From: Jianfeng Tan <jianfeng.tan@intel.com>
To: dev@dpdk.org
Cc: yuanhan.liu@linux.intel.com, stephen@networkplumber.org,
Jianfeng Tan <jianfeng.tan@intel.com>
Subject: [dpdk-dev] [PATCH v2 5/9] net/virtio: setup rxq interrupts
Date: Thu, 29 Dec 2016 07:30:39 +0000 [thread overview]
Message-ID: <1482996643-113253-6-git-send-email-jianfeng.tan@intel.com> (raw)
In-Reply-To: <1482996643-113253-1-git-send-email-jianfeng.tan@intel.com>
This patch mainly allocates structure to store queue/irq mapping,
and configure queue/irq mapping down through PCI ops. It also creates
eventfds for each Rx queue and tell the kernel about the eventfd/intr
binding.
Mostly importantly, different from previous NICs (usually implements
these logic in dev_start()), virtio's interrupt settings should be
configured down to QEMU before sending DRIVER_OK notification.
Note: We only support 1:1 queue/irq mapping so far, which means, each
rx queue has one exclusive interrupt (corresponding to irqfd in the
qemu/kvm) to get notified when packets are available on that queue.
Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
---
drivers/net/virtio/virtio_ethdev.c | 89 ++++++++++++++++++++++++++++++++++++++
1 file changed, 89 insertions(+)
diff --git a/drivers/net/virtio/virtio_ethdev.c b/drivers/net/virtio/virtio_ethdev.c
index 3f8b90c..082346b 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1206,6 +1206,76 @@ rx_func_get(struct rte_eth_dev *eth_dev)
eth_dev->rx_pkt_burst = &virtio_recv_pkts;
}
+/* Only support 1:1 queue/interrupt mapping so far.
+ * TODO: under below cases, lsc and rxq interrupt share one interrupt.
+ * a) binded to uio, igb_uio, vfio (type1);
+ * b) device only has one vec, see _vectors_ option in -device virtio-net-pci.
+ * TODO: support n:1 queue/interrupt mapping.
+ */
+static int
+virtio_queues_bind_intr(struct rte_eth_dev *dev)
+{
+ uint32_t i;
+ struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+ struct virtio_hw *hw = dev->data->dev_private;
+
+ PMD_INIT_LOG(INFO, "queue/interrupt binding\n");
+ for (i = 0; i < dev->data->nb_rx_queues; ++i) {
+ intr_handle->intr_vec[i] = i + 1;
+ if (vtpci_irq_queue(hw->vqs[i * VTNET_CQ], i + 1) ==
+ VIRTIO_MSI_NO_VECTOR) {
+ PMD_DRV_LOG(ERR, "failed to set queue vector");
+ return -EBUSY;
+ }
+ }
+
+ return 0;
+}
+
+static int
+virtio_configure_intr(struct rte_eth_dev *dev)
+{
+ uint32_t intr_vector;
+ struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
+
+ /* check if rxq interrupt is enabled */
+ if (!rte_intr_cap_multiple(intr_handle)) {
+ PMD_INIT_LOG(ERR, "Multiple intr vector not supported");
+ return -ENOTSUP;
+ }
+
+ intr_vector = dev->data->nb_rx_queues;
+ if (rte_intr_efd_enable(intr_handle, intr_vector)) {
+ PMD_INIT_LOG(ERR, "Fail to create eventfd");
+ return -1;
+ }
+
+ if (!intr_handle->intr_vec) {
+ intr_handle->intr_vec =
+ rte_zmalloc("intr_vec", intr_vector * sizeof(int), 0);
+ if (!intr_handle->intr_vec) {
+ PMD_INIT_LOG(ERR, "Failed to allocate %d rxq vectors",
+ intr_vector);
+ return -ENOMEM;
+ }
+ }
+
+ if (virtio_queues_bind_intr(dev) < 0) {
+ PMD_INIT_LOG(ERR, "Failed to bind queue/interrupt");
+ return -1;
+ }
+
+ /* DO NOT try remove this! This function will enable msix, or QEMU
+ * will encounter SIGSEGV.
+ */
+ if (rte_intr_enable(intr_handle) < 0) {
+ PMD_DRV_LOG(ERR, "interrupt enable failed");
+ return -1;
+ }
+
+ return 0;
+}
+
/* reset device and renegotiate features if needed */
static int
virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t req_features)
@@ -1299,6 +1369,17 @@ virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t req_features)
ret = virtio_alloc_queues(eth_dev);
if (ret < 0)
return ret;
+
+ /* Make sure rxq interrupt is configured before sending DRIVER_OK,
+ * so that QEMU can properly set those irq into kvm.
+ */
+ if (eth_dev->data->dev_conf.intr_conf.rxq) {
+ if (virtio_configure_intr(eth_dev) < 0) {
+ PMD_INIT_LOG(ERR, "failed to configure interrupt");
+ return -1;
+ }
+ }
+
vtpci_reinit_complete(hw);
if (pci_dev)
@@ -1503,7 +1584,15 @@ virtio_dev_start(struct rte_eth_dev *dev)
PMD_DRV_LOG(ERR, "link status not supported by host");
return -ENOTSUP;
}
+ }
+ /* Enable uio/vfio intr/eventfd mapping: althrough we already did that
+ * in device configure, but it could be unmapped when device is
+ * stopped.
+ */
+ if (dev->data->dev_conf.intr_conf.lsc ||
+ dev->data->dev_conf.intr_conf.rxq) {
+ rte_intr_disable(&dev->pci_dev->intr_handle);
if (rte_intr_enable(&dev->pci_dev->intr_handle) < 0) {
PMD_DRV_LOG(ERR, "interrupt enable failed");
return -EIO;
--
2.7.4
next prev parent reply other threads:[~2016-12-29 7:30 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-12-04 0:18 [dpdk-dev] [PATCH 0/5] Interrupt mode for virtio PMD Jianfeng Tan
2016-12-04 0:18 ` [dpdk-dev] [PATCH 1/5] net/virtio: add Rx descriptor check Jianfeng Tan
2016-12-04 0:18 ` [dpdk-dev] [PATCH 2/5] net/virtio: setup Rx fastpath interrupts Jianfeng Tan
2016-12-04 0:18 ` [dpdk-dev] [PATCH 3/5] net/virtio: remove Rx queue interrupts when stopping Jianfeng Tan
2016-12-04 0:18 ` [dpdk-dev] [PATCH 4/5] net/virtio: add Rx queue intr enable/disable functions Jianfeng Tan
2016-12-04 0:18 ` [dpdk-dev] [PATCH 5/5] examples/l3fwd: add parse-ptype option Jianfeng Tan
2016-12-08 6:59 ` Yuanhan Liu
2016-12-08 7:40 ` Tan, Jianfeng
2016-12-29 7:30 ` [dpdk-dev] [PATCH v2 0/9] rxq interrupt mode for virtio PMD Jianfeng Tan
2016-12-29 7:30 ` [dpdk-dev] [PATCH v2 1/9] net/virtio: fix rewriting LSC flag Jianfeng Tan
2016-12-29 7:30 ` [dpdk-dev] [PATCH v2 2/9] net/virtio: add Rx descriptor check Jianfeng Tan
2016-12-29 7:30 ` [dpdk-dev] [PATCH v2 3/9] net/virtio: add PCI ops for queue/irq binding Jianfeng Tan
2016-12-30 5:46 ` Yuanhan Liu
2016-12-29 7:30 ` [dpdk-dev] [PATCH v2 4/9] net/virtio: add Rx queue intr enable/disable functions Jianfeng Tan
2016-12-30 5:59 ` Yuanhan Liu
2016-12-29 7:30 ` Jianfeng Tan [this message]
2016-12-30 6:27 ` [dpdk-dev] [PATCH v2 5/9] net/virtio: setup rxq interrupts Yuanhan Liu
2017-01-04 6:56 ` Tan, Jianfeng
2017-01-04 7:22 ` Yuanhan Liu
2017-01-04 7:30 ` Tan, Jianfeng
2016-12-29 7:30 ` [dpdk-dev] [PATCH v2 6/9] net/virtio: unbind intr/eventfd when stop device Jianfeng Tan
2016-12-29 7:30 ` [dpdk-dev] [PATCH v2 7/9] net/virtio: unmapping queue/irq when close device Jianfeng Tan
2016-12-30 6:30 ` Yuanhan Liu
2016-12-29 7:30 ` [dpdk-dev] [PATCH v2 8/9] examples/l3fwd: add parse-ptype option Jianfeng Tan
2016-12-30 6:39 ` Yuanhan Liu
2016-12-30 7:30 ` Tan, Jianfeng
2017-01-03 6:59 ` Yuanhan Liu
2016-12-29 7:30 ` [dpdk-dev] [PATCH v2 9/9] examples/l3fwd-power: fix not stop and close device Jianfeng Tan
2016-12-30 6:44 ` Yuanhan Liu
2016-12-30 6:56 ` Tan, Jianfeng
2016-12-29 7:42 ` [dpdk-dev] [PATCH v2 0/9] rxq interrupt mode for virtio PMD Tan, Jianfeng
2016-12-30 5:41 ` Yuanhan Liu
2016-12-30 6:57 ` Yuanhan Liu
2017-01-16 14:46 ` [dpdk-dev] [PATCH v3 00/10] " Jianfeng Tan
2017-01-16 14:46 ` [dpdk-dev] [PATCH v3 01/10] net/virtio: fix rewriting LSC flag Jianfeng Tan
2017-01-16 14:46 ` [dpdk-dev] [PATCH v3 02/10] net/virtio: clean up wrapper of set_config_irq Jianfeng Tan
2017-01-16 14:46 ` [dpdk-dev] [PATCH v3 03/10] net/virtio: add Rx descriptor check Jianfeng Tan
2017-01-16 19:16 ` Stephen Hemminger
2017-01-17 4:09 ` Yuanhan Liu
2017-01-16 14:46 ` [dpdk-dev] [PATCH v3 04/10] net/virtio: add PCI ops for queue/irq binding Jianfeng Tan
2017-01-16 14:46 ` [dpdk-dev] [PATCH v3 05/10] net/virtio: add Rx queue intr enable/disable functions Jianfeng Tan
2017-01-17 2:30 ` Jason Wang
2017-01-16 14:46 ` [dpdk-dev] [PATCH v3 06/10] net/virtio: setup rxq interrupts Jianfeng Tan
2017-01-16 14:46 ` [dpdk-dev] [PATCH v3 07/10] net/virtio: unbind intr/eventfd when stop device Jianfeng Tan
2017-01-16 14:47 ` [dpdk-dev] [PATCH v3 08/10] net/virtio: unmapping queue/irq when close device Jianfeng Tan
2017-01-16 14:47 ` [dpdk-dev] [PATCH v3 09/10] examples/l3fwd-power: add parse-ptype option Jianfeng Tan
2017-01-17 5:16 ` Yuanhan Liu
2017-01-17 5:23 ` Tan, Jianfeng
2017-01-16 14:47 ` [dpdk-dev] [PATCH v3 10/10] examples/l3fwd-power: fix not stop and close device Jianfeng Tan
2017-01-17 2:14 ` [dpdk-dev] [PATCH v3 00/10] rxq interrupt mode for virtio PMD Yao, Lei A
2017-01-17 7:10 ` [dpdk-dev] [PATCH v4 " Jianfeng Tan
2017-01-17 7:10 ` [dpdk-dev] [PATCH v4 01/10] net/virtio: fix rewriting LSC flag Jianfeng Tan
2017-01-17 7:10 ` [dpdk-dev] [PATCH v4 02/10] net/virtio: clean up wrapper of set_config_irq Jianfeng Tan
2017-01-17 7:10 ` [dpdk-dev] [PATCH v4 03/10] net/virtio: add Rx descriptor check Jianfeng Tan
2017-01-17 7:10 ` [dpdk-dev] [PATCH v4 04/10] net/virtio: add PCI ops for queue/irq binding Jianfeng Tan
2017-01-17 7:10 ` [dpdk-dev] [PATCH v4 05/10] net/virtio: add Rx queue intr enable/disable functions Jianfeng Tan
2017-01-17 7:10 ` [dpdk-dev] [PATCH v4 06/10] net/virtio: setup rxq interrupts Jianfeng Tan
2017-01-17 7:10 ` [dpdk-dev] [PATCH v4 07/10] net/virtio: unbind intr/eventfd when stop device Jianfeng Tan
2017-01-17 7:10 ` [dpdk-dev] [PATCH v4 08/10] net/virtio: unmapping queue/irq when close device Jianfeng Tan
2017-01-17 7:10 ` [dpdk-dev] [PATCH v4 09/10] examples/l3fwd-power: add parse-ptype option Jianfeng Tan
2017-01-17 7:10 ` [dpdk-dev] [PATCH v4 10/10] examples/l3fwd-power: fix not stop and close device Jianfeng Tan
2017-01-17 8:28 ` [dpdk-dev] [PATCH v4 00/10] rxq interrupt mode for virtio PMD Yuanhan Liu
2017-01-17 8:00 ` [dpdk-dev] [PATCH v5 06/10] net/virtio: setup rxq interrupts Jianfeng Tan
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=1482996643-113253-6-git-send-email-jianfeng.tan@intel.com \
--to=jianfeng.tan@intel.com \
--cc=dev@dpdk.org \
--cc=stephen@networkplumber.org \
--cc=yuanhan.liu@linux.intel.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).