From: "Michael S. Tsirkin" <mst@redhat.com>
To: qemu-devel@nongnu.org
Cc: "Peter Maydell" <peter.maydell@linaro.org>,
"Hawkins Jiawei" <yin31149@gmail.com>,
"Eugenio Pérez" <eperezma@redhat.com>,
"Jason Wang" <jasowang@redhat.com>
Subject: [PULL v3 01/62] vdpa: Use iovec for vhost_vdpa_net_cvq_add()
Date: Sun, 22 Oct 2023 05:21:58 -0400 [thread overview]
Message-ID: <0e6bff0d43bf04c6e7a16c2775879816ca056b3d.1697966402.git.mst@redhat.com> (raw)
In-Reply-To: <cover.1697966402.git.mst@redhat.com>
From: Hawkins Jiawei <yin31149@gmail.com>
Next patches in this series will no longer perform an
immediate poll and check of the device's used buffers
for each CVQ state load command. Consequently, there
will be multiple pending buffers in the shadow VirtQueue,
making it a must for every control command to have its
own buffer.
To achieve this, this patch refactor vhost_vdpa_net_cvq_add()
to accept `struct iovec`, which eliminates the coupling of
control commands to `s->cvq_cmd_out_buffer` and `s->status`,
allowing them to use their own buffer.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <8a328f146fb043f34edb75ba6d043d2d6de88f99.1697165821.git.yin31149@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
net/vhost-vdpa.c | 39 ++++++++++++++++++++++-----------------
1 file changed, 22 insertions(+), 17 deletions(-)
diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c
index 939c984d5b..618758596a 100644
--- a/net/vhost-vdpa.c
+++ b/net/vhost-vdpa.c
@@ -618,22 +618,14 @@ static void vhost_vdpa_net_cvq_stop(NetClientState *nc)
vhost_vdpa_net_client_stop(nc);
}
-static ssize_t vhost_vdpa_net_cvq_add(VhostVDPAState *s, size_t out_len,
- size_t in_len)
+static ssize_t vhost_vdpa_net_cvq_add(VhostVDPAState *s,
+ const struct iovec *out_sg, size_t out_num,
+ const struct iovec *in_sg, size_t in_num)
{
- /* Buffers for the device */
- const struct iovec out = {
- .iov_base = s->cvq_cmd_out_buffer,
- .iov_len = out_len,
- };
- const struct iovec in = {
- .iov_base = s->status,
- .iov_len = sizeof(virtio_net_ctrl_ack),
- };
VhostShadowVirtqueue *svq = g_ptr_array_index(s->vhost_vdpa.shadow_vqs, 0);
int r;
- r = vhost_svq_add(svq, &out, 1, &in, 1, NULL);
+ r = vhost_svq_add(svq, out_sg, out_num, in_sg, in_num, NULL);
if (unlikely(r != 0)) {
if (unlikely(r == -ENOSPC)) {
qemu_log_mask(LOG_GUEST_ERROR, "%s: No space on device queue\n",
@@ -659,6 +651,15 @@ static ssize_t vhost_vdpa_net_load_cmd(VhostVDPAState *s, uint8_t class,
.cmd = cmd,
};
size_t data_size = iov_size(data_sg, data_num);
+ /* Buffers for the device */
+ const struct iovec out = {
+ .iov_base = s->cvq_cmd_out_buffer,
+ .iov_len = sizeof(ctrl) + data_size,
+ };
+ const struct iovec in = {
+ .iov_base = s->status,
+ .iov_len = sizeof(*s->status),
+ };
assert(data_size < vhost_vdpa_net_cvq_cmd_page_len() - sizeof(ctrl));
@@ -669,8 +670,7 @@ static ssize_t vhost_vdpa_net_load_cmd(VhostVDPAState *s, uint8_t class,
iov_to_buf(data_sg, data_num, 0,
s->cvq_cmd_out_buffer + sizeof(ctrl), data_size);
- return vhost_vdpa_net_cvq_add(s, data_size + sizeof(ctrl),
- sizeof(virtio_net_ctrl_ack));
+ return vhost_vdpa_net_cvq_add(s, &out, 1, &in, 1);
}
static int vhost_vdpa_net_load_mac(VhostVDPAState *s, const VirtIONet *n)
@@ -1248,10 +1248,15 @@ static int vhost_vdpa_net_handle_ctrl_avail(VhostShadowVirtqueue *svq,
.iov_base = s->cvq_cmd_out_buffer,
};
/* in buffer used for device model */
- const struct iovec in = {
+ const struct iovec model_in = {
.iov_base = &status,
.iov_len = sizeof(status),
};
+ /* in buffer used for vdpa device */
+ const struct iovec vdpa_in = {
+ .iov_base = s->status,
+ .iov_len = sizeof(*s->status),
+ };
ssize_t dev_written = -EINVAL;
out.iov_len = iov_to_buf(elem->out_sg, elem->out_num, 0,
@@ -1285,7 +1290,7 @@ static int vhost_vdpa_net_handle_ctrl_avail(VhostShadowVirtqueue *svq,
goto out;
}
} else {
- dev_written = vhost_vdpa_net_cvq_add(s, out.iov_len, sizeof(status));
+ dev_written = vhost_vdpa_net_cvq_add(s, &out, 1, &vdpa_in, 1);
if (unlikely(dev_written < 0)) {
goto out;
}
@@ -1301,7 +1306,7 @@ static int vhost_vdpa_net_handle_ctrl_avail(VhostShadowVirtqueue *svq,
}
status = VIRTIO_NET_ERR;
- virtio_net_handle_ctrl_iov(svq->vdev, &in, 1, &out, 1);
+ virtio_net_handle_ctrl_iov(svq->vdev, &model_in, 1, &out, 1);
if (status != VIRTIO_NET_OK) {
error_report("Bad CVQ processing in model");
}
--
MST
next prev parent reply other threads:[~2023-10-22 9:22 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-22 9:21 [PULL v3 00/62] virtio,pc,pci: features, cleanups Michael S. Tsirkin
2023-10-22 9:21 ` Michael S. Tsirkin [this message]
2023-10-22 9:22 ` [PULL v3 02/62] vdpa: Avoid using vhost_vdpa_net_load_*() outside vhost_vdpa_net_load() Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 03/62] vdpa: Check device ack in vhost_vdpa_net_load_rx_mode() Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 04/62] vdpa: Move vhost_svq_poll() to the caller of vhost_vdpa_net_cvq_add() Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 05/62] vdpa: Introduce cursors to vhost_vdpa_net_loadx() Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 06/62] vhost: Expose vhost_svq_available_slots() Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 07/62] vdpa: Send cvq state load commands in parallel Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 08/62] vhost-user: strip superfluous whitespace Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 09/62] vhost-user: tighten "reply_supported" scope in "set_vring_addr" Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 10/62] vhost-user: factor out "vhost_user_write_sync" Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 11/62] vhost-user: flatten "enforce_reply" into "vhost_user_write_sync" Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 12/62] vhost-user: hoist "write_sync", "get_features", "get_u64" Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 13/62] vhost-user: allow "vhost_set_vring" to wait for a reply Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 14/62] vhost-user: call VHOST_USER_SET_VRING_ENABLE synchronously Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 15/62] memory: initialize 'fv' in MemoryRegionCache to make Coverity happy Michael S. Tsirkin
2023-10-22 9:22 ` [PULL v3 16/62] vhost-user: do not send RESET_OWNER on device reset Michael S. Tsirkin
2023-10-22 9:23 ` [PULL v3 17/62] vhost-backend: remove vhost_kernel_reset_device() Michael S. Tsirkin
2023-10-22 9:23 ` [PULL v3 18/62] virtio: call ->vhost_reset_device() during reset Michael S. Tsirkin
2023-10-22 9:23 ` [PULL v3 19/62] hw/i386/acpi-build: Remove build-time assertion on PIIX/ICH9 reset registers being identical Michael S. Tsirkin
2023-10-22 9:23 ` [PULL v3 20/62] timer/i8254: Fix one shot PIT mode Michael S. Tsirkin
2023-10-22 9:23 ` [PULL v3 21/62] hw/display: fix memleak from virtio_add_resource Michael S. Tsirkin
2023-10-24 6:19 ` Michael Tokarev
2023-10-22 9:23 ` [PULL v3 22/62] hw/i386/pc: Merge two if statements into one Michael S. Tsirkin
2023-10-22 9:23 ` [PULL v3 23/62] hw/i386/pc_piix: Allow for setting properties before realizing PIIX3 south bridge Michael S. Tsirkin
2023-10-22 9:23 ` [PULL v3 24/62] hw/i386/pc_piix: Assign PIIX3's ISA interrupts before its realize() Michael S. Tsirkin
2023-10-22 9:23 ` [PULL v3 25/62] hw/isa/piix3: Resolve redundant PIIX_NUM_PIC_IRQS Michael S. Tsirkin
2023-10-22 9:23 ` [PULL v3 26/62] hw/i386/pc_piix: Wire PIIX3's ISA interrupts by new "isa-irqs" property Michael S. Tsirkin
2023-10-22 9:23 ` [PULL v3 27/62] hw/i386/pc_piix: Remove redundant "piix3" variable Michael S. Tsirkin
2023-10-22 9:23 ` [PULL v3 28/62] hw/isa/piix3: Rename "pic" attribute to "isa_irqs_in" Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 29/62] hw/i386/pc_q35: Wire ICH9 LPC function's interrupts before its realize() Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 30/62] hw/isa/piix3: Wire PIC IRQs to ISA bus in host device Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 31/62] hw/i386/pc: Wire RTC ISA IRQs in south bridges Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 32/62] hw/isa/piix3: Create IDE controller in host device Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 33/62] hw/isa/piix3: Create USB " Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 34/62] hw/isa/piix3: Create power management " Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 35/62] hw/isa/piix3: Drop the "3" from PIIX base class name Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 36/62] hw/isa/piix4: Remove unused inbound ISA interrupt lines Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 37/62] hw/isa/piix4: Rename "isa" attribute to "isa_irqs_in" Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 38/62] hw/isa/piix4: Rename reset control operations to match PIIX3 Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 39/62] hw/isa/piix4: Reuse struct PIIXState from PIIX3 Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 40/62] hw/isa/piix3: Merge hw/isa/piix4.c Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 41/62] hw/isa/piix: Allow for optional PIC creation in PIIX3 Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 42/62] hw/isa/piix: Allow for optional PIT " Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 43/62] hw/isa/piix: Harmonize names of reset control memory regions Michael S. Tsirkin
2023-10-22 9:24 ` [PULL v3 44/62] hw/isa/piix: Share PIIX3's base class with PIIX4 Michael S. Tsirkin
2023-10-22 9:25 ` [PULL v3 45/62] hw/isa/piix: Reuse PIIX3 base class' realize method in PIIX4 Michael S. Tsirkin
2023-10-22 9:25 ` [PULL v3 46/62] hw/isa/piix: Rename functions to be shared for PCI interrupt triggering Michael S. Tsirkin
2023-10-22 9:25 ` [PULL v3 47/62] hw/isa/piix: Reuse PIIX3's PCI interrupt triggering in PIIX4 Michael S. Tsirkin
2023-10-22 9:25 ` [PULL v3 48/62] hw/isa/piix: Resolve duplicate code regarding PCI interrupt wiring Michael S. Tsirkin
2023-10-22 9:25 ` [PULL v3 49/62] hw/isa/piix: Implement multi-process QEMU support also for PIIX4 Michael S. Tsirkin
2023-10-22 9:25 ` [PULL v3 50/62] hw/i386/pc_piix: Make PIIX4 south bridge usable in PC machine Michael S. Tsirkin
2023-10-22 9:25 ` [PULL v3 51/62] vhost-user-common: send get_inflight_fd once Michael S. Tsirkin
2023-10-22 9:25 ` [PULL v3 52/62] vhost: move and rename the conn retry times Michael S. Tsirkin
2023-10-22 9:25 ` [PULL v3 53/62] vhost-user-scsi: support reconnect to backend Michael S. Tsirkin
2023-10-22 9:25 ` [PULL v3 54/62] vhost-user-scsi: start vhost when guest kicks Michael S. Tsirkin
2023-10-22 9:26 ` [PULL v3 55/62] vhost-user: fix lost reconnect Michael S. Tsirkin
2023-10-22 9:26 ` [PULL v3 56/62] hw/i386/cxl: ensure maxram is greater than ram size for calculating cxl range Michael S. Tsirkin
2023-10-22 9:26 ` [PULL v3 57/62] tests/acpi: Allow update of DSDT.cxl Michael S. Tsirkin
2023-10-22 9:26 ` [PULL v3 58/62] hw/cxl: Add QTG _DSM support for ACPI0017 device Michael S. Tsirkin
2023-10-22 9:26 ` [PULL v3 59/62] tests/acpi: Update DSDT.cxl with QTG DSM Michael S. Tsirkin
2023-10-22 9:26 ` [PULL v3 60/62] vhost-user: Fix protocol feature bit conflict Michael S. Tsirkin
2023-10-22 9:26 ` [PULL v3 61/62] MAINTAINERS: Add include/hw/intc/i8259.h to the PC chip section Michael S. Tsirkin
2023-10-22 9:26 ` [PULL v3 62/62] intel-iommu: Report interrupt remapping faults, fix return value Michael S. Tsirkin
2023-10-24 1:15 ` [PULL v3 00/62] virtio,pc,pci: features, cleanups Stefan Hajnoczi
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=0e6bff0d43bf04c6e7a16c2775879816ca056b3d.1697966402.git.mst@redhat.com \
--to=mst@redhat.com \
--cc=eperezma@redhat.com \
--cc=jasowang@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=yin31149@gmail.com \
/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).