]> www.infradead.org Git - users/jedix/linux-maple.git/log
users/jedix/linux-maple.git
9 years agosdp: Do not ignore scope_id in IPv6
Amir Vadai [Thu, 6 Jan 2011 12:38:34 +0000 (14:38 +0200)]
sdp: Do not ignore scope_id in IPv6

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: make backport sk_inet()->xxx simpler
Amir Vadai [Wed, 5 Jan 2011 11:26:32 +0000 (13:26 +0200)]
sdp: make backport sk_inet()->xxx simpler

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: Fix get getsockname/getpeername in IPv6
Amir Vadai [Wed, 5 Jan 2011 09:33:21 +0000 (11:33 +0200)]
sdp: Fix get getsockname/getpeername in IPv6

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: Allow bind to address with family AF_INETx_SDP
Amir Vadai [Sun, 2 Jan 2011 11:31:12 +0000 (13:31 +0200)]
sdp: Allow bind to address with family AF_INETx_SDP

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: Use %pI4 + %pI6 in new kernels
Amir Vadai [Thu, 30 Dec 2010 12:38:48 +0000 (14:38 +0200)]
sdp: Use %pI4 + %pI6 in new kernels

A patch in this subject was in the wrong backport. Moving it to a dedicated patch.

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: fix connect to IPv4 over IPv6
Amir Vadai [Wed, 29 Dec 2010 15:23:07 +0000 (17:23 +0200)]
sdp: fix connect to IPv4 over IPv6

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: Fix some issues in ipv6 support
Amir Vadai [Wed, 29 Dec 2010 10:40:52 +0000 (12:40 +0200)]
sdp: Fix some issues in ipv6 support

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: ipv6 support
Amir Vadai [Mon, 27 Dec 2010 09:58:26 +0000 (11:58 +0200)]
sdp: ipv6 support

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: move histogram allocation from stack to heap
Amir Vadai [Tue, 21 Dec 2010 11:38:37 +0000 (13:38 +0200)]
sdp: move histogram allocation from stack to heap

- Cause a kernel crash when accessing /proc/net/sdpstats on RH6

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: print error value when ib_umem_get fails
Amir Vadai [Tue, 21 Dec 2010 12:27:06 +0000 (14:27 +0200)]
sdp: print error value when ib_umem_get fails

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: remove 'reading beyond SKB' warning
Amir Vadai [Tue, 14 Dec 2010 12:41:12 +0000 (14:41 +0200)]
sdp: remove 'reading beyond SKB' warning

This is a good sanity check, but could print a warning when a
partially used SrcAvail skb is cancelled.
This should be fixed in a way that will leave the sanity check,
but need to make minimal changes before the GA.

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: RdmaRdCompl not sent sometimes
Amir Vadai [Tue, 14 Dec 2010 06:48:42 +0000 (08:48 +0200)]
sdp: RdmaRdCompl not sent sometimes

When SrcAvailCancel is handled after RDMA finshed and before sending
RdmaRdCompl, RdmaRdCompl won't be sent, and a data corruption will occur.
Made sure that all sdp_abort_rx_srcavail will send RdmaRdCompl if needed.

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: make sdp_prf index atomic
Amir Vadai [Tue, 14 Dec 2010 07:55:54 +0000 (09:55 +0200)]
sdp: make sdp_prf index atomic

Will make sdpprf more reliable - records won't override other records.

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: handle failed RDMA read
Amir Vadai [Tue, 7 Dec 2010 07:32:50 +0000 (09:32 +0200)]
sdp: handle failed RDMA read

When RDMA read is failed, do not report those bytes as read, and
return error code to user.

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: Fix compilation error when SDP_DEBUG_DATA is off
Amir Vadai [Tue, 30 Nov 2010 06:48:33 +0000 (08:48 +0200)]
sdp: Fix compilation error when SDP_DEBUG_DATA is off

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: Enable set zcopy threshold to 0 using setsockopt
Amir Vadai [Wed, 17 Nov 2010 09:03:03 +0000 (11:03 +0200)]
sdp: Enable set zcopy threshold to 0 using setsockopt

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: BUG2161 - hanging sockets are left
Amir Vadai [Tue, 16 Nov 2010 11:32:14 +0000 (13:32 +0200)]
sdp: BUG2161 - hanging sockets are left

When getting DREQ at TCP_CLOSE_WAIT state, CMA ref count is taken
and not returned. This happens because CMA timer is not started.

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: Take into account HW inline capabilities
Amir Vadai [Sun, 14 Nov 2010 12:57:15 +0000 (14:57 +0200)]
sdp: Take into account HW inline capabilities

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: Send small sends using inline
Amir Vadai [Mon, 8 Nov 2010 11:49:12 +0000 (13:49 +0200)]
sdp: Send small sends using inline

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: access socket history from debugfs
Amir Vadai [Wed, 3 Nov 2010 13:47:01 +0000 (15:47 +0200)]
sdp: access socket history from debugfs

Also moved sdpprf to /sys/kernel/debug/sdp - seems to be a better place
than /proc/net

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: use a macro to convert ssk into sk
Amir Vadai [Tue, 2 Nov 2010 10:43:21 +0000 (12:43 +0200)]
sdp: use a macro to convert ssk into sk

Instead of accessing &ssk->isk.sk cast the sdp socket into
sock. Do it from a macro sk_ssk(ssk)

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: Initialize remote credits when sending Hello
Amir Vadai [Wed, 27 Oct 2010 13:56:23 +0000 (15:56 +0200)]
sdp: Initialize remote credits when sending Hello

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: BUG2158 - do not send SrcAvail too small
Amir Vadai [Wed, 27 Oct 2010 12:24:07 +0000 (14:24 +0200)]
sdp: BUG2158 - do not send SrcAvail too small

- Shouldn't send SrcAvail with no RDMA payload - could happen when RLIMIT is
  too small.

- Do not crash when getting 2 SrcAvail.

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: add a sanity check for sg_len
Eldad Zinger [Mon, 25 Oct 2010 13:02:43 +0000 (15:02 +0200)]
sdp: add a sanity check for sg_len

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fix socket_allocated counter
Eldad Zinger [Sun, 24 Oct 2010 15:15:06 +0000 (17:15 +0200)]
sdp: fix socket_allocated counter

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: no point of waiting for data if remote host can't send
Eldad Zinger [Sun, 24 Oct 2010 09:24:27 +0000 (11:24 +0200)]
sdp: no point of waiting for data if remote host can't send

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fix support for a case of no locking capabilities
Eldad Zinger [Sun, 24 Oct 2010 09:19:24 +0000 (11:19 +0200)]
sdp: fix support for a case of no locking capabilities

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: if can not allocate memory - no point of waiting
Eldad Zinger [Sun, 24 Oct 2010 09:16:56 +0000 (11:16 +0200)]
sdp: if can not allocate memory - no point of waiting

We don't wake up sleeper when we free memory, so mem sleeper will sleep forever.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: cosmetics
Eldad Zinger [Sun, 24 Oct 2010 08:50:08 +0000 (10:50 +0200)]
sdp: cosmetics

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fix for CMA reference count
Eldad Zinger [Wed, 20 Oct 2010 15:14:35 +0000 (17:14 +0200)]
sdp: fix for CMA reference count

When sdp_destroy_work() is invoked (due to the cma handler),
but "ssk->cma_timewait_timeout == 0", a socket reference was putted even though
it was never taken. This can happen in 2 scenarios:
1. sdp_destroy_work() locked the socket before sdp_close()
2. sdp_destroy_work() locked the socket after sdp_cma_timewait_timeout_work()

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: add ability to set a maximum memory usage for the entire module
Eldad Zinger [Wed, 20 Oct 2010 15:08:53 +0000 (17:08 +0200)]
sdp: add ability to set a maximum memory usage for the entire module

use top_mem_usage module parameter for this.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: Accept AF_INET_SDP in address supplied to connect()
Amir Vadai [Sun, 24 Oct 2010 15:03:49 +0000 (17:03 +0200)]
sdp: Accept AF_INET_SDP in address supplied to connect()

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: fix RDMA read completion with error
Amir Vadai [Mon, 18 Oct 2010 08:48:16 +0000 (10:48 +0200)]
sdp: fix RDMA read completion with error

- set socket to error state for all wr_id's
- no need in sdp_handle_send_comp()

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: sdp_post_rdma_read() should clean up its mess
Amir Vadai [Mon, 18 Oct 2010 12:20:12 +0000 (14:20 +0200)]
sdp: sdp_post_rdma_read() should clean up its mess

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: fix compilation error when compiling without debug flags
Eldad Zinger [Tue, 19 Oct 2010 14:49:53 +0000 (16:49 +0200)]
sdp: fix compilation error when compiling without debug flags

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: treat unfinished RDMA operation as a fatal error
Eldad Zinger [Tue, 19 Oct 2010 15:26:34 +0000 (17:26 +0200)]
sdp: treat unfinished RDMA operation as a fatal error

mark socket with EIO error code.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fix for race condition with SrcAvailCancel handling
Eldad Zinger [Tue, 19 Oct 2010 14:52:09 +0000 (16:52 +0200)]
sdp: fix for race condition with SrcAvailCancel handling

Tx complition resets "rdma_inflight->busy" and wakes up the recvmsg().
But rx_workqueue might get the lock of the socket before recvmsg(), so
SrcAvailCancel might be polled from the rx_cq, and since "busy = 0",
rx_sa wil be freed (w/o freeing umem&fmr).

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: call ib_umem_get with the right access
Amir Vadai [Sun, 17 Oct 2010 17:48:38 +0000 (19:48 +0200)]
sdp: call ib_umem_get with the right access

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: Improve the look of packet dump
Amir Vadai [Sun, 17 Oct 2010 17:45:43 +0000 (19:45 +0200)]
sdp: Improve the look of packet dump

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: cosmetics, debug messages, error codes
Eldad Zinger [Sun, 17 Oct 2010 17:44:06 +0000 (19:44 +0200)]
sdp: cosmetics, debug messages, error codes

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: SrcAvailCancel should not be processed during RDMA read
Eldad Zinger [Sun, 17 Oct 2010 17:26:01 +0000 (19:26 +0200)]
sdp: SrcAvailCancel should not be processed during RDMA read

fmr should be freed only after tx completion, regardless if SrcAvailCancel arrived.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: BUG2082 - fix orphan counter reading
Eldad Zinger [Sun, 17 Oct 2010 10:57:38 +0000 (12:57 +0200)]
sdp: BUG2082 - fix orphan counter reading

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: when aborting SrcAvail, should check if it wasn't aborted already
Eldad Zinger [Tue, 12 Oct 2010 14:06:37 +0000 (16:06 +0200)]
sdp: when aborting SrcAvail, should check if it wasn't aborted already

Check is relevant only after sleeping.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: ZCopy doesn't support multithreading - warning & kernel panic protection
Eldad Zinger [Tue, 12 Oct 2010 08:22:05 +0000 (10:22 +0200)]
sdp: ZCopy doesn't support multithreading - warning & kernel panic protection

Correspond to BUG 2147.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: Enable RoCE by default
Amir Vadai [Sun, 10 Oct 2010 09:10:38 +0000 (11:10 +0200)]
sdp: Enable RoCE by default

Could led to long delays when trying to establish connection to ethernet
peer, and libsdp is set to 'both'. Should be fixed by letting user set
both/sdp/tcp per interface and not only by subnet/port/process in libsdp.conf.

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: SrcAvailCancel should be processed even if SrcAvail was partly processed
Eldad Zinger [Mon, 11 Oct 2010 15:50:59 +0000 (17:50 +0200)]
sdp: SrcAvailCancel should be processed even if SrcAvail was partly processed

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: BUG2144 - first free rx_sa before sending SendSM
Eldad Zinger [Sun, 10 Oct 2010 09:50:12 +0000 (11:50 +0200)]
sdp: BUG2144 - first free rx_sa before sending SendSM

otherwise a new rx_sa might be alloced before the old one is removed.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fix code readability
Eldad Zinger [Sun, 10 Oct 2010 09:47:06 +0000 (11:47 +0200)]
sdp: fix code readability

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: BUG2141 - fix refcnt bug
Eldad Zinger [Wed, 6 Oct 2010 09:52:10 +0000 (11:52 +0200)]
sdp: BUG2141 - fix refcnt bug

This bug is reproducable when sdpnetstat(1) is run while a socket is being
destructed. sdp_proc utilities might try to hold a refcnt for a socket that its
destruction already began. This is possible because sdp_proc utilities uses
sock_list to scan for sockets, but a socket is removed from that list only
after all its resourcs are freed.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: refcnt debug tool
Eldad Zinger [Wed, 6 Oct 2010 07:47:27 +0000 (09:47 +0200)]
sdp: refcnt debug tool

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: cosmetics & add/remove warning messages
Eldad Zinger [Tue, 5 Oct 2010 11:53:34 +0000 (13:53 +0200)]
sdp: cosmetics & add/remove warning messages

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: extend socket locking scope in dreq timeout function
Eldad Zinger [Sun, 3 Oct 2010 09:58:06 +0000 (11:58 +0200)]
sdp: extend socket locking scope in dreq timeout function

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: sdp_poll() should not excessively poll rx_cq
Eldad Zinger [Tue, 28 Sep 2010 11:39:02 +0000 (13:39 +0200)]
sdp: sdp_poll() should not excessively poll rx_cq

sdp_poll() might be called before send(), and if there is no data in rx_cq,
polling rx_cq will waste 700 usec (recv_poll).

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: postpone rx timer when arming rx_cq
Eldad Zinger [Tue, 28 Sep 2010 11:32:38 +0000 (13:32 +0200)]
sdp: postpone rx timer when arming rx_cq

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fix for timestamping values in debug messages
Eldad Zinger [Tue, 28 Sep 2010 11:28:50 +0000 (13:28 +0200)]
sdp: fix for timestamping values in debug messages

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: better sequence-number handling and cosmetic updates
Eldad Zinger [Tue, 28 Sep 2010 10:21:58 +0000 (12:21 +0200)]
sdp: better sequence-number handling and cosmetic updates

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: add some checking & protection for enum values in debug utilities
Eldad Zinger [Mon, 27 Sep 2010 08:45:29 +0000 (10:45 +0200)]
sdp: add some checking & protection for enum values in debug utilities

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: BUG2138 - 32bit hosts can't devide 64bit variables
Eldad Zinger [Tue, 21 Sep 2010 17:21:50 +0000 (19:21 +0200)]
sdp: BUG2138 - 32bit hosts can't devide 64bit variables

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: remove recursion in tx_ring processing
Eldad Zinger [Mon, 20 Sep 2010 08:50:57 +0000 (10:50 +0200)]
sdp: remove recursion in tx_ring processing

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fix keepalive timer setup for server-sockets
Eldad Zinger [Mon, 13 Sep 2010 13:20:28 +0000 (15:20 +0200)]
sdp: fix keepalive timer setup for server-sockets

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fix logarithmic histogram index
Eldad Zinger [Mon, 13 Sep 2010 13:13:38 +0000 (15:13 +0200)]
sdp: fix logarithmic histogram index

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fix compilation on ia64 and ppc
Eldad Zinger [Tue, 14 Sep 2010 15:32:20 +0000 (17:32 +0200)]
sdp: fix compilation on ia64 and ppc

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: Some improvements to multistream BW
Amir Vadai [Sun, 12 Sep 2010 16:24:35 +0000 (18:24 +0200)]
sdp: Some improvements to multistream BW

- Removed almost all locks in data path - using lock_sock only
- CPU Affinity for SKB handling
- Don't do any RX proccessing from IRQ/timer
- recv_poll value in usec instead of msec

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
9 years agosdp: add support for no recv polling at all
Eldad Zinger [Thu, 26 Aug 2010 07:52:21 +0000 (10:52 +0300)]
sdp: add support for no recv polling at all

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: BUG1923 - fix support for MSG_OOB
Eldad Zinger [Sun, 22 Aug 2010 05:21:09 +0000 (08:21 +0300)]
sdp: BUG1923 - fix support for MSG_OOB

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: remove unused variables
Eldad Zinger [Thu, 19 Aug 2010 13:25:00 +0000 (16:25 +0300)]
sdp: remove unused variables

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fixed for error code in sendmsg-BZCopy
Eldad Zinger [Sun, 15 Aug 2010 08:21:59 +0000 (11:21 +0300)]
sdp: fixed for error code in sendmsg-BZCopy

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fixed skbs-control-queue memory leak
Eldad Zinger [Sun, 15 Aug 2010 08:11:03 +0000 (11:11 +0300)]
sdp: fixed skbs-control-queue memory leak

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fixed device removal issues
Eldad Zinger [Sun, 15 Aug 2010 07:48:05 +0000 (10:48 +0300)]
sdp: fixed device removal issues

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: added the ability to use sdpprf for any debug message
Eldad Zinger [Sun, 15 Aug 2010 07:40:02 +0000 (10:40 +0300)]
sdp: added the ability to use sdpprf for any debug message

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fixed a deadlock when tx_timer calls sdp_reset that tries to del the timer
Eldad Zinger [Sun, 15 Aug 2010 07:17:42 +0000 (10:17 +0300)]
sdp: fixed a deadlock when tx_timer calls sdp_reset that tries to del the timer

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fixed SrcAvail memory leak
Eldad Zinger [Sun, 15 Aug 2010 07:05:40 +0000 (10:05 +0300)]
sdp: fixed SrcAvail memory leak

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: removed unnecessary variable 'vm_wait'
Eldad Zinger [Mon, 9 Aug 2010 14:11:35 +0000 (17:11 +0300)]
sdp: removed unnecessary variable 'vm_wait'

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: stability improvements for ZCopy
Eldad Zinger [Mon, 9 Aug 2010 14:05:44 +0000 (17:05 +0300)]
sdp: stability improvements for ZCopy

- beter handling of ZCopy errors and SrcAvailCancel requests
- fixed SrcAvailCancel sequence number handling

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: properly kill nagle_timer on socket reset
Eldad Zinger [Thu, 5 Aug 2010 07:16:58 +0000 (10:16 +0300)]
sdp: properly kill nagle_timer on socket reset

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: BUG2092 - ib_device field in sdp_sock is reset not in user-context
Eldad Zinger [Thu, 5 Aug 2010 06:31:27 +0000 (09:31 +0300)]
sdp: BUG2092 - ib_device field in sdp_sock is reset not in user-context

ib_device field of sdp_sock is NULL after either cases:
- cma handler resetting the socket
- device removal
Both cases are not in user-context.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fix for stopping tx timer/tasklet when socket state is TCP_CLOSE
Eldad Zinger [Thu, 5 Aug 2010 05:39:58 +0000 (08:39 +0300)]
sdp: fix for stopping tx timer/tasklet when socket state is TCP_CLOSE

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: BUG1403 - last sk_refcnt is called while cma handler is invoked
Eldad Zinger [Tue, 3 Aug 2010 16:14:59 +0000 (19:14 +0300)]
sdp: BUG1403 - last sk_refcnt is called while cma handler is invoked

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: bug fix for a case of no memory to allocate for rx_sa
Eldad Zinger [Sun, 1 Aug 2010 11:19:26 +0000 (14:19 +0300)]
sdp: bug fix for a case of no memory to allocate for rx_sa

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: rx_irq should use tasklet instead of timer due to latency issue
Eldad Zinger [Mon, 2 Aug 2010 12:05:12 +0000 (15:05 +0300)]
sdp: rx_irq should use tasklet instead of timer due to latency issue

mod_timer(..., 0) measured to add 4ms delay.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: better handling of page-allocation-failure
Eldad Zinger [Sun, 1 Aug 2010 16:16:20 +0000 (19:16 +0300)]
sdp: better handling of page-allocation-failure

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: reduce size of sdp_buf to what is really being used
Eldad Zinger [Sun, 1 Aug 2010 15:46:49 +0000 (18:46 +0300)]
sdp: reduce size of sdp_buf to what is really being used

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: remove the relation between qp_active and sdp_free_fmr()
Eldad Zinger [Sun, 1 Aug 2010 15:08:11 +0000 (18:08 +0300)]
sdp: remove the relation between qp_active and sdp_free_fmr()

This will fix the warning message from ib_destroy_fmr_pool().

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: small fix to support device removal during traffic
Eldad Zinger [Sun, 1 Aug 2010 15:00:11 +0000 (18:00 +0300)]
sdp: small fix to support device removal during traffic

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: bug fix for a case of no memory to allocate for tx_sa
Eldad Zinger [Sun, 1 Aug 2010 11:30:54 +0000 (14:30 +0300)]
sdp: bug fix for a case of no memory to allocate for tx_sa

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fix behavior when a skb allocation fails
Eldad Zinger [Mon, 26 Jul 2010 07:24:05 +0000 (10:24 +0300)]
sdp: fix behavior when a skb allocation fails

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: some small non-behavioral changes in sdp_dreq_wait_timeout_work()
Eldad Zinger [Thu, 22 Jul 2010 09:08:52 +0000 (12:08 +0300)]
sdp: some small non-behavioral changes in sdp_dreq_wait_timeout_work()

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: rewrite orphan count logic
Eldad Zinger [Thu, 22 Jul 2010 05:03:49 +0000 (08:03 +0300)]
sdp: rewrite orphan count logic

orphan count is increased on every call to sdp_common_release and decreased on
every call to sdp_destruct (just before the socket is freed).

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: change socket reference semantics: keepalive != alive
Eldad Zinger [Wed, 21 Jul 2010 14:40:45 +0000 (17:40 +0300)]
sdp: change socket reference semantics: keepalive != alive

keepalive timer manages a socket reference count regardless to the initial
socket reference (SOCK_REF_ALIVE).

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: remove unnecessary argument from sdp_connected_handler.
Eldad Zinger [Wed, 21 Jul 2010 14:05:33 +0000 (17:05 +0300)]
sdp: remove unnecessary argument from sdp_connected_handler.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: removed extra debug message and comment.
Eldad Zinger [Tue, 20 Jul 2010 11:43:26 +0000 (14:43 +0300)]
sdp: removed extra debug message and comment.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fix for socket refcnt when error is marked while in TCP_TIME_WAIT state
Eldad Zinger [Tue, 20 Jul 2010 11:32:44 +0000 (14:32 +0300)]
sdp: fix for socket refcnt when error is marked while in TCP_TIME_WAIT state

When socket is in TCP_TIME_WAIT state, the next sdp_destroy_work() call will
put the cma_refcnt, so we should not change the socket state even if there is
an error.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: enable support for ib devices that do not support fmr
Eldad Zinger [Tue, 20 Jul 2010 09:45:02 +0000 (12:45 +0300)]
sdp: enable support for ib devices that do not support fmr

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: before activating rx_ring timer, need to check that qp is still active.
Eldad Zinger [Wed, 14 Jul 2010 08:20:19 +0000 (11:20 +0300)]
sdp: before activating rx_ring timer, need to check that qp is still active.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: fix compilation warnings
Eldad Zinger [Wed, 14 Jul 2010 07:14:14 +0000 (10:14 +0300)]
sdp: fix compilation warnings

use %z for size_t in printk's

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: define SDP_MAX_PAYLOAD as ulong instead of int, to comply with PAGE_SIZE
Eldad Zinger [Wed, 14 Jul 2010 04:50:43 +0000 (07:50 +0300)]
sdp: define SDP_MAX_PAYLOAD as ulong instead of int, to comply with PAGE_SIZE

SDP_MAX_PAYLOAD is compared a lot to PAGE_SIZE by min(), and on some kernels a
compilation warning of "comparison of distinct pointer types lacks a cast"
appears.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: some small non-functional changes.
Eldad Zinger [Mon, 12 Jul 2010 13:18:07 +0000 (16:18 +0300)]
sdp: some small non-functional changes.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: enable rx_cq arming when no one polls.
Eldad Zinger [Mon, 12 Jul 2010 12:59:57 +0000 (15:59 +0300)]
sdp: enable rx_cq arming when no one polls.

No need to explicitly arm the cq, posts_handler_put will arm it.

9 years agosdp: before arming cq, need to check if cq was not destroyed already.
Eldad Zinger [Sun, 11 Jul 2010 09:14:20 +0000 (12:14 +0300)]
sdp: before arming cq, need to check if cq was not destroyed already.

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
9 years agosdp: When purging tx_ring, rdma_inflight accountings should be disregarded, so the...
Eldad Zinger [Thu, 8 Jul 2010 13:37:24 +0000 (16:37 +0300)]
sdp: When purging tx_ring, rdma_inflight accountings should be disregarded, so the number of skbs to free is just (posted=head-tail).

Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>