_iommu_handle_command(VirtIODevice *vdev,
Quoted 1 line

VirtQueue

*vq) {

Quoted 10 lines
Quoted 10 lines
     VirtIOIOMMU *s = VIRTIO_IOMMU(vdev);
     struct virtio_iommu_req_head head;
     struct virtio_iommu_req_tail tail = {};
hw/virtio/virtio-iommu.c +2 -1
+    size_t output_size = sizeof(tail), sz;
     VirtQueueElement *elem;
     unsigned int iov_cnt;
     struct iovec *iov;
-    size_t sz;
+    void *buf = NULL;
Quoted 4 lines
Quoted 4 lines

for (;;) {
elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); @@
-452,6 +524,17 @@ static void virtio_iommu_handle_command(VirtIODevice

*vdev, VirtQueue *vq)

Quoted 8 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
         case VIRTIO_IOMMU_T_UNMAP:
             tail.status = virtio_iommu_handle_unmap(s, iov, iov_cnt);
             break;

+ case VIRTIO_IOMMU_T_PROBE:

As per spec
"
If the device does not offer the VIRTIO_IOMMU_F_PROBE feature, and if the

driver sends a VIRTIO_-

Quoted 8 lines
Quoted 1 line

IOMMU_T_PROBE request, then the device SHOULD NOT write the buffer and

SHOULD set the used

Quoted 3 lines

length to zero.
"
So we should check if device supports "VIRTIO_IOMMU_F_PROBE" before

proceed?
But are the device and from that patch onwards we do support the VIRTIO_IOMMU_F_PROBE feature, right?

Yes I agree, do you think if for debugging one wants to try out without this feature then he should just disable VIRTIO_IOMMU_F_PROBE.

Thanks
-Bharat

Quoted 79 lines

Thanks

Eric

Quoted 75 lines

Thanks
-Bharat

Quoted 70 lines
Quoted 70 lines
Quoted 68 lines
hw/virtio/virtio-iommu.c +15 -3
+        {
+            struct virtio_iommu_req_tail *ptail;
+
+            output_size = s->config.probe_size + sizeof(tail);
+            buf = g_malloc0(output_size);
+
+            ptail = (struct virtio_iommu_req_tail *)
+                        (buf + s->config.probe_size);
+            ptail->status = virtio_iommu_handle_probe(s, iov, iov_cnt, buf);
+        }
         default:
             tail.status = VIRTIO_IOMMU_S_UNSUPP;
         }
@@ -459,12 +542,13 @@ static void
 virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue *vq)
 
 out:
         sz = iov_from_buf(elem->in_sg, elem->in_num, 0,
-                          &tail, sizeof(tail));
-        assert(sz == sizeof(tail));
+                          buf ? buf : &tail, output_size);
+        assert(sz == output_size);
 
-        virtqueue_push(vq, elem, sizeof(tail));
+        virtqueue_push(vq, elem, sz);
         virtio_notify(vdev, vq);
         g_free(elem);
+        g_free(buf);
     }
 }
@@ -667,6 +751,7 @@ static void
 virtio_iommu_device_realize(DeviceState *dev, Error **errp)
     s->config.page_size_mask = TARGET_PAGE_MASK;
     s->config.input_range.end = -1UL;
     s->config.domain_range.end = 32;
+    s->config.probe_size = VIOMMU_PROBE_SIZE;
 
     virtio_add_feature(&s->features, VIRTIO_RING_F_EVENT_IDX);
     virtio_add_feature(&s->features, VIRTIO_RING_F_INDIRECT_DESC);

@@ -676,6
+761,7 @@ static void virtio_iommu_device_realize(DeviceState *dev, +Error
**errp)

     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MAP_UNMAP);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_BYPASS);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MMIO);
hw/virtio/virtio-iommu.c +1 -0
+    virtio_add_feature(&s->features, VIRTIO_IOMMU_F_PROBE);
 
     qemu_mutex_init(&s->mutex);
hw/virtio/trace-events index renamed from hw/virtio/trace-events +0 -0

e83500bee9..5550475691 100644

hw/virtio/trace-events +0 -0
@@ -73,3 +73,4 @@ virtio_iommu_get_domain(uint32_t domain_id)

"Alloc domain=%d"
virtio_iommu_put_domain(uint32_t domain_id) "Free domain=%d" virtio_iommu_translate_out(uint64_t virt_addr, uint64_t phys_addr,
uint32_t sid) "0x%"PRIx64" -> 0x%"PRIx64 " for sid=%d"
virtio_iommu_report_fault(uint8_t reason, uint32_t flags, uint32_t
endpoint, uint64_t addr) "FAULT reason=%d flags=%d endpoint=%d address =0x%"PRIx64

hw/virtio/trace-events +3 -1
+virtio_iommu_fill_resv_property(uint32_t devid, uint8_t subtype,
+uint64_t start, uint64_t end) "dev= %d, type=%d start=0x%"PRIx64"
+end=0x%"PRIx64
--

2.20.1

author
#8 lore raw

Re: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Bharat,

Quoted 223 lines · Bharat Bhushan

Hi Eric,

Quoted 221 lines

-----Original Message-----
From: Auger Eric <[email protected]>
Sent: Monday, May 11, 2020 2:19 PM
To: Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Bharat,

Quoted 83 lines · Bharat Bhushan

Hi Eric,

Quoted 81 lines

-----Original Message-----
From: Auger Eric <[email protected]>
Sent: Monday, May 11, 2020 12:26 PM
To: Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Bharat,

Quoted 69 lines · Bharat Bhushan

Hi Eric,

Quoted 67 lines

-----Original Message-----
From: Eric Auger <[email protected]>
Sent: Friday, May 8, 2020 11:01 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; jean- [email protected]; Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]
Subject: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

External Email

-------------------------------------------------------------------
--
- This patch implements the PROBE request. At the moment, only THE RESV_MEM property is handled. The first goal is to report iommu wide reserved regions such as the MSI regions set by the machine code. On
x86 this will be the IOAPIC MSI region,
[0xFEE00000 - 0xFEEFFFFF], on ARM this may be the ITS doorbell.

In the future we may introduce per device reserved regions. This will be useful when protecting host assigned devices which may expose their own reserved regions

---

v1 -> v2:
- move the unlock back to the same place
- remove the push label and factorize the code after the out label
- fix a bunch of cpu_to_leX according to the latest spec revision
- do not remove sizeof(last) from free space
- check the ep exists

---
 include/hw/virtio/virtio-iommu.h |  2 +
 hw/virtio/virtio-iommu.c         | 94 ++++++++++++++++++++++++++++++--
 hw/virtio/trace-events           |  1 +
 3 files changed, 93 insertions(+), 4 deletions(-)
include/hw/virtio/virtio-iommu.h +2 -0
index e653004d7c..49eb105cd8 100644
@@ -53,6 +53,8 @@ typedef struct VirtIOIOMMU {
     GHashTable *as_by_busptr;
     IOMMUPciBus *iommu_pcibus_by_bus_num[PCI_BUS_MAX];
     PCIBus *primary_bus;
+    ReservedRegion *reserved_regions;
+    uint32_t nb_reserved_regions;
     GTree *domains;
     QemuMutex mutex;
     GTree *endpoints;
hw/virtio/virtio-iommu.c +0 -0

index
22ba8848c2..35d772e021 100644

hw/virtio/virtio-iommu.c +1 -0
@@ -38,6 +38,7 @@
 
 /* Max size */
 #define VIOMMU_DEFAULT_QUEUE_SIZE 256
+#define VIOMMU_PROBE_SIZE 512
 
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines

hw/virtio/virtio-iommu.c static int virtio_iommu_unmap(VirtIOIOMMU *s, · line 382

Quoted 7 lines
hw/virtio/virtio-iommu.c +1 -0
 typedef struct VirtIOIOMMUDomain {
     uint32_t id;
@@ -378,6 +379,65 @@ static int virtio_iommu_unmap(VirtIOIOMMU *s,
     return ret;
 }
 
+static ssize_t virtio_iommu_fill_resv_mem_prop(VirtIOIOMMU *s, uint32_t

ep,

Quoted 106 lines
Quoted 106 lines
Quoted 48 lines
Quoted 48 lines
hw/virtio/virtio-iommu.c +48 -0
+                                               uint8_t *buf,
+size_t
+free) {
+    struct virtio_iommu_probe_resv_mem prop = {};
+    size_t size = sizeof(prop), length = size - sizeof(prop.head), total;
+    int i;
+
+    total = size * s->nb_reserved_regions;
+
+    if (total > free) {
+        return -ENOSPC;
+    }
+
+    for (i = 0; i < s->nb_reserved_regions; i++) {
+        prop.head.type = cpu_to_le16(VIRTIO_IOMMU_PROBE_T_RESV_MEM);
+        prop.head.length = cpu_to_le16(length);
+        prop.subtype = s->reserved_regions[i].type;
+        prop.start = cpu_to_le64(s->reserved_regions[i].low);
+        prop.end = cpu_to_le64(s->reserved_regions[i].high);
+
+        memcpy(buf, &prop, size);
+
+        trace_virtio_iommu_fill_resv_property(ep, prop.subtype,
+                                              prop.start, prop.end);
+        buf += size;
+    }
+    return total;
+}
+
+/**
+ * virtio_iommu_probe - Fill the probe request buffer with
+ * the properties the device is able to return and add a NONE
+ * property at the end.
+ */
+static int virtio_iommu_probe(VirtIOIOMMU *s,
+                              struct virtio_iommu_req_probe *req,
+                              uint8_t *buf) {
+    uint32_t ep_id = le32_to_cpu(req->endpoint);
+    size_t free = VIOMMU_PROBE_SIZE;
+    ssize_t count;
+
+    if (!virtio_iommu_mr(s, ep_id)) {
+        return VIRTIO_IOMMU_S_NOENT;
+    }
+
+    count = virtio_iommu_fill_resv_mem_prop(s, ep_id, buf, free);
+    if (count < 0) {
+        return VIRTIO_IOMMU_S_INVAL;
Quoted 8 lines

hw/virtio/virtio-iommu.c static int virtio_iommu_unmap(VirtIOIOMMU *s, · line 436

Quoted 6 lines
hw/virtio/virtio-iommu.c +6 -0
+    }
+    buf += count;
+    free -= count;
+
+    /* Fill the rest with zeroes */
+    memset(buf, 0, free);

No need to fill with zero here as "buf" is set to zero on allocation, no?

You're right. I will remove this in the next version.

Thanks

Eric

Quoted 16 lines

Thanks
-Bharat

Quoted 12 lines
hw/virtio/virtio-iommu.c +4 -1
+
+    return VIRTIO_IOMMU_S_OK;
+}
+
 static int virtio_iommu_iov_to_req(struct iovec *iov,
                                    unsigned int iov_cnt,
                                    void *req, size_t req_sz) @@
-407,15 +467,27 @@

virtio_iommu_handle_req(detach)
virtio_iommu_handle_req(map)
virtio_iommu_handle_req(unmap)

Quoted 12 lines
Quoted 11 lines
hw/virtio/virtio-iommu.c +10 -0
+static int virtio_iommu_handle_probe(VirtIOIOMMU *s,
+                                     struct iovec *iov,
+                                     unsigned int iov_cnt,
+                                     uint8_t *buf) {
+    struct virtio_iommu_req_probe req;
+    int ret = virtio_iommu_iov_to_req(iov, iov_cnt, &req,
+sizeof(req));
+
+    return ret ? ret : virtio_iommu_probe(s, &req, buf); }
+
 static void virtio_iommu_handle_command(VirtIODevice *vdev,
Quoted 1 line

VirtQueue

*vq) {

Quoted 10 lines
Quoted 10 lines
     VirtIOIOMMU *s = VIRTIO_IOMMU(vdev);
     struct virtio_iommu_req_head head;
     struct virtio_iommu_req_tail tail = {};
hw/virtio/virtio-iommu.c +2 -1
+    size_t output_size = sizeof(tail), sz;
     VirtQueueElement *elem;
     unsigned int iov_cnt;
     struct iovec *iov;
-    size_t sz;
+    void *buf = NULL;
Quoted 4 lines
Quoted 4 lines

for (;;) {
elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); @@
-452,6 +524,17 @@ static void virtio_iommu_handle_command(VirtIODevice

*vdev, VirtQueue *vq)

Quoted 8 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
         case VIRTIO_IOMMU_T_UNMAP:
             tail.status = virtio_iommu_handle_unmap(s, iov, iov_cnt);
             break;

+ case VIRTIO_IOMMU_T_PROBE:

As per spec
"
If the device does not offer the VIRTIO_IOMMU_F_PROBE feature, and if the

driver sends a VIRTIO_-

Quoted 1 line

IOMMU_T_PROBE request, then the device SHOULD NOT write the buffer and

SHOULD set the used

Quoted 8 lines
Quoted 6 lines
Quoted 3 lines

length to zero.
"
So we should check if device supports "VIRTIO_IOMMU_F_PROBE" before

proceed?
But are the device and from that patch onwards we do support the VIRTIO_IOMMU_F_PROBE feature, right?

Yes I agree, do you think if for debugging one wants to try out without this feature then he should just disable VIRTIO_IOMMU_F_PROBE.

You mean for debugging the driver? I don't think this is the purpose of this device.

Thanks

Eric

Quoted 84 lines

Thanks
-Bharat

Quoted 79 lines

Thanks

Eric

Quoted 75 lines

Thanks
-Bharat

Quoted 70 lines
Quoted 70 lines
Quoted 68 lines
hw/virtio/virtio-iommu.c +15 -3
+        {
+            struct virtio_iommu_req_tail *ptail;
+
+            output_size = s->config.probe_size + sizeof(tail);
+            buf = g_malloc0(output_size);
+
+            ptail = (struct virtio_iommu_req_tail *)
+                        (buf + s->config.probe_size);
+            ptail->status = virtio_iommu_handle_probe(s, iov, iov_cnt, buf);
+        }
         default:
             tail.status = VIRTIO_IOMMU_S_UNSUPP;
         }
@@ -459,12 +542,13 @@ static void
 virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue *vq)
 
 out:
         sz = iov_from_buf(elem->in_sg, elem->in_num, 0,
-                          &tail, sizeof(tail));
-        assert(sz == sizeof(tail));
+                          buf ? buf : &tail, output_size);
+        assert(sz == output_size);
 
-        virtqueue_push(vq, elem, sizeof(tail));
+        virtqueue_push(vq, elem, sz);
         virtio_notify(vdev, vq);
         g_free(elem);
+        g_free(buf);
     }
 }
@@ -667,6 +751,7 @@ static void
 virtio_iommu_device_realize(DeviceState *dev, Error **errp)
     s->config.page_size_mask = TARGET_PAGE_MASK;
     s->config.input_range.end = -1UL;
     s->config.domain_range.end = 32;
+    s->config.probe_size = VIOMMU_PROBE_SIZE;
 
     virtio_add_feature(&s->features, VIRTIO_RING_F_EVENT_IDX);
     virtio_add_feature(&s->features, VIRTIO_RING_F_INDIRECT_DESC);

@@ -676,6
+761,7 @@ static void virtio_iommu_device_realize(DeviceState *dev, +Error
**errp)

     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MAP_UNMAP);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_BYPASS);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MMIO);
hw/virtio/virtio-iommu.c +1 -0
+    virtio_add_feature(&s->features, VIRTIO_IOMMU_F_PROBE);
 
     qemu_mutex_init(&s->mutex);
hw/virtio/trace-events index renamed from hw/virtio/trace-events +0 -0

e83500bee9..5550475691 100644

hw/virtio/trace-events +0 -0
@@ -73,3 +73,4 @@ virtio_iommu_get_domain(uint32_t domain_id)

"Alloc domain=%d"
virtio_iommu_put_domain(uint32_t domain_id) "Free domain=%d" virtio_iommu_translate_out(uint64_t virt_addr, uint64_t phys_addr,
uint32_t sid) "0x%"PRIx64" -> 0x%"PRIx64 " for sid=%d"
virtio_iommu_report_fault(uint8_t reason, uint32_t flags, uint32_t
endpoint, uint64_t addr) "FAULT reason=%d flags=%d endpoint=%d address =0x%"PRIx64

hw/virtio/trace-events +3 -1
+virtio_iommu_fill_resv_property(uint32_t devid, uint8_t subtype,
+uint64_t start, uint64_t end) "dev= %d, type=%d start=0x%"PRIx64"
+end=0x%"PRIx64
--

2.20.1

#9 lore raw

RE: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Eric,

Quoted 245 lines

-----Original Message-----
From: Auger Eric <[email protected]>
Sent: Tuesday, May 12, 2020 8:39 AM
To: Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Bharat,

Quoted 110 lines · Bharat Bhushan

Hi Eric,

Quoted 108 lines

-----Original Message-----
From: Auger Eric <[email protected]>
Sent: Monday, May 11, 2020 2:19 PM
To: Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Bharat,

Quoted 86 lines · Bharat Bhushan

Hi Eric,

Quoted 84 lines

-----Original Message-----
From: Auger Eric <[email protected]>
Sent: Monday, May 11, 2020 12:26 PM
To: Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

Hi Bharat,

Quoted 71 lines · Bharat Bhushan

Hi Eric,

Quoted 69 lines

-----Original Message-----
From: Eric Auger <[email protected]>
Sent: Friday, May 8, 2020 11:01 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; jean- [email protected]; Bharat Bhushan <[email protected]>; [email protected]; [email protected]; [email protected]
Subject: [EXT] [PATCH v2 2/5] virtio-iommu: Implement RESV_MEM probe request

External Email

-----------------------------------------------------------------
--
--
- This patch implements the PROBE request. At the moment, only THE RESV_MEM property is handled. The first goal is to report iommu wide reserved regions such as the MSI regions set by the machine code. On
x86 this will be the IOAPIC MSI region,
[0xFEE00000 - 0xFEEFFFFF], on ARM this may be the ITS doorbell.

In the future we may introduce per device reserved regions. This will be useful when protecting host assigned devices which may expose their own reserved regions

---

v1 -> v2:
- move the unlock back to the same place
- remove the push label and factorize the code after the out label
- fix a bunch of cpu_to_leX according to the latest spec revision
- do not remove sizeof(last) from free space
- check the ep exists

---
 include/hw/virtio/virtio-iommu.h |  2 +
 hw/virtio/virtio-iommu.c         | 94 ++++++++++++++++++++++++++++++--
 hw/virtio/trace-events           |  1 +
 3 files changed, 93 insertions(+), 4 deletions(-)
include/hw/virtio/virtio-iommu.h +2 -0
index e653004d7c..49eb105cd8 100644
@@ -53,6 +53,8 @@ typedef struct VirtIOIOMMU {
     GHashTable *as_by_busptr;
     IOMMUPciBus *iommu_pcibus_by_bus_num[PCI_BUS_MAX];
     PCIBus *primary_bus;
+    ReservedRegion *reserved_regions;
+    uint32_t nb_reserved_regions;
     GTree *domains;
     QemuMutex mutex;
     GTree *endpoints;
hw/virtio/virtio-iommu.c +0 -0

index
22ba8848c2..35d772e021 100644

hw/virtio/virtio-iommu.c +1 -0
@@ -38,6 +38,7 @@
 
 /* Max size */
 #define VIOMMU_DEFAULT_QUEUE_SIZE 256
+#define VIOMMU_PROBE_SIZE 512
 
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

hw/virtio/virtio-iommu.c static int virtio_iommu_unmap(VirtIOIOMMU *s, · line 383

Quoted 8 lines
hw/virtio/virtio-iommu.c +2 -0
 typedef struct VirtIOIOMMUDomain {
     uint32_t id;
@@ -378,6 +379,65 @@ static int virtio_iommu_unmap(VirtIOIOMMU *s,
     return ret;
 }
 
+static ssize_t virtio_iommu_fill_resv_mem_prop(VirtIOIOMMU *s,
+uint32_t

ep,

Quoted 15 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
hw/virtio/virtio-iommu.c +7 -0
+                                               uint8_t *buf,
+size_t
+free) {
+    struct virtio_iommu_probe_resv_mem prop = {};
+    size_t size = sizeof(prop), length = size - sizeof(prop.head), total;
+    int i;
+
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

hw/virtio/virtio-iommu.c static int virtio_iommu_unmap(VirtIOIOMMU *s, · line 398

Quoted 8 lines
hw/virtio/virtio-iommu.c +8 -0
+    total = size * s->nb_reserved_regions;
+
+    if (total > free) {
+        return -ENOSPC;
+    }
+
+    for (i = 0; i < s->nb_reserved_regions; i++) {
+        prop.head.type =

cpu_to_le16(VIRTIO_IOMMU_PROBE_T_RESV_MEM);

Quoted 107 lines
Quoted 107 lines
Quoted 96 lines
Quoted 86 lines
Quoted 33 lines
Quoted 33 lines
hw/virtio/virtio-iommu.c +33 -0
+        prop.head.length = cpu_to_le16(length);
+        prop.subtype = s->reserved_regions[i].type;
+        prop.start = cpu_to_le64(s->reserved_regions[i].low);
+        prop.end = cpu_to_le64(s->reserved_regions[i].high);
+
+        memcpy(buf, &prop, size);
+
+        trace_virtio_iommu_fill_resv_property(ep, prop.subtype,
+                                              prop.start, prop.end);
+        buf += size;
+    }
+    return total;
+}
+
+/**
+ * virtio_iommu_probe - Fill the probe request buffer with
+ * the properties the device is able to return and add a NONE
+ * property at the end.
+ */
+static int virtio_iommu_probe(VirtIOIOMMU *s,
+                              struct virtio_iommu_req_probe *req,
+                              uint8_t *buf) {
+    uint32_t ep_id = le32_to_cpu(req->endpoint);
+    size_t free = VIOMMU_PROBE_SIZE;
+    ssize_t count;
+
+    if (!virtio_iommu_mr(s, ep_id)) {
+        return VIRTIO_IOMMU_S_NOENT;
+    }
+
+    count = virtio_iommu_fill_resv_mem_prop(s, ep_id, buf, free);
+    if (count < 0) {
+        return VIRTIO_IOMMU_S_INVAL;
Quoted 8 lines

hw/virtio/virtio-iommu.c static int virtio_iommu_unmap(VirtIOIOMMU *s, · line 437

Quoted 6 lines
hw/virtio/virtio-iommu.c +6 -0
+    }
+    buf += count;
+    free -= count;
+
+    /* Fill the rest with zeroes */
+    memset(buf, 0, free);

No need to fill with zero here as "buf" is set to zero on allocation, no?

You're right. I will remove this in the next version.

Thanks

Eric

Quoted 16 lines

Thanks
-Bharat

Quoted 12 lines
hw/virtio/virtio-iommu.c +4 -1
+
+    return VIRTIO_IOMMU_S_OK;
+}
+
 static int virtio_iommu_iov_to_req(struct iovec *iov,
                                    unsigned int iov_cnt,
                                    void *req, size_t req_sz) @@
-407,15 +467,27 @@

virtio_iommu_handle_req(detach)
virtio_iommu_handle_req(map)
virtio_iommu_handle_req(unmap)

Quoted 12 lines
Quoted 11 lines
hw/virtio/virtio-iommu.c +10 -0
+static int virtio_iommu_handle_probe(VirtIOIOMMU *s,
+                                     struct iovec *iov,
+                                     unsigned int iov_cnt,
+                                     uint8_t *buf) {
+    struct virtio_iommu_req_probe req;
+    int ret = virtio_iommu_iov_to_req(iov, iov_cnt, &req,
+sizeof(req));
+
+    return ret ? ret : virtio_iommu_probe(s, &req, buf); }
+
 static void virtio_iommu_handle_command(VirtIODevice *vdev,
Quoted 1 line

VirtQueue

*vq) {

Quoted 10 lines
Quoted 10 lines
     VirtIOIOMMU *s = VIRTIO_IOMMU(vdev);
     struct virtio_iommu_req_head head;
     struct virtio_iommu_req_tail tail = {};
hw/virtio/virtio-iommu.c +2 -1
+    size_t output_size = sizeof(tail), sz;
     VirtQueueElement *elem;
     unsigned int iov_cnt;
     struct iovec *iov;
-    size_t sz;
+    void *buf = NULL;
Quoted 9 lines
Quoted 4 lines
Quoted 4 lines

for (;;) {
elem = virtqueue_pop(vq, sizeof(VirtQueueElement)); @@
-452,6 +524,17 @@ static void virtio_iommu_handle_command(VirtIODevice

*vdev, VirtQueue *vq)

Quoted 4 lines
Quoted 4 lines
         case VIRTIO_IOMMU_T_UNMAP:
             tail.status = virtio_iommu_handle_unmap(s, iov, iov_cnt);
             break;

+ case VIRTIO_IOMMU_T_PROBE:

Quoted 4 lines

As per spec
"
If the device does not offer the VIRTIO_IOMMU_F_PROBE feature,
and if the

driver sends a VIRTIO_-

Quoted 2 lines

IOMMU_T_PROBE request, then the device SHOULD NOT write the
buffer and

SHOULD set the used

Quoted 3 lines

length to zero.
"
So we should check if device supports "VIRTIO_IOMMU_F_PROBE" before

Quoted 8 lines
Quoted 5 lines
Quoted 3 lines

proceed?
But are the device and from that patch onwards we do support the VIRTIO_IOMMU_F_PROBE feature, right?

Yes I agree, do you think if for debugging one wants to try out without this

feature then he should just disable VIRTIO_IOMMU_F_PROBE.

You mean for debugging the driver? I don't think this is the purpose of this device.

Yes for debugging purpose only.

Thanks
-Bharat

Quoted 89 lines

Thanks

Eric

Quoted 85 lines

Thanks
-Bharat

Quoted 80 lines

Thanks

Eric

Quoted 76 lines

Thanks
-Bharat

Quoted 71 lines
Quoted 71 lines
Quoted 69 lines
hw/virtio/virtio-iommu.c +15 -3
+        {
+            struct virtio_iommu_req_tail *ptail;
+
+            output_size = s->config.probe_size + sizeof(tail);
+            buf = g_malloc0(output_size);
+
+            ptail = (struct virtio_iommu_req_tail *)
+                        (buf + s->config.probe_size);
+            ptail->status = virtio_iommu_handle_probe(s, iov, iov_cnt, buf);
+        }
         default:
             tail.status = VIRTIO_IOMMU_S_UNSUPP;
         }
@@ -459,12 +542,13 @@ static void
 virtio_iommu_handle_command(VirtIODevice *vdev, VirtQueue *vq)
 
 out:
         sz = iov_from_buf(elem->in_sg, elem->in_num, 0,
-                          &tail, sizeof(tail));
-        assert(sz == sizeof(tail));
+                          buf ? buf : &tail, output_size);
+        assert(sz == output_size);
 
-        virtqueue_push(vq, elem, sizeof(tail));
+        virtqueue_push(vq, elem, sz);
         virtio_notify(vdev, vq);
         g_free(elem);
+        g_free(buf);
     }
 }
@@ -667,6 +751,7 @@ static void
 virtio_iommu_device_realize(DeviceState *dev, Error **errp)
     s->config.page_size_mask = TARGET_PAGE_MASK;
     s->config.input_range.end = -1UL;
     s->config.domain_range.end = 32;
+    s->config.probe_size = VIOMMU_PROBE_SIZE;
 
     virtio_add_feature(&s->features, VIRTIO_RING_F_EVENT_IDX);
     virtio_add_feature(&s->features,

VIRTIO_RING_F_INDIRECT_DESC); @@ -676,6
+761,7 @@ static void virtio_iommu_device_realize(DeviceState +*dev, Error
**errp)

     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MAP_UNMAP);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_BYPASS);
     virtio_add_feature(&s->features, VIRTIO_IOMMU_F_MMIO);
hw/virtio/virtio-iommu.c +1 -0
+    virtio_add_feature(&s->features, VIRTIO_IOMMU_F_PROBE);
 
     qemu_mutex_init(&s->mutex);
hw/virtio/trace-events +0 -0

index
e83500bee9..5550475691 100644

hw/virtio/trace-events +0 -0
@@ -73,3 +73,4 @@ virtio_iommu_get_domain(uint32_t domain_id)

"Alloc domain=%d"
virtio_iommu_put_domain(uint32_t domain_id) "Free domain=%d" virtio_iommu_translate_out(uint64_t virt_addr, uint64_t
phys_addr, uint32_t sid) "0x%"PRIx64" -> 0x%"PRIx64 " for sid=%d"
virtio_iommu_report_fault(uint8_t reason, uint32_t flags,
uint32_t endpoint, uint64_t addr) "FAULT reason=%d flags=%d endpoint=%d address =0x%"PRIx64

hw/virtio/trace-events +3 -1
+virtio_iommu_fill_resv_property(uint32_t devid, uint8_t subtype,
+uint64_t start, uint64_t end) "dev= %d, type=%d start=0x%"PRIx64"
+end=0x%"PRIx64
--

2.20.1

author
#10 lore raw

PATCH v2 3/5 virtio-iommu: Handle reserved regions in the translation process

In reply to Eric Auger

When translating an address we need to check if it belongs to a reserved virtual address range. If it does, there are 2 cases:

- it belongs to a RESERVED region: the guest should neither use this address in a MAP not instruct the end-point to DMA on them. We report an error

- It belongs to an MSI region: we bypass the translation.

---

v1 -> v2:
- use addr when testing addr belongs to the reserved region and use a block local variable

---
 hw/virtio/virtio-iommu.c | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)
hw/virtio/virtio-iommu.c +20 -0
index 35d772e021..ba72cfaa63 100644
@@ -607,6 +607,7 @@ static IOMMUTLBEntry virtio_iommu_translate(IOMMUMemoryRegion *mr, hwaddr addr,
     uint32_t sid, flags;
     bool bypass_allowed;
     bool found;
+    int i;
 
     interval.low = addr;
     interval.high = addr + 1;
@@ -640,6 +641,25 @@ static IOMMUTLBEntry virtio_iommu_translate(IOMMUMemoryRegion *mr, hwaddr addr,
         goto unlock;
     }
 
+    for (i = 0; i < s->nb_reserved_regions; i++) {
+        ReservedRegion *reg = &s->reserved_regions[i];
+
+        if (addr >= reg->low && addr <= reg->high) {
+            switch (reg->type) {
+            case VIRTIO_IOMMU_RESV_MEM_T_MSI:
+                entry.perm = flag;
+                break;
+            case VIRTIO_IOMMU_RESV_MEM_T_RESERVED:
+            default:
+                virtio_iommu_report_fault(s, VIRTIO_IOMMU_FAULT_R_MAPPING,
+                                          VIRTIO_IOMMU_FAULT_F_ADDRESS,
+                                          sid, addr);
+                break;
+            }
+            goto unlock;
+        }
+    }
+
     if (!ep->domain) {
         if (!bypass_allowed) {
             error_report_once("%s %02x:%02x.%01x not attached to any domain",
Signature · 1 line
2.20.1
Reviewed-by on v2 3/5
#11 lore raw

Re: [PATCH v2 3/5] virtio-iommu: Handle reserved regions in the translation process

Quoted 10 lines · Eric Auger

When translating an address we need to check if it belongs to a reserved virtual address range. If it does, there are 2 cases:

- it belongs to a RESERVED region: the guest should neither use this address in a MAP not instruct the end-point to DMA on them. We report an error

- It belongs to an MSI region: we bypass the translation.

Signature · 1 line
Peter Xu
author
#12 lore raw

PATCH v2 4/5 virtio-iommu-pci: Add array of Interval properties

In reply to Eric Auger

The machine may need to pass reserved regions to the virtio-iommu-pci device (such as the MSI window on x86). So let's add an array of Interval properties.

---

v12 -> v12:
- added Jean's R-b

---
 hw/virtio/virtio-iommu-pci.c | 3 +++
 1 file changed, 3 insertions(+)
hw/virtio/virtio-iommu-pci.c +3 -0
index 3dfbf55b47..44ae9ebc11 100644
@@ -33,6 +33,9 @@ struct VirtIOIOMMUPCI {
 
 static Property virtio_iommu_pci_properties[] = {
     DEFINE_PROP_UINT32("class", VirtIOPCIProxy, class_code, 0),
+    DEFINE_PROP_ARRAY("reserved-regions", VirtIOIOMMUPCI,
+                      vdev.nb_reserved_regions, vdev.reserved_regions,
+                      qdev_prop_reserved_region, ReservedRegion),
     DEFINE_PROP_END_OF_LIST(),
 };
 
Signature · 1 line
2.20.1
author
#13 lore raw

PATCH v2 5/5 hw/arm/virt: Let the virtio-iommu bypass MSIs

In reply to Eric Auger

At the moment the virtio-iommu translates MSI transactions. This behavior is inherited from ARM SMMU. The virt machine code knows where the guest MSI doorbells are so we can easily declare those regions as VIRTIO_IOMMU_RESV_MEM_T_MSI. With that setting the guest will not map MSIs through the IOMMU and those transactions will be simply bypassed.

Depending on which MSI controller is in use (ITS or GICV2M), we declare either:
- the ITS interrupt translation space (ITS_base + 0x10000), containing the GITS_TRANSLATOR or
- The GICV2M single frame, containing the MSI_SETSP_NS register.

---

v1 -> v2:
- Test which MSI controller is instantiated
- If GICV2M is in use, declare its doorbell as an MSI doorbell too

---
 include/hw/arm/virt.h |  6 ++++++
 hw/arm/virt.c         | 18 ++++++++++++++++++
 2 files changed, 24 insertions(+)
include/hw/arm/virt.h +6 -0
index 6d67ace76e..ad20cb6e15 100644
@@ -96,6 +96,11 @@ typedef enum VirtIOMMUType {
     VIRT_IOMMU_VIRTIO,
 } VirtIOMMUType;
 
+typedef enum VirtMSIControllerType {
+    VIRT_GICV2M,
+    VIRT_ITS,
+} VirtMSIControllerType;
+
 typedef enum VirtGICType {
     VIRT_GIC_VERSION_MAX,
     VIRT_GIC_VERSION_HOST,
@@ -135,6 +140,7 @@ typedef struct {
     OnOffAuto acpi;
     VirtGICType gic_version;
     VirtIOMMUType iommu;
+    VirtMSIControllerType msi_controller;
     uint16_t virtio_iommu_bdf;
     struct arm_boot_info bootinfo;
     MemMapEntry *memmap;
hw/arm/virt.c +18 -0
index 634db0cfe9..d2dd07885b 100644
@@ -602,6 +602,7 @@ static void create_its(VirtMachineState *vms)
     sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, vms->memmap[VIRT_GIC_ITS].base);
 
     fdt_add_its_gic_node(vms);
+    vms->msi_controller = VIRT_ITS;
 }
 
 static void create_v2m(VirtMachineState *vms)
@@ -622,6 +623,7 @@ static void create_v2m(VirtMachineState *vms)
     }
 
     fdt_add_v2m_gic_node(vms);
+    vms->msi_controller = VIRT_GICV2M;
 }
 
 static void create_gic(VirtMachineState *vms)
@@ -2136,8 +2138,24 @@ out:
 static void virt_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev,
                                             DeviceState *dev, Error **errp)
 {
+    VirtMachineState *vms = VIRT_MACHINE(hotplug_dev);
+
     if (object_dynamic_cast(OBJECT(dev), TYPE_PC_DIMM)) {
         virt_memory_pre_plug(hotplug_dev, dev, errp);
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) {
+        /* we declare a VIRTIO_IOMMU_RESV_MEM_T_MSI region */
+
+        if (vms->msi_controller == VIRT_ITS) {
+            /* GITS_TRANSLATER page */
+            qdev_prop_set_uint32(dev, "len-reserved-regions", 1);
+            qdev_prop_set_string(dev, "reserved-regions[0]",
+                                 "0x8090000, 0x809FFFF, 1");
+        } else if (vms->msi_controller == VIRT_GICV2M) {
+            /* MSI_SETSPI_NS page */
+            qdev_prop_set_uint32(dev, "len-reserved-regions", 1);
+            qdev_prop_set_string(dev, "reserved-regions[0]",
+                                 "0x8020000, 0x8020FFF, 1");
+        }
     }
 }
 
Signature · 1 line
2.20.1
#14 lore raw

Re: [PATCH v2 5/5] hw/arm/virt: Let the virtio-iommu bypass MSIs

Quoted 30 lines · Eric Auger

At the moment the virtio-iommu translates MSI transactions. This behavior is inherited from ARM SMMU. The virt machine code knows where the guest MSI doorbells are so we can easily declare those regions as VIRTIO_IOMMU_RESV_MEM_T_MSI. With that setting the guest will not map MSIs through the IOMMU and those transactions will be simply bypassed.

Depending on which MSI controller is in use (ITS or GICV2M), we declare either:
- the ITS interrupt translation space (ITS_base + 0x10000), containing the GITS_TRANSLATOR or
- The GICV2M single frame, containing the MSI_SETSP_NS register.

---

v1 -> v2:
- Test which MSI controller is instantiated
- If GICV2M is in use, declare its doorbell as an MSI doorbell too

---
 include/hw/arm/virt.h |  6 ++++++
 hw/arm/virt.c         | 18 ++++++++++++++++++
 2 files changed, 24 insertions(+)
include/hw/arm/virt.h +0 -0
index 6d67ace76e..ad20cb6e15 100644

include/hw/arm/virt.h typedef enum VirtIOMMUType { · line 102

Quoted 7 lines
include/hw/arm/virt.h +4 -0
@@ -96,6 +96,11 @@ typedef enum VirtIOMMUType {
     VIRT_IOMMU_VIRTIO,
 } VirtIOMMUType;
 
+typedef enum VirtMSIControllerType {
+    VIRT_GICV2M,
+    VIRT_ITS,
+} VirtMSIControllerType;

I think you need a third value for msi_controller == 0. If I instantiate a GICv3 without ITS at the moment the V2M region gets reserved. Not a big deal since MSIs aren't supported at all in this case, but it would be cleaner to skip any reservation.

Thanks,
Jean

Quoted 60 lines
include/hw/arm/virt.h +2 -0
+
 typedef enum VirtGICType {
     VIRT_GIC_VERSION_MAX,
     VIRT_GIC_VERSION_HOST,
@@ -135,6 +140,7 @@ typedef struct {
     OnOffAuto acpi;
     VirtGICType gic_version;
     VirtIOMMUType iommu;
+    VirtMSIControllerType msi_controller;
     uint16_t virtio_iommu_bdf;
     struct arm_boot_info bootinfo;
     MemMapEntry *memmap;
hw/arm/virt.c +18 -0
index 634db0cfe9..d2dd07885b 100644
@@ -602,6 +602,7 @@ static void create_its(VirtMachineState *vms)
     sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, vms->memmap[VIRT_GIC_ITS].base);
 
     fdt_add_its_gic_node(vms);
+    vms->msi_controller = VIRT_ITS;
 }
 
 static void create_v2m(VirtMachineState *vms)
@@ -622,6 +623,7 @@ static void create_v2m(VirtMachineState *vms)
     }
 
     fdt_add_v2m_gic_node(vms);
+    vms->msi_controller = VIRT_GICV2M;
 }
 
 static void create_gic(VirtMachineState *vms)
@@ -2136,8 +2138,24 @@ out:
 static void virt_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev,
                                             DeviceState *dev, Error **errp)
 {
+    VirtMachineState *vms = VIRT_MACHINE(hotplug_dev);
+
     if (object_dynamic_cast(OBJECT(dev), TYPE_PC_DIMM)) {
         virt_memory_pre_plug(hotplug_dev, dev, errp);
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_VIRTIO_IOMMU_PCI)) {
+        /* we declare a VIRTIO_IOMMU_RESV_MEM_T_MSI region */
+
+        if (vms->msi_controller == VIRT_ITS) {
+            /* GITS_TRANSLATER page */
+            qdev_prop_set_uint32(dev, "len-reserved-regions", 1);
+            qdev_prop_set_string(dev, "reserved-regions[0]",
+                                 "0x8090000, 0x809FFFF, 1");
+        } else if (vms->msi_controller == VIRT_GICV2M) {
+            /* MSI_SETSPI_NS page */
+            qdev_prop_set_uint32(dev, "len-reserved-regions", 1);
+            qdev_prop_set_string(dev, "reserved-regions[0]",
+                                 "0x8020000, 0x8020FFF, 1");
+        }
     }
 }
 
Signature · 1 line
2.20.1