Chuck Lever 5c166bef4f xprtrdma: Re-write rpcrdma_flush_cqs() ... Currently rpcrdma_flush_cqs() attempts to avoid code duplication, and simply invokes rpcrdma_recvcq_upcall and rpcrdma_sendcq_upcall. 1. rpcrdma_flush_cqs() can run concurrently with provider upcalls. Both flush_cqs() and the upcalls were invoking ib_poll_cq() in different threads using the same wc buffers (ep->rep_recv_wcs and ep->rep_send_wcs), added by commit 1c00dd0776 ("xprtrmda: Reduce calls to ib_poll_cq() in completion handlers"). During transport disconnect processing, this sometimes resulted in the same reply getting added to the rpcrdma_tasklets_g list more than once, which corrupted the list. 2. The upcall functions drain only a limited number of CQEs, thanks to the poll budget added by commit 8301a2c047 ("xprtrdma: Limit work done by completion handler"). Fixes: a7bc211ac9 ("xprtrdma: On disconnect, don't ignore ... ") BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=276 Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>		2014-11-25 13:39:20 -05:00
..
Makefile	xprtrdma: add separate Kconfig options for NFSoRDMA client and server support	2014-03-28 18:02:12 -04:00
rpc_rdma.c	xprtrdma: Don't invalidate FRMRs if registration fails	2014-07-31 16:22:53 -04:00
svc_rdma_marshal.c	svcrdma: underflow issue in decode_write_list()	2013-07-15 11:46:23 -04:00
svc_rdma_recvfrom.c	svcrdma: Add zero padding if the client doesn't send it	2014-07-22 16:40:21 -04:00
svc_rdma_sendto.c	svcrdma: send_write() must not overflow the device's max sge	2014-07-11 15:03:48 -04:00
svc_rdma_transport.c	svcrdma: advertise the correct max payload	2014-09-29 14:35:18 -04:00
svc_rdma.c	net: Convert uses of typedef ctl_table to struct ctl_table	2013-06-13 02:36:09 -07:00
transport.c	xprtrdma: unmap all FMRs during transport disconnect	2014-11-25 13:39:20 -05:00
verbs.c	xprtrdma: Re-write rpcrdma_flush_cqs()	2014-11-25 13:39:20 -05:00
xprt_rdma.h	xprtrdma: Cap req_cqinit	2014-11-25 13:39:20 -05:00