From: Auger Eric <eric.auger@redhat.com>
To: Alex Williamson <alex.williamson@redhat.com>, qemu-devel@nongnu.org
Cc: peterx@redhat.com, kvm@vger.kernel.org
Subject: Re: [PATCH v3 6/6] vfio/quirks: Enable ioeventfd quirks to be handled by vfio directly
Date: Fri, 4 May 2018 09:51:56 +0200 [thread overview]
Message-ID: <0700cc1f-ea20-c01d-5c6a-3e3fb1439730@redhat.com> (raw)
In-Reply-To: <20180503214513.2749.30627.stgit@gimli.home>
Hi,
On 05/03/2018 11:45 PM, Alex Williamson wrote:
> With vfio ioeventfd support, we can program vfio-pci to perform a
> specified BAR write when an eventfd is triggered. This allows the
> KVM ioeventfd to be wired directly to vfio-pci, entirely avoiding
> userspace handling for these events. On the same micro-benchmark
> where the ioeventfd got us to almost 90% of performance versus
> disabling the GeForce quirks, this gets us to within 95%.
>
> Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Thanks
Eric
> ---
> hw/vfio/pci-quirks.c | 53 +++++++++++++++++++++++++++++++++++++++++++-------
> hw/vfio/pci.c | 2 ++
> hw/vfio/pci.h | 2 ++
> hw/vfio/trace-events | 2 +-
> 4 files changed, 51 insertions(+), 8 deletions(-)
>
> diff --git a/hw/vfio/pci-quirks.c b/hw/vfio/pci-quirks.c
> index f7886487744e..061259b86b71 100644
> --- a/hw/vfio/pci-quirks.c
> +++ b/hw/vfio/pci-quirks.c
> @@ -16,6 +16,7 @@
> #include "qemu/range.h"
> #include "qapi/error.h"
> #include "qapi/visitor.h"
> +#include <sys/ioctl.h>
> #include "hw/nvram/fw_cfg.h"
> #include "pci.h"
> #include "trace.h"
> @@ -287,12 +288,33 @@ static VFIOQuirk *vfio_quirk_alloc(int nr_mem)
> return quirk;
> }
>
> -static void vfio_ioeventfd_exit(VFIOIOEventFD *ioeventfd)
> +static void vfio_ioeventfd_exit(VFIOPCIDevice *vdev, VFIOIOEventFD *ioeventfd)
> {
> QLIST_REMOVE(ioeventfd, next);
> memory_region_del_eventfd(ioeventfd->mr, ioeventfd->addr, ioeventfd->size,
> true, ioeventfd->data, &ioeventfd->e);
> - qemu_set_fd_handler(event_notifier_get_fd(&ioeventfd->e), NULL, NULL, NULL);
> +
> + if (ioeventfd->vfio) {
> + struct vfio_device_ioeventfd vfio_ioeventfd;
> +
> + vfio_ioeventfd.argsz = sizeof(vfio_ioeventfd);
> + vfio_ioeventfd.flags = ioeventfd->size;
> + vfio_ioeventfd.data = ioeventfd->data;
> + vfio_ioeventfd.offset = ioeventfd->region->fd_offset +
> + ioeventfd->region_addr;
> + vfio_ioeventfd.fd = -1;
> +
> + if (ioctl(vdev->vbasedev.fd, VFIO_DEVICE_IOEVENTFD, &vfio_ioeventfd)) {
> + error_report("Failed to remove vfio ioeventfd for %s+0x%"
> + HWADDR_PRIx"[%d]:0x%"PRIx64" (%m)",
> + memory_region_name(ioeventfd->mr), ioeventfd->addr,
> + ioeventfd->size, ioeventfd->data);
> + }
> + } else {
> + qemu_set_fd_handler(event_notifier_get_fd(&ioeventfd->e),
> + NULL, NULL, NULL);
> + }
> +
> event_notifier_cleanup(&ioeventfd->e);
> trace_vfio_ioeventfd_exit(memory_region_name(ioeventfd->mr),
> (uint64_t)ioeventfd->addr, ioeventfd->size,
> @@ -306,7 +328,7 @@ static void vfio_drop_dynamic_eventfds(VFIOPCIDevice *vdev, VFIOQuirk *quirk)
>
> QLIST_FOREACH_SAFE(ioeventfd, &quirk->ioeventfds, next, tmp) {
> if (ioeventfd->dynamic) {
> - vfio_ioeventfd_exit(ioeventfd);
> + vfio_ioeventfd_exit(vdev, ioeventfd);
> }
> }
> }
> @@ -359,12 +381,29 @@ static VFIOIOEventFD *vfio_ioeventfd_init(VFIOPCIDevice *vdev,
> ioeventfd->region = region;
> ioeventfd->region_addr = region_addr;
>
> - qemu_set_fd_handler(event_notifier_get_fd(&ioeventfd->e),
> - vfio_ioeventfd_handler, NULL, ioeventfd);
> + if (!vdev->no_vfio_ioeventfd) {
> + struct vfio_device_ioeventfd vfio_ioeventfd;
> +
> + vfio_ioeventfd.argsz = sizeof(vfio_ioeventfd);
> + vfio_ioeventfd.flags = ioeventfd->size;
> + vfio_ioeventfd.data = ioeventfd->data;
> + vfio_ioeventfd.offset = ioeventfd->region->fd_offset +
> + ioeventfd->region_addr;
> + vfio_ioeventfd.fd = event_notifier_get_fd(&ioeventfd->e);
> +
> + ioeventfd->vfio = !ioctl(vdev->vbasedev.fd,
> + VFIO_DEVICE_IOEVENTFD, &vfio_ioeventfd);
> + }
> +
> + if (!ioeventfd->vfio) {
> + qemu_set_fd_handler(event_notifier_get_fd(&ioeventfd->e),
> + vfio_ioeventfd_handler, NULL, ioeventfd);
> + }
> +
> memory_region_add_eventfd(ioeventfd->mr, ioeventfd->addr, ioeventfd->size,
> true, ioeventfd->data, &ioeventfd->e);
> trace_vfio_ioeventfd_init(memory_region_name(mr), (uint64_t)addr,
> - size, data);
> + size, data, ioeventfd->vfio);
>
> return ioeventfd;
> }
> @@ -1827,7 +1866,7 @@ void vfio_bar_quirk_exit(VFIOPCIDevice *vdev, int nr)
>
> QLIST_FOREACH(quirk, &bar->quirks, next) {
> while (!QLIST_EMPTY(&quirk->ioeventfds)) {
> - vfio_ioeventfd_exit(QLIST_FIRST(&quirk->ioeventfds));
> + vfio_ioeventfd_exit(vdev, QLIST_FIRST(&quirk->ioeventfds));
> }
>
> for (i = 0; i < quirk->nr_mem; i++) {
> diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
> index ba1239551115..84e27c7bb2d1 100644
> --- a/hw/vfio/pci.c
> +++ b/hw/vfio/pci.c
> @@ -3177,6 +3177,8 @@ static Property vfio_pci_dev_properties[] = {
> no_geforce_quirks, false),
> DEFINE_PROP_BOOL("x-no-kvm-ioeventfd", VFIOPCIDevice, no_kvm_ioeventfd,
> false),
> + DEFINE_PROP_BOOL("x-no-vfio-ioeventfd", VFIOPCIDevice, no_vfio_ioeventfd,
> + false),
> DEFINE_PROP_UINT32("x-pci-vendor-id", VFIOPCIDevice, vendor_id, PCI_ANY_ID),
> DEFINE_PROP_UINT32("x-pci-device-id", VFIOPCIDevice, device_id, PCI_ANY_ID),
> DEFINE_PROP_UINT32("x-pci-sub-vendor-id", VFIOPCIDevice,
> diff --git a/hw/vfio/pci.h b/hw/vfio/pci.h
> index a4ac583fbd6e..52b065421a68 100644
> --- a/hw/vfio/pci.h
> +++ b/hw/vfio/pci.h
> @@ -34,6 +34,7 @@ typedef struct VFIOIOEventFD {
> VFIORegion *region;
> hwaddr region_addr;
> bool dynamic; /* Added runtime, removed on device reset */
> + bool vfio;
> } VFIOIOEventFD;
>
> typedef struct VFIOQuirk {
> @@ -163,6 +164,7 @@ typedef struct VFIOPCIDevice {
> bool no_kvm_msix;
> bool no_geforce_quirks;
> bool no_kvm_ioeventfd;
> + bool no_vfio_ioeventfd;
> VFIODisplay *dpy;
> } VFIOPCIDevice;
>
> diff --git a/hw/vfio/trace-events b/hw/vfio/trace-events
> index f8f97d1ff90c..d2a74952e389 100644
> --- a/hw/vfio/trace-events
> +++ b/hw/vfio/trace-events
> @@ -79,7 +79,7 @@ vfio_quirk_ati_bonaire_reset_done(const char *name) "%s"
> vfio_quirk_ati_bonaire_reset(const char *name) "%s"
> vfio_ioeventfd_exit(const char *name, uint64_t addr, unsigned size, uint64_t data) "%s+0x%"PRIx64"[%d]:0x%"PRIx64
> vfio_ioeventfd_handler(const char *name, uint64_t addr, unsigned size, uint64_t data) "%s+0x%"PRIx64"[%d] -> 0x%"PRIx64
> -vfio_ioeventfd_init(const char *name, uint64_t addr, unsigned size, uint64_t data) "%s+0x%"PRIx64"[%d]:0x%"PRIx64
> +vfio_ioeventfd_init(const char *name, uint64_t addr, unsigned size, uint64_t data, bool vfio) "%s+0x%"PRIx64"[%d]:0x%"PRIx64" vfio:%d"
> vfio_pci_igd_bar4_write(const char *name, uint32_t index, uint32_t data, uint32_t base) "%s [0x%03x] 0x%08x -> 0x%08x"
> vfio_pci_igd_bdsm_enabled(const char *name, int size) "%s %dMB"
> vfio_pci_igd_opregion_enabled(const char *name) "%s"
>
prev parent reply other threads:[~2018-05-04 7:51 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-05-03 21:44 [PATCH v3 0/6] vfio/quirks: ioeventfd support Alex Williamson
2018-05-03 21:44 ` [PATCH v3 1/6] scripts/update-linux-headers: Update handling of __aligned_u64 Alex Williamson
2018-05-22 14:04 ` Peter Maydell
2018-05-03 21:44 ` [PATCH v3 2/6] linux-headers: Update to v4.17-rc3 Alex Williamson
2018-05-03 21:44 ` [PATCH v3 3/6] vfio/quirks: Add common quirk alloc helper Alex Williamson
2018-05-03 21:44 ` [PATCH v3 4/6] vfio/quirks: Add quirk reset callback Alex Williamson
2018-05-03 21:45 ` [PATCH v3 5/6] vfio/quirks: ioeventfd quirk acceleration Alex Williamson
2018-05-04 2:17 ` Peter Xu
2018-05-04 7:51 ` Auger Eric
2018-05-03 21:45 ` [PATCH v3 6/6] vfio/quirks: Enable ioeventfd quirks to be handled by vfio directly Alex Williamson
2018-05-04 2:17 ` Peter Xu
2018-05-04 7:51 ` Auger Eric [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=0700cc1f-ea20-c01d-5c6a-3e3fb1439730@redhat.com \
--to=eric.auger@redhat.com \
--cc=alex.williamson@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=peterx@redhat.com \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox