public inbox for linux-rdma@vger.kernel.org
 help / color / mirror / Atom feed
From: Salil Mehta <salil.mehta@huawei.com>
To: dledford@redhat.com
Cc: salil.mehta@huawei.com, xavier.huwei@huawei.com,
	oulijun@huawei.com, yisen.zhuang@huawei.com,
	mehta.salil.lnk@gmail.com, linux-rdma@vger.kernel.org,
	linux-kernel@vger.kernel.org, linuxarm@huawei.com,
	lipeng <lipeng321@huawei.com>
Subject: [PATCH for-next 1/8] net: hns: fix port unavailable after hnae_reserve_buffer_map fail
Date: Sat, 10 Sep 2016 05:09:23 +0100	[thread overview]
Message-ID: <20160910040930.25988-2-salil.mehta@huawei.com> (raw)
In-Reply-To: <20160910040930.25988-1-salil.mehta@huawei.com>

From: lipeng <lipeng321@huawei.com>

When hnae_reserve_buffer_map fail, it will break cycle and some
buffer description has no available memory, therefore the port will
be unavailable.

Signed-off-by: Peng Li <lipeng321@huawei.com>
Reviewed-by: Yisen Zhuang <yisen.zhuang@huawei.com>
Signed-off-by: Salil Mehta <salil.mehta@huawei.com>
---
 drivers/net/ethernet/hisilicon/hns/hns_enet.c | 23 ++++++++++++++++++-----
 1 file changed, 18 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
index d7e1f8c..32ff270 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
@@ -747,6 +747,14 @@ static void hns_nic_rx_up_pro(struct hns_nic_ring_data *ring_data,
 	ndev->last_rx = jiffies;
 }
 
+static int hns_desc_unused(struct hnae_ring *ring)
+{
+	int ntc = ring->next_to_clean;
+	int ntu = ring->next_to_use;
+
+	return ((ntc >= ntu) ? 0 : ring->desc_num) + ntc - ntu;
+}
+
 static int hns_nic_rx_poll_one(struct hns_nic_ring_data *ring_data,
 			       int budget, void *v)
 {
@@ -755,17 +763,21 @@ static int hns_nic_rx_poll_one(struct hns_nic_ring_data *ring_data,
 	int num, bnum, ex_num;
 #define RCB_NOF_ALLOC_RX_BUFF_ONCE 16
 	int recv_pkts, recv_bds, clean_count, err;
+	int unused_count = hns_desc_unused(ring);
 
 	num = readl_relaxed(ring->io_base + RCB_REG_FBDNUM);
 	rmb(); /* make sure num taken effect before the other data is touched */
 
 	recv_pkts = 0, recv_bds = 0, clean_count = 0;
+	num -= unused_count;
 recv:
 	while (recv_pkts < budget && recv_bds < num) {
 		/* reuse or realloc buffers */
-		if (clean_count >= RCB_NOF_ALLOC_RX_BUFF_ONCE) {
-			hns_nic_alloc_rx_buffers(ring_data, clean_count);
+		if (clean_count + unused_count >= RCB_NOF_ALLOC_RX_BUFF_ONCE) {
+			hns_nic_alloc_rx_buffers(ring_data,
+						 clean_count + unused_count);
 			clean_count = 0;
+			unused_count = hns_desc_unused(ring);
 		}
 
 		/* poll one pkt */
@@ -789,7 +801,7 @@ recv:
 	/* make all data has been write before submit */
 	if (recv_pkts < budget) {
 		ex_num = readl_relaxed(ring->io_base + RCB_REG_FBDNUM);
-
+		ex_num -= unused_count;
 		if (ex_num > clean_count) {
 			num += ex_num - clean_count;
 			rmb(); /*complete read rx ring bd number*/
@@ -799,8 +811,9 @@ recv:
 
 out:
 	/* make all data has been write before submit */
-	if (clean_count > 0)
-		hns_nic_alloc_rx_buffers(ring_data, clean_count);
+	if (clean_count + unused_count > 0)
+		hns_nic_alloc_rx_buffers(ring_data,
+					 clean_count + unused_count);
 
 	return recv_pkts;
 }
-- 
1.9.1

  reply	other threads:[~2016-09-10  4:09 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-09-10  4:09 [PATCH for-next 0/8] Bug Fixes and Code Improvement in HNS driver Salil Mehta
2016-09-10  4:09 ` Salil Mehta [this message]
2016-09-10  4:09 ` [PATCH for-next 2/8] net: hns: bug fix about setting coalsecs-usecs to 0 Salil Mehta
2016-09-10  4:09 ` [PATCH for-next 3/8] net: hns: add fini_process for v2 napi process Salil Mehta
     [not found] ` <20160910040930.25988-1-salil.mehta-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>
2016-09-10  4:09   ` [PATCH for-next 4/8] net: hns: delete repeat read fbd num after while Salil Mehta
2016-09-10  4:09   ` [PATCH for-next 8/8] net: hns: delete redundant broadcast packet filter process Salil Mehta
2016-09-28 14:57   ` [PATCH for-next 0/8] Bug Fixes and Code Improvement in HNS driver Doug Ledford
     [not found]     ` <57EBDA43.3010708-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2016-09-28 16:06       ` Salil Mehta
2016-09-29 17:16       ` Salil Mehta
2016-09-10  4:09 ` [PATCH for-next 5/8] net: hns: fix port not available after testing loopback Salil Mehta
2016-09-10  4:09 ` [PATCH for-next 6/8] net: hns: fix the bug of forwarding table Salil Mehta
2016-09-10  4:09 ` [PATCH for-next 7/8] net: hns: bug fix about broadcast/multicast packets Salil Mehta

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=20160910040930.25988-2-salil.mehta@huawei.com \
    --to=salil.mehta@huawei.com \
    --cc=dledford@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=linuxarm@huawei.com \
    --cc=lipeng321@huawei.com \
    --cc=mehta.salil.lnk@gmail.com \
    --cc=oulijun@huawei.com \
    --cc=xavier.huwei@huawei.com \
    --cc=yisen.zhuang@huawei.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