DPDK patches and discussions
 help / color / mirror / Atom feed
From: Wang Xiao W <xiao.w.wang@intel.com>
To: dev@dpdk.org
Cc: Wang Xiao W <xiao.w.wang@intel.com>
Subject: [dpdk-dev] [PATCH 17/28] fm10k: store actual count of DWORDS pulled/pushed from mbmem
Date: Thu, 10 Sep 2015 12:38:26 +0800	[thread overview]
Message-ID: <1441859917-26475-18-git-send-email-xiao.w.wang@intel.com> (raw)
In-Reply-To: <1441859917-26475-1-git-send-email-xiao.w.wang@intel.com>

A previous bug was uncovered by addition of a debug stat to indicate the
actual number of DWORDS we pulled from the mbmem. It turned out this was
not the same as the tx_dwords counter. While the previous bug fix should
have corrected this in all cases, add some debug stats that count the
number of DWORDs pushed or pulled from the mbmem. Base drivers can use
this in debug builds to help detect this problem in the future.

Signed-off-by: Wang Xiao W <xiao.w.wang@intel.com>
---
 drivers/net/fm10k/base/fm10k_mbx.c | 4 ++++
 drivers/net/fm10k/base/fm10k_mbx.h | 2 ++
 2 files changed, 6 insertions(+)

diff --git a/drivers/net/fm10k/base/fm10k_mbx.c b/drivers/net/fm10k/base/fm10k_mbx.c
index ad9d81a..c6d34eb 100644
--- a/drivers/net/fm10k/base/fm10k_mbx.c
+++ b/drivers/net/fm10k/base/fm10k_mbx.c
@@ -394,6 +394,8 @@ STATIC void fm10k_mbx_write_copy(struct fm10k_hw *hw,
 			if (!tail)
 				tail++;
 
+			mbx->tx_mbmem_pulled++;
+
 			/* write message to hardware FIFO */
 			FM10K_WRITE_MBX(hw, mbmem + tail++, *(head++));
 		} while (--len && --end);
@@ -480,6 +482,8 @@ STATIC void fm10k_mbx_read_copy(struct fm10k_hw *hw,
 			if (!head)
 				head++;
 
+			mbx->rx_mbmem_pushed++;
+
 			/* read message from hardware FIFO */
 			*(tail++) = FM10K_READ_MBX(hw, mbmem + head++);
 		} while (--len && --end);
diff --git a/drivers/net/fm10k/base/fm10k_mbx.h b/drivers/net/fm10k/base/fm10k_mbx.h
index 6332584..4b22f0e 100644
--- a/drivers/net/fm10k/base/fm10k_mbx.h
+++ b/drivers/net/fm10k/base/fm10k_mbx.h
@@ -313,8 +313,10 @@ struct fm10k_mbx_info {
 	u64 tx_dropped;
 	u64 tx_messages;
 	u64 tx_dwords;
+	u64 tx_mbmem_pulled;
 	u64 rx_messages;
 	u64 rx_dwords;
+	u64 rx_mbmem_pushed;
 	u64 rx_parse_err;
 
 	/* Buffer to store messages */
-- 
1.9.3

  parent reply	other threads:[~2015-09-10  4:39 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-09-10  4:38 [dpdk-dev] [PATCH 00/28] fm10k: update shared code from ND team Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 01/28] fm10k: add PF Tx Timestamp mode handler function Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 02/28] fm10k: add no-op pointer for VF request_tx_timestamp_mode Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 03/28] fm10k: Set PF queues to unlimited bandwidth Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 04/28] fm10k: fix fm10k_mbx_write_copy header comment Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 05/28] fm10k: Add support for ITR scaling based on PCIe link speed Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 06/28] fm10k: reset head instead of calling update_max_size Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 07/28] fm10k: mbx_update_max_size does not drop all Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 08/28] fm10k: ensure VF restores itr_scale on stop_hw Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 09/28] fm10k: ensure itr_scale is set even if we don't know speed Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 10/28] fm10k: correct VF multicast update Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 11/28] fm10k: Re-map all possible VF queues after a VFLR Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 12/28] fm10k: pack TLV overlay structures correctly Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 13/28] fm10k: 1558 DIR_NEGATIVE bit is actually DIR_POSITIVE Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 14/28] fm10k: remove err_no reference Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 15/28] fm10k: fix iov_msg_lport_state_pf re-enable bug Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 16/28] fm10k: add macro definitions about valid ether addr Wang Xiao W
2015-09-10  4:38 ` Wang Xiao W [this message]
2015-09-10  4:38 ` [dpdk-dev] [PATCH 18/28] fm10k: fix iov_msg_mac_vlan_pf VID checks Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 19/28] fm10k: Fix Solaris build issue Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 20/28] fm10k: fix Tx FIFO clearing for phantom messages Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 21/28] fm10k: create "correct" header for the remote end on connect Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 22/28] fm10k: do not assume VF always has 1 queue Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 23/28] fm10k: Add support for Boulder Rapids and Atwood Channel Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 24/28] fm10k: remove report_timestamp PF<->VF message Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 25/28] fm10k: remove request_tx_timestamp_mode call Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 26/28] fm10k: add 1588 clock owner message support Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 27/28] fm10k: TRIVIAL fix typo in DEBUGFUNC Wang Xiao W
2015-09-10  4:38 ` [dpdk-dev] [PATCH 28/28] fm10k: add support for MASTER_CLK_OFFSET message Wang Xiao W
2015-10-07 11:40 ` [dpdk-dev] [PATCH 00/28] fm10k: update shared code from ND team Thomas Monjalon

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=1441859917-26475-18-git-send-email-xiao.w.wang@intel.com \
    --to=xiao.w.wang@intel.com \
    --cc=dev@dpdk.org \
    /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).