From: "gavin.liu" <gavin.liu@jaguarmicro.com>
To: jasowang@redhat.com, mst@redhat.com
Cc: xuanzhuo@linux.alibaba.com, virtualization@lists.linux.dev,
angus.chen@jaguarmicro.com, yuxue.liu@jaguarmicro.com
Subject: [PATCH v2] vp_vdpa: fix the method of calculating vectors
Date: Mon, 18 Mar 2024 21:00:08 +0800 [thread overview]
Message-ID: <20240318130008.1928-1-gavin.liu@jaguarmicro.com> (raw)
From: Yuxue Liu <yuxue.liu@jaguarmicro.com>
When there is a ctlq and it doesn't require interrupt
callbacks,the original method of calculating vectors
wastes hardware MSI or MSI-X resources as well as system
IRQ resources. Referencing the per_vq_vectors mode in the
vp_find_vqs_msix function, calculate the required number
of vectors based on whether the callback is set.
Signed-off-by: Yuxue Liu <yuxue.liu@jaguarmicro.com>
---
V1 -> V2: fix when allocating IRQs, scan all queues.
V1: https://lore.kernel.org/all/20240318030121.1873-1-gavin.liu@jaguarmicro.com/
---
drivers/vdpa/virtio_pci/vp_vdpa.c | 24 ++++++++++++++++++------
1 file changed, 18 insertions(+), 6 deletions(-)
diff --git a/drivers/vdpa/virtio_pci/vp_vdpa.c b/drivers/vdpa/virtio_pci/vp_vdpa.c
index 281287fae89f..87329d4358ce 100644
--- a/drivers/vdpa/virtio_pci/vp_vdpa.c
+++ b/drivers/vdpa/virtio_pci/vp_vdpa.c
@@ -160,8 +160,15 @@ static int vp_vdpa_request_irq(struct vp_vdpa *vp_vdpa)
struct pci_dev *pdev = mdev->pci_dev;
int i, ret, irq;
int queues = vp_vdpa->queues;
- int vectors = queues + 1;
+ int allocated_vectors, vectors = 0;
+ u16 msix_vec;
+ for (i = 0; i < queues; i++) {
+ if (vp_vdpa->vring[i].cb.callback != NULL)
+ vectors++;
+ }
+ /*By default, config interrupts request a single vector*/
+ vectors = vectors + 1;
ret = pci_alloc_irq_vectors(pdev, vectors, vectors, PCI_IRQ_MSIX);
if (ret != vectors) {
dev_err(&pdev->dev,
@@ -169,13 +176,17 @@ static int vp_vdpa_request_irq(struct vp_vdpa *vp_vdpa)
vectors, ret);
return ret;
}
-
vp_vdpa->vectors = vectors;
for (i = 0; i < queues; i++) {
+ if (vp_vdpa->vring[i].cb.callback == NULL)
+ continue;
+ else
+ msix_vec = allocated_vectors++;
+
snprintf(vp_vdpa->vring[i].msix_name, VP_VDPA_NAME_SIZE,
"vp-vdpa[%s]-%d\n", pci_name(pdev), i);
- irq = pci_irq_vector(pdev, i);
+ irq = pci_irq_vector(pdev, msix_vec);
ret = devm_request_irq(&pdev->dev, irq,
vp_vdpa_vq_handler,
0, vp_vdpa->vring[i].msix_name,
@@ -185,13 +196,14 @@ static int vp_vdpa_request_irq(struct vp_vdpa *vp_vdpa)
"vp_vdpa: fail to request irq for vq %d\n", i);
goto err;
}
- vp_modern_queue_vector(mdev, i, i);
+ vp_modern_queue_vector(mdev, i, msix_vec);
vp_vdpa->vring[i].irq = irq;
}
+ msix_vec = allocated_vectors;
snprintf(vp_vdpa->msix_name, VP_VDPA_NAME_SIZE, "vp-vdpa[%s]-config\n",
pci_name(pdev));
- irq = pci_irq_vector(pdev, queues);
+ irq = pci_irq_vector(pdev, msix_vec);
ret = devm_request_irq(&pdev->dev, irq, vp_vdpa_config_handler, 0,
vp_vdpa->msix_name, vp_vdpa);
if (ret) {
@@ -199,7 +211,7 @@ static int vp_vdpa_request_irq(struct vp_vdpa *vp_vdpa)
"vp_vdpa: fail to request irq for vq %d\n", i);
goto err;
}
- vp_modern_config_vector(mdev, queues);
+ vp_modern_config_vector(mdev, msix_vec);
vp_vdpa->config_irq = irq;
return 0;
--
2.43.0
next reply other threads:[~2024-03-18 13:00 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-03-18 13:00 gavin.liu [this message]
2024-04-08 8:02 ` [PATCH v2] vp_vdpa: fix the method of calculating vectors Michael S. Tsirkin
2024-04-09 1:49 ` [PATCH v3] " lyx634449800
2024-04-09 3:53 ` Jason Wang
2024-04-09 5:40 ` Michael S. Tsirkin
2024-04-09 8:58 ` [PATCH v4] vp_vdpa: don't allocate unused msix vectors lyx634449800
2024-04-09 9:26 ` Michael S. Tsirkin
2024-04-09 9:56 ` Heng Qi
2024-04-10 3:30 ` [PATCH v5] " lyx634449800
2024-04-22 12:08 ` Michael S. Tsirkin
2024-04-23 1:39 ` 回复: " Gavin Liu
2024-04-23 8:35 ` Michael S. Tsirkin
2024-04-23 8:42 ` Angus Chen
2024-04-23 9:25 ` 回复: " Gavin Liu
2024-04-25 22:09 ` Michael S. Tsirkin
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=20240318130008.1928-1-gavin.liu@jaguarmicro.com \
--to=gavin.liu@jaguarmicro.com \
--cc=angus.chen@jaguarmicro.com \
--cc=jasowang@redhat.com \
--cc=mst@redhat.com \
--cc=virtualization@lists.linux.dev \
--cc=xuanzhuo@linux.alibaba.com \
--cc=yuxue.liu@jaguarmicro.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 read-only IMAP folder(s) and NNTP newsgroup(s).