From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Eli Cohen <eli@mellanox.com>,
Leon Romanovsky <leon@kernel.org>,
Doug Ledford <dledford@redhat.com>
Subject: [PATCH 3.14 35/46] IB/mlx5: Fix post send fence logic
Date: Thu, 18 Aug 2016 15:54:57 +0200 [thread overview]
Message-ID: <20160818135447.600782512@linuxfoundation.org> (raw)
In-Reply-To: <20160818135442.457400364@linuxfoundation.org>
3.14-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eli Cohen <eli@mellanox.com>
commit c9b254955b9f8814966f5dabd34c39d0e0a2b437 upstream.
If the caller specified IB_SEND_FENCE in the send flags of the work
request and no previous work request stated that the successive one
should be fenced, the work request would be executed without a fence.
This could result in RDMA read or atomic operations failure due to a MR
being invalidated. Fix this by adding the mlx5 enumeration for fencing
RDMA/atomic operations and fix the logic to apply this.
Fixes: e126ba97dba9 ('mlx5: Add driver for Mellanox Connect-IB adapters')
Signed-off-by: Eli Cohen <eli@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/infiniband/hw/mlx5/qp.c | 7 ++++---
include/linux/mlx5/qp.h | 1 +
2 files changed, 5 insertions(+), 3 deletions(-)
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -2037,10 +2037,11 @@ static u8 get_fence(u8 fence, struct ib_
return MLX5_FENCE_MODE_SMALL_AND_FENCE;
else
return fence;
-
- } else {
- return 0;
+ } else if (unlikely(wr->send_flags & IB_SEND_FENCE)) {
+ return MLX5_FENCE_MODE_FENCE;
}
+
+ return 0;
}
int mlx5_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
--- a/include/linux/mlx5/qp.h
+++ b/include/linux/mlx5/qp.h
@@ -137,6 +137,7 @@ enum {
enum {
MLX5_FENCE_MODE_NONE = 0 << 5,
MLX5_FENCE_MODE_INITIATOR_SMALL = 1 << 5,
+ MLX5_FENCE_MODE_FENCE = 2 << 5,
MLX5_FENCE_MODE_STRONG_ORDERING = 3 << 5,
MLX5_FENCE_MODE_SMALL_AND_FENCE = 4 << 5,
};
next prev parent reply other threads:[~2016-08-18 13:54 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CGME20160818135526uscas1p209dc90415100838cf7c73a2d19ca32a3@uscas1p2.samsung.com>
2016-08-18 13:54 ` [PATCH 3.14 00/46] 3.14.77-stable review Greg Kroah-Hartman
2016-08-18 13:54 ` [PATCH 3.14 01/46] tcp: make challenge acks faster Greg Kroah-Hartman
2016-08-18 13:54 ` [PATCH 3.14 02/46] usb: renesas_usbhs: protect the CFIFOSEL setting in usbhsg_ep_enable() Greg Kroah-Hartman
2016-08-18 13:54 ` [PATCH 3.14 11/46] cifs: Check for existing directory when opening file with O_CREAT Greg Kroah-Hartman
2016-08-18 13:54 ` [PATCH 3.14 12/46] cifs: fix crash due to race in hmac(md5) handling Greg Kroah-Hartman
2016-08-18 13:54 ` [PATCH 3.14 14/46] random: properly align get_random_int_hash Greg Kroah-Hartman
2016-08-19 3:14 ` Eric Biggers
2016-08-19 7:33 ` Greg Kroah-Hartman
2016-08-18 13:54 ` [PATCH 3.14 15/46] random: print a warning for the first ten uninitialized random users Greg Kroah-Hartman
2016-08-18 13:54 ` [PATCH 3.14 20/46] nfs: dont create zero-length requests Greg Kroah-Hartman
2016-08-18 13:54 ` [PATCH 3.14 30/46] balloon: check the number of available pages in leak balloon Greg Kroah-Hartman
2016-08-18 13:54 ` [PATCH 3.14 31/46] ftrace/recordmcount: Work around for addition of metag magic but not relocations Greg Kroah-Hartman
2016-08-18 13:54 ` [PATCH 3.14 32/46] metag: Fix __cmpxchg_u32 asm constraint for CMP Greg Kroah-Hartman
2016-08-18 13:54 ` [PATCH 3.14 33/46] IB/mlx5: Fix MODIFY_QP command input structure Greg Kroah-Hartman
2016-08-18 13:54 ` [PATCH 3.14 34/46] IB/mlx5: Fix returned values of query QP Greg Kroah-Hartman
2016-08-18 13:54 ` Greg Kroah-Hartman [this message]
2016-08-18 13:54 ` [PATCH 3.14 36/46] IB/IPoIB: Dont update neigh validity for unresolved entries Greg Kroah-Hartman
2016-08-18 13:54 ` [PATCH 3.14 37/46] IB/mlx4: Fix the SQ size of an RC QP Greg Kroah-Hartman
2016-08-18 13:55 ` [PATCH 3.14 38/46] ubi: Make volume resize power cut aware Greg Kroah-Hartman
2016-08-18 13:55 ` [PATCH 3.14 39/46] ubi: Fix race condition between ubi device creation and udev Greg Kroah-Hartman
2016-08-18 13:55 ` [PATCH 3.14 40/46] target: Fix race between iscsi-target connection shutdown + ABORT_TASK Greg Kroah-Hartman
2016-08-18 13:55 ` [PATCH 3.14 41/46] target: Fix max_unmap_lba_count calc overflow Greg Kroah-Hartman
2016-08-18 13:55 ` [PATCH 3.14 42/46] Input: i8042 - break load dependency between atkbd/psmouse and i8042 Greg Kroah-Hartman
2016-08-18 13:55 ` [PATCH 3.14 43/46] PCI: Mark Atheros AR9485 and QCA9882 to avoid bus reset Greg Kroah-Hartman
2016-08-18 13:55 ` [PATCH 3.14 44/46] dm flakey: error READ bios during the down_interval Greg Kroah-Hartman
2016-08-18 13:55 ` [PATCH 3.14 45/46] module: Invalidate signatures on force-loaded modules Greg Kroah-Hartman
2016-08-18 13:55 ` [PATCH 3.14 46/46] Documentation/module-signing.txt: Note need for version info if reusing a key Greg Kroah-Hartman
2016-08-18 20:05 ` [PATCH 3.14 00/46] 3.14.77-stable review Guenter Roeck
2016-08-19 7:38 ` Greg Kroah-Hartman
2016-08-18 21:34 ` Shuah Khan
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=20160818135447.600782512@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=dledford@redhat.com \
--cc=eli@mellanox.com \
--cc=leon@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=stable@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).