From: Allen Hubbe <Allen.Hubbe@emc.com>
To: linux-ntb@googlegroups.com
Cc: linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org,
Jon Mason <jdmason@kudzu.us>, Dave Jiang <dave.jiang@intel.com>,
Allen Hubbe <Allen.Hubbe@emc.com>
Subject: [PATCH 11/16] ntb_transport: reset qp link stats on down
Date: Wed, 20 May 2015 11:41:12 -0400 [thread overview]
Message-ID: <574bb412dc6ae6b3bcb5dd0823bd76f3f18854dd.1432135632.git.Allen.Hubbe@emc.com> (raw)
In-Reply-To: <cover.1432135631.git.Allen.Hubbe@emc.com>
In-Reply-To: <cover.1432135631.git.Allen.Hubbe@emc.com>
Reset the link stats when the link goes down. In particular, the tx and
rx index and count must be reset, or else the tx side will be sending
packets to the rx side, where the rx side is not expecting them. Reset
all the stats, to be consistent.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
---
drivers/ntb/ntb_transport.c | 36 ++++++++++++++++++++++++++++--------
1 file changed, 28 insertions(+), 8 deletions(-)
diff --git a/drivers/ntb/ntb_transport.c b/drivers/ntb/ntb_transport.c
index 9b4e240..c76d8b9 100644
--- a/drivers/ntb/ntb_transport.c
+++ b/drivers/ntb/ntb_transport.c
@@ -647,18 +647,37 @@ static int ntb_set_mw(struct ntb_transport_ctx *nt, int num_mw,
return 0;
}
+static void ntb_qp_link_down_reset(struct ntb_transport_qp *qp)
+{
+ qp->link_is_up = false;
+
+ qp->tx_index = 0;
+ qp->rx_index = 0;
+ qp->rx_bytes = 0;
+ qp->rx_pkts = 0;
+ qp->rx_ring_empty = 0;
+ qp->rx_err_no_buf = 0;
+ qp->rx_err_oflow = 0;
+ qp->rx_err_ver = 0;
+ qp->rx_memcpy = 0;
+ qp->rx_async = 0;
+ qp->tx_bytes = 0;
+ qp->tx_pkts = 0;
+ qp->tx_ring_full = 0;
+ qp->tx_err_no_buf = 0;
+ qp->tx_memcpy = 0;
+ qp->tx_async = 0;
+}
+
static void ntb_qp_link_cleanup(struct ntb_transport_qp *qp)
{
struct ntb_transport_ctx *nt = qp->transport;
struct pci_dev *pdev = nt->ndev->pdev;
- if (qp->link_is_up) {
- cancel_delayed_work_sync(&qp->link_work);
- return;
- }
-
dev_info(&pdev->dev, "qp %d: Link Cleanup\n", qp->qp_num);
- qp->link_is_up = false;
+
+ cancel_delayed_work_sync(&qp->link_work);
+ ntb_qp_link_down_reset(qp);
if (qp->event_handler)
qp->event_handler(qp->cb_data, qp->link_is_up);
@@ -865,9 +884,9 @@ static int ntb_transport_init_queue(struct ntb_transport_ctx *nt,
qp->qp_num = qp_num;
qp->transport = nt;
qp->ndev = nt->ndev;
- qp->link_is_up = false;
qp->client_ready = false;
qp->event_handler = NULL;
+ ntb_qp_link_down_reset(qp);
if (qp_count % mw_count && mw_num + 1 < qp_count / mw_count)
num_qps_mw = qp_count / mw_count + 1;
@@ -1453,7 +1472,6 @@ static void ntb_send_link_down(struct ntb_transport_qp *qp)
if (!qp->link_is_up)
return;
- qp->link_is_up = false;
dev_info(&pdev->dev, "qp %d: Send Link Down\n", qp->qp_num);
for (i = 0; i < NTB_LINK_DOWN_TIMEOUT; i++) {
@@ -1475,6 +1493,8 @@ static void ntb_send_link_down(struct ntb_transport_qp *qp)
if (rc)
dev_err(&pdev->dev, "ntb: QP%d unable to send linkdown msg\n",
qp->qp_num);
+
+ ntb_qp_link_down_reset(qp);
}
/**
--
2.4.0.rc0.43.gcf8a8c6
next prev parent reply other threads:[~2015-05-20 20:44 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-05-20 15:41 [PATCH 00/16] ntb: NTB Abstraction Layer Allen Hubbe
2015-05-20 15:41 ` [PATCH 01/16] Move files in preparation for NTB Abstraction Allen Hubbe
2015-05-20 15:41 ` [PATCH 02/16] NTB Abstraction Layer Allen Hubbe
2015-05-21 8:51 ` Paul Bolle
2015-05-21 11:32 ` Hubbe, Allen
2015-05-20 15:41 ` [PATCH 03/16] ntb: Enable link training for RP mode in the driver probe Allen Hubbe
2015-05-20 21:21 ` Bjorn Helgaas
2015-05-20 21:46 ` Hubbe, Allen
2015-05-20 22:20 ` Bjorn Helgaas
2015-05-20 23:00 ` Hubbe, Allen
2015-05-21 12:06 ` Bjorn Helgaas
2015-05-21 12:49 ` Hubbe, Allen
2015-05-20 15:41 ` [PATCH 04/16] Check the DID for certain workaround error flags to be set Allen Hubbe
2015-05-20 21:11 ` Bjorn Helgaas
2015-05-20 21:15 ` Jiang, Dave
2015-05-20 21:26 ` Hubbe, Allen
2015-05-20 15:41 ` [PATCH 05/16] Intel NTB params for snb b2b addresses Allen Hubbe
2015-05-20 15:41 ` [PATCH 06/16] NTB Pingpong Client Allen Hubbe
2015-05-21 8:54 ` Paul Bolle
2015-05-20 15:41 ` [PATCH 07/16] NTB Tool Client Allen Hubbe
2015-05-21 9:02 ` Paul Bolle
2015-05-20 15:41 ` [PATCH 08/16] ntb_transport: rate limit ntb_qp_link_work Allen Hubbe
2015-05-20 15:41 ` [PATCH 09/16] ntb_transport: differentiate link down messages Allen Hubbe
2015-05-20 15:41 ` [PATCH 10/16] ntb_transport: don't advance rx on link down Allen Hubbe
2015-05-20 15:41 ` Allen Hubbe [this message]
2015-05-20 15:41 ` [PATCH 12/16] ntb_transport: numa aware memory and dma chan Allen Hubbe
2015-05-20 15:41 ` [PATCH 13/16] ntb_hw_intel: numa aware memory allocation Allen Hubbe
2015-05-20 15:41 ` [PATCH 14/16] ntb: performance improvement by write combining Allen Hubbe
2015-05-20 15:41 ` [PATCH 15/16] ntb: default to cpu memcpy for performance Allen Hubbe
2015-05-20 15:41 ` [PATCH 16/16] ntb_transport: fix small code format issues Allen Hubbe
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=574bb412dc6ae6b3bcb5dd0823bd76f3f18854dd.1432135632.git.Allen.Hubbe@emc.com \
--to=allen.hubbe@emc.com \
--cc=dave.jiang@intel.com \
--cc=jdmason@kudzu.us \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-ntb@googlegroups.com \
--cc=linux-pci@vger.kernel.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
all inboxes | Powered by JetHome®