From: Sagi Grimberg <sagi@grimberg.me>
To: Arnd Bergmann <arnd@arndb.de>
Cc: Binoy Jayan <binoy.jayan@linaro.org>,
Doug Ledford <dledford@redhat.com>,
Sean Hefty <sean.hefty@intel.com>,
Hal Rosenstock <hal.rosenstock@gmail.com>,
Leon Romanovsky <leonro@mellanox.com>,
Sagi Grimberg <sagig@mellanox.com>,
Bart Van Assche <bart.vanassche@sandisk.com>,
Nicholas Bellinger <nab@linux-iscsi.org>,
Jenny Derzhavetz <jennyf@mellanox.com>,
Ira Weiny <ira.weiny@intel.com>,
Steve Wise <swise@opengridcomputing.com>,
Mark Bloch <markb@mellanox.com>,
Tatyana E Nikolova <tatyana.e.nikolova@intel.com>,
Matan Barak <matanb@mellanox.com>, Lijun Ou <oulijun@huawei.com>,
"Wei Hu(Xavier)" <xavier.huwei@huawei.com>,
Faisal Latif <faisal.latif@intel.com>,
Mustafa Ismail <mustafa.ismail@intel.com>,
Mark Brown <broonie@kernel.org>,
linux-rdma@vger.kernel.org, linux-kernel@vger.kernel.org,
target-devel@vger.kernel.org
Subject: Re: [PATCH v5 5/9] IB/isert: Replace semaphore sem with completion
Date: Mon, 21 Nov 2016 14:33:02 +0200 [thread overview]
Message-ID: <5489f4ef-02fb-5207-6261-c08ee805acae@grimberg.me> (raw)
In-Reply-To: <2625189.HB0S3sfnY7@wuerfel>
>>> diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c
>>> index 6dd43f6..de80f56 100644
>>> --- a/drivers/infiniband/ulp/isert/ib_isert.c
>>> +++ b/drivers/infiniband/ulp/isert/ib_isert.c
>>> @@ -619,7 +619,7 @@
>>> mutex_unlock(&isert_np->mutex);
>>>
>>> isert_info("np %p: Allow accept_np to continue\n", isert_np);
>>> - up(&isert_np->sem);
>>> + complete(&isert_np->comp);
>>> }
>>>
>>> static void
>>> @@ -2311,7 +2311,7 @@ struct rdma_cm_id *
>>> isert_err("Unable to allocate struct isert_np\n");
>>> return -ENOMEM;
>>> }
>>> - sema_init(&isert_np->sem, 0);
>>> + init_completion(&isert_np->comp);
>>
>> This is still racy, a connect event can complete just before we
>> init the completion and *will* get lost...
>>
>> This code started off with a waitqueue which exposes the same
>> problem, see:
>> 531b7bf4bd79 Target/iser: Fix iscsit_accept_np and rdma_cm racy flow
>>
>> So, still NAK from me...
>
> I don't see what you mean here. The code using a waitqueue had no
> counter but the completion does.
The problem here is that init_completion sets the counter to zero
and between this thread wakes up and until it initializes comp->done
we might have another connect event completing it and I fail to
see how this is handled correctly.
next prev parent reply other threads:[~2016-11-21 12:33 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-11-21 6:08 [PATCH v5 0/9] infiniband: Remove semaphores Binoy Jayan
2016-11-21 6:08 ` [PATCH v5 1/9] IB/core: iwpm_nlmsg_request: Replace semaphore with completion Binoy Jayan
2016-11-21 6:08 ` [PATCH v5 2/9] IB/core: Replace semaphore sm_sem with an atomic wait Binoy Jayan
[not found] ` <CA+55aFxGjaqduhRCyk0mVxEA7aqQ-omdG8SBreZ=x5cW2ovngQ@mail.gmail.com>
2016-11-21 16:52 ` Arnd Bergmann
2016-11-21 16:57 ` Christoph Hellwig
2016-11-21 17:57 ` Linus Torvalds
2016-11-21 22:51 ` Arnd Bergmann
2016-11-21 6:08 ` [PATCH v5 3/9] IB/hns: Replace semaphore poll_sem with mutex Binoy Jayan
2016-11-21 6:08 ` [PATCH v5 4/9] IB/mthca: " Binoy Jayan
2016-11-21 6:08 ` [PATCH v5 5/9] IB/isert: Replace semaphore sem with completion Binoy Jayan
2016-11-21 7:36 ` Sagi Grimberg
2016-11-21 10:22 ` Arnd Bergmann
2016-11-21 12:33 ` Sagi Grimberg [this message]
2016-11-21 14:50 ` Arnd Bergmann
2016-11-21 6:08 ` [PATCH v5 6/9] IB/hns: Replace counting semaphore event_sem with wait_event Binoy Jayan
2016-11-21 6:08 ` [PATCH v5 7/9] IB/mthca: " Binoy Jayan
2016-11-21 6:08 ` [PATCH v5 8/9] IB/mlx5: Add helper mlx5_ib_post_send_wait Binoy Jayan
2016-11-21 6:08 ` [PATCH v5 9/9] IB/mlx5: Replace semaphore umr_common:sem with wait_event Binoy Jayan
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=5489f4ef-02fb-5207-6261-c08ee805acae@grimberg.me \
--to=sagi@grimberg.me \
--cc=arnd@arndb.de \
--cc=bart.vanassche@sandisk.com \
--cc=binoy.jayan@linaro.org \
--cc=broonie@kernel.org \
--cc=dledford@redhat.com \
--cc=faisal.latif@intel.com \
--cc=hal.rosenstock@gmail.com \
--cc=ira.weiny@intel.com \
--cc=jennyf@mellanox.com \
--cc=leonro@mellanox.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=markb@mellanox.com \
--cc=matanb@mellanox.com \
--cc=mustafa.ismail@intel.com \
--cc=nab@linux-iscsi.org \
--cc=oulijun@huawei.com \
--cc=sagig@mellanox.com \
--cc=sean.hefty@intel.com \
--cc=swise@opengridcomputing.com \
--cc=target-devel@vger.kernel.org \
--cc=tatyana.e.nikolova@intel.com \
--cc=xavier.huwei@huawei.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).