DPDK patches and discussions
 help / color / mirror / Atom feed
From: Somnath Kotur <somnath.kotur@broadcom.com>
To: dev@dpdk.org
Cc: ferruh.yigit@intel.com
Subject: [dpdk-dev] [PATCH 2/6] net/bnxt: fix to use first valid profile if lossy profile not found
Date: Tue, 17 Dec 2019 09:47:51 +0530	[thread overview]
Message-ID: <20191217041755.29232-3-somnath.kotur@broadcom.com> (raw)
In-Reply-To: <20191217041755.29232-1-somnath.kotur@broadcom.com>

In the case when CoS classification is disabled, driver was iterating
looking for only lossy profiles as that is what is expected to be used
for regular NIC operations. But in certain custom profiles, there were
no lossy profiles configured, only lossless profiles instead.
To handle such cases, it is better to fallback to using the first valid
profile.

Fixes: 698aa7e95 ("net/bnxt: add code to determine the Tx COS queue")

Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Ajit Kumar Khaparde <ajit.khaparde@broadcom.com>
---
 drivers/net/bnxt/bnxt_hwrm.c | 44 ++++++++++++++++++++++++++++++++++++--------
 drivers/net/bnxt/bnxt_hwrm.h |  3 +++
 2 files changed, 39 insertions(+), 8 deletions(-)

diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index 64dc78a..694d2d0 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -1211,6 +1211,35 @@ static int bnxt_hwrm_port_phy_qcfg(struct bnxt *bp,
 	return rc;
 }
 
+static bool bnxt_find_lossy_profile(struct bnxt *bp)
+{
+	int i = 0;
+
+	for (i = BNXT_COS_QUEUE_COUNT - 1; i >= 0; i--) {
+		if (bp->tx_cos_queue[i].profile ==
+		    HWRM_QUEUE_SERVICE_PROFILE_LOSSY) {
+			bp->tx_cosq_id[0] = bp->tx_cos_queue[i].id;
+			return true;
+		}
+	}
+	return false;
+}
+
+static void bnxt_find_first_valid_profile(struct bnxt *bp)
+{
+	int i = 0;
+
+	for (i = BNXT_COS_QUEUE_COUNT - 1; i >= 0; i--) {
+		if (bp->tx_cos_queue[i].profile !=
+		    HWRM_QUEUE_SERVICE_PROFILE_UNKNOWN &&
+		    bp->tx_cos_queue[i].id !=
+		    HWRM_QUEUE_SERVICE_PROFILE_UNKNOWN) {
+			bp->tx_cosq_id[0] = bp->tx_cos_queue[i].id;
+			break;
+		}
+	}
+}
+
 int bnxt_hwrm_queue_qportcfg(struct bnxt *bp)
 {
 	int rc = 0;
@@ -1270,14 +1299,13 @@ int bnxt_hwrm_queue_qportcfg(struct bnxt *bp)
 						bp->tx_cos_queue[i].id;
 			}
 		} else {
-			for (i = BNXT_COS_QUEUE_COUNT - 1; i >= 0; i--) {
-				if (bp->tx_cos_queue[i].profile ==
-					HWRM_QUEUE_SERVICE_PROFILE_LOSSY) {
-					bp->tx_cosq_id[0] =
-						bp->tx_cos_queue[i].id;
-					break;
-				}
-			}
+			/* When CoS classification is disabled, for normal NIC
+			 * operations, ideally we should look to use LOSSY.
+			 * If not found, fallback to the first valid profile
+			 */
+			if (!bnxt_find_lossy_profile(bp))
+				bnxt_find_first_valid_profile(bp);
+
 		}
 	}
 
diff --git a/drivers/net/bnxt/bnxt_hwrm.h b/drivers/net/bnxt/bnxt_hwrm.h
index abe5de9..d8d1360 100644
--- a/drivers/net/bnxt/bnxt_hwrm.h
+++ b/drivers/net/bnxt/bnxt_hwrm.h
@@ -35,6 +35,9 @@
 #define HWRM_QUEUE_SERVICE_PROFILE_LOSSY \
 	HWRM_QUEUE_QPORTCFG_OUTPUT_QUEUE_ID0_SERVICE_PROFILE_LOSSY
 
+#define HWRM_QUEUE_SERVICE_PROFILE_UNKNOWN \
+	HWRM_QUEUE_QPORTCFG_OUTPUT_QUEUE_ID0_SERVICE_PROFILE_UNKNOWN
+
 #define HWRM_FUNC_RESOURCE_QCAPS_OUTPUT_VF_RESV_STRATEGY_MINIMAL_STATIC \
 	HWRM_FUNC_RESOURCE_QCAPS_OUTPUT_VF_RESERVATION_STRATEGY_MINIMAL_STATIC
 #define HWRM_FUNC_RESOURCE_QCAPS_OUTPUT_VF_RESV_STRATEGY_MAXIMAL \
-- 
1.8.3.1


  parent reply	other threads:[~2019-12-17  4:18 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-12-17  4:17 [dpdk-dev] [PATCH 0/6] bnxt patchset Somnath Kotur
2019-12-17  4:17 ` [dpdk-dev] [PATCH 1/6] net/bnxt: fix link failure during port toggle by increasing link wait time Somnath Kotur
2019-12-17  4:17 ` Somnath Kotur [this message]
2019-12-17  4:17 ` [dpdk-dev] [PATCH 3/6] net/bnxt: fix flow flush to sync with flow destroy routine Somnath Kotur
2019-12-17  4:17 ` [dpdk-dev] [PATCH 4/6] net/bnxt: fix non matching flow hitting filter rule Somnath Kotur
2019-12-17  4:17 ` [dpdk-dev] [PATCH 5/6] net/bnxt: fix to keep the L2 filter intact so it can be reused Somnath Kotur
2019-12-17  4:17 ` [dpdk-dev] [PATCH 6/6] net/bnxt: fix to free l2 filters while clearing vnic flows/filters Somnath Kotur
2019-12-21  2:29 ` [dpdk-dev] [PATCH v2 0/5] bnxt patchset Ajit Khaparde
2019-12-21  2:50   ` Ajit Khaparde
2020-01-07  0:37   ` [dpdk-dev] [PATCH v3 0/7] " Ajit Khaparde
2020-01-07  0:37     ` [dpdk-dev] [PATCH v3 1/7] net/bnxt: fix link failure during port toggle Ajit Khaparde
2020-01-07  0:37     ` [dpdk-dev] [PATCH v3 2/7] net/bnxt: fix to use first valid profile Ajit Khaparde
2020-01-07  0:37     ` [dpdk-dev] [PATCH v3 3/7] net/bnxt: fix flow flush to sync with flow destroy Ajit Khaparde
2020-01-07  0:37     ` [dpdk-dev] [PATCH v3 4/7] net/bnxt: fix non matching flow hitting filter rule Ajit Khaparde
2020-01-07  0:37     ` [dpdk-dev] [PATCH v3 5/7] net/bnxt: fix to reuse an L2 filter Ajit Khaparde
2020-01-07  0:37     ` [dpdk-dev] [PATCH v3 6/7] net/bnxt: add support for flow mark action Ajit Khaparde
2020-01-07  0:37     ` [dpdk-dev] [PATCH v3 7/7] net/bnxt: fix to not overwrite error message Ajit Khaparde
2020-01-07  9:18       ` Ferruh Yigit
2020-01-07 21:50         ` Ajit Khaparde
2020-01-07 23:04     ` [dpdk-dev] [PATCH v3 0/7] bnxt patchset Ajit Khaparde
2019-12-21  2:29 ` [dpdk-dev] [PATCH v2 1/5] net/bnxt: fix link failure during port toggle Ajit Khaparde
2019-12-21  2:29 ` [dpdk-dev] [PATCH v2 2/5] net/bnxt: fix to use first valid profile Ajit Khaparde
2019-12-21  2:29 ` [dpdk-dev] [PATCH v2 3/5] net/bnxt: fix flow flush to sync with flow destroy Ajit Khaparde
2019-12-21  2:29 ` [dpdk-dev] [PATCH v2 4/5] net/bnxt: fix non matching flow hitting filter rule Ajit Khaparde
2019-12-21  2:29 ` [dpdk-dev] [PATCH v2 5/5] net/bnxt: fix to reuse an L2 filter Ajit Khaparde

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=20191217041755.29232-3-somnath.kotur@broadcom.com \
    --to=somnath.kotur@broadcom.com \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@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).