tzachid [Wed, 10 Mar 2010 10:04:03 +0000 (10:04 +0000)]
[nd] Don't use inline for RDMA reads. (mlnx 5666)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2727
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Wed, 10 Mar 2010 09:45:19 +0000 (09:45 +0000)]
[ibal] MAD completion callbacks are expected to be at DISPATCH_LEVEL. (mlnx 5654)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2726
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Mon, 8 Mar 2010 14:23:15 +0000 (14:23 +0000)]
[AL] Replace spaces with tabs. (mlnx 5652)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2725
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Mon, 8 Mar 2010 14:01:46 +0000 (14:01 +0000)]
[AL] Replace spaces with tabs. (mlnx 5650)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2724
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Fri, 5 Mar 2010 00:49:41 +0000 (00:49 +0000)]
[DAPL2] recreate WinOF -->OFED mods which were recently overwritten.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2723
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Thu, 4 Mar 2010 22:13:53 +0000 (22:13 +0000)]
dapl2: update
Update to commit
4b939076aa32bb52957fcc6791e187c9433d4c24 + ibal provider fixes for WOF2-2 support. This synchronizes the trunk with the WOF2-2 branch.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2722
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Wed, 3 Mar 2010 23:39:36 +0000 (23:39 +0000)]
[TOOLS] make -h option work by adding h to getopt() options string.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2717
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Fri, 19 Feb 2010 18:13:20 +0000 (18:13 +0000)]
[IPoIB_ndis6_cm] migrate IPoIB shutter code into the trunk - match 2.2 release.
shutter code stops IPoIB from receiving packets during device shutdown.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2710
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Thu, 18 Feb 2010 23:02:12 +0000 (23:02 +0000)]
ibat: replace MAXINT64 with value
MAXINT64 is not defined for all windows platforms. Use a hard-coded value instead.
Signed-off-by: Sean Hefty <sean.hefty@intel.com.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2704
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Thu, 18 Feb 2010 21:51:19 +0000 (21:51 +0000)]
ibat/resolve: retry ibat resolution
Winverbs ND scale out testing showed that IBAT::Resolve() can
return E_PENDING, which requires that the resolution be retried.
A similar issue to this was seen when testing with the librdmacm.
Rather than duplicating retry logic in the winverbs ND provider,
add new functionality to ibat, with retry capability. To
avoid breaking the ibat.dll interface, extend the API with a
new call ResolvePath() that takes a timeout value.
ResolvePath() automatically retries Resolve() while the result
is E_PENDING, until the request times out. Modify the winverbs
ND provider to call ResolvePath(). Also update other places
where Resolve() is called in a loop: the librdmacm and wsd.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2703
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Thu, 18 Feb 2010 21:50:12 +0000 (21:50 +0000)]
libibverbs: release WVProvider reference from ibvw_acquire_windata
ibvw_release_windata should release the WVProvider reference taken
in ibvw_acquire_windata.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2702
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Thu, 18 Feb 2010 21:49:14 +0000 (21:49 +0000)]
dapl: locking cleanup and fixes
Cleanup allocated completion channels. Destroy cm_ptr locks before freeing the cm_ptr to avoid memory leaks. And avoid accessing the cm_ptr after queuing it for destruction with the cr_thread to avoid use after free errors.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2701
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Tue, 2 Feb 2010 01:12:41 +0000 (01:12 +0000)]
[OPENSM] OFED changes - rename %ProgramFiles%\WinOF --> %ProgramFiles%\OFED.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2695
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Mon, 1 Feb 2010 18:13:47 +0000 (18:13 +0000)]
[OFED] switch references to WinOF -- >OFED.
Support target-OS & target-arch in build scripts & etc\clean-build.bat.
Migrate WIX package definitions from ofed.wxs files to new file common\Package.inc (easier maintenance).
DAT.conf,dt-cli.bat,dt-svr.bat: WinOF --> OFED as binaries are now installed into %ProgramFiles%\OFED\ .
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2694
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Mon, 1 Feb 2010 02:43:51 +0000 (02:43 +0000)]
[OFED] OFED name change wof.wxs --> ofed.wxs, WinOF_cfg.inc --> Config.inc
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2693
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Mon, 1 Feb 2010 02:31:33 +0000 (02:31 +0000)]
[NDINSTALL] utilize getopt() changes to implement '/?' help request.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2692
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Mon, 1 Feb 2010 02:29:19 +0000 (02:29 +0000)]
[OPENSM] whitespace cleanup
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2691
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Mon, 1 Feb 2010 02:25:09 +0000 (02:25 +0000)]
[OFED] rename trunk\WinOF trunk\OFED
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2690
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Fri, 29 Jan 2010 05:07:01 +0000 (05:07 +0000)]
libibverbs/device: destroy completion channel when closing device
From: Sean Hefty <sean.hefty@intel.com>
ibv_close_device should destroy the completion channel
when closing the device and freeing the memory.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2689
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Fri, 29 Jan 2010 05:06:11 +0000 (05:06 +0000)]
dapl: use private_data_len for mem copies
From: Sean Hefty <sean.hefty@intel.com>
When copying private_data out of rdma_cm events, use the
reported private_data_len for the size, and not IB maximums.
This fixes a bug running over the librdmacm on windows, where
DAPL accessed invalid memory.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2688
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Fri, 29 Jan 2010 05:05:03 +0000 (05:05 +0000)]
dapl/cma: fix referencing freed address
From: Sean Hefty <sean.hefty@intel.com>
DAPL uses a pointer to reference the local and remote addresses
of an endpoint. It expects that those addresses are located
in memory that is always accessible. Typically, for the local
address, the pointer references the address stored with the DAPL
HCA device. However, for the cma provider, it changes this pointer
to reference the address stored with the rdma_cm_id.
This causes a problem when that endpoint is connected on the
passive side of a connection. When connect requests are given
to DAPL, a new rdma_cm_id is associated with the request. The
DAPL code replaces the current rdma_cm_id associated with a
user's endpoint with the new rdma_cm_id. The old rdma_cm_id is
then deleted. But the endpoint's local address pointer still
references the address stored with the old rdma_cm_id. The
result is that any reference to the address will access freed
memory.
Fix this by keeping the local address pointer always pointing
to the address associated with the DAPL HCA device. This is about
the best that can be done given the DAPL interface design.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2687
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Fri, 29 Jan 2010 05:04:08 +0000 (05:04 +0000)]
librdmacm: set private_data_len
From: Sean Hefty <sean.hefty@intel.com>
Set the private_data_len for reported events. This allows
DAPL to use the value when copying private data.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2686
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Fri, 29 Jan 2010 04:58:32 +0000 (04:58 +0000)]
dapl: move close device after async thread is done using it
Before calling ibv_close_device, wait for the asynchronous
processing thread to finish using the device. This prevents
a use after free error.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2685
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Tue, 26 Jan 2010 10:09:53 +0000 (10:09 +0000)]
[core] Improved error message. (mlnx 5511)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2683
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Tue, 26 Jan 2010 10:01:33 +0000 (10:01 +0000)]
[core] Release the RDMA interface if ib_register_ca fails. (mlnx 5510)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2682
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
(\18¬ [Thu, 21 Jan 2010 06:09:33 +0000 (06:09 +0000)]
ib/mad: fix routing of vendor mads
SVN commit 2174 introduced an error that resulted in all
vendor MADs being routed to the local HCA driver.
This results in the ib-diag vendstat failing to receive
a response when trying to gather statistics about a remote
device.
We should only route vendor mads to the local HCA if the
mad is one of the mellanox vendor classes, the mad is not
a response, and the local HCA is the destination for the
mad.
Problem reported by Mohammad Sawalha.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2675
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
(\18¬ [Thu, 21 Jan 2010 06:09:19 +0000 (06:09 +0000)]
winverbs/ep: handle receiving REQ then DREQ
For fast connections, it's possible to receive a DREQ immediately
after receiving a REQ, without an RTU coming in between. If we've
sent a REP to the REQ, then the DREQ should be treated as if the
connection had been fully established. (The RTU could be delayed,
and the communication established event is processed asynchronously,
so there's no way to tell for certain.)
This fixes an issue where the passive side Accept() call fails
waiting for the RTU, but receives a DREQ instead.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2674
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
port.2.tmp [Thu, 21 Jan 2010 06:09:06 +0000 (06:09 +0000)]
docs/interfaces.txt: add documentation to describe interface differences
Provide a simple documentation that defines the main userspace
verbs interfaces and the differences between them.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2673
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
(\18¬ [Thu, 21 Jan 2010 06:08:53 +0000 (06:08 +0000)]
winverbs/lib: cache pkey table
Connection establishment rates take a huge performance hit
as a result of using FindPkey, which results in querying the
HCA to lookup current pkey information. Add a cache to the
userspace library for pkey information.
The cache is updated using the CWVDevice:Notify() routine.
Before any access is made to the cache, the caller checks
for changes to the partition tables. If a change is found,
then the cache is updated. Otherwise, the cache is accessed
to map pkey values to indices.
This results in the connection rate over winverbs almost
doubling.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2672
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Thu, 21 Jan 2010 06:08:38 +0000 (06:08 +0000)]
libibumad: fix freeing wrong memory on failure
umad_get_ca() should free the allocated 'ports' memory on failure,
not the ca structure, which is provided by the user.
Bug reported by Leonid Keller.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2671
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
port.2.tmp [Fri, 15 Jan 2010 17:14:46 +0000 (17:14 +0000)]
winverbs: delete critical sections when no longer needed
To avoid leaking memory, we need to call DeleteCriticalSection
for all critical sections once they are no longer needed.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2669
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Fri, 15 Jan 2010 17:14:20 +0000 (17:14 +0000)]
etc/getopt: allow use of '/' for command line args
Windows uses '/' to specify command line arguments. Change
getopt to allow using '-' or '/'.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2668
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
LOCALS~1\Temp/report.7.tmp [Wed, 13 Jan 2010 19:34:38 +0000 (19:34 +0000)]
winverbs/nd: map completion status to ND values
Convert flushed completion status values
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2665
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
LOCALS~1\Temp/report.7.tmp [Wed, 13 Jan 2010 19:34:27 +0000 (19:34 +0000)]
winverbs/netdirect: add support for inline sends
By default, use inline sends, and allow user to limit
inline send size through an environment variable.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2664
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
LOCALS~1\Temp/report.7.tmp [Wed, 13 Jan 2010 19:34:17 +0000 (19:34 +0000)]
ib/cm: poll for CM REQ events
Replace the callback mechanism for reporting connection
requests with one that requires the user to poll for the
events. This allows queuing REQs in the CM until the user
is ready to process the events.
Still provide a callback mechanism to notify the user that
REQ events are ready to be retrieved.
This change improves the connection rate for winverbs when
the user retrieves only a small number of requests at a time.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2663
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
LOCALS~1\Temp/report.7.tmp [Wed, 13 Jan 2010 19:34:07 +0000 (19:34 +0000)]
ib/cm: separate listen callback into own handler
Extract listen handling in cm_cep_handler into a separate
handler. This will simplify changes to support queuing
connection requests and allow connection requests to move
to a fetch model, rather than a callback model.
Since connection requests require allocating new resources,
connection rates can be improved by queuing the requests
until the user has allocated the resources, then fetches
the request.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2662
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Wed, 13 Jan 2010 19:33:56 +0000 (19:33 +0000)]
ib/cm: fix handling of REJ in REQ_RCVD state
If a REJ is received immediately after receiving a REQ,
then a callback can be queued to the user for the REJ before
the user has had a chance to see or process the REQ. The
result is that the user will not have had a chance to set
any context associated with the REQ, which can lead to a
crash processing the REJ.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2661
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Wed, 13 Jan 2010 18:40:25 +0000 (18:40 +0000)]
[IPOIB_NDIS6_CM] fix the improper reporting of IPoIB port rate and removes obsolete OID_GEN_LINK_SPEED
signed-off by: Alexander Naslednikov (xalex at mellanox.co.il)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2658
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Wed, 13 Jan 2010 17:39:56 +0000 (17:39 +0000)]
[IPOIB/IPOIB_NDIS6_CM] fix the problem when IPoIB increments the receive statistics twice.
Signed-off by: Tzachi Dar (tzachid at mellanox.co.il)
Signed-off by: Alex Naslednikov (xalex@mellanox.co.il)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2657
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Wed, 13 Jan 2010 00:39:37 +0000 (00:39 +0000)]
[IPOIB] install the ND/winverbs (wvndprov.dll) provider along with the ND/ibal provider.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2656
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Tue, 5 Jan 2010 18:51:23 +0000 (18:51 +0000)]
[IPoIB_NDIS6_CM] fix the wrong length calculation of
first MDL within Net Buffer object.
The actual length of first MDL should be incremented by it's offset
signed-off by: Alexander Naslednikov (xalex at mellanox.co.il)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2651
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
E~1\svn\LOCALS~1\Temp/report.7.tmp [Mon, 4 Jan 2010 19:01:09 +0000 (19:01 +0000)]
ib/cm: fix handling failed send completions
__cep_mad_send_cb() assumes that the mad being processed is
associated with the current state of the CEP. This may not be
the case.
For example, for a short lived connection, it was observed that
a REP mad completed with status canceled. This is normal. However,
the user already attempted to disconnect the connection by sending
a DREQ. This left the cep in the DREQ_SENT state by the time that
the REP mad completed. Since the REP failed, but the state was
DREQ_SENT, the code assumed that the DREQ had failed and transitioned
the cep into TIMEWAIT. The result is that the DREQ is never
matched with a DREP or canceled, but holds a reference on the CEP.
Until the DREQ times out (time depends on connection, but easily
up to a minute), attempts to destroy the CEP are blocked.
Fix this by simply discarding any completed sends that were not
sent from the current state of the cep when the completion handler
is invoked.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2650
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
KMDF tracks all requests that pass through an IO queue. Even after a
request has been removed from the queue, the request still maintains a
reference on the queue. Any attempt to delete the queue will block
until all requests holding references on the queue have completed.
To avoid deadlock conditions during cleanup, we need to ensure that
all requests can complete during cleanup. Modify the asynchronous
handling code to queue a separate data structure, so that all requests
can remain on the IO queues.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2649
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Mon, 4 Jan 2010 19:00:38 +0000 (19:00 +0000)]
etc/bld: allow specifying x32 for x86 builds
To make it easier on the user, allow specifying a 32-bit build
using x32 or x86. x32 is more natural with x64 being the
64-bit build.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2648
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Mon, 4 Jan 2010 17:34:19 +0000 (17:34 +0000)]
[mlx4_bus] Add support for RoCEE to the low level driver.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2647
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Tue, 29 Dec 2009 21:31:52 +0000 (21:31 +0000)]
[DOCS] manual.htm - Corrected uDAT/uDAPL discussion as uDAPL v1.1 provider is no longer supported. Fixed lingering %ProgramFiles(x86)% install path problems; removed (x86) as WinOF is installed as %ProgramFIles%\ on all systems. Updated SRP discussion to include a how to configure a Linx/OFED SRP target (vdisk+BLOCKIO) accessible from Windows.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2646
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Thu, 24 Dec 2009 00:40:46 +0000 (00:40 +0000)]
[OPENSM] Temp fix for broken SRP operation; revert back to working code. Real fix touches numerous files, will be delayed until after 2.2 release.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2645
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Thu, 24 Dec 2009 00:36:08 +0000 (00:36 +0000)]
[OPENSM] added syslog() call to record Windows Service control calls (128,129) [clear log file & start heavy sweep].
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2644
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Thu, 24 Dec 2009 00:26:19 +0000 (00:26 +0000)]
[DAPL2] tolerate a disconnect callback racing ahead during EP disconnect(), do not return an error on invalid handle.
update .rc file to reflect IBAL provider.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2643
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Tue, 15 Dec 2009 20:48:55 +0000 (20:48 +0000)]
[WINOF] reference/install OpenSM related files from a single file instead of OpenSM_service.inc & tools.inc.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2641
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Tue, 15 Dec 2009 20:39:53 +0000 (20:39 +0000)]
[OPENSM] added OpenSM version to osmtest.rc file description string for easy ID. osmtest.rc was not being compiled in, added osmtest.rc file to SOURCES file.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2640
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Sat, 12 Dec 2009 00:57:11 +0000 (00:57 +0000)]
[OPENSM] set opensm cache & config folder as %ProgramFiles%\WinOF\OpenSM; integrated with WinOF installers.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2633
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Fri, 11 Dec 2009 21:14:34 +0000 (21:14 +0000)]
[WINOF] introduce new cmds 'allnf path' & 'compnf path' to work around a bug in the WDK build env. If one forces a build (compf ulp\ipoib) for IPoIB or IPoIB_NDIS6_CM then 'all' instances of ipoib.sys are deleted and only the specified path component is rebuild leaving an incomplete build. 'compnf/allnf ulp\ipoib' allows nmake dependancies to make the correct rebuild descision without deleting all instances of ipoib.sys.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2632
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Fri, 11 Dec 2009 21:07:33 +0000 (21:07 +0000)]
[WINOF] fully support Windows7 builds; reanme windows7 distribution folder to include Server 2008 R2
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2631
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Fri, 11 Dec 2009 21:05:02 +0000 (21:05 +0000)]
[WINOF] optionally sign drivers only for a specific OS & processor architecture.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2630
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Thu, 10 Dec 2009 00:15:41 +0000 (00:15 +0000)]
[OPENSM] Windows variants of printf/snprintf do not support '%zu', condense to '%u'.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2629
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Mon, 7 Dec 2009 23:13:43 +0000 (23:13 +0000)]
[OPENSM] shed some light on the 'how and why' of having seemingly duplicate files in the opensm src tree.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2628
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Mon, 7 Dec 2009 23:01:20 +0000 (23:01 +0000)]
[OpenSM] OFED version 3.3.3 ported to Windows - vendor == IBAL for now; umad in the works.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2627
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Mon, 7 Dec 2009 22:29:57 +0000 (22:29 +0000)]
[OPENSM] remove files no longer used in opensm build.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2626
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Mon, 7 Dec 2009 18:06:48 +0000 (18:06 +0000)]
[OPENSM] added version number to FileDescription string for easy identification.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2624
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Mon, 7 Dec 2009 11:24:16 +0000 (11:24 +0000)]
[IPoIB NDIS 6.0 CM] (mlnx 5232)
Bug at s_buf->p_endpt initialization
Debug prints for Endpoint Communication
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2623
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Mon, 7 Dec 2009 11:23:12 +0000 (11:23 +0000)]
[IPoIB NDIS 6.0 CM] (mlnx 5220)
Access to all CL lists should be lock-protected.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2622
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Thu, 3 Dec 2009 20:32:45 +0000 (20:32 +0000)]
winverbs/ND: fix calculating RDMA read/write address
The RDMA address should be base + offset, in network byte order. Rather
than trying to convert the offset into network byte order, then add that
to the base, which is already in network byte order, we need to convert
the base to host order, add the offset, then convert the results back.
This allows the winverbs ND provider to pass the ndmw test.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2619
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Thu, 3 Dec 2009 16:12:28 +0000 (16:12 +0000)]
[mlx4_bus] Fix for vlan handling. (mlnx 5160)
sign off by: galina
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2618
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Wed, 2 Dec 2009 09:51:33 +0000 (09:51 +0000)]
[IPoIB NDIS 6.0 CM] Use correct branding for inx file.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2617
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Wed, 2 Dec 2009 09:41:13 +0000 (09:41 +0000)]
[IPoIB NDIS 6.0 CM] (mlnx 5191)
Fixing 2 bugs:
1. Bug caused by NDIS 6.5 offloadmisc WHQL test: it assumes that the driver version is 6.2 and sends structs with bigger size
2. Internal bug at offload handling.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2616
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Wed, 2 Dec 2009 09:35:40 +0000 (09:35 +0000)]
[IPOIB_NDIS6_CM] Build fixes (mlnx 5193)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2615
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Wed, 2 Dec 2009 09:28:59 +0000 (09:28 +0000)]
[IPOIB_NDIS6_CM] g_stat: added to the adapter structure two previously global counters for multi-homed machines. (mlnx 5192)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2614
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Tue, 1 Dec 2009 16:43:52 +0000 (16:43 +0000)]
[IPoIB NDIS 6.0 CM] Return correct status for an unsupported OIDs. (mlnx 5214)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2613
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Tue, 1 Dec 2009 16:42:37 +0000 (16:42 +0000)]
[IPoIB NDIS 6.0 CM] Return success value even if NdisReadNetworkAddress fails (the default)(mlnx 5212)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2612
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Tue, 1 Dec 2009 16:40:42 +0000 (16:40 +0000)]
[IPoIB NDIS 6.0 CM] Added logging facilities. (mlnx 5211).
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2611
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Tue, 1 Dec 2009 16:39:27 +0000 (16:39 +0000)]
[IPoIB] [IPoIB NDIS 6.0 CM] Added support for GUID and fix debug 00:25:90 (mlnx 5210)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2610
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Tue, 1 Dec 2009 16:37:01 +0000 (16:37 +0000)]
[IPoIB NDIS 6.0 CM]
Bug at LSO flow.
When header is contained in 2 first SG Elements, or the total number of the SG List > MAX_SGE_NUM,
we enter send_copy flow. So, the p_send_buf should be at least 64K. (mlnx 5209)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2609
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Tue, 1 Dec 2009 16:35:45 +0000 (16:35 +0000)]
[IPOIB] fixing bugs in previous commit. (mlnx 5205)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2608
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Tue, 1 Dec 2009 16:32:43 +0000 (16:32 +0000)]
[IPoIB NDIS 6.0 CM] Tracing facilities (mlnx 5208)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2607
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
leonidk [Tue, 1 Dec 2009 13:08:20 +0000 (13:08 +0000)]
[SRP] revert the changes, occasionally made in 2605
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2606
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Mon, 30 Nov 2009 15:46:11 +0000 (15:46 +0000)]
[IPoIB NDIS 6.0 CM] Fixing a bug when ipoib_get_adapter_params fails reading parameter but returns NDIS_STATUS_SUCCESS. (mlnx 5190)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2605
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Mon, 30 Nov 2009 15:41:53 +0000 (15:41 +0000)]
[IPOIB] g_stat: added more fields: prev_port and halt thread.
When IPoIB is stuck, it is usually found in the halt thread, waiting for resources reclamation. (mlnx 5188)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2604
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Mon, 30 Nov 2009 15:39:35 +0000 (15:39 +0000)]
[IPoIB NDIS 6.0 CM] Fixing a bug when IPoIB does not dereference port object during shutdown progress. (mlnx 5185)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2603
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Mon, 30 Nov 2009 15:36:34 +0000 (15:36 +0000)]
[IPoIB NDIS 6.0 CM] Fixing a possible race when we get PORT_REMOVE before PORT_DOWN. (mlnx 5184)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2602
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Mon, 30 Nov 2009 15:34:19 +0000 (15:34 +0000)]
[IPoIB NDIS 6.0 CM] Fixing the problem when MCast packet stuck the pending list. (mlnx 5162)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2601
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Mon, 30 Nov 2009 15:32:24 +0000 (15:32 +0000)]
[IPoIB NDIS 6.0 CM] Fixing a possible race when we can destroy IPoIB without destroying and completing its pending list. (mlnx 5166)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2600
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Mon, 30 Nov 2009 15:22:32 +0000 (15:22 +0000)]
[IPoIB NDIS 6.0 CM]
Upgrade revision to NDIS 6.1 because of bug at WHQL tests:
When running NDIS6.5 test on 6.0 driver, it starts to execute VMQ tests that not supported at IPoIB driver.
(mlnx 5136)
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2599
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Wed, 25 Nov 2009 00:01:00 +0000 (00:01 +0000)]
dapl: sync with 2.0.25
commit
0983c66cbd9511128c1fa221470c4c983903e420
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2598
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Mon, 23 Nov 2009 20:13:48 +0000 (20:13 +0000)]
[DAPL,ETC,WINOF] Remove DAPL 1.2 from WinOF installer: do not copy DAPL 1.2 files in makebin; remove DAPL 1.2 options from DAT config file.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2596
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Mon, 23 Nov 2009 20:10:16 +0000 (20:10 +0000)]
[WINOF] switch from WIX 2.0 to use the stable release of WIX 3.0 for building WinOF installers.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2595
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Mon, 23 Nov 2009 18:59:00 +0000 (18:59 +0000)]
[DAPL2] synchronize with OFED DAPL git tree.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2594
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
shefty [Mon, 23 Nov 2009 18:22:11 +0000 (18:22 +0000)]
etc/.bat: update pkg and sign batch files
include dpal pkg and timestamp with signing
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2591
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
leonidk [Wed, 18 Nov 2009 09:27:30 +0000 (09:27 +0000)]
[MLX4] fix a bug entered by incorrect merge
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2588
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Wed, 18 Nov 2009 07:45:36 +0000 (07:45 +0000)]
[mlx4] Use all entries in the mac table.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2587
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
stansmith [Wed, 18 Nov 2009 01:10:54 +0000 (01:10 +0000)]
[COMPLIB] Cleanup code by using C99 standard type uintptr_t in preference over complib defined non-standard data type uintn_t.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2586
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
leonidk [Tue, 17 Nov 2009 15:03:51 +0000 (15:03 +0000)]
[MLX4] g_stat: added more prints and flags
added support for g_stat structure for MLX4_BUS driver + MAD tracing mechanism.
MAD tracing mechanism works under the control of flags, set in new Registry mlx4_bus parameter StatFlags.
The mechanism prints to debugger the IB headers of packets sent over MLX transport, i.e. on QP0 and QP1. [mlnx: 5138]
The flags are:
0x0001 - print LRH
0x0002 - print BTH
0x0004 - print DETH
0x0008 - print GRH (it won't print id the GRH is absent)
0x0010 - print some WQE info
0x0020 - print some more UD header info
0x0040 - print some send WR info
0x0080 - MLX WQE dump
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2585
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
leonidk [Tue, 17 Nov 2009 14:37:35 +0000 (14:37 +0000)]
[MLX4] added events for Registry handling error flow. [mlnx: 5058]
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2584
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
leonidk [Tue, 17 Nov 2009 14:32:57 +0000 (14:32 +0000)]
[MLX4] syntax improvements
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2583
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
leonidk [Tue, 17 Nov 2009 14:29:08 +0000 (14:29 +0000)]
[MLX4] improve debug in error flow of command interface. [mlnx: 5039]
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2582
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
leonidk [Tue, 17 Nov 2009 14:26:45 +0000 (14:26 +0000)]
[MLX4] added initialization of port type to default values and improved debug prints. [mlnx: 5025]
We come across a case, when Port type from Registry has not been read.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2581
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
leonidk [Tue, 17 Nov 2009 14:12:21 +0000 (14:12 +0000)]
[IPOIB, IBBUS] g_stat: first patch of several, adding and populating global g_stat structure to the drivers of IB stack
For debug purposes.
Usable in both checked and free version.
Usage: open Watch Window in Debugger and add ipoib!g_stat or ibbus!g_stat.
An example of a real problem: MLX4 driver is stuck on unload.
It can happen when IBAL is stuck on resource reclamation.
The reclamation is being done in one of IBAL threads, which are started on IBAL's start up.
So you'd like to look into this thread.
But where it is ?
You can find it by printing ALL the threads of System process and looking for ibbus in their stack traces.
It can take minutes.
After this patch you can do that in 3 seconds this way:
1. open Watch Window in WinDbg and add ibbus!g_stat.
2. open
ibbus!g_stat->Drv->Gp_async_obj_mgr->Thread_pool->P_thread[0]->Osd->P_thread
This field contains the address of the thread, you wanted.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2580
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Tue, 17 Nov 2009 13:38:42 +0000 (13:38 +0000)]
[wsd] Fix a typo.
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2579
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Tue, 17 Nov 2009 13:13:40 +0000 (13:13 +0000)]
[bus] fix bug in __read_dev_params.
signed off by: Firas@mellanox.co.il
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2578
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86
tzachid [Tue, 17 Nov 2009 13:05:20 +0000 (13:05 +0000)]
[bus] fix minor issue found in prefast.
Signed off by: urih@mellanox.co.il
git-svn-id: svn://openib.tc.cornell.edu/gen1/trunk@2577
ad392aa1-c5ef-ae45-8dd8-
e69d62a5ef86