From: dennis.dalessandro@cornelisnetworks.com
To: dledford@redhat.com, jgg@ziepe.ca
Cc: linux-rdma@vger.kernel.org,
Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>,
Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Subject: [PATCH for-next 03/10] IB/hfi1: Correct oversized ring allocation
Date: Mon, 29 Mar 2021 09:54:09 -0400 [thread overview]
Message-ID: <1617026056-50483-4-git-send-email-dennis.dalessandro@cornelisnetworks.com> (raw)
In-Reply-To: <1617026056-50483-1-git-send-email-dennis.dalessandro@cornelisnetworks.com>
From: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
The completion ring for tx is using the wrong size to size the
ring, oversizing the ring by two orders of magniture.
Correct the allocation size and use kcalloc_node() to allocate
the ring. Fix mistaken GFP defines in similar allocations.
Reviewed-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
---
drivers/infiniband/hw/hfi1/ipoib.h | 3 ++-
drivers/infiniband/hw/hfi1/ipoib_tx.c | 14 +++++++-------
2 files changed, 9 insertions(+), 8 deletions(-)
diff --git a/drivers/infiniband/hw/hfi1/ipoib.h b/drivers/infiniband/hw/hfi1/ipoib.h
index 1659beb..39c8909 100644
--- a/drivers/infiniband/hw/hfi1/ipoib.h
+++ b/drivers/infiniband/hw/hfi1/ipoib.h
@@ -52,8 +52,9 @@
* @producer_lock: producer sync lock
* @consumer_lock: consumer sync lock
*/
+struct ipoib_txreq;
struct hfi1_ipoib_circ_buf {
- void **items;
+ struct ipoib_txreq **items;
unsigned long head;
unsigned long tail;
unsigned long max_items;
diff --git a/drivers/infiniband/hw/hfi1/ipoib_tx.c b/drivers/infiniband/hw/hfi1/ipoib_tx.c
index e8dece5..1c38c38 100644
--- a/drivers/infiniband/hw/hfi1/ipoib_tx.c
+++ b/drivers/infiniband/hw/hfi1/ipoib_tx.c
@@ -714,14 +714,14 @@ int hfi1_ipoib_txreq_init(struct hfi1_ipoib_dev_priv *priv)
priv->tx_napis = kcalloc_node(dev->num_tx_queues,
sizeof(struct napi_struct),
- GFP_ATOMIC,
+ GFP_KERNEL,
priv->dd->node);
if (!priv->tx_napis)
goto free_txreq_cache;
priv->txqs = kcalloc_node(dev->num_tx_queues,
sizeof(struct hfi1_ipoib_txq),
- GFP_ATOMIC,
+ GFP_KERNEL,
priv->dd->node);
if (!priv->txqs)
goto free_tx_napis;
@@ -753,9 +753,9 @@ int hfi1_ipoib_txreq_init(struct hfi1_ipoib_dev_priv *priv)
priv->dd->node);
txq->tx_ring.items =
- vzalloc_node(array_size(tx_ring_size,
- sizeof(struct ipoib_txreq)),
- priv->dd->node);
+ kcalloc_node(tx_ring_size,
+ sizeof(struct ipoib_txreq *),
+ GFP_KERNEL, priv->dd->node);
if (!txq->tx_ring.items)
goto free_txqs;
@@ -776,7 +776,7 @@ int hfi1_ipoib_txreq_init(struct hfi1_ipoib_dev_priv *priv)
struct hfi1_ipoib_txq *txq = &priv->txqs[i];
netif_napi_del(txq->napi);
- vfree(txq->tx_ring.items);
+ kfree(txq->tx_ring.items);
}
kfree(priv->txqs);
@@ -829,7 +829,7 @@ void hfi1_ipoib_txreq_deinit(struct hfi1_ipoib_dev_priv *priv)
hfi1_ipoib_drain_tx_list(txq);
netif_napi_del(txq->napi);
(void)hfi1_ipoib_drain_tx_ring(txq, txq->tx_ring.max_items);
- vfree(txq->tx_ring.items);
+ kfree(txq->tx_ring.items);
}
kfree(priv->txqs);
--
1.8.3.1
next prev parent reply other threads:[~2021-03-29 13:55 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-29 13:54 [PATCH for-next 00/10] Fixes for 5.13 dennis.dalessandro
2021-03-29 13:54 ` [PATCH for-next 01/10] IB/hfi1: Add AIP tx traces dennis.dalessandro
2021-03-29 13:54 ` [PATCH for-next 02/10] IB/{ipoib,hfi1}: Add a timeout handler for rdma_netdev dennis.dalessandro
2021-03-29 13:54 ` dennis.dalessandro [this message]
2021-03-29 13:54 ` [PATCH for-next 04/10] IB/hfi1: Use napi_schedule_irqoff() for tx napi dennis.dalessandro
2021-03-29 13:54 ` [PATCH for-next 05/10] IB/hfi1: Remove indirect call to hfi1_ipoib_send_dma() dennis.dalessandro
2021-03-29 13:54 ` [PATCH for-next 06/10] rdma: Set physical MTU for query_port function dennis.dalessandro
2021-04-01 8:42 ` Leon Romanovsky
2021-04-07 23:13 ` Jason Gunthorpe
2021-04-08 12:06 ` Dennis Dalessandro
2021-04-08 12:14 ` Leon Romanovsky
2021-04-08 12:31 ` Dennis Dalessandro
2021-04-19 12:20 ` Wan, Kaike
2021-04-19 12:29 ` Leon Romanovsky
2021-04-19 13:08 ` Dennis Dalessandro
2021-04-20 5:11 ` Leon Romanovsky
2021-04-20 12:17 ` Dennis Dalessandro
2021-04-20 12:31 ` Leon Romanovsky
2021-04-20 12:34 ` Dennis Dalessandro
2021-04-19 13:09 ` Wan, Kaike
2021-04-20 5:15 ` Leon Romanovsky
2021-03-29 13:54 ` [PATCH for-next 07/10] IB/hfi1: Add additional usdma traces dennis.dalessandro
2021-03-29 13:54 ` [PATCH for-next 08/10] IB/hfi1: Use kzalloc() for mmu_rb_handler allocation dennis.dalessandro
2021-03-29 13:54 ` [PATCH for-next 09/10] IB/hfi1: Remove unused function dennis.dalessandro
2021-03-29 14:06 ` [PATCH for-next v2 " dennis.dalessandro
2021-03-29 13:54 ` [PATCH for-next 10/10] IB/hfi1: Rework AIP and VNIC dummy netdev usage dennis.dalessandro
2021-04-13 16:59 ` Jason Gunthorpe
2021-04-07 23:21 ` [PATCH for-next 00/10] Fixes for 5.13 Jason Gunthorpe
2021-04-08 12:19 ` Dennis Dalessandro
2021-04-08 12:20 ` Jason Gunthorpe
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=1617026056-50483-4-git-send-email-dennis.dalessandro@cornelisnetworks.com \
--to=dennis.dalessandro@cornelisnetworks.com \
--cc=dledford@redhat.com \
--cc=jgg@ziepe.ca \
--cc=linux-rdma@vger.kernel.org \
--cc=mike.marciniszyn@cornelisnetworks.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).