Merge tag 'pull-vfio-20231009' of https://github.com/legoater/qemu into staging

vfio queue:

* Fix for VFIO display when using Intel vGPUs
* Support for dynamic MSI-X

# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCAAdFiEEoPZlSPBIlev+awtgUaNDx8/77KEFAmUjoLIACgkQUaNDx8/7
# 7KE+gw/9FTQFRkmlkSMlqRGjINF/VmfX6TsX+dy3ZB+aJia6qahco+u9hd3yQxiA
# /KI4FZnQCH/ZFizjR7hJdsxLnd+l989RFmoy+NTEXfgBMSLu4aU1UlVC1pyuhJ5L
# xadGQ2UIclD1Gz70laa9ketebLHdyc/Pku2xt9oreR6kRRFHZ3V4QhMNhcwGapO1
# 0wytLFXPVyGa7YYTB5qQPHPWyY9sM0n6E4E7jVnhfOw75cUVNvSr+9HlJbR1FN3Z
# 4klNMXayKGAZmh9oKpQWBsf4aUwLDu//eCk64TkQHp0pNrvRAJJBwgkhsI1FigeW
# SJ2JjQsIg/vLu2oyUhp2PJ59cQSMFZPgEqRhhRQ2RKhIfwOZY4kgfvKFtSHvWijV
# u0r8/HMIJE0fNffigyDlfLCsUEYu3OuJXMlU+5xrwi77hWlPrGb8D1J7LhwUnldk
# kZaw9VEranlbMQT773cMA7f/pgS1Sc6CkdqfJLGIHA4PsEk44Lzen2BzRroz8+Km
# tn8hHt+GQK/ZGKmOPXWm44Bd48Be08cMz/pOI2cqoScEKKEQ8HUul3H1/k8sqauh
# 1gPo1hIPXo/GaGRvUvPsj4cK8oQm77EHksEQ4Nxvn+ZWTW2FnMQkb9QFbF8bTmEo
# KiJJ6s8qbd1CWGYbO0GSE8ss3NUZq1YbWsMXmUP0JccEgvjeL2M=
# =QRhQ
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 09 Oct 2023 02:41:54 EDT
# gpg:                using RSA key A0F66548F04895EBFE6B0B6051A343C7CFFBECA1
# gpg: Good signature from "Cédric Le Goater <clg@redhat.com>" [unknown]
# gpg:                 aka "Cédric Le Goater <clg@kaod.org>" [unknown]
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: A0F6 6548 F048 95EB FE6B  0B60 51A3 43C7 CFFB ECA1

* tag 'pull-vfio-20231009' of https://github.com/legoater/qemu:
  vfio/pci: enable MSI-X in interrupt restoring on dynamic allocation
  vfio/pci: use an invalid fd to enable MSI-X
  vfio/pci: enable vector on dynamic MSI-X allocation
  vfio/pci: detect the support of dynamic MSI-X allocation
  vfio/pci: rename vfio_put_device to vfio_pci_put_device
  vfio/display: Fix missing update to set backing fields

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

hw/vfio/display.c

@@ -243,6 +243,8 @@ static VFIODMABuf *vfio_display_get_dmabuf(VFIOPCIDevice *vdev,
     dmabuf->dmabuf_id  = plane.dmabuf_id;
     dmabuf->buf.width  = plane.width;
     dmabuf->buf.height = plane.height;
+    dmabuf->buf.backing_width = plane.width;
+    dmabuf->buf.backing_height = plane.height;
     dmabuf->buf.stride = plane.stride;
     dmabuf->buf.fourcc = plane.drm_format;
     dmabuf->buf.modifier = plane.drm_format_mod;

hw/vfio/pci.c

@@ -369,12 +369,56 @@ static void vfio_msi_interrupt(void *opaque)
     notify(&vdev->pdev, nr);
 }
 
+/*
+ * Get MSI-X enabled, but no vector enabled, by setting vector 0 with an invalid
+ * fd to kernel.
+ */
+static int vfio_enable_msix_no_vec(VFIOPCIDevice *vdev)
+{
+    g_autofree struct vfio_irq_set *irq_set = NULL;
+    int ret = 0, argsz;
+    int32_t *fd;
+
+    argsz = sizeof(*irq_set) + sizeof(*fd);
+
+    irq_set = g_malloc0(argsz);
+    irq_set->argsz = argsz;
+    irq_set->flags = VFIO_IRQ_SET_DATA_EVENTFD |
+                     VFIO_IRQ_SET_ACTION_TRIGGER;
+    irq_set->index = VFIO_PCI_MSIX_IRQ_INDEX;
+    irq_set->start = 0;
+    irq_set->count = 1;
+    fd = (int32_t *)&irq_set->data;
+    *fd = -1;
+
+    ret = ioctl(vdev->vbasedev.fd, VFIO_DEVICE_SET_IRQS, irq_set);
+
+    return ret;
+}
+
 static int vfio_enable_vectors(VFIOPCIDevice *vdev, bool msix)
 {
     struct vfio_irq_set *irq_set;
     int ret = 0, i, argsz;
     int32_t *fds;
 
+    /*
+     * If dynamic MSI-X allocation is supported, the vectors to be allocated
+     * and enabled can be scattered. Before kernel enabling MSI-X, setting
+     * nr_vectors causes all these vectors to be allocated on host.
+     *
+     * To keep allocation as needed, use vector 0 with an invalid fd to get
+     * MSI-X enabled first, then set vectors with a potentially sparse set of
+     * eventfds to enable interrupts only when enabled in guest.
+     */
+    if (msix && !vdev->msix->noresize) {
+        ret = vfio_enable_msix_no_vec(vdev);
+
+        if (ret) {
+            return ret;
+        }
+    }
+
     argsz = sizeof(*irq_set) + (vdev->nr_vectors * sizeof(*fds));
 
     irq_set = g_malloc0(argsz);
@@ -470,6 +514,7 @@ static int vfio_msix_vector_do_use(PCIDevice *pdev, unsigned int nr,
     VFIOPCIDevice *vdev = VFIO_PCI(pdev);
     VFIOMSIVector *vector;
     int ret;
+    bool resizing = !!(vdev->nr_vectors < nr + 1);
 
     trace_vfio_msix_vector_do_use(vdev->vbasedev.name, nr);
 
@@ -512,33 +557,42 @@ static int vfio_msix_vector_do_use(PCIDevice *pdev, unsigned int nr,
     }
 
     /*
-     * We don't want to have the host allocate all possible MSI vectors
-     * for a device if they're not in use, so we shutdown and incrementally
-     * increase them as needed.
+     * When dynamic allocation is not supported, we don't want to have the
+     * host allocate all possible MSI vectors for a device if they're not
+     * in use, so we shutdown and incrementally increase them as needed.
+     * nr_vectors represents the total number of vectors allocated.
+     *
+     * When dynamic allocation is supported, let the host only allocate
+     * and enable a vector when it is in use in guest. nr_vectors represents
+     * the upper bound of vectors being enabled (but not all of the ranges
+     * is allocated or enabled).
      */
-    if (vdev->nr_vectors < nr + 1) {
+    if (resizing) {
         vdev->nr_vectors = nr + 1;
-        if (!vdev->defer_kvm_irq_routing) {
+    }
+
+    if (!vdev->defer_kvm_irq_routing) {
+        if (vdev->msix->noresize && resizing) {
             vfio_disable_irqindex(&vdev->vbasedev, VFIO_PCI_MSIX_IRQ_INDEX);
             ret = vfio_enable_vectors(vdev, true);
             if (ret) {
                 error_report("vfio: failed to enable vectors, %d", ret);
             }
-        }
-    } else {
-        Error *err = NULL;
-        int32_t fd;
-
-        if (vector->virq >= 0) {
-            fd = event_notifier_get_fd(&vector->kvm_interrupt);
         } else {
-            fd = event_notifier_get_fd(&vector->interrupt);
-        }
+            Error *err = NULL;
+            int32_t fd;
 
-        if (vfio_set_irq_signaling(&vdev->vbasedev,
-                                     VFIO_PCI_MSIX_IRQ_INDEX, nr,
-                                     VFIO_IRQ_SET_ACTION_TRIGGER, fd, &err)) {
-            error_reportf_err(err, VFIO_MSG_PREFIX, vdev->vbasedev.name);
+            if (vector->virq >= 0) {
+                fd = event_notifier_get_fd(&vector->kvm_interrupt);
+            } else {
+                fd = event_notifier_get_fd(&vector->interrupt);
+            }
+
+            if (vfio_set_irq_signaling(&vdev->vbasedev,
+                                       VFIO_PCI_MSIX_IRQ_INDEX, nr,
+                                       VFIO_IRQ_SET_ACTION_TRIGGER, fd, &err)) {
+                error_reportf_err(err, VFIO_MSG_PREFIX, vdev->vbasedev.name);
+            }
         }
     }
 
@@ -608,6 +662,8 @@ static void vfio_commit_kvm_msi_virq_batch(VFIOPCIDevice *vdev)
 
 static void vfio_msix_enable(VFIOPCIDevice *vdev)
 {
+    int ret;
+
     vfio_disable_interrupts(vdev);
 
     vdev->msi_vectors = g_new0(VFIOMSIVector, vdev->msix->entries);
@@ -630,8 +686,6 @@ static void vfio_msix_enable(VFIOPCIDevice *vdev)
     vfio_commit_kvm_msi_virq_batch(vdev);
 
     if (vdev->nr_vectors) {
-        int ret;
-
         ret = vfio_enable_vectors(vdev, true);
         if (ret) {
             error_report("vfio: failed to enable vectors, %d", ret);
@@ -645,13 +699,14 @@ static void vfio_msix_enable(VFIOPCIDevice *vdev)
          * MSI-X capability, but leaves the vector table masked.  We therefore
          * can't rely on a vector_use callback (from request_irq() in the guest)
          * to switch the physical device into MSI-X mode because that may come a
-         * long time after pci_enable_msix().  This code enables vector 0 with
-         * triggering to userspace, then immediately release the vector, leaving
-         * the physical device with no vectors enabled, but MSI-X enabled, just
-         * like the guest view.
+         * long time after pci_enable_msix().  This code sets vector 0 with an
+         * invalid fd to make the physical device MSI-X enabled, but with no
+         * vectors enabled, just like the guest view.
          */
-        vfio_msix_vector_do_use(&vdev->pdev, 0, NULL, NULL);
-        vfio_msix_vector_release(&vdev->pdev, 0);
+        ret = vfio_enable_msix_no_vec(vdev);
+        if (ret) {
+            error_report("vfio: failed to enable MSI-X, %d", ret);
+        }
     }
 
     trace_vfio_msix_enable(vdev->vbasedev.name);
@@ -1493,7 +1548,9 @@ static void vfio_msix_early_setup(VFIOPCIDevice *vdev, Error **errp)
     uint8_t pos;
     uint16_t ctrl;
     uint32_t table, pba;
-    int fd = vdev->vbasedev.fd;
+    int ret, fd = vdev->vbasedev.fd;
+    struct vfio_irq_info irq_info = { .argsz = sizeof(irq_info),
+                                      .index = VFIO_PCI_MSIX_IRQ_INDEX };
     VFIOMSIXInfo *msix;
 
     pos = pci_find_capability(&vdev->pdev, PCI_CAP_ID_MSIX);
@@ -1530,6 +1587,15 @@ static void vfio_msix_early_setup(VFIOPCIDevice *vdev, Error **errp)
     msix->pba_offset = pba & ~PCI_MSIX_FLAGS_BIRMASK;
     msix->entries = (ctrl & PCI_MSIX_FLAGS_QSIZE) + 1;
 
+    ret = ioctl(vdev->vbasedev.fd, VFIO_DEVICE_GET_IRQ_INFO, &irq_info);
+    if (ret < 0) {
+        error_setg_errno(errp, -ret, "failed to get MSI-X irq info");
+        g_free(msix);
+        return;
+    }
+
+    msix->noresize = !!(irq_info.flags & VFIO_IRQ_INFO_NORESIZE);
+
     /*
      * Test the size of the pba_offset variable and catch if it extends outside
      * of the specified BAR. If it is the case, we need to apply a hardware
@@ -1562,7 +1628,8 @@ static void vfio_msix_early_setup(VFIOPCIDevice *vdev, Error **errp)
     }
 
     trace_vfio_msix_early_setup(vdev->vbasedev.name, pos, msix->table_bar,
-                                msix->table_offset, msix->entries);
+                                msix->table_offset, msix->entries,
+                                msix->noresize);
     vdev->msix = msix;
 
     vfio_pci_fixup_msix_region(vdev);
@@ -2826,7 +2893,7 @@ static void vfio_populate_device(VFIOPCIDevice *vdev, Error **errp)
     }
 }
 
-static void vfio_put_device(VFIOPCIDevice *vdev)
+static void vfio_pci_put_device(VFIOPCIDevice *vdev)
 {
     g_free(vdev->vbasedev.name);
     g_free(vdev->msix);
@@ -3317,7 +3384,7 @@ static void vfio_instance_finalize(Object *obj)
      *
      * g_free(vdev->igd_opregion);
      */
-    vfio_put_device(vdev);
+    vfio_pci_put_device(vdev);
     vfio_put_group(group);
 }
 

hw/vfio/pci.h

@@ -113,6 +113,7 @@ typedef struct VFIOMSIXInfo {
     uint32_t table_offset;
     uint32_t pba_offset;
     unsigned long *pending;
+    bool noresize;
 } VFIOMSIXInfo;
 
 #define TYPE_VFIO_PCI "vfio-pci"

hw/vfio/trace-events

@@ -27,7 +27,7 @@ vfio_vga_read(uint64_t addr, int size, uint64_t data) " (0x%"PRIx64", %d) = 0x%"
 vfio_pci_read_config(const char *name, int addr, int len, int val) " (%s, @0x%x, len=0x%x) 0x%x"
 vfio_pci_write_config(const char *name, int addr, int val, int len) " (%s, @0x%x, 0x%x, len=0x%x)"
 vfio_msi_setup(const char *name, int pos) "%s PCI MSI CAP @0x%x"
-vfio_msix_early_setup(const char *name, int pos, int table_bar, int offset, int entries) "%s PCI MSI-X CAP @0x%x, BAR %d, offset 0x%x, entries %d"
+vfio_msix_early_setup(const char *name, int pos, int table_bar, int offset, int entries, bool noresize) "%s PCI MSI-X CAP @0x%x, BAR %d, offset 0x%x, entries %d, noresize %d"
 vfio_check_pcie_flr(const char *name) "%s Supports FLR via PCIe cap"
 vfio_check_pm_reset(const char *name) "%s Supports PM reset"
 vfio_check_af_flr(const char *name) "%s Supports FLR via AF cap"