d/numam-spdk
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

vhost: restart session after KICK socket message

Browse Source 
Change-Id: I13fbfcde617a963503a479769374cf8dba5fa278
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/4024
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This commit is contained in:
Changpeng Liu 2020-09-02 13:43:45 +08:00 committed by Tomasz Zawadzki
parent 7b16ba29fa
commit d653a65eaf
1 changed files with 6 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
lib/vhost/rte_vhost_compat.c
View File

@ -175,7 +175,6 @@ extern_vhost_pre_msg_handler(int vid, void *_msg)
case VHOST_USER_SET_VRING_BASE:
case VHOST_USER_SET_VRING_ADDR:
case VHOST_USER_SET_VRING_NUM:
case VHOST_USER_SET_VRING_KICK:
if (vsession->forced_polling && vsession->started) {
/* Additional queues are being initialized, so we either processed
* enough I/Os and are switching from SeaBIOS to the OS now, or
@ -186,6 +185,12 @@ extern_vhost_pre_msg_handler(int vid, void *_msg)
vsession->forced_polling = false;
}
break;
case VHOST_USER_SET_VRING_KICK:
/* rte_vhost(after 20.08) will call new_device after one active vring is
* configured, we will start the session before all vrings are available,
* so for each new vring, if the session is started, we need to restart it
* again.
*/
case VHOST_USER_SET_VRING_CALL:
/* rte_vhost will close the previous callfd and won't notify
* us about any change. This will effectively make SPDK fail