The vector_use callback is used to enable vector that is unmasked in
guest. The kernel used to only support static MSI-X allocation. When
allocating a new interrupt using "static MSI-X allocation" kernels,
QEMU first disables all previously allocated vectors and then
re-allocates all including the new one. The nr_vectors of VFIOPCIDevice
indicates that all vectors from 0 to nr_vectors are allocated (and may
be enabled), which is used to to loop all the possibly used vectors
When, e.g., disabling MSI-X interrupts.
Extend the vector_use function to support dynamic MSI-X allocation when
host supports the capability. QEMU therefore can individually allocate
and enable a new interrupt without affecting others or causing interrupts
lost during runtime.
Utilize nr_vectors to calculate the upper bound of enabled vectors in
dynamic MSI-X allocation mode since looping all msix_entries_nr is not
efficient and unnecessary.
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Tested-by: Reinette Chatre <reinette.chatre@intel.com>
---
Changes since v1:
- Revise Qemu to QEMU.
Changes since RFC v1:
- Test vdev->msix->noresize to identify the allocation mode. (Alex)
- Move defer_kvm_irq_routing test out and update nr_vectors in a
common place before vfio_enable_vectors(). (Alex)
- Revise the comments. (Alex)
---
hw/vfio/pci.c | 44 +++++++++++++++++++++++++++-----------------
1 file changed, 27 insertions(+), 17 deletions(-)
diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
index 60654ca28ab8..84987e46fd7a 100644
--- a/hw/vfio/pci.c
+++ b/hw/vfio/pci.c
@@ -470,6 +470,7 @@ static int vfio_msix_vector_do_use(PCIDevice *pdev, unsigned int nr,
VFIOPCIDevice *vdev = VFIO_PCI(pdev);
VFIOMSIVector *vector;
int ret;
+ int old_nr_vecs = vdev->nr_vectors;
trace_vfio_msix_vector_do_use(vdev->vbasedev.name, nr);
@@ -512,33 +513,42 @@ 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) {
vdev->nr_vectors = nr + 1;
- if (!vdev->defer_kvm_irq_routing) {
+ }
+
+ if (!vdev->defer_kvm_irq_routing) {
+ if (vdev->msix->noresize && (old_nr_vecs < nr + 1)) {
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);
+ }
}
}
--
2.27.0On Mon, 18 Sep 2023 05:45:05 -0400
Jing Liu <jing2.liu@intel.com> wrote:
> The vector_use callback is used to enable vector that is unmasked in
> guest. The kernel used to only support static MSI-X allocation. When
> allocating a new interrupt using "static MSI-X allocation" kernels,
> QEMU first disables all previously allocated vectors and then
> re-allocates all including the new one. The nr_vectors of VFIOPCIDevice
> indicates that all vectors from 0 to nr_vectors are allocated (and may
> be enabled), which is used to to loop all the possibly used vectors
^^ ^^
s/to to/to/
> When, e.g., disabling MSI-X interrupts.
>
> Extend the vector_use function to support dynamic MSI-X allocation when
> host supports the capability. QEMU therefore can individually allocate
> and enable a new interrupt without affecting others or causing interrupts
> lost during runtime.
>
> Utilize nr_vectors to calculate the upper bound of enabled vectors in
> dynamic MSI-X allocation mode since looping all msix_entries_nr is not
> efficient and unnecessary.
>
> Signed-off-by: Jing Liu <jing2.liu@intel.com>
> Tested-by: Reinette Chatre <reinette.chatre@intel.com>
> ---
> Changes since v1:
> - Revise Qemu to QEMU.
>
> Changes since RFC v1:
> - Test vdev->msix->noresize to identify the allocation mode. (Alex)
> - Move defer_kvm_irq_routing test out and update nr_vectors in a
> common place before vfio_enable_vectors(). (Alex)
> - Revise the comments. (Alex)
> ---
> hw/vfio/pci.c | 44 +++++++++++++++++++++++++++-----------------
> 1 file changed, 27 insertions(+), 17 deletions(-)
>
> diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
> index 60654ca28ab8..84987e46fd7a 100644
> --- a/hw/vfio/pci.c
> +++ b/hw/vfio/pci.c
> @@ -470,6 +470,7 @@ static int vfio_msix_vector_do_use(PCIDevice *pdev, unsigned int nr,
> VFIOPCIDevice *vdev = VFIO_PCI(pdev);
> VFIOMSIVector *vector;
> int ret;
> + int old_nr_vecs = vdev->nr_vectors;
Minor suggestion, it reads slightly better below if this were something
like:
bool resizing = !!(vdev->nr_vectors < nr + 1);
Then use the bool in place of the nr+1 tests below. Thanks,
Alex
>
> trace_vfio_msix_vector_do_use(vdev->vbasedev.name, nr);
>
> @@ -512,33 +513,42 @@ 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) {
> vdev->nr_vectors = nr + 1;
> - if (!vdev->defer_kvm_irq_routing) {
> + }
> +
> + if (!vdev->defer_kvm_irq_routing) {
> + if (vdev->msix->noresize && (old_nr_vecs < nr + 1)) {
> 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);
> + }
> }
> }
>
Hi Alex,
> On Sat, 9/23/2023 4:55 AM, Alex Williamson <alex.williamson@redhat.com> wrote:
> On Mon, 18 Sep 2023 05:45:05 -0400
> Jing Liu <jing2.liu@intel.com> wrote:
>
> > The vector_use callback is used to enable vector that is unmasked in
> > guest. The kernel used to only support static MSI-X allocation. When
> > allocating a new interrupt using "static MSI-X allocation" kernels,
> > QEMU first disables all previously allocated vectors and then
> > re-allocates all including the new one. The nr_vectors of
> > VFIOPCIDevice indicates that all vectors from 0 to nr_vectors are
> > allocated (and may be enabled), which is used to to loop all the
> > possibly used vectors
> ^^ ^^
>
> s/to to/to/
Will change.
>
> > When, e.g., disabling MSI-X interrupts.
> >
> > Extend the vector_use function to support dynamic MSI-X allocation
> > when host supports the capability. QEMU therefore can individually
> > allocate and enable a new interrupt without affecting others or
> > causing interrupts lost during runtime.
> >
> > Utilize nr_vectors to calculate the upper bound of enabled vectors in
> > dynamic MSI-X allocation mode since looping all msix_entries_nr is not
> > efficient and unnecessary.
> >
> > Signed-off-by: Jing Liu <jing2.liu@intel.com>
> > Tested-by: Reinette Chatre <reinette.chatre@intel.com>
> > ---
> > Changes since v1:
> > - Revise Qemu to QEMU.
> >
> > Changes since RFC v1:
> > - Test vdev->msix->noresize to identify the allocation mode. (Alex)
> > - Move defer_kvm_irq_routing test out and update nr_vectors in a
> > common place before vfio_enable_vectors(). (Alex)
> > - Revise the comments. (Alex)
> > ---
> > hw/vfio/pci.c | 44 +++++++++++++++++++++++++++-----------------
> > 1 file changed, 27 insertions(+), 17 deletions(-)
> >
> > diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c index
> > 60654ca28ab8..84987e46fd7a 100644
> > --- a/hw/vfio/pci.c
> > +++ b/hw/vfio/pci.c
> > @@ -470,6 +470,7 @@ static int vfio_msix_vector_do_use(PCIDevice *pdev,
> unsigned int nr,
> > VFIOPCIDevice *vdev = VFIO_PCI(pdev);
> > VFIOMSIVector *vector;
> > int ret;
> > + int old_nr_vecs = vdev->nr_vectors;
>
> Minor suggestion, it reads slightly better below if this were something
> like:
>
> bool resizing = !!(vdev->nr_vectors < nr + 1);
>
> Then use the bool in place of the nr+1 tests below. Thanks,
>
Got it. This change makes it nice to read. Thanks for the advice. Will send v3 later.
Thanks,
Jing
> Alex
>
> >
> > trace_vfio_msix_vector_do_use(vdev->vbasedev.name, nr);
> >
> > @@ -512,33 +513,42 @@ 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) {
> > vdev->nr_vectors = nr + 1;
> > - if (!vdev->defer_kvm_irq_routing) {
> > + }
> > +
> > + if (!vdev->defer_kvm_irq_routing) {
> > + if (vdev->msix->noresize && (old_nr_vecs < nr + 1)) {
> > 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);
> > + }
> > }
> > }
> >
© 2016 - 2024 Red Hat, Inc.