netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Lijun Pan <ljp@linux.ibm.com>
To: netdev@vger.kernel.org
Cc: Lijun Pan <ljp@linux.ibm.com>
Subject: [PATCH net 1/3] ibmvnic: rework to ensure SCRQ entry reads are properly ordered
Date: Thu, 21 Jan 2021 00:17:08 -0600	[thread overview]
Message-ID: <20210121061710.53217-2-ljp@linux.ibm.com> (raw)
In-Reply-To: <20210121061710.53217-1-ljp@linux.ibm.com>

Move the dma_rmb() between pending_scrq() and ibmvnic_next_scrq()
into the end of pending_scrq(), and explain why.
Explain in detail why the dma_rmb() is placed at the end of
ibmvnic_next_scrq().

Fixes: b71ec9522346 ("ibmvnic: Ensure that SCRQ entry reads are correctly ordered")
Signed-off-by: Lijun Pan <ljp@linux.ibm.com>
---
 drivers/net/ethernet/ibm/ibmvnic.c | 41 +++++++++++++++++-------------
 1 file changed, 24 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c
index 9778c83150f1..8e043683610f 100644
--- a/drivers/net/ethernet/ibm/ibmvnic.c
+++ b/drivers/net/ethernet/ibm/ibmvnic.c
@@ -2511,12 +2511,6 @@ static int ibmvnic_poll(struct napi_struct *napi, int budget)
 
 		if (!pending_scrq(adapter, rx_scrq))
 			break;
-		/* The queue entry at the current index is peeked at above
-		 * to determine that there is a valid descriptor awaiting
-		 * processing. We want to be sure that the current slot
-		 * holds a valid descriptor before reading its contents.
-		 */
-		dma_rmb();
 		next = ibmvnic_next_scrq(adapter, rx_scrq);
 		rx_buff =
 		    (struct ibmvnic_rx_buff *)be64_to_cpu(next->
@@ -3256,13 +3250,6 @@ static int ibmvnic_complete_tx(struct ibmvnic_adapter *adapter,
 		int total_bytes = 0;
 		int num_packets = 0;
 
-		/* The queue entry at the current index is peeked at above
-		 * to determine that there is a valid descriptor awaiting
-		 * processing. We want to be sure that the current slot
-		 * holds a valid descriptor before reading its contents.
-		 */
-		dma_rmb();
-
 		next = ibmvnic_next_scrq(adapter, scrq);
 		for (i = 0; i < next->tx_comp.num_comps; i++) {
 			if (next->tx_comp.rcs[i])
@@ -3636,11 +3623,25 @@ static int pending_scrq(struct ibmvnic_adapter *adapter,
 			struct ibmvnic_sub_crq_queue *scrq)
 {
 	union sub_crq *entry = &scrq->msgs[scrq->cur];
+	int rc;
 
 	if (entry->generic.first & IBMVNIC_CRQ_CMD_RSP)
-		return 1;
+		rc = 1;
 	else
-		return 0;
+		rc = 0;
+
+	/* Ensure that the entire SCRQ descriptor scrq->msgs
+	 * has been loaded before reading its contents.
+	 * This barrier makes sure this function's entry, esp.
+	 * entry->generic.first & IBMVNIC_CRQ_CMD_RSP
+	 * 1. is loaded before ibmvnic_next_scrq()'s
+	 * entry->generic.first & IBMVNIC_CRQ_CMD_RSP;
+	 * 2. OR is loaded before ibmvnic_poll()'s
+	 * disable_scrq_irq()'s scrq->hw_irq.
+	 */
+	dma_rmb();
+
+	return rc;
 }
 
 static union sub_crq *ibmvnic_next_scrq(struct ibmvnic_adapter *adapter,
@@ -3659,8 +3660,14 @@ static union sub_crq *ibmvnic_next_scrq(struct ibmvnic_adapter *adapter,
 	}
 	spin_unlock_irqrestore(&scrq->lock, flags);
 
-	/* Ensure that the entire buffer descriptor has been
-	 * loaded before reading its contents
+	/* Ensure that the entire SCRQ descriptor scrq->msgs
+	 * has been loaded before reading its contents.
+	 * This barrier makes sure this function's entry, esp.
+	 * entry->generic.first & IBMVNIC_CRQ_CMD_RSP
+	 * 1. is loaded before ibmvnic_poll()'s
+	 * be64_to_cpu(next->rx_comp.correlator);
+	 * 2. OR is loaded before ibmvnic_complet_tx()'s
+	 * be32_to_cpu(next->tx_comp.correlators[i]).
 	 */
 	dma_rmb();
 
-- 
2.23.0


  reply	other threads:[~2021-01-21  6:24 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-21  6:17 [PATCH net 0/3] fixes the memory barrier for SCRQ/CRQ entry Lijun Pan
2021-01-21  6:17 ` Lijun Pan [this message]
2021-01-24  5:08   ` [PATCH net 1/3] ibmvnic: rework to ensure SCRQ entry reads are properly ordered Jakub Kicinski
2021-01-25  4:34     ` Lijun Pan
2021-01-21  6:17 ` [PATCH net 2/3] ibmvnic: remove unnecessary rmb() inside ibmvnic_poll Lijun Pan
2021-01-24  5:09   ` Jakub Kicinski
2021-01-25  4:38     ` Lijun Pan
2021-01-25 20:35       ` Jakub Kicinski
2021-01-21  6:17 ` [PATCH net 3/3] ibmvnic: Ensure that CRQ entry read/write are correctly ordered Lijun Pan

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=20210121061710.53217-2-ljp@linux.ibm.com \
    --to=ljp@linux.ibm.com \
    --cc=netdev@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;
as well as URLs for NNTP newsgroup(s).