From: Thomas Falcon <tlfalcon@linux.vnet.ibm.com>
To: netdev@vger.kernel.org
Cc: jallen@linux.vnet.ibm.com, nfont@linux.vnet.ibm.com,
tlfalcon@linux.vnet.ibm.com, wvoigt@us.ibm.com
Subject: [PATCH 1/2 net] ibmvnic: Fix overflowing firmware/hardware TX queue
Date: Sun, 5 Mar 2017 12:18:41 -0600 [thread overview]
Message-ID: <1488737922-17305-1-git-send-email-tlfalcon@linux.vnet.ibm.com> (raw)
Use a counter to track the number of outstanding transmissions sent
that have not received completions. If the counter reaches the maximum
number of queue entries, stop transmissions on that queue. As we receive
more completions from firmware, wake the queue once the counter reaches
an acceptable level.
This patch prevents hardware/firmware TX queue from filling up and
and generating errors. Since incorporating this fix, internal testing
has reported that these firmware errors have stopped.
Signed-off-by: Thomas Falcon <tlfalcon@linux.vnet.ibm.com>
---
drivers/net/ethernet/ibm/ibmvnic.c | 27 ++++++++++++++++++++++++++-
drivers/net/ethernet/ibm/ibmvnic.h | 1 +
2 files changed, 27 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c
index 9198e6b..0e87b83 100644
--- a/drivers/net/ethernet/ibm/ibmvnic.c
+++ b/drivers/net/ethernet/ibm/ibmvnic.c
@@ -705,6 +705,7 @@ static int ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev)
u8 *hdrs = (u8 *)&adapter->tx_rx_desc_req;
struct device *dev = &adapter->vdev->dev;
struct ibmvnic_tx_buff *tx_buff = NULL;
+ struct ibmvnic_sub_crq_queue *tx_scrq;
struct ibmvnic_tx_pool *tx_pool;
unsigned int tx_send_failed = 0;
unsigned int tx_map_failed = 0;
@@ -724,6 +725,7 @@ static int ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev)
int ret = 0;
tx_pool = &adapter->tx_pool[queue_num];
+ tx_scrq = adapter->tx_scrq[queue_num];
txq = netdev_get_tx_queue(netdev, skb_get_queue_mapping(skb));
handle_array = (u64 *)((u8 *)(adapter->login_rsp_buf) +
be32_to_cpu(adapter->login_rsp_buf->
@@ -826,6 +828,14 @@ static int ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev)
ret = NETDEV_TX_BUSY;
goto out;
}
+
+ atomic_inc(&tx_scrq->used);
+
+ if (atomic_read(&tx_scrq->used) >= adapter->req_tx_entries_per_subcrq) {
+ netdev_info(netdev, "Stopping queue %d\n", queue_num);
+ netif_stop_subqueue(netdev, queue_num);
+ }
+
tx_packets++;
tx_bytes += skb->len;
txq->trans_start = jiffies;
@@ -1213,6 +1223,7 @@ static struct ibmvnic_sub_crq_queue *init_sub_crq_queue(struct ibmvnic_adapter
scrq->adapter = adapter;
scrq->size = 4 * PAGE_SIZE / sizeof(*scrq->msgs);
scrq->cur = 0;
+ atomic_set(&scrq->used, 0);
scrq->rx_skb_top = NULL;
spin_lock_init(&scrq->lock);
@@ -1355,8 +1366,22 @@ static int ibmvnic_complete_tx(struct ibmvnic_adapter *adapter,
DMA_TO_DEVICE);
}
- if (txbuff->last_frag)
+ if (txbuff->last_frag) {
+ atomic_dec(&scrq->used);
+
+ if (atomic_read(&scrq->used) <=
+ (adapter->req_tx_entries_per_subcrq / 2) &&
+ netif_subqueue_stopped(adapter->netdev,
+ txbuff->skb)) {
+ netif_wake_subqueue(adapter->netdev,
+ scrq->pool_index);
+ netdev_dbg(adapter->netdev,
+ "Started queue %d\n",
+ scrq->pool_index);
+ }
+
dev_kfree_skb_any(txbuff->skb);
+ }
adapter->tx_pool[pool].free_map[adapter->tx_pool[pool].
producer_index] = index;
diff --git a/drivers/net/ethernet/ibm/ibmvnic.h b/drivers/net/ethernet/ibm/ibmvnic.h
index 422824f..1993b42 100644
--- a/drivers/net/ethernet/ibm/ibmvnic.h
+++ b/drivers/net/ethernet/ibm/ibmvnic.h
@@ -863,6 +863,7 @@ struct ibmvnic_sub_crq_queue {
spinlock_t lock;
struct sk_buff *rx_skb_top;
struct ibmvnic_adapter *adapter;
+ atomic_t used;
};
struct ibmvnic_long_term_buff {
--
1.8.3.1
next reply other threads:[~2017-03-05 18:18 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-05 18:18 Thomas Falcon [this message]
2017-03-05 18:18 ` [PATCH 2/2 net] ibmvnic: Allocate number of rx/tx buffers agreed on by firmware Thomas Falcon
2017-03-07 22:14 ` David Miller
2017-03-07 22:14 ` [PATCH 1/2 net] ibmvnic: Fix overflowing firmware/hardware TX queue David Miller
-- strict thread matches above, loose matches on Subject: below --
2017-03-05 18:10 Thomas Falcon
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=1488737922-17305-1-git-send-email-tlfalcon@linux.vnet.ibm.com \
--to=tlfalcon@linux.vnet.ibm.com \
--cc=jallen@linux.vnet.ibm.com \
--cc=netdev@vger.kernel.org \
--cc=nfont@linux.vnet.ibm.com \
--cc=wvoigt@us.ibm.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).