Skip to content

cloud-kernel
cloud-kernel

openanolis / cloud-kernel
1 年多 前同步成功

通知 160
Star 36
Fork 7
cloud-kernel
cloud-kernel
切换分支/标签
  1. 09 9月, 2010 9 次提交
    • A
      RDS: Explicitly allocate rm in sendmsg() · fc445084
      由 Andy Grover 提交于 
      r_m_copy_from_user used to allocate the rm as well as kernel
buffers for the data, and then copy the data in. Now, sendmsg()
allocates the rm, although the data buffer alloc still happens
in r_m_copy_from_user.

SGs are still allocated with rm, but now r_m_alloc_sgs() is
used to reserve them. This allows multiple SG lists to be
allocated from the one rm -- this is important once we also
want to alloc our rdma sgl from this pool.
Signed-off-by: NAndy Grover <andy.grover@oracle.com>
      fc445084
    • A
      RDS: cleanup/fix rds_rdma_unuse · 3ef13f3c
      由 Andy Grover 提交于 
      First, it looks to me like the atomic_inc is wrong.
We should be decrementing refcount only once here, no? It's
already being done by the mr_put() at the end.

Second, simplify the logic a bit by bailing early (with a warning)
if !mr.
Signed-off-by: NAndy Grover <andy.grover@oracle.com>
      3ef13f3c
    • A
      RDS: break out rdma and data ops into nested structs in rds_message · e779137a
      由 Andy Grover 提交于 
      Clearly separate rdma-related variables in rm from data-related ones.
This is in anticipation of adding atomic support.
Signed-off-by: NAndy Grover <andy.grover@oracle.com>
      e779137a
    • A
      RDS: cleanup: remove "== NULL"s and "!= NULL"s in ptr comparisons · 8690bfa1
      由 Andy Grover 提交于 
      Favor "if (foo)" style over "if (foo != NULL)".
Signed-off-by: NAndy Grover <andy.grover@oracle.com>
      8690bfa1
    • A
      RDS: move rds_shutdown_worker impl. to rds_conn_shutdown · 2dc39357
      由 Andy Grover 提交于 
      This fits better in connection.c, rather than threads.c.
Signed-off-by: NAndy Grover <andy.grover@oracle.com>
      2dc39357
    • A
      RDS: Fix locking in send on m_rs_lock · 9de0864c
      由 Andy Grover 提交于 
      Do not nest m_rs_lock under c_lock

Disable interrupts in {rdma,atomic}_send_complete
Signed-off-by: NAndy Grover <andy.grover@oracle.com>
      9de0864c
    • A
      RDS: Rewrite rds_send_drop_to() for clarity · 7c82eaf0
      由 Andy Grover 提交于 
      This function has been the source of numerous bugs; it's just
too complicated. Simplified to nest spinlocks cleanly within
the second loop body, and kick out early if there are no
rms to drop.

This will be a little slower because conn lock is grabbed for
each entry instead of "caching" the lock across rms, but this
should be entirely irrelevant to fastpath performance.
Signed-off-by: NAndy Grover <andy.grover@oracle.com>
      7c82eaf0
    • T
      RDS: Fix corrupted rds_mrs · 35b52c70
      由 Tina Yang 提交于 
      On second look at this bug (OFED #2002), it seems that the
collision is not with the retransmission queue (packet acked
by the peer), but with the local send completion.  A theoretical
sequence of events (from time t0 to t3) is thought to be as
follows,

Thread #1
t0:
    sock_release
    rds_release
    rds_send_drop_to /* wait on send completion */
t2:
    rds_rdma_drop_keys()   /* destroy & free all mrs */

Thread #2
t1:
    rds_ib_send_cq_comp_handler
    rds_ib_send_unmap_rm
    rds_message_unmapped   /* wake up #1 @ t0 */
t3:
    rds_message_put
    rds_message_purge
    rds_mr_put   /* memory corruption detected */

The problem with the rds_rdma_drop_keys() is it could
remove a mr's refcount more than its due (i.e. repeatedly
as long as it still remains in the tree (mr->r_refcount > 0)).
Theoretically it should remove only one reference - reference
by the tree.

        /* Release any MRs associated with this socket */
        while ((node = rb_first(&rs->rs_rdma_keys))) {
                mr = container_of(node, struct rds_mr, r_rb_node);
                if (mr->r_trans == rs->rs_transport)
                        mr->r_invalidate = 0;
                rds_mr_put(mr);
        }

I think the correct way of doing it is to remove the mr from
the tree and rds_destroy_mr it first, then a rds_mr_put()
to decrement its reference count by one.  Whichever thread
holds the last reference will free the mr via rds_mr_put().
Signed-off-by: NTina Yang <tina.yang@oracle.com>
Signed-off-by: NAndy Grover <andy.grover@oracle.com>
      35b52c70
    • A
      RDS: Fix BUG_ONs to not fire when in a tasklet · 9e2effba
      由 Andy Grover 提交于 
      in_interrupt() is true in softirqs. The BUG_ONs are supposed
to check for if irqs are disabled, so we should use
BUG_ON(irqs_disabled()) instead, duh.
Signed-off-by: NAndy Grover <andy.grover@oracle.com>
      9e2effba
  3. 07 9月, 2010 31 次提交