aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/infiniband/hw/mlx4/mlx4_ib.h
diff options
context:
space:
mode:
authorRoland Dreier <rolandd@cisco.com>2007-04-18 20:20:28 -0700
committerRoland Dreier <rolandd@cisco.com>2007-05-08 18:00:37 -0700
commit1bf66a30421ca772820f489d88c16d0c430d6a67 (patch)
treeb1ab223e6908d772bcad7f9bc3382c33ad5a4490 /drivers/infiniband/hw/mlx4/mlx4_ib.h
parentIB/uverbs: Export ib_umem_get()/ib_umem_release() to modules (diff)
downloadlinux-dev-1bf66a30421ca772820f489d88c16d0c430d6a67.tar.xz
linux-dev-1bf66a30421ca772820f489d88c16d0c430d6a67.zip
IB: Put rlimit accounting struct in struct ib_umem
When memory pinned with ib_umem_get() is released, ib_umem_release() needs to subtract the amount of memory being unpinned from mm->locked_vm. However, ib_umem_release() may be called with mm->mmap_sem already held for writing if the memory is being released as part of an munmap() call, so it is sometimes necessary to defer this accounting into a workqueue. However, the work struct used to defer this accounting is dynamically allocated before it is queued, so there is the possibility of failing that allocation. If the allocation fails, then ib_umem_release has no choice except to bail out and leave the process with a permanently elevated locked_vm. Fix this by allocating the structure to defer accounting as part of the original struct ib_umem, so there's no possibility of failing a later allocation if creating the struct ib_umem and pinning memory succeeds. Signed-off-by: Roland Dreier <rolandd@cisco.com>
Diffstat (limited to 'drivers/infiniband/hw/mlx4/mlx4_ib.h')
0 files changed, 0 insertions, 0 deletions