linux/include/rdma
Shawn Bohrer 87773dd56d IB: ib_umem_release() should decrement mm->pinned_vm from ib_umem_get
In debugging an application that receives -ENOMEM from ib_reg_mr(), I
found that ib_umem_get() can fail because the pinned_vm count has
wrapped causing it to always be larger than the lock limit even with
RLIMIT_MEMLOCK set to RLIM_INFINITY.

The wrapping of pinned_vm occurs because the process that calls
ib_reg_mr() will have its mm->pinned_vm count incremented.  Later a
different process with a different mm_struct than the one that
allocated the ib_umem struct ends up releasing it which results in
decrementing the new processes mm->pinned_vm count past zero and
wrapping.

I'm not entirely sure what circumstances cause a different process to
release the ib_umem than the one that allocated it but the kernel
stack trace of the freeing process from my situation looks like the
following:

    Call Trace:
     [<ffffffff814d64b1>] dump_stack+0x19/0x1b
     [<ffffffffa0b522a5>] ib_umem_release+0x1f5/0x200 [ib_core]
     [<ffffffffa0b90681>] mlx4_ib_destroy_qp+0x241/0x440 [mlx4_ib]
     [<ffffffffa0b4d93c>] ib_destroy_qp+0x12c/0x170 [ib_core]
     [<ffffffffa0cc7129>] ib_uverbs_close+0x259/0x4e0 [ib_uverbs]
     [<ffffffff81141cba>] __fput+0xba/0x240
     [<ffffffff81141e4e>] ____fput+0xe/0x10
     [<ffffffff81060894>] task_work_run+0xc4/0xe0
     [<ffffffff810029e5>] do_notify_resume+0x95/0xa0
     [<ffffffff814e3dd0>] int_signal+0x12/0x17

The following patch fixes the issue by storing the pid struct of the
process that calls ib_umem_get() so that ib_umem_release and/or
ib_umem_account() can properly decrement the pinned_vm count of the
correct mm_struct.

Signed-off-by: Shawn Bohrer <sbohrer@rgmadvisors.com>
Reviewed-by: Shachar Raindel <raindel@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-09-19 09:55:42 -07:00
..
ib_addr.h IB/cma: IBoE (RoCE) IP-based GID addressing 2014-01-18 14:12:35 -08:00
ib_cache.h IB/core: Add ib_find_exact_cached_pkey() 2012-09-30 20:33:30 -07:00
ib_cm.h IB/core: Don't resolve passive side RoCE L2 address in CMA REQ handler 2014-04-01 14:05:26 -07:00
ib_fmr_pool.h RDMA: Improve include file coding style 2008-07-14 23:48:44 -07:00
ib_mad.h IB/mad: Add user space RMPP support 2014-08-10 20:36:00 -07:00
ib_marshall.h RDMA/cma: Export rdma cm interface to userspace 2006-12-12 11:50:22 -08:00
ib_pack.h IB/core: Ethernet L2 attributes in verbs/cm structures 2014-01-14 14:20:54 -08:00
ib_pma.h IB/pma: Add include file for IBA performance counters definitions 2011-07-18 21:04:35 -07:00
ib_sa.h IB/core: Ethernet L2 attributes in verbs/cm structures 2014-01-14 14:20:54 -08:00
ib_smi.h IB: Remove __constant_{endian} uses 2009-01-17 17:11:57 -08:00
ib_umem.h IB: ib_umem_release() should decrement mm->pinned_vm from ib_umem_get 2014-09-19 09:55:42 -07:00
ib_verbs.h IB/core: Add user MR re-registration support 2014-08-01 15:11:13 -07:00
ib.h RDMA/cma: Define native IB address 2013-06-20 13:08:01 -07:00
iw_cm.h RDMA/cma: Add IPv6 support for iWARP 2013-08-12 12:32:31 -07:00
iw_portmap.h RDMA/core: Add support for iWARP Port Mapper user space service 2014-06-10 10:11:45 -07:00
rdma_cm_ib.h RDMA: Fix license text 2008-07-14 23:48:43 -07:00
rdma_cm.h RDMA/cma: Export cma_get_service_id() 2013-06-20 23:35:41 -07:00
rdma_netlink.h RDMA/core: Add support for iWARP Port Mapper user space service 2014-06-10 10:11:45 -07:00