Add PF driver and LM commands #5

GavinLi-NV · 2022-05-27T08:25:27Z

Add PF driver
Add Admin Q and message sending command API
Add LM commands for suspend/resume, save/restore state and save pending bytes

kailiangz1 · 2022-06-08T01:24:38Z

drivers/common/virtio/virtio_admin.h

+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE. */
+#include <linux/types.h>


we should not rely on linux types, use DPDK types

kailiangz1 · 2022-06-08T01:29:03Z

drivers/vdpa/virtio/virtio_vdpa.c

@@ -62,6 +64,7 @@ struct virtio_vdpa_priv {

 #define VIRTIO_VDPA_INTR_RETRIES_USEC 1000
 #define VIRTIO_VDPA_INTR_RETRIES 256
+#include <virtio_lm.h>


this file had included in line 14

kailiangz1 · 2022-06-08T01:30:04Z

drivers/vdpa/virtio/virtio_vdpa.c

+	.get_mi_by_bdf = NULL,
+};
+
+void


kailiangz1 · 2022-06-08T01:35:36Z

drivers/vdpa/virtio/virtio_vdpa.c

+	/* check pf_priv before use it, might be null if not set */
+	priv->pf_priv = pf_priv;
+	if (!priv->pf_priv)
+		DRV_LOG(WARNING, "PF was not set");


return error if no pf
DRV_LOG(ERR, "dev:%s no pf driver found",
devname);
rte_errno = EINVAL;
return -rte_errno;

kailiangz1 · 2022-06-08T01:40:31Z

drivers/vdpa/virtio_mi/lm.c

+	return 0;
+}
+
+#define VIRTIO_ARG_VDPA 	  "vdpa"


we can move this marco to virtio_api.h, both pf driver and vpda driver use this parameter

kailiangz1 · 2022-06-08T01:41:56Z

drivers/vdpa/virtio_mi/lm.c

+{
+	char devname[RTE_DEV_NAME_MAX_LEN] = {0};
+	struct virtio_vdpa_pf_priv *priv = NULL;
+	int vdpa = 0, ret = 0;


just ret;
no need =0

kailiangz1 · 2022-06-08T01:45:56Z

drivers/vdpa/virtio_mi/lm.c

+	}
+
+	virtio_pci_dev_features_get(priv->vpdev, &priv->guest_features);
+	if (!(priv->guest_features & (VIRTIO_VDPA_MI_SUPPORTED_FEATURE))) {


guest_features should name device_features

kailiangz1 · 2022-06-08T01:48:09Z

drivers/vdpa/virtio_mi/lm.c

+		virtio_pci_dev_reset(priv->vpdev);
+
+		/* Tell the host we've noticed this device. */
+		virtio_pci_dev_set_status(priv->vpdev, VIRTIO_CONFIG_STATUS_ACK);


remove needn't set ack and driver status

kailiangz1 · 2022-06-08T02:05:03Z

drivers/vdpa/virtio_mi/lm.c

+	uint16_t i, queue_idx;
+	int ret;
+
+	hw->max_queue_pairs = priv->vpdev->common_cfg->num_queues / 2;


max_queue_pairs can be get from virtio_net_config, should not caculate like this

What about blk device?

kailiangz1 · 2022-06-08T08:24:17Z

drivers/vdpa/virtio_mi/lm.c

+	int ret;
+
+	hw->max_queue_pairs = virtio_pci_dev_nr_vq_get(priv->vpdev) / 2;
+	priv->hw_nr_virtqs = 1;


use callback function to net and blk. get admin queue indx.
then init admin queue

…ures These commands will be used to control the migration process and the dirty page tracking process by the parent device for its managed devices. For now, there is only one type of managed devices in the PCI transport - the VFs. Signed-off-by: Gavin Li <gavinl@nvidia.com>

Signed-off-by: Gavin Li <gavinl@nvidia.com>

Add a new vdpa PF driver for net devices Signed-off-by: Gavin Li <gavinl@nvidia.com>

Signed-off-by: Gavin Li <gavinl@nvidia.com>

Admin virtqueues are used to send administrative commands to manipulate various features of the device which would not easily map into the configuration space. Use of Admin virtqueues is negotiated by the VIRTIO_F_ADMIN_VQ feature bit. Admin queue will be initialized by PF driver if the virtio device support admin queue Signed-off-by: Gavin Li <gavinl@nvidia.com>

Expose adminq command sending functions. The message format will be like below: struct virtio_admin_cmd { /* Device-readable part */ u8 class; u8 command; u8 command-specific-data[]; /* Device-writable part */ u8 command-specific-result[]; u8 status; }; Signed-off-by: Gavin Li <gavinl@nvidia.com>

kailiangz1 · 2022-07-08T06:26:45Z

drivers/common/virtio_mi/lm.c

+	uint16_t hw_nr_virtqs; /* number of vq device supported*/
+};
+
+struct sge_iova {


align to spec
diff --git a/content.tex b/content.tex
index c1cb60c..d82d399 100644
--- a/content.tex
+++ b/content.tex
@@ -2116,6 +2116,7 @@ \subsubsection{Dirty page track ctrl}\label{sec:Virtio Transport Options / Virti
struct virtio_sge {
le64 addr;
le32 len;

le32 reserved;

};

kailiangz1 · 2022-07-08T11:18:07Z

drivers/common/virtio_mi/lm.c

+		enum virtio_internal_status status)
+{
+	struct virtio_hw *hw;
+


why not expose set status, suspend and resume had no different

kailiangz1 · 2022-07-08T11:20:02Z

drivers/common/virtio_mi/lm.c

+}
+
+int
+virtio_vdpa_cmd_suspend(struct virtio_vdpa_pf_priv *priv, uint16_t vdev_id,


we should also have get status API
VIRTIO_ADMIN_PCI_MIGRATION_GET_INTERNAL_STATUS

Signed-off-by: Gavin Li <gavinl@nvidia.com>

Add PRC APIs for adding/removing/listing PF Signed-off-by: Gavin Li <gavinl@nvidia.com>

Signed-off-by: Gavin Li <gavinl@nvidia.com>

When MSIX configured less then queue number, quit testpmd in VM, cause vDPA crash. Mellanox#3 0x00007fbc8421b489 in _int_free () from /lib64/libc.so.6 Mellanox#4 0x0000000001a471c5 in virtio_vdpa_virtq_doorbell_relay_disable (vq_idx=vq_idx@entry=11, priv=<optimized out>, priv=<optimized out>) at ../drivers/vdpa/virtio/virtio_vdpa.c:349 Mellanox#5 0x0000000001a47275 in virtio_vdpa_virtq_disable () at ../drivers/vdpa/virtio/virtio_vdpa.c:413 Mellanox#6 0x0000000001a47a5a in virtio_vdpa_vring_state_set () at ../drivers/vdpa/virtio/virtio_vdpa.c:588 Mellanox#7 0x00000000005ad8af in vhost_user_notify_queue_state (dev=0x17ffcd000, index=11, enable=0) at ../lib/vhost/vhost_user.c:283 Mellanox#8 0x00000000005b0414 in vhost_user_msg_handler (vid=<optimized out>, fd=<optimized out>) at ../lib/vhost/vhost_user.c:3164 Mellanox#9 0x00000000012f812f in vhost_user_read_cb () at ../lib/vhost/socket.c:310 When callfd == -1, virtio_pci_dev_interrupt_enable is skilled. But in virtio_vdpa_virtq_disable, no such check to skip virtio_pci_dev_interrupt_disable. virtio_vdpa_virtq_disable return error without changing queue state to disable. Double free is caused by this wrong queue state. The fix is to add/check vector_enable variable for virtio_pci_dev_interrupt_disable. And remove error return in virtio_vdpa_virtq_disable. RM: 3587409 Signed-off-by: Yajun Wu <yajunw@nvidia.com>

When MSIX configured less then queue number, quit testpmd in VM, cause vDPA crash. Mellanox#3 0x00007fbc8421b489 in _int_free () from /lib64/libc.so.6 Mellanox#4 0x0000000001a471c5 in virtio_vdpa_virtq_doorbell_relay_disable (vq_idx=vq_idx@entry=11, priv=<optimized out>, priv=<optimized out>) at ../drivers/vdpa/virtio/virtio_vdpa.c:349 Mellanox#5 0x0000000001a47275 in virtio_vdpa_virtq_disable () at ../drivers/vdpa/virtio/virtio_vdpa.c:413 Mellanox#6 0x0000000001a47a5a in virtio_vdpa_vring_state_set () at ../drivers/vdpa/virtio/virtio_vdpa.c:588 Mellanox#7 0x00000000005ad8af in vhost_user_notify_queue_state (dev=0x17ffcd000, index=11, enable=0) at ../lib/vhost/vhost_user.c:283 Mellanox#8 0x00000000005b0414 in vhost_user_msg_handler (vid=<optimized out>, fd=<optimized out>) at ../lib/vhost/vhost_user.c:3164 Mellanox#9 0x00000000012f812f in vhost_user_read_cb () at ../lib/vhost/socket.c:310 When callfd == -1, virtio_pci_dev_interrupt_enable is skipped. But in virtio_vdpa_virtq_disable, no such check to skip virtio_pci_dev_interrupt_disable. virtio_vdpa_virtq_disable return error without changing queue state to disable. Double free is caused by this wrong queue state. The fix is to add/check vector_enable variable for virtio_pci_dev_interrupt_disable. And remove error return in virtio_vdpa_virtq_disable. RM: 3587409 Signed-off-by: Yajun Wu <yajunw@nvidia.com>

When MSIX configured less then queue number, quit testpmd in VM, cause vDPA crash. #3 0x00007fbc8421b489 in _int_free () from /lib64/libc.so.6 #4 0x0000000001a471c5 in virtio_vdpa_virtq_doorbell_relay_disable (vq_idx=vq_idx@entry=11, priv=<optimized out>, priv=<optimized out>) at ../drivers/vdpa/virtio/virtio_vdpa.c:349 #5 0x0000000001a47275 in virtio_vdpa_virtq_disable () at ../drivers/vdpa/virtio/virtio_vdpa.c:413 #6 0x0000000001a47a5a in virtio_vdpa_vring_state_set () at ../drivers/vdpa/virtio/virtio_vdpa.c:588 #7 0x00000000005ad8af in vhost_user_notify_queue_state (dev=0x17ffcd000, index=11, enable=0) at ../lib/vhost/vhost_user.c:283 #8 0x00000000005b0414 in vhost_user_msg_handler (vid=<optimized out>, fd=<optimized out>) at ../lib/vhost/vhost_user.c:3164 #9 0x00000000012f812f in vhost_user_read_cb () at ../lib/vhost/socket.c:310 When callfd == -1, virtio_pci_dev_interrupt_enable is skipped. But in virtio_vdpa_virtq_disable, no such check to skip virtio_pci_dev_interrupt_disable. virtio_vdpa_virtq_disable return error without changing queue state to disable. Double free is caused by this wrong queue state. The fix is to add/check vector_enable variable for virtio_pci_dev_interrupt_disable. And remove error return in virtio_vdpa_virtq_disable. RM: 3587409 Signed-off-by: Yajun Wu <yajunw@nvidia.com>

[ upstream commit 1c80a40 ] The net/vhost pmd currently provides a -1 vid when disabling interrupt after a virtio port got disconnected. This can be caught when running with ASan. First, start dpdk-l3fwd-power in interrupt mode with a net/vhost port. $ ./build-clang/examples/dpdk-l3fwd-power -l0,1 --in-memory \ -a 0000:00:00.0 \ --vdev net_vhost0,iface=plop.sock,client=1\ -- \ -p 0x1 \ --interrupt-only \ --config '(0,0,1)' \ --parse-ptype 0 Then start testpmd with virtio-user. $ ./build-clang/app/dpdk-testpmd -l0,2 --single-file-segment --in-memory \ -a 0000:00:00.0 \ --vdev net_virtio_user0,path=plop.sock,server=1 \ -- \ -i Finally stop testpmd. ASan then splats in dpdk-l3fwd-power: ================================================================= ==3641005==ERROR: AddressSanitizer: global-buffer-overflow on address 0x000005ed0778 at pc 0x000001270f81 bp 0x7fddbd2eee20 sp 0x7fddbd2eee18 READ of size 8 at 0x000005ed0778 thread T2 #0 0x1270f80 in get_device .../lib/vhost/vhost.h:801:27 Mellanox#1 0x1270f80 in rte_vhost_get_vhost_vring .../lib/vhost/vhost.c:951:8 Mellanox#2 0x3ac95cb in eth_rxq_intr_disable .../drivers/net/vhost/rte_eth_vhost.c:647:8 Mellanox#3 0x170e0bf in rte_eth_dev_rx_intr_disable .../lib/ethdev/rte_ethdev.c:5443:25 Mellanox#4 0xf72ba7 in turn_on_off_intr .../examples/l3fwd-power/main.c:881:4 Mellanox#5 0xf71045 in main_intr_loop .../examples/l3fwd-power/main.c:1061:6 Mellanox#6 0x17f9292 in eal_thread_loop .../lib/eal/common/eal_common_thread.c:210:9 Mellanox#7 0x18373f5 in eal_worker_thread_loop .../lib/eal/linux/eal.c:915:2 Mellanox#8 0x7fddc16ae12c in start_thread (/lib64/libc.so.6+0x8b12c) (BuildId: 81daba31ee66dbd63efdc4252a872949d874d136) Mellanox#9 0x7fddc172fbbf in __GI___clone3 (/lib64/libc.so.6+0x10cbbf) (BuildId: 81daba31ee66dbd63efdc4252a872949d874d136) 0x000005ed0778 is located 8 bytes to the left of global variable 'vhost_devices' defined in '.../lib/vhost/vhost.c:24' (0x5ed0780) of size 8192 0x000005ed0778 is located 20 bytes to the right of global variable 'vhost_config_log_level' defined in '.../lib/vhost/vhost.c:2174' (0x5ed0760) of size 4 SUMMARY: AddressSanitizer: global-buffer-overflow .../lib/vhost/vhost.h:801:27 in get_device Shadow bytes around the buggy address: 0x000080bd2090: f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 0x000080bd20a0: f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 f9 0x000080bd20b0: f9 f9 f9 f9 00 f9 f9 f9 00 f9 f9 f9 00 f9 f9 f9 0x000080bd20c0: 00 00 00 00 00 00 00 f9 f9 f9 f9 f9 04 f9 f9 f9 0x000080bd20d0: 00 00 00 00 00 f9 f9 f9 f9 f9 f9 f9 00 00 00 00 =>0x000080bd20e0: 00 f9 f9 f9 f9 f9 f9 f9 04 f9 f9 f9 04 f9 f9[f9] 0x000080bd20f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x000080bd2100: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x000080bd2110: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x000080bd2120: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x000080bd2130: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Shadow byte legend (one shadow byte represents 8 application bytes): Addressable: 00 Partially addressable: 01 02 03 04 05 06 07 Heap left redzone: fa Freed heap region: fd Stack left redzone: f1 Stack mid redzone: f2 Stack right redzone: f3 Stack after return: f5 Stack use after scope: f8 Global redzone: f9 Global init order: f6 Poisoned by user: f7 Container overflow: fc Array cookie: ac Intra object redzone: bb ASan internal: fe Left alloca redzone: ca Right alloca redzone: cb Thread T2 created by T0 here: #0 0xe98996 in __interceptor_pthread_create (.examples/dpdk-l3fwd-power+0xe98996) (BuildId: d0b984a3b0287b9e0f301b73426fa921aeecca3a) Mellanox#1 0x1836767 in eal_worker_thread_create .../lib/eal/linux/eal.c:952:6 Mellanox#2 0x1834b83 in rte_eal_init .../lib/eal/linux/eal.c:1257:9 Mellanox#3 0xf68902 in main .../examples/l3fwd-power/main.c:2496:8 Mellanox#4 0x7fddc164a50f in __libc_start_call_main (/lib64/libc.so.6+0x2750f) (BuildId: 81daba31ee66dbd63efdc4252a872949d874d136) ==3641005==ABORTING More generally, any application passing an incorrect vid would trigger such an OOB access. Fixes: 4796ad6 ("examples/vhost: import userspace vhost application") Signed-off-by: David Marchand <david.marchand@redhat.com> Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>

GavinLi-NV force-pushed the main branch 3 times, most recently from 5da14fa to b1a2687 Compare May 31, 2022 01:37

kailiangz1 reviewed Jun 8, 2022

View reviewed changes

kailiangz1 requested changes Jun 8, 2022

View reviewed changes

GavinLi-NV force-pushed the main branch 3 times, most recently from 82edfc4 to 60ffbc7 Compare June 8, 2022 07:51

kailiangz1 reviewed Jun 8, 2022

View reviewed changes

GavinLi-NV force-pushed the main branch 7 times, most recently from d6a6a52 to b3d2ff3 Compare June 15, 2022 08:00

GavinLi-NV force-pushed the main branch from b3d2ff3 to 79bba55 Compare June 22, 2022 09:12

GavinLi-NV force-pushed the main branch 5 times, most recently from 188c2a6 to 3e3059a Compare July 7, 2022 03:11

GavinLi-NV added 7 commits July 7, 2022 09:34

common/virtio: Add admin queue control in virtio hw

d7f2c67

Signed-off-by: Gavin Li <gavinl@nvidia.com>

net/virtio: Support more VDPA driver types

987f50a

Signed-off-by: Gavin Li <gavinl@nvidia.com>

vdpa/virtio_mi: Add virtio vdpa pf device driver for net devices

0abcb85

Add a new vdpa PF driver for net devices Signed-off-by: Gavin Li <gavinl@nvidia.com>

common/virtio: Expose get_queue_size

7356001

Signed-off-by: Gavin Li <gavinl@nvidia.com>

GavinLi-NV force-pushed the main branch from d179325 to 61dcab1 Compare July 7, 2022 06:35

kailiangz1 reviewed Jul 8, 2022

View reviewed changes

GavinLi-NV force-pushed the main branch from 61dcab1 to d848951 Compare July 8, 2022 07:00

kailiangz1 reviewed Jul 8, 2022

View reviewed changes

GavinLi-NV added 3 commits July 13, 2022 06:23

common/virtio_mi: Live migration commands

20df6c8

Signed-off-by: Gavin Li <gavinl@nvidia.com>

common/virtio_mi: Add PRC APIs

763d2e7

Add PRC APIs for adding/removing/listing PF Signed-off-by: Gavin Li <gavinl@nvidia.com>

common/virtio_mi: vdpa blk device support

81c5efa

Signed-off-by: Gavin Li <gavinl@nvidia.com>

GavinLi-NV force-pushed the main branch from d848951 to 81c5efa Compare July 13, 2022 03:25

kailiangz1 approved these changes Jul 15, 2022

View reviewed changes

kailiangz1 merged commit a9dfb12 into Mellanox:main Jul 19, 2022

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add PF driver and LM commands #5

Add PF driver and LM commands #5

GavinLi-NV commented May 27, 2022

kailiangz1 Jun 8, 2022

GavinLi-NV Jun 8, 2022

kailiangz1 Jun 8, 2022

GavinLi-NV Jun 8, 2022

kailiangz1 Jun 8, 2022

GavinLi-NV Jun 8, 2022

kailiangz1 Jun 8, 2022

GavinLi-NV Jun 8, 2022

kailiangz1 Jun 8, 2022

GavinLi-NV Jun 8, 2022

kailiangz1 Jun 8, 2022

GavinLi-NV Jun 8, 2022

kailiangz1 Jun 8, 2022

GavinLi-NV Jun 8, 2022

kailiangz1 Jun 8, 2022

GavinLi-NV Jun 8, 2022

kailiangz1 Jun 8, 2022

GavinLi-NV Jun 8, 2022

GavinLi-NV Jun 8, 2022

kailiangz1 Jun 8, 2022

GavinLi-NV Jun 8, 2022

kailiangz1 Jul 8, 2022

kailiangz1 Jul 8, 2022

kailiangz1 Jul 8, 2022

Add PF driver and LM commands #5

Add PF driver and LM commands #5

Conversation

GavinLi-NV commented May 27, 2022

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment