qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Michael S. Tsirkin" <mst@redhat.com>
To: qemu-devel@nongnu.org
Cc: "Peter Maydell" <peter.maydell@linaro.org>,
	"Jonah Palmer" <jonah.palmer@oracle.com>,
	"Eugenio Pérez" <eperezma@redhat.com>
Subject: [PULL 25/63] virtio: virtqueue_ordered_flush - VIRTIO_F_IN_ORDER support
Date: Sun, 21 Jul 2024 20:17:41 -0400	[thread overview]
Message-ID: <844619147c5110f335cbbbad51cda6a898c7ef33.1721607331.git.mst@redhat.com> (raw)
In-Reply-To: <cover.1721607331.git.mst@redhat.com>

From: Jonah Palmer <jonah.palmer@oracle.com>

Add VIRTIO_F_IN_ORDER feature support for the virtqueue_flush operation.

The goal of the virtqueue_ordered_flush operation when the
VIRTIO_F_IN_ORDER feature has been negotiated is to write elements to
the used/descriptor ring in-order and then update used_idx.

The function iterates through the VirtQueueElement used_elems array
in-order starting at vq->used_idx. If the element is valid (filled), the
element is written to the used/descriptor ring. This process continues
until we find an invalid (not filled) element.

For packed VQs, the first entry (at vq->used_idx) is written to the
descriptor ring last so the guest doesn't see any invalid descriptors.

If any elements were written, the used_idx is updated.

Signed-off-by: Jonah Palmer <jonah.palmer@oracle.com>
Message-Id: <20240710125522.4168043-5-jonah.palmer@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
---
 hw/virtio/virtio.c | 71 +++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 70 insertions(+), 1 deletion(-)

diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index 0000a7b41c..397c261c3c 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -1023,6 +1023,73 @@ static void virtqueue_packed_flush(VirtQueue *vq, unsigned int count)
     }
 }
 
+static void virtqueue_ordered_flush(VirtQueue *vq)
+{
+    unsigned int i = vq->used_idx % vq->vring.num;
+    unsigned int ndescs = 0;
+    uint16_t old = vq->used_idx;
+    uint16_t new;
+    bool packed;
+    VRingUsedElem uelem;
+
+    packed = virtio_vdev_has_feature(vq->vdev, VIRTIO_F_RING_PACKED);
+
+    if (packed) {
+        if (unlikely(!vq->vring.desc)) {
+            return;
+        }
+    } else if (unlikely(!vq->vring.used)) {
+        return;
+    }
+
+    /* First expected in-order element isn't ready, nothing to do */
+    if (!vq->used_elems[i].in_order_filled) {
+        return;
+    }
+
+    /* Search for filled elements in-order */
+    while (vq->used_elems[i].in_order_filled) {
+        /*
+         * First entry for packed VQs is written last so the guest
+         * doesn't see invalid descriptors.
+         */
+        if (packed && i != vq->used_idx) {
+            virtqueue_packed_fill_desc(vq, &vq->used_elems[i], ndescs, false);
+        } else if (!packed) {
+            uelem.id = vq->used_elems[i].index;
+            uelem.len = vq->used_elems[i].len;
+            vring_used_write(vq, &uelem, i);
+        }
+
+        vq->used_elems[i].in_order_filled = false;
+        ndescs += vq->used_elems[i].ndescs;
+        i += vq->used_elems[i].ndescs;
+        if (i >= vq->vring.num) {
+            i -= vq->vring.num;
+        }
+    }
+
+    if (packed) {
+        virtqueue_packed_fill_desc(vq, &vq->used_elems[vq->used_idx], 0, true);
+        vq->used_idx += ndescs;
+        if (vq->used_idx >= vq->vring.num) {
+            vq->used_idx -= vq->vring.num;
+            vq->used_wrap_counter ^= 1;
+            vq->signalled_used_valid = false;
+        }
+    } else {
+        /* Make sure buffer is written before we update index. */
+        smp_wmb();
+        new = old + ndescs;
+        vring_used_idx_set(vq, new);
+        if (unlikely((int16_t)(new - vq->signalled_used) <
+                     (uint16_t)(new - old))) {
+            vq->signalled_used_valid = false;
+        }
+    }
+    vq->inuse -= ndescs;
+}
+
 void virtqueue_flush(VirtQueue *vq, unsigned int count)
 {
     if (virtio_device_disabled(vq->vdev)) {
@@ -1030,7 +1097,9 @@ void virtqueue_flush(VirtQueue *vq, unsigned int count)
         return;
     }
 
-    if (virtio_vdev_has_feature(vq->vdev, VIRTIO_F_RING_PACKED)) {
+    if (virtio_vdev_has_feature(vq->vdev, VIRTIO_F_IN_ORDER)) {
+        virtqueue_ordered_flush(vq);
+    } else if (virtio_vdev_has_feature(vq->vdev, VIRTIO_F_RING_PACKED)) {
         virtqueue_packed_flush(vq, count);
     } else {
         virtqueue_split_flush(vq, count);
-- 
MST



  parent reply	other threads:[~2024-07-22  0:19 UTC|newest]

Thread overview: 73+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-07-22  0:16 [PULL 00/63] virtio,pci,pc: features,fixes Michael S. Tsirkin
2024-07-22  0:16 ` [PULL 01/63] hw/virtio/virtio-crypto: Fix op_code assignment in virtio_crypto_create_asym_session Michael S. Tsirkin
2024-07-22  0:16 ` [PULL 02/63] MAINTAINERS: add Stefano Garzarella as vhost/vhost-user reviewer Michael S. Tsirkin
2024-07-22  0:16 ` [PULL 03/63] hw/cxl/cxl-mailbox-utils: remove unneeded mailbox output payload space zeroing Michael S. Tsirkin
2024-07-22  0:16 ` [PULL 04/63] hw/cxl: Check for multiple mappings of memory backends Michael S. Tsirkin
2024-07-22  0:16 ` [PULL 05/63] hw/cxl/cxl-host: Fix segmentation fault when getting cxl-fmw property Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 06/63] hw/cxl: Add get scan media capabilities cmd support Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 07/63] hw/cxl/mbox: replace sanitize_running() with cxl_dev_media_disabled() Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 08/63] hw/cxl/events: discard all event records during sanitation Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 09/63] hw/cxl: Add get scan media results cmd support Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 10/63] cxl/mailbox: move mailbox effect definitions to a header Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 11/63] hw/cxl/cxl-mailbox-utils: Add support for feature commands (8.2.9.6) Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 12/63] hw/cxl/cxl-mailbox-utils: Add device patrol scrub control feature Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 13/63] hw/cxl/cxl-mailbox-utils: Add device DDR5 ECS " Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 14/63] hw/cxl: Support firmware updates Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 15/63] MAINTAINERS: Add myself as a VT-d reviewer Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 16/63] virtio-snd: add max size bounds check in input cb Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 17/63] virtio-snd: check for invalid param shift operands Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 18/63] intel_iommu: fix FRCD construction macro Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 19/63] intel_iommu: move VTD_FRCD_PV and VTD_FRCD_PP declarations Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 20/63] intel_iommu: fix type of the mask field in VTDIOTLBPageInvInfo Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 21/63] intel_iommu: make type match Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 22/63] virtio: Add bool to VirtQueueElement Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 23/63] virtio: virtqueue_pop - VIRTIO_F_IN_ORDER support Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 24/63] virtio: virtqueue_ordered_fill " Michael S. Tsirkin
2024-07-22  0:17 ` Michael S. Tsirkin [this message]
2024-07-22  0:17 ` [PULL 26/63] vhost,vhost-user: Add VIRTIO_F_IN_ORDER to vhost feature bits Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 27/63] virtio: Add VIRTIO_F_IN_ORDER property definition Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 28/63] contrib/vhost-user-blk: fix overflowing expression Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 29/63] hw/pci: Do not add ROM BAR for SR-IOV VF Michael S. Tsirkin
2024-07-22 14:21   ` Akihiko Odaki
2024-07-23  0:15     ` Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 30/63] hw/pci: Fix SR-IOV VF number calculation Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 31/63] pcie_sriov: Ensure PF and VF are mutually exclusive Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 32/63] pcie_sriov: Check PCI Express for SR-IOV PF Michael S. Tsirkin
2024-07-22  0:17 ` [PULL 33/63] pcie_sriov: Allow user to create SR-IOV device Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 34/63] virtio-pci: Implement SR-IOV PF Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 35/63] virtio-net: Implement SR-IOV VF Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 36/63] docs: Document composable SR-IOV device Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 37/63] smbios: make memory device size configurable per Machine Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 38/63] accel/kvm: Extract common KVM vCPU {creation,parking} code Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 39/63] hw/acpi: Move CPU ctrl-dev MMIO region len macro to common header file Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 40/63] hw/acpi: Update ACPI GED framework to support vCPU Hotplug Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 41/63] hw/acpi: Update GED _EVT method AML with CPU scan Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 42/63] hw/acpi: Update CPUs AML with cpu-(ctrl)dev change Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 43/63] physmem: Add helper function to destroy CPU AddressSpace Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 44/63] gdbstub: Add helper function to unregister GDB register space Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 45/63] Revert "virtio-iommu: Clear IOMMUDevice when VFIO device is unplugged" Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 46/63] virtio-iommu: Remove probe_done Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 47/63] virtio-iommu: Free [host_]resv_ranges on unset_iommu_devices Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 48/63] virtio-iommu: Remove the end point on detach Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 49/63] hw/vfio/common: Add vfio_listener_region_del_iommu trace event Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 50/63] virtio-iommu: Add trace point on virtio_iommu_detach_endpoint_from_domain Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 51/63] hw/riscv/virt-acpi-build.c: Add namespace devices for PLIC and APLIC Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 52/63] hw/riscv/virt-acpi-build.c: Update the HID of RISC-V UART Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 53/63] tests/acpi: Allow DSDT acpi table changes for aarch64 Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 54/63] acpi/gpex: Create PCI link devices outside PCI root bridge Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 55/63] tests/acpi: update expected DSDT blob for aarch64 and microvm Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 56/63] tests/qtest/bios-tables-test.c: Remove the fall back path Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 57/63] tests/acpi: Add empty ACPI data files for RISC-V Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 58/63] tests/qtest/bios-tables-test.c: Enable basic testing " Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 59/63] tests/acpi: Add expected ACPI AML files " Michael S. Tsirkin
2024-07-22  0:18 ` [PULL 60/63] hw/pci: Add all Data Object Types defined in PCIe r6.0 Michael S. Tsirkin
2024-07-22  0:19 ` [PULL 61/63] backends: Initial support for SPDM socket support Michael S. Tsirkin
2024-07-22  0:19 ` [PULL 62/63] hw/nvme: Add SPDM over DOE support Michael S. Tsirkin
2024-07-22  0:19 ` [PULL 63/63] virtio: Always reset vhost devices Michael S. Tsirkin
2024-07-22 21:32 ` [PULL 00/63] virtio,pci,pc: features,fixes Richard Henderson
2024-07-23  0:20   ` Michael S. Tsirkin
2024-07-23  0:44     ` Richard Henderson
2024-07-23 11:36       ` Michael S. Tsirkin
2024-07-23 10:18   ` Hanna Czenczek
2024-07-23 10:45     ` Michael S. Tsirkin
2024-07-23 11:06       ` Hanna Czenczek

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=844619147c5110f335cbbbad51cda6a898c7ef33.1721607331.git.mst@redhat.com \
    --to=mst@redhat.com \
    --cc=eperezma@redhat.com \
    --cc=jonah.palmer@oracle.com \
    --cc=peter.maydell@linaro.org \
    --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;
as well as URLs for NNTP newsgroup(s).