diff options
| -rw-r--r-- | hw/vfio_pci.c | 34 | ||||
| -rw-r--r-- | main-loop.c | 27 |
2 files changed, 41 insertions, 20 deletions
diff --git a/hw/vfio_pci.c b/hw/vfio_pci.c index 28c83031d0..c51ae6761b 100644 --- a/hw/vfio_pci.c +++ b/hw/vfio_pci.c @@ -562,8 +562,8 @@ static int vfio_enable_vectors(VFIODevice *vdev, bool msix) return ret; } -static int vfio_msix_vector_use(PCIDevice *pdev, - unsigned int nr, MSIMessage msg) +static int vfio_msix_vector_do_use(PCIDevice *pdev, unsigned int nr, + MSIMessage *msg, IOHandler *handler) { VFIODevice *vdev = DO_UPCAST(VFIODevice, pdev, pdev); VFIOMSIVector *vector; @@ -587,7 +587,7 @@ static int vfio_msix_vector_use(PCIDevice *pdev, * Attempt to enable route through KVM irqchip, * default to userspace handling if unavailable. */ - vector->virq = kvm_irqchip_add_msi_route(kvm_state, msg); + vector->virq = msg ? kvm_irqchip_add_msi_route(kvm_state, *msg) : -1; if (vector->virq < 0 || kvm_irqchip_add_irqfd_notifier(kvm_state, &vector->interrupt, vector->virq) < 0) { @@ -596,7 +596,7 @@ static int vfio_msix_vector_use(PCIDevice *pdev, vector->virq = -1; } qemu_set_fd_handler(event_notifier_get_fd(&vector->interrupt), - vfio_msi_interrupt, NULL, vector); + handler, NULL, vector); } /* @@ -639,6 +639,12 @@ static int vfio_msix_vector_use(PCIDevice *pdev, return 0; } +static int vfio_msix_vector_use(PCIDevice *pdev, + unsigned int nr, MSIMessage msg) +{ + return vfio_msix_vector_do_use(pdev, nr, &msg, vfio_msi_interrupt); +} + static void vfio_msix_vector_release(PCIDevice *pdev, unsigned int nr) { VFIODevice *vdev = DO_UPCAST(VFIODevice, pdev, pdev); @@ -697,6 +703,22 @@ static void vfio_enable_msix(VFIODevice *vdev) vdev->interrupt = VFIO_INT_MSIX; + /* + * Some communication channels between VF & PF or PF & fw rely on the + * physical state of the device and expect that enabling MSI-X from the + * guest enables the same on the host. When our guest is Linux, the + * guest driver call to pci_enable_msix() sets the enabling bit in the + * 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. + */ + vfio_msix_vector_do_use(&vdev->pdev, 0, NULL, NULL); + vfio_msix_vector_release(&vdev->pdev, 0); + if (msix_set_vector_notifiers(&vdev->pdev, vfio_msix_vector_use, vfio_msix_vector_release, NULL)) { error_report("vfio: msix_set_vector_notifiers failed\n"); @@ -1815,13 +1837,13 @@ static int vfio_get_device(VFIOGroup *group, const char *name, VFIODevice *vdev) error_report("Warning, device %s does not support reset\n", name); } - if (dev_info.num_regions != VFIO_PCI_NUM_REGIONS) { + if (dev_info.num_regions < VFIO_PCI_CONFIG_REGION_INDEX + 1) { error_report("vfio: unexpected number of io regions %u\n", dev_info.num_regions); goto error; } - if (dev_info.num_irqs != VFIO_PCI_NUM_IRQS) { + if (dev_info.num_irqs < VFIO_PCI_MSIX_IRQ_INDEX + 1) { error_report("vfio: unexpected number of irqs %u\n", dev_info.num_irqs); goto error; } diff --git a/main-loop.c b/main-loop.c index 54f38ae1ae..6f52ac39bc 100644 --- a/main-loop.c +++ b/main-loop.c @@ -330,7 +330,7 @@ void qemu_fd_register(int fd) static int os_host_main_loop_wait(uint32_t timeout) { GMainContext *context = g_main_context_default(); - int ret, i; + int select_ret, g_poll_ret, ret, i; PollingEntry *pe; WaitObjects *w = &wait_objects; gint poll_timeout; @@ -345,13 +345,6 @@ static int os_host_main_loop_wait(uint32_t timeout) return ret; } - if (nfds >= 0) { - ret = select(nfds + 1, &rfds, &wfds, &xfds, &tv0); - if (ret != 0) { - timeout = 0; - } - } - g_main_context_prepare(context, &max_priority); n_poll_fds = g_main_context_query(context, max_priority, &poll_timeout, poll_fds, ARRAY_SIZE(poll_fds)); @@ -367,9 +360,9 @@ static int os_host_main_loop_wait(uint32_t timeout) } qemu_mutex_unlock_iothread(); - ret = g_poll(poll_fds, n_poll_fds + w->num, poll_timeout); + g_poll_ret = g_poll(poll_fds, n_poll_fds + w->num, poll_timeout); qemu_mutex_lock_iothread(); - if (ret > 0) { + if (g_poll_ret > 0) { for (i = 0; i < w->num; i++) { w->revents[i] = poll_fds[n_poll_fds + i].revents; } @@ -384,12 +377,18 @@ static int os_host_main_loop_wait(uint32_t timeout) g_main_context_dispatch(context); } - /* If an edge-triggered socket event occurred, select will return a - * positive result on the next iteration. We do not need to do anything - * here. + /* Call select after g_poll to avoid a useless iteration and therefore + * improve socket latency. */ - return ret; + if (nfds >= 0) { + select_ret = select(nfds + 1, &rfds, &wfds, &xfds, &tv0); + if (select_ret != 0) { + timeout = 0; + } + } + + return select_ret || g_poll_ret; } #endif |