ivampiresp/Spdk
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

nvmf/vfio-user: avoid doorbell reads in cq_is_full()

Browse Source 
Profiling data showed the deference of the CQ head in cq_is_full() was a
significant contributor to the CPU cost of post_completion(). Use the
cached ->last_head value instead of a doorbell read every time.

Signed-off-by: John Levon <john.levon@nutanix.com>
Change-Id: Ib8c92ce4fa79683950555d7b0c235449e457b844
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11848
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
This commit is contained in:
John Levon 2022-03-08 19:18:43 +00:00 committed by Tomasz Zawadzki
parent 14ecc7787d
commit a8326f8155
1 changed files with 18 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
lib/nvmf/vfio_user.c
View File

@ -509,6 +509,17 @@ cq_tail_advance(struct nvmf_vfio_user_cq *cq)
}
}
/*
* As per NVMe Base spec 3.3.1.2.1, we are supposed to implement CQ flow
* control: if there is no space in the CQ, we should wait until there is.
*
* In practice, we just fail the controller instead: as it happens, all host
* implementations we care about right-size the CQ: this is required anyway for
* NVMEoF support (see 3.3.2.8).
*
* Since reading the head doorbell is relatively expensive, we use the cached
* value, so we only have to read it for real if it appears that we are full.
*/
static inline bool
cq_is_full(struct nvmf_vfio_user_cq *cq)
{
@ -521,7 +532,13 @@ cq_is_full(struct nvmf_vfio_user_cq *cq)
qindex = 0;
}
return qindex == *cq_dbl_headp(cq);
if (qindex != cq->last_head) {
return false;
}
cq->last_head = *cq_dbl_headp(cq);
return qindex == cq->last_head;
}
static bool