From: Bart Van Assche <bart.vanassche@sandisk.com>
To: Christoph Hellwig <hch@lst.de>, <linux-rdma@vger.kernel.org>
Cc: <sagig@dev.mellanox.co.il>, <bart.vanassche@sandisk.com>,
<axboe@fb.com>, <linux-scsi@vger.kernel.org>,
<linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 08/13] IB/srpt: chain RDMA READ/WRITE requests
Date: Tue, 29 Dec 2015 10:58:24 +0100 [thread overview]
Message-ID: <56825940.5070404@sandisk.com> (raw)
In-Reply-To: <1449521512-22921-9-git-send-email-hch@lst.de>
On 12/07/2015 09:51 PM, Christoph Hellwig wrote:
> Remove struct rdma_iu and instead allocate the struct ib_rdma_wr array
> early and fill out directly. This allows us to chain the WRs, and thus
> archive both less lock contention on the HCA workqueue as well as much
> simpler error handling.
Please consider folding the patch below into this patch.
Thanks,
Bart.
[PATCH] IB/srpt: Fix a recently introduced kernel crash
BUG: unable to handle kernel paging request at 0000000100000198
IP: [<ffffffff810a4ea2>] __lock_acquire+0xa2/0x560
Call Trace:
[<ffffffff810a53c2>] lock_acquire+0x62/0x80
[<ffffffff8151bd33>] _raw_spin_lock_irqsave+0x43/0x60
[<ffffffffa04fd437>] srpt_rdma_read_done+0x57/0x120 [ib_srpt]
[<ffffffffa0144dd3>] __ib_process_cq+0x43/0xc0 [ib_core]
[<ffffffffa0145115>] ib_cq_poll_work+0x25/0x70 [ib_core]
[<ffffffff8107184d>] process_one_work+0x1bd/0x460
[<ffffffff81073148>] worker_thread+0x118/0x420
[<ffffffff81078454>] kthread+0xe4/0x100
[<ffffffff8151cbbf>] ret_from_fork+0x3f/0x70
---
drivers/infiniband/ulp/srpt/ib_srpt.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c
index 8068aff..3daab39 100644
--- a/drivers/infiniband/ulp/srpt/ib_srpt.c
+++ b/drivers/infiniband/ulp/srpt/ib_srpt.c
@@ -1395,7 +1395,7 @@ static void srpt_rdma_read_done(struct ib_cq *cq, struct ib_wc *wc)
{
struct srpt_rdma_ch *ch = cq->cq_context;
struct srpt_send_ioctx *ioctx =
- container_of(wc->wr_cqe, struct srpt_send_ioctx, ioctx.cqe);
+ container_of(wc->wr_cqe, struct srpt_send_ioctx, rdma_cqe);
WARN_ON(ioctx->n_rdma <= 0);
atomic_add(ioctx->n_rdma, &ch->sq_wr_avail);
@@ -1418,7 +1418,7 @@ static void srpt_rdma_read_done(struct ib_cq *cq, struct ib_wc *wc)
static void srpt_rdma_write_done(struct ib_cq *cq, struct ib_wc *wc)
{
struct srpt_send_ioctx *ioctx =
- container_of(wc->wr_cqe, struct srpt_send_ioctx, ioctx.cqe);
+ container_of(wc->wr_cqe, struct srpt_send_ioctx, rdma_cqe);
if (unlikely(wc->status != IB_WC_SUCCESS)) {
pr_info("RDMA_WRITE for ioctx 0x%p failed with status %d\n",
--
2.1.4
next prev parent reply other threads:[~2015-12-29 9:58 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-12-07 20:51 completion queue abstraction V2 Christoph Hellwig
2015-12-07 20:51 ` [PATCH 01/13] irq_poll: make blk-iopoll available outside the block layer Christoph Hellwig
2015-12-10 18:41 ` Bart Van Assche
2015-12-07 20:51 ` [PATCH 02/13] irq_poll: don't disable new irq_poll instances Christoph Hellwig
2015-12-10 18:41 ` Bart Van Assche
2015-12-07 20:51 ` [PATCH 03/13] irq_poll: fold irq_poll_sched_prep into irq_poll_sched Christoph Hellwig
2015-12-10 18:41 ` Bart Van Assche
2015-12-29 9:54 ` Bart Van Assche
2015-12-30 9:42 ` Christoph Hellwig
2016-01-20 7:02 ` Andrew Donnellan
2016-01-20 7:15 ` Andrew Donnellan
2015-12-07 20:51 ` [PATCH 04/13] irq_poll: fold irq_poll_disable_pending into irq_poll_softirq Christoph Hellwig
2015-12-10 18:41 ` Bart Van Assche
2015-12-07 20:51 ` [PATCH 05/13] irq_poll: mark __irq_poll_complete static Christoph Hellwig
2015-12-10 18:42 ` Bart Van Assche
2015-12-07 20:51 ` [PATCH 06/13] irq_poll: remove unused data and max fields Christoph Hellwig
2015-12-10 18:42 ` Bart Van Assche
2015-12-07 20:51 ` [PATCH 07/13] IB: add a proper completion queue abstraction Christoph Hellwig
2015-12-10 18:42 ` Bart Van Assche
2015-12-11 14:17 ` Christoph Hellwig
2016-01-15 13:54 ` Parav Pandit
2016-01-17 9:24 ` Sagi Grimberg
2016-01-17 11:01 ` Parav Pandit
2016-01-17 11:06 ` Sagi Grimberg
2016-01-17 11:09 ` Parav Pandit
2015-12-07 20:51 ` [PATCH 08/13] IB/srpt: chain RDMA READ/WRITE requests Christoph Hellwig
2015-12-10 18:42 ` Bart Van Assche
2015-12-29 9:58 ` Bart Van Assche [this message]
2015-12-30 9:43 ` Christoph Hellwig
2015-12-07 20:51 ` [PATCH 09/13] IB/srpt: use the new CQ API Christoph Hellwig
2015-12-10 18:42 ` Bart Van Assche
2015-12-07 20:51 ` [PATCH 10/13] IB/srp: " Christoph Hellwig
2015-12-10 18:42 ` Bart Van Assche
2015-12-11 14:22 ` Christoph Hellwig
2015-12-11 17:59 ` Doug Ledford
2015-12-12 8:08 ` Christoph Hellwig
2015-12-13 10:26 ` Sagi Grimberg
2015-12-14 16:26 ` Doug Ledford
2015-12-07 20:51 ` [PATCH 11/13] IB/iser: Use a dedicated descriptor for login Christoph Hellwig
2015-12-07 20:51 ` [PATCH 12/13] IB/iser: Use helper for container_of Christoph Hellwig
2015-12-07 20:51 ` [PATCH 13/13] IB/iser: Convert to CQ abstraction Christoph Hellwig
2015-12-13 10:25 ` completion queue abstraction V2 Sagi Grimberg
2015-12-23 19:44 ` Doug Ledford
2015-12-29 9:51 ` Bart Van Assche
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=56825940.5070404@sandisk.com \
--to=bart.vanassche@sandisk.com \
--cc=axboe@fb.com \
--cc=hch@lst.de \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=sagig@dev.mellanox.co.il \
/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).