]>
www.infradead.org Git - users/jedix/linux-maple.git/log
Eldad Zinger [Wed, 31 Mar 2010 13:02:03 +0000 (16:02 +0300)]
sdp: BUG1992 - enable transmission of credits update when tx_credits == 1
According to spec, if one credit is available, we can only send messages that
provide additional credits and also do not contain data payload.
Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
Eldad Zinger [Wed, 31 Mar 2010 12:27:38 +0000 (15:27 +0300)]
sdp: unnecessary 'if' statement canceled.
Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
Eldad Zinger [Thu, 18 Mar 2010 11:31:08 +0000 (13:31 +0200)]
sdp: BUG1727 - fixed select(2) behavior on a new nonblocking socket.
when calling for select(2) after socket(2) on a nonblocking socket,
select(2) should return 'writable'.
Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
Eldad Zinger [Wed, 17 Mar 2010 13:24:28 +0000 (15:24 +0200)]
sdp: BUG1727 - sdp_destroy_work() and sdp_connect() interfere with each other.
This fix was done in order to make sure that sdp_destroy_work() is done
before sdp_connect() is active.
Signed-off-by: Eldad Zinger <eldadz@mellanox.co.il>
Amir Vadai [Sun, 14 Mar 2010 14:36:32 +0000 (16:36 +0200)]
sdp: support iovlen > 1 in zcopy
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 14 Mar 2010 14:35:38 +0000 (16:35 +0200)]
sdp: make sdp_socket.h available to user applications
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 1 Apr 2010 07:28:37 +0000 (10:28 +0300)]
sdp: enable FMR pool cache
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 25 Feb 2010 22:50:29 +0000 (00:50 +0200)]
sdp: Stop SA Cancel timeout when getting SendSM/RdmaRdCompl
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 25 Feb 2010 09:43:03 +0000 (11:43 +0200)]
sdp: SendSM wasn't sent sometimes after getting SrcAvailCancel
* skb was freed if rx_sa is aborted - preventing SendSM
to be sent.
* Didn't update rx_sa->used in case of SrcAvailCancel
and therefore not sending RdmaRdCompl.
This also caused the next read to fail because offset
wasn't updated
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 24 Feb 2010 11:58:35 +0000 (13:58 +0200)]
sdp: Send SendSM from recvmsg context and not from interrupt
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 24 Feb 2010 08:59:31 +0000 (10:59 +0200)]
sdp: Fix bug in crossing SrcAvail
* Handle RdmaRdCompl in interrupt, before splitted to two Q's
This way the handling is sequencial, and no race could occure
between RdmaRdCompl and SrcAvailCancel
* Fixed an error when checking that RdmaRdCompl is not for
old SrcAvail
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 24 Feb 2010 08:39:54 +0000 (10:39 +0200)]
sdp: Add detailed ZCopy aborted send statistics
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 23 Feb 2010 08:02:59 +0000 (10:02 +0200)]
sdp: Prevent kernel crash if device init fails (plus bonus fix)
If sdp_add_device() fails, there is no client data stored in the IB device,
leading to a kernel crash when a connection is being established. Fix this
by rejecting connections when the device is not initialized.
Also, fix a bad goto target in an error case early in sdp_init_qp().
Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 18 Feb 2010 12:03:27 +0000 (14:03 +0200)]
sdp: Fix crossing SrcAvail handling
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 16 Feb 2010 15:36:00 +0000 (17:36 +0200)]
sdp: Fix bugs in huge paged HW's
* Protect some constants that are based on PAGE_SIZE:
- FMR size
- xmit_goal
* renamed SDP_HEAD_SIZE => SDP_SKB_HEAD_SIZE
* removed unneeded special IA64 code due to changes here
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 4 Feb 2010 15:31:52 +0000 (17:31 +0200)]
sdp: BUG1899 - fix warnings on RH4.8 by avoiding multiple deletions on the same timer.
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 24 Jan 2010 15:12:34 +0000 (17:12 +0200)]
sdp: must use ib_sg_dma_*, not sg_dma_* for mapping
This fixes OFED bug 1895, althoug some warnings are still generated,
when running qperf sdp_bw with large sizes (using zcopy), on the
truescale adapters.
Signed-off-by: Dave Olson <dave.olson@qlogic.com>
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 31 Dec 2009 09:25:39 +0000 (11:25 +0200)]
sdp: use IB_CQ_VECTOR_LEAST_ATTACHED for cq's
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 30 Dec 2009 13:43:10 +0000 (15:43 +0200)]
sdp: make statistics per cpu
- Caused lots of cache misses during multi stream traffic
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 30 Dec 2009 08:10:09 +0000 (10:10 +0200)]
sdp: added statistics instead of prints
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 27 Dec 2009 09:08:26 +0000 (11:08 +0200)]
sdp: Fix partial ZCopy send bug + recvmsg with MSG_PEEK support
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 27 Dec 2009 10:51:16 +0000 (12:51 +0200)]
sdp: Set a lower limit to ZCopy threshold
No need to use ZCopy, when the whole packet is sent as SrcAvail payload
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 22 Dec 2009 16:13:20 +0000 (18:13 +0200)]
sdp: Fix ZCopy sink not working
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 20 Dec 2009 11:23:28 +0000 (13:23 +0200)]
sdp: fixed compilation warning
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 20 Dec 2009 09:45:37 +0000 (11:45 +0200)]
sdp: ZCopy SrcAvail payload size limit fixed + fix ZCopy rx for small packets
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 13 Dec 2009 07:52:16 +0000 (09:52 +0200)]
sdp: small spec compliancy fixes in code
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Mon, 7 Dec 2009 13:31:58 +0000 (15:31 +0200)]
sdp: fix lockup on mthca cards
- rx_irq called sdp_sock_queue_rcv_skb() who calls
sk->sk_data_ready() == sock_def_readable
- sock_def_readable() can't be called from interrupt context
because it takes read_lock (the writer to the lock uses
write_lock_bh)
- moved rx processing to tasklet
sock_def_readable was trying to take read_lock
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 2 Dec 2009 14:35:51 +0000 (16:35 +0200)]
sdp: fix a warning in RH4.0
never del uninitialized timer
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 2 Dec 2009 13:32:03 +0000 (15:32 +0200)]
sdp: Fix Hello Ack Header to be according to spec
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 2 Dec 2009 07:06:37 +0000 (09:06 +0200)]
sdp: make /proc/net/sdpprf on only if debug data is on
This way, in performance measurements it is off and on regression
it is on.
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Mon, 30 Nov 2009 15:45:13 +0000 (17:45 +0200)]
sdp: Cleanup ZCopy page registrations
- use ib_umem_get instead of get_user_pages
- Use FMR for RDMA reads
- ZCopy sender side to use ib_umem
- send remainder of ZCopy with BCopy
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Mon, 30 Nov 2009 14:14:14 +0000 (16:14 +0200)]
sdp: cancel_work_sync on 2.6.22 didn't return a value
Since the return value was used for debug message only,
gave up this message to make backports easier.
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 29 Nov 2009 12:23:58 +0000 (14:23 +0200)]
sdp: changed important prints in zcopy to warning instead of debug
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 26 Nov 2009 15:10:57 +0000 (17:10 +0200)]
sdp: flush rx_comp_work before destroying socket + make dreq_wait_timeout_work flushing synced
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 26 Nov 2009 15:28:58 +0000 (17:28 +0200)]
sdp: cleaned up debug prints
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 25 Nov 2009 12:59:46 +0000 (14:59 +0200)]
sdp: take a reference during zcopy_send
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 25 Nov 2009 13:06:01 +0000 (15:06 +0200)]
sdp: Disable BZcopy + Enable ZCopy
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 25 Nov 2009 08:26:59 +0000 (10:26 +0200)]
sdp: changed some warnings into debug prints
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 24 Nov 2009 09:01:05 +0000 (11:01 +0200)]
sdp: fixed BUG1826 part 1 - schedule while atomic
Allocate skb according to context
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 24 Nov 2009 07:33:53 +0000 (09:33 +0200)]
sdp: Fixed annoying warning by memtrack
kzalloc done in sdp_seq_open is freed by the system in a function
that memtrack can't see. Therefore it printed false warning.
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 24 Nov 2009 07:32:39 +0000 (09:32 +0200)]
sdp: fixed BUG1796 - running out of memory on rx
rcv queue could grow endlessly because minimal RX buffers in QP
was set to SDP_MIN_TX_CREDITS + 1 - so there always were credits
available for the sender.
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Mon, 23 Nov 2009 11:29:41 +0000 (13:29 +0200)]
sdp: fixed sparse warnings
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 11 Oct 2009 10:16:27 +0000 (12:16 +0200)]
sdp: removed unneeded list initialization - percpu_counter might not have this memeber
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Mon, 7 Sep 2009 13:02:27 +0000 (16:02 +0300)]
sdp: get max send sge from device capabilities instead of hard coded
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Jack Morgenstein [Sun, 30 Aug 2009 14:16:13 +0000 (17:16 +0300)]
sdp: incorrect SDP_FMR_SIZE on 32-bit machines
On 32-bit machines, sizeof (u64 *) is 4 bytes (size of a ***pointer***).
However, the max SDP FMR pool size should be PAGE_SIZE / sizeof(an mtt entry) --
and mtt entries are u64's (or __be64's).
This resulted in SDP requesting twice as many entries per pool on 32-bit machines
as could fit on a single page -- with the result that the fmr pool allocation failed
at driver startup.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Jack Morgenstein [Sun, 30 Aug 2009 14:24:07 +0000 (17:24 +0300)]
sdp: check if sdp device is actually present in sdp_remove_one
If sdp fails to initialize at driver startup for any reason,
the device is still registered with the ib_core, but there will be
no client data (i.e., ib_set_client_data() will not be called, and all
kernel resources are de-allocated).
On removal, ib_get_client_data() will return NULL in this case -- and this
must be tested for -- or we will get a kernel Oops for a NULL pointer
dereference.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Amir Vadai [Sun, 23 Aug 2009 15:28:39 +0000 (18:28 +0300)]
sdp: Process tx completions from sendmsg context. arm tx cq when needed
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 20 Aug 2009 11:20:57 +0000 (14:20 +0300)]
sdp: More code cleanup and ZCopy bugs fixes
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Mon, 17 Aug 2009 10:37:18 +0000 (13:37 +0300)]
sdp: code cleanup
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 13 Aug 2009 13:00:40 +0000 (16:00 +0300)]
sdp: fix cross SrcAvail deadlock
The sending SrcAvail will give up and send SrcAvailCancel.
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 11 Aug 2009 10:57:48 +0000 (13:57 +0300)]
sdp: Fix ZCopy compatability issues
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 16 Aug 2009 13:07:28 +0000 (16:07 +0300)]
sdp: fix memory leak in bzcopy
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 9 Aug 2009 08:15:44 +0000 (11:15 +0300)]
sdp: fixed signedness warning in compilation + don't use getnstimeofday
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 29 Jul 2009 15:20:10 +0000 (18:20 +0300)]
sdp: split very big tx buffer into smaller sends
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 29 Jul 2009 15:08:43 +0000 (18:08 +0300)]
sdp: QP should be destroyed before its CQs
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Mon, 27 Jul 2009 10:34:35 +0000 (13:34 +0300)]
sdp: fix driver to accept credit updates after RCV_SHUTDOWN
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 12 Jul 2009 12:32:36 +0000 (15:32 +0300)]
sdp: Add support for ZCopy combined mode - RDMA Read
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 22 Jul 2009 07:36:01 +0000 (10:36 +0300)]
sdp: removed unnecessary statistics that caused compilation errors on powerpc
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 15 Jul 2009 09:08:06 +0000 (12:08 +0300)]
sdp: fix some warning and bugs in porting to ofed 1.5
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 1 Jul 2009 16:38:06 +0000 (19:38 +0300)]
sdp: fix bad credits advertised when connection initiated
This enables removing the ugly post credits after connection
establishment
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 1 Jul 2009 16:33:07 +0000 (19:33 +0300)]
sdp: Fix memory leak in bzcopy
Thanks to Lars Ellenberg for finding the leak
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 1 Jul 2009 14:00:29 +0000 (17:00 +0300)]
sdp: fix wrong credit advertised in Hello MID
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 28 Jun 2009 20:33:46 +0000 (23:33 +0300)]
sdp: fix compilation error on 2.6.30
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 21 Jun 2009 14:13:35 +0000 (17:13 +0300)]
sdp: fixed coding style
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 16 Jun 2009 08:04:50 +0000 (11:04 +0300)]
sdp: IB_CQ_VECTOR_LEAST_ATTACHED is not supported yet in 1.5 tree
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 9 Jun 2009 06:25:12 +0000 (09:25 +0300)]
sdp: fixed div by zero in sdpstats
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Mon, 8 Jun 2009 14:02:20 +0000 (17:02 +0300)]
sdp: make interrupt moderation adaptive
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 4 Jun 2009 14:46:19 +0000 (17:46 +0300)]
sdp: arm nagle timer on not sent packet instead of on sent packet
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 4 Jun 2009 14:42:13 +0000 (17:42 +0300)]
sdp: two bug fixes
fix bad HELLO/HELLO_ACK buffer size matching
let recvmsg do posts on data packets too
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 4 Jun 2009 10:52:34 +0000 (13:52 +0300)]
sdp: make bzcopy poll timeout in jiffies instead of iterations count
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 4 Jun 2009 09:28:33 +0000 (12:28 +0300)]
sdp: fix bad handling for not aligned buffers in bzcopy + removed poll at end of send
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 3 Jun 2009 06:34:49 +0000 (09:34 +0300)]
sdp: fix RX to work well on sink side + cosmetics changes
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 26 May 2009 16:16:56 +0000 (19:16 +0300)]
sdp: TX from 1 context only. RX with minimal context switches
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 24 May 2009 07:26:08 +0000 (10:26 +0300)]
sdp: don't arm nagle timer for every sent packet
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Mon, 18 May 2009 12:29:54 +0000 (15:29 +0300)]
sdp: remove leftover from debugging
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Mon, 18 May 2009 11:45:18 +0000 (14:45 +0300)]
sdp: Do not nagle BZCopy packets
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 14 May 2009 14:10:33 +0000 (17:10 +0300)]
sdp: don't do nagle on first packet
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 14 May 2009 11:54:07 +0000 (14:54 +0300)]
sdp: fix backports
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Wed, 13 May 2009 11:37:13 +0000 (14:37 +0300)]
sdp: process RX CQ from interrupt
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 26 Apr 2009 12:31:27 +0000 (15:31 +0300)]
sdp: created sdp_rx and sdp_tx
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 26 Apr 2009 07:55:26 +0000 (10:55 +0300)]
sdp: /proc/net/sdpprf - performance utilities
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 21 Apr 2009 14:18:37 +0000 (17:18 +0300)]
sdp: no tx interrupts
poll tx cq with timer instead of interrupts
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 21 Apr 2009 13:56:37 +0000 (16:56 +0300)]
sdp: move tx_ring into dedicated structre + many cosmetic fixes
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 21 Apr 2009 13:23:25 +0000 (16:23 +0300)]
sdp: fixed compilation error when statistics turned off
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 21 Apr 2009 13:07:23 +0000 (16:07 +0300)]
sdp: cosmetics changes
* renamed ssk->bufs into ssk->tx_credits
* renamed SDP_MIN_BUFS into SDP_MIN_TX_CREDITS
* get bz struct from skb in a macro
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 21 Apr 2009 12:38:27 +0000 (15:38 +0300)]
sdp: Interrupts performance fixes
* use rcvbuf_initial - fix server to use rcvbuf_initial_size as initial
* hw interrupt moderation
* fix interrupts statistics
* fix max frags bug
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Sun, 5 Apr 2009 15:22:45 +0000 (18:22 +0300)]
sdp: added /proc/net/sdpstats + packets dump
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 28 Apr 2009 06:40:10 +0000 (09:40 +0300)]
sdp: BUG1311 Netpipe fails with a IB_WC_LOC_LEN_ERR.
This problem is seen when the receive buffer or the receive buffer fragments
are smaller than the senders buffer. If the sender is using 64KB pages
and supports a sk fragment of 64KB it may send fragments that are a full 64KB
in length causing the receiver to generate an IB_WC_LOC_LEN_ERR. This patch
makes two changes:
If the kernel does not support a full 64KB fragment it will reject resize
requests over 32K. (On older kernels a fragment size is defined as a U16)
If a kernel supports a 64KB fragment then it allows a full 64KB receive
fragment to be used.
Signed-off-by: David Wilder <dwilder@us.ibm.com>
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Nicolas Morey-Chaisemartin [Wed, 29 Apr 2009 14:23:04 +0000 (16:23 +0200)]
sdp: change orphan_count and sockets_allocated from atomic_t to percpu_counter
Fixed SDP to work on 2.6.29+
As percpu_counter are huge they can be allocated on the stack without causing sdp module to crash.
Both variable are now dynamically allocated at module init.
Signed-off-by: Nicolas Morey-Chaisemartin <nicolas.morey-chaisemartin@ext.bull.net>
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 17 Mar 2009 15:38:48 +0000 (17:38 +0200)]
sdp: BUG1472 - clean socket timeouts and refcount when device is removed
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 10 Mar 2009 07:15:57 +0000 (09:15 +0200)]
sdp: fixed typo Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 17 Feb 2009 09:04:09 +0000 (11:04 +0200)]
sdp: BUG1502 - scheduling while atomic
Don't destory qp from interrupt context - do it in work queue
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 29 Jan 2009 06:40:09 +0000 (08:40 +0200)]
sdp: small typo fixed
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 8 Jan 2009 11:00:16 +0000 (13:00 +0200)]
SDP: BUG1309 - SDP close is slow + fix recv buffer initial size setting
sdp_post_recv is called when the QP is closed. therefore it tries to post
recv buffers without success till the socket is destructed.
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Mon, 8 Dec 2008 15:09:01 +0000 (17:09 +0200)]
SDP: BUG1087 - fixed recovery from failing rdma_create_qp()
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 2 Dec 2008 13:01:26 +0000 (15:01 +0200)]
SDP: Fix to limit max buffer size in sdp_resize_buffers on IA64
Fix for bug 1311
https://bugs.openfabrics.org/show_bug.cgi?id=1311
Signed-off-by: Nicolas Morey-Chaisemartin <nicolas.morey-chaisemartin@ext.bull.net>
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Tue, 2 Dec 2008 11:34:09 +0000 (13:34 +0200)]
sdp: BUG1429 - Sdp doesnt close resources
When called from sdp_dreq_timeout() - not holding ref count for
timewait exit - when IB is already teared down.
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Mon, 1 Dec 2008 12:49:07 +0000 (14:49 +0200)]
sdp: BUG1047 - crash in sdp_destroy_qp() when no memory
fix initialization of sdp socket special attributes after sk_clone()
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Mon, 17 Nov 2008 08:11:27 +0000 (10:11 +0200)]
SDP: BUG1391 - bugs in the zero-copy send code
* fix sdp_bz_setup() code to handle the case of kernel data segment correctly (kernel sockets)
* make sdp_bz_setup() pass ENOMEM, EFAULT or other errors to sendmsg().
* Fix: the deallocation of bz descriptor in sendmsg() is not handled properly -- it is allocated many times, but freed once
* Fix: sdp_bzcopy_get() code does not raise reference count for all pages in the bz descriptor (only the "partial" pages will get the count raised).
However, the send completion code will call put_page() on all entries, leading to a crash for page-aligned transfers.
Signed-off-by: Constantine Gavrilov <constantine.gavrilov@gmail.com>
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Amir Vadai [Thu, 20 Nov 2008 13:02:46 +0000 (15:02 +0200)]
SDP: BUG1348 - a socket is left after netper on the server side
Removed unneccessary sock_hold from sdp_shutdown. This ref count
will be taken by close(). no need to take it twice.
Signed-off-by: Amir Vadai <amirv@mellanox.co.il>