[PATCH] nvmet: don't check iosqes,iocqes for discovery controllers

[Sagi Grimberg]

>From the base spec, Figure 78:
"Controller Configuration, these fields are defined as parameters to
configure an "I/O Controller (IOC)" and not to configure a "Discovery
Controller (DC)".
...
"If the controller does not support I/O queues, then this field shall
be read-only with a value of 0h"

Just have this check for I/O controllers.

Fixes: a07b4970f464 ("nvmet: add a generic NVMe target")
Reported-by: Belanger, Martin <Martin.Belanger at dell.com>
Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
---
 drivers/nvme/target/core.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c
index be6fcdaf51a7..a027433b8be8 100644
--- a/drivers/nvme/target/core.c
+++ b/drivers/nvme/target/core.c
@@ -1118,9 +1118,20 @@ static void nvmet_start_ctrl(struct nvmet_ctrl *ctrl)
 {
 	lockdep_assert_held(&ctrl->lock);
 
-	if (nvmet_cc_iosqes(ctrl->cc) != NVME_NVM_IOSQES ||
-	    nvmet_cc_iocqes(ctrl->cc) != NVME_NVM_IOCQES ||
-	    nvmet_cc_mps(ctrl->cc) != 0 ||
+	/*
+	 * Only I/O controllers should verify iosqes,iocqes.
+	 * Strictly speaking, the spec says a discovery controller
+	 * should verify iosqes,iocqes are zeroed, however that
+	 * would break backwards compatibility, so don't enforce it.
+	 */
+	if (ctrl->subsys->type != NVME_NQN_DISC &&
+	    (nvmet_cc_iosqes(ctrl->cc) != NVME_NVM_IOSQES ||
+	     nvmet_cc_iocqes(ctrl->cc) != NVME_NVM_IOCQES)) {
+		ctrl->csts = NVME_CSTS_CFS;
+		return;
+	}
+
+	if (nvmet_cc_mps(ctrl->cc) != 0 ||
 	    nvmet_cc_ams(ctrl->cc) != 0 ||
 	    nvmet_cc_css(ctrl->cc) != 0) {
 		ctrl->csts = NVME_CSTS_CFS;
-- 
2.27.0