stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Chuck Lever <chuck.lever@oracle.com>,
	Dan Aloni <dan@kernelim.com>,
	Anna Schumaker <Anna.Schumaker@Netapp.com>,
	Sasha Levin <sashal@kernel.org>,
	linux-nfs@vger.kernel.org, netdev@vger.kernel.org
Subject: [PATCH AUTOSEL 5.8 01/29] xprtrdma: Release in-flight MRs on disconnect
Date: Mon, 14 Sep 2020 09:03:30 -0400	[thread overview]
Message-ID: <20200914130358.1804194-1-sashal@kernel.org> (raw)

From: Chuck Lever <chuck.lever@oracle.com>

[ Upstream commit 5de55ce951a1466e31ff68a7bc6b0a7ce3cb5947 ]

Dan Aloni reports that when a server disconnects abruptly, a few
memory regions are left DMA mapped. Over time this leak could pin
enough I/O resources to slow or even deadlock an NFS/RDMA client.

I found that if a transport disconnects before pending Send and
FastReg WRs can be posted, the to-be-registered MRs are stranded on
the req's rl_registered list and never released -- since they
weren't posted, there's no Send completion to DMA unmap them.

Reported-by: Dan Aloni <dan@kernelim.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/sunrpc/xprtrdma/verbs.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index 75c646743df3e..ca89f24a1590b 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -933,6 +933,8 @@ static void rpcrdma_req_reset(struct rpcrdma_req *req)
 
 	rpcrdma_regbuf_dma_unmap(req->rl_sendbuf);
 	rpcrdma_regbuf_dma_unmap(req->rl_recvbuf);
+
+	frwr_reset(req);
 }
 
 /* ASSUMPTION: the rb_allreqs list is stable for the duration,
-- 
2.25.1


             reply	other threads:[~2020-09-14 17:14 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-09-14 13:03 Sasha Levin [this message]
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 02/29] NFSv4.1 handle ERR_DELAY error reclaiming locking state on delegation recall Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 03/29] phy: omap-usb2-phy: disable PHY charger detect Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 04/29] habanalabs: prevent user buff overflow Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 05/29] habanalabs: fix report of RAZWI initiator coordinates Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 06/29] scsi: pm8001: Fix memleak in pm8001_exec_internal_task_abort Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 07/29] scsi: libfc: Fix for double free() Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 08/29] scsi: lpfc: Fix FLOGI/PLOGI receive race condition in pt2pt discovery Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 09/29] scsi: lpfc: Extend the RDF FPIN Registration descriptor for additional events Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 10/29] regulator: pwm: Fix machine constraints application Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 11/29] spi: spi-loopback-test: Fix out-of-bounds read Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 12/29] interconnect: Show bandwidth for disabled paths as zero in debugfs Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 13/29] NFS: Zero-stateid SETATTR should first return delegation Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 14/29] SUNRPC: stop printk reading past end of string Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 15/29] rapidio: Replace 'select' DMAENGINES 'with depends on' Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 16/29] cifs: fix DFS mount with cifsacl/modefromsid Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 17/29] kobject: Drop unneeded conditional in __kobject_del() Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 18/29] openrisc: Fix cache API compile issue when not inlining Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 19/29] nvme-fc: cancel async events before freeing event struct Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 20/29] nvme-rdma: " Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 21/29] nvme-tcp: " Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 22/29] block: only call sched requeue_request() for scheduled requests Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 23/29] f2fs: fix indefinite loop scanning for free nid Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 24/29] f2fs: Return EOF on unaligned end of file DIO read Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 25/29] i2c: algo: pca: Reapply i2c bus settings after reset Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 26/29] spi: Fix memory leak on splited transfers Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 27/29] gcov: add support for GCC 10.1 Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 28/29] KVM: Check the allocation of pv cpu mask Sasha Levin
2020-09-14 13:03 ` [PATCH AUTOSEL 5.8 29/29] KVM: MIPS: Change the definition of kvm type Sasha Levin

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=20200914130358.1804194-1-sashal@kernel.org \
    --to=sashal@kernel.org \
    --cc=Anna.Schumaker@Netapp.com \
    --cc=chuck.lever@oracle.com \
    --cc=dan@kernelim.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=netdev@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).