From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Steve Wise <swise@opengridcomputing.com>,
Chuck Lever <chuck.lever@oracle.com>,
"J. Bruce Fields" <bfields@redhat.com>
Subject: [PATCH 4.1 42/46] svcrdma: handle rdma read with a non-zero initial page offset
Date: Fri, 23 Oct 2015 10:46:28 -0700 [thread overview]
Message-ID: <20151023174622.036366641@linuxfoundation.org> (raw)
In-Reply-To: <20151023174620.779720995@linuxfoundation.org>
4.1-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steve Wise <swise@opengridcomputing.com>
commit c91aed9896946721bb30705ea2904edb3725dd61 upstream.
The server rdma_read_chunk_lcl() and rdma_read_chunk_frmr() functions
were not taking into account the initial page_offset when determining
the rdma read length. This resulted in a read who's starting address
and length exceeded the base/bounds of the frmr.
The server gets an async error from the rdma device and kills the
connection, and the client then reconnects and resends. This repeats
indefinitely, and the application hangs.
Most work loads don't tickle this bug apparently, but one test hit it
every time: building the linux kernel on a 16 core node with 'make -j
16 O=/mnt/0' where /mnt/0 is a ramdisk mounted via NFSRDMA.
This bug seems to only be tripped with devices having small fastreg page
list depths. I didn't see it with mlx4, for instance.
Fixes: 0bf4828983df ('svcrdma: refactor marshalling logic')
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Tested-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
net/sunrpc/xprtrdma/svc_rdma_recvfrom.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
--- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
@@ -146,7 +146,8 @@ int rdma_read_chunk_lcl(struct svcxprt_r
ctxt->read_hdr = head;
pages_needed =
min_t(int, pages_needed, rdma_read_max_sge(xprt, pages_needed));
- read = min_t(int, pages_needed << PAGE_SHIFT, rs_length);
+ read = min_t(int, (pages_needed << PAGE_SHIFT) - *page_offset,
+ rs_length);
for (pno = 0; pno < pages_needed; pno++) {
int len = min_t(int, rs_length, PAGE_SIZE - pg_off);
@@ -245,7 +246,8 @@ int rdma_read_chunk_frmr(struct svcxprt_
ctxt->direction = DMA_FROM_DEVICE;
ctxt->frmr = frmr;
pages_needed = min_t(int, pages_needed, xprt->sc_frmr_pg_list_len);
- read = min_t(int, pages_needed << PAGE_SHIFT, rs_length);
+ read = min_t(int, (pages_needed << PAGE_SHIFT) - *page_offset,
+ rs_length);
frmr->kva = page_address(rqstp->rq_arg.pages[pg_no]);
frmr->direction = DMA_FROM_DEVICE;
next prev parent reply other threads:[~2015-10-23 17:55 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-23 17:45 [PATCH 4.1 00/46] 4.1.12-stable review Greg Kroah-Hartman
2015-10-23 17:45 ` [PATCH 4.1 01/46] net/ibm/emac: bump version numbers for correct work with ethtool Greg Kroah-Hartman
2015-10-23 17:45 ` [PATCH 4.1 02/46] l2tp: protect tunnel->del_work by ref_count Greg Kroah-Hartman
2015-10-23 17:45 ` [PATCH 4.1 03/46] skbuff: Fix skb checksum flag on skb pull Greg Kroah-Hartman
2015-10-23 17:45 ` [PATCH 4.1 04/46] skbuff: Fix skb checksum partial check Greg Kroah-Hartman
2015-10-23 17:45 ` [PATCH 4.1 05/46] inet: fix races in reqsk_queue_hash_req() Greg Kroah-Hartman
2015-10-23 17:45 ` [PATCH 4.1 06/46] net: add pfmemalloc check in sk_add_backlog() Greg Kroah-Hartman
2015-10-23 17:45 ` [PATCH 4.1 07/46] ppp: dont override sk->sk_state in pppoe_flush_dev() Greg Kroah-Hartman
2015-10-23 17:45 ` [PATCH 4.1 08/46] inet: fix race in reqsk_queue_unlink() Greg Kroah-Hartman
2015-10-23 17:45 ` [PATCH 4.1 09/46] bpf: fix panic in SO_GET_FILTER with native ebpf programs Greg Kroah-Hartman
2015-10-23 17:45 ` [PATCH 4.1 10/46] ovs: do not allocate memory from offline numa node Greg Kroah-Hartman
2015-10-23 17:45 ` [PATCH 4.1 11/46] act_mirred: clear sender cpu before sending to tx Greg Kroah-Hartman
2015-10-23 17:45 ` [PATCH 4.1 12/46] ethtool: Use kcalloc instead of kmalloc for ethtool_get_strings Greg Kroah-Hartman
2015-10-23 17:45 ` [PATCH 4.1 13/46] tipc: move fragment importance field to new header position Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 14/46] netlink: Trim skb to alloc size to avoid MSG_TRUNC Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 15/46] af_unix: Convert the unix_sk macro to an inline function for type safety Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 16/46] af_unix: return data from multiple SKBs on recv() with MSG_PEEK flag Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 17/46] net/unix: fix logic about sk_peek_offset Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 18/46] drm: Fix locking for sysfs dpms file Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 19/46] crypto: sparc - initialize blkcipher.ivsize Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 20/46] crypto: ahash - ensure statesize is non-zero Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 21/46] memcg: convert threshold to bytes Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 22/46] btrfs: check unsupported filters in balance arguments Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 23/46] btrfs: fix use after free iterating extrefs Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 24/46] arm64: errata: use KBUILD_CFLAGS_MODULE for erratum #843419 Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 25/46] nfsd/blocklayout: accept any minlength Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 26/46] mfd: max77843: Fix max77843_chg_init() return on error Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 27/46] i2c: rcar: enable RuntimePM before registering to the core Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 28/46] i2c: s3c2410: " Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 29/46] i2c: designware: Do not use parameters from ACPI on Dell Inspiron 7348 Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 30/46] i2c: designware-platdrv: enable RuntimePM before registering to the core Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 31/46] workqueue: make sure delayed work run in local cpu Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 32/46] drm/nouveau/fbcon: take runpm reference when userspace has an open fd Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 33/46] drm/dp/mst: make mst i2c transfer code more robust Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 34/46] drm/radeon: attach tile property to mst connector Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 35/46] drm/radeon: add pm sysfs files late Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 36/46] dm thin: fix missing pool reference count decrement in pool_ctr error path Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 37/46] rbd: fix double free on rbd_dev->header_name Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 38/46] sched/preempt: Rename PREEMPT_CHECK_OFFSET to PREEMPT_DISABLE_OFFSET Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 39/46] sched/preempt: Fix cond_resched_lock() and cond_resched_softirq() Greg Kroah-Hartman
2015-10-23 20:14 ` Thomas Backlund
2015-10-23 23:21 ` Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 41/46] arm64: Fix THP protection change logic Greg Kroah-Hartman
2015-10-23 17:46 ` Greg Kroah-Hartman [this message]
2015-10-23 17:46 ` [PATCH 4.1 43/46] locks: have flock_lock_file take an inode pointer instead of a filp Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 44/46] locks: new helpers - flock_lock_inode_wait and posix_lock_inode_wait Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 45/46] locks: inline posix_lock_file_wait and flock_lock_file_wait Greg Kroah-Hartman
2015-10-23 17:46 ` [PATCH 4.1 46/46] nfs4: have do_vfs_lock take an inode pointer Greg Kroah-Hartman
2015-10-23 20:34 ` [PATCH 4.1 00/46] 4.1.12-stable review Shuah Khan
2015-10-23 23:22 ` Greg Kroah-Hartman
2015-10-24 1:11 ` Guenter Roeck
2015-10-24 3:15 ` Guenter Roeck
2015-10-24 13:20 ` Greg Kroah-Hartman
[not found] ` <562b9ffc.e8acc20a.c45a5.08fd@mx.google.com>
2015-10-24 15:16 ` Kevin Hilman
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=20151023174622.036366641@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=bfields@redhat.com \
--cc=chuck.lever@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=stable@vger.kernel.org \
--cc=swise@opengridcomputing.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).