- 11 Mar, 2016 40 commits
-
-
Jubin John authored
Fill in srq function stubs with code derived from hfi1 and qib. Move necessary functions and data structure members as well. Reviewed-by:
Dennis Dalessandro <dennis.dalessandro@intel.com> Reviewed-by:
Harish Chegondi <harish.chegondi@intel.com> Signed-off-by:
Jubin John <jubin.john@intel.com> Signed-off-by:
Doug Ledford <dledford@redhat.com>
-
Harish Chegondi authored
Rely on rvt_query_qp function defined in rdmavt Reviewed-by:
-
Harish Chegondi authored
Multicast is now supported by rdmavt. Remove the verbs multicast functions and use that. Reviewed-by:
-
Harish Chegondi authored
This patch removes the simple post recv function in favor of using rdmavt. The packet receive processing still lives in the driver though. Reviewed-by:
-
Harish Chegondi authored
This patch removes the post_send and post_one_send from the qib driver. The "posting" of sends will be done by rdmavt which will walk a WQE and queue work. This patch will still provide the capability to schedule that work as well as kick the progress. These are provided to the rdmavt layer. Reviewed-by:
-
Harish Chegondi authored
Use the completion queue functionality provided by rdmavt. Reviewed-by:
-
Dennis Dalessandro authored
Get rid of create and free mad agent from the driver and use rdmavt version. Reviewed-by:
Ira Weiny <ira.weiny@intel.com> Signed-off-by:
-
Dennis Dalessandro authored
No longer do drivers need to call into the IB core to allocate the verbs device. Use the functionality provided by rdmavt. Reviewed-by:
-
Harish Chegondi authored
Drivers using rdmavt can rely on rvt_query_qp instead of defining their own query_qp functions. Reviewed-by:
-
Dennis Dalessandro authored
Update all files added by rdmavt which do not yet have 2016 as the copyright year. Reviewed-by:
-
Dennis Dalessandro authored
This patch adds mad agent create and free to rdmavt. Reviewed-by:
-
Dennis Dalessandro authored
This patch adds rdmavt device structure allocation in rdamvt. The ib_device alloc is now done in rdmavt instead of the driver. Drivers need to tell rdmavt the number of ports when calling. A side of effect of this patch is fixing a bug with port initialization where the device structure port array was allocated over top of an existing one. Reviewed-by:
-
Ira Weiny authored
Low level drivers need to be able to check incoming attributes as well as be able to adjust their private data on queue pair modification. Add 2 driver callbacks, check_modify_qp and modify_qp, to facilitate this. Signed-off-by:
-
Ira Weiny authored
s_sde should be in the low level driver QP private data. Remove the definition from rvt_qp. Signed-off-by:
-
Dennis Dalessandro authored
There are a number of minor things that should be set by rdmavt rather than by the drivers. Now that rdmavt has solidified in its design we can go ahead and clean up this stuff. Reviewed-by:
-
Dennis Dalessandro authored
This patch adds in the multicast add and remove functions as well as the ancillary infrastructure needed. Reviewed-by:
Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by:
-
Dennis Dalessandro authored
This patch adds the simple post receive verbs call to rdmavt. The actual interrupt handling and packet processing is still done in the low level driver. Reviewed-by:
-
Dennis Dalessandro authored
This patch adds in support the qp destroy verb call. Reviewed-by:
-
Dennis Dalessandro authored
Add modify qp and supporting functions. Reviewed-by:
-
Dennis Dalessandro authored
This patch adds support of tracing events using the kernels built-in event tracing infrastructure. This can be extended to provide a wide range of trace and debug capabilities which have a negligible impact on performance when enabled. These should be preferred over the use of the rvt_pr* functions. Reviewed-by:
-
Dennis Dalessandro authored
Add in a post_send and post_one_send to rdmavt. The ULP will provide a WQE to rdmavt which will then walk and queue each element. Rdmavt will then queue the work to be done in the driver or kick the driver's progress routine. There needs to be a follow on patch which adds in another lock for the head of the queue so that it can be added to and read from in parallel. This will touch protocol handlers and require other changes in the drivers. This will be done separately. Reviewed-by:
-
Dennis Dalessandro authored
Brings in completion queue functionality. A kthread worker is added to the rvt_dev_info to serve as a worker for completion queues. Reviewed-by:
-
Harish Chegondi authored
Rely on rdmavt functions for creation of qp and qp table. Function to allocate a qpn is still being provided by qib as the algorithm to allocate a qpn in qib is different from that of the algorithm in rdmavt. Reviewed-by:
-
Harish Chegondi authored
Use the definitions of the s_flags and r_flags which are now in rdmavt. Reviewed-by:
-
Harish Chegondi authored
Removed qib_query_device function to use rdmavt rvt_query_device function The device attributes still need to be filled in by the driver. Reviewed-by:
-
Harish Chegondi authored
IB user context alloc and dealloc functions have been added to rdmavt. Delete the QIB user context alloc/dealloc functions and use the ones in rdmavt. Reviewed-by:
-
Harish Chegondi authored
This patch removes the private queue pair structure and the table which holds the queue pair numbers in favor of using what is provided by rdmavt. Reviewed-by:
-
Harish Chegondi authored
Remove qib query pkey function which is no longer needed as this is now being done in rdmavt. The allocation and maintenance of the list still resides in the driver. Reviewed-by:
-
Harish Chegondi authored
Since mmap functionality has been moved into rdmavt, its time for qib to use that. Reviewed-by:
-
Harish Chegondi authored
Additional work is required to create an AH. This patch adds support to set the VL correctly. Reviewed-by:
-
Harish Chegondi authored
Remove several ibport members from qib and use the rdmavt version. rc_acks, rc_qacks, and rc_delayed_comp are defined as per CPU variables in rdmavt. Add support for these rdmavt per CPU variables which were not per cpu variables in qib ibport structure. Reviewed-by:
-
Mike Marciniszyn authored
The current code is problematic when the QP creation and ipoib is used to support NFS and NFS desires to do IO for paging purposes. In that case, the GFP_KERNEL allocation within create_qp causes a deadlock in tight memory situations. This fix adds support to create queue pair with GFP_NOIO flag for connected mode only to cleanly fail the create queue pair in those situations. This was previously fixed in qib but needed to get ported to hfi1. This patch handles that for both hardwares in the new rdmavt common layer. Reviewed-by:
-
Harish Chegondi authored
With this commit, the drivers using rdmavt need not define query_device function. But they should fill in the IB device attributes structure rvt_dev_info.dparms.props Reviewed-by:
-
Harish Chegondi authored
qib needs to reserve only one qpn for non-verbs stuff. Also fixed the for loop to reserve the end qpn. Reviewed-by:
-
Dennis Dalessandro authored
Until all queue pair functionality is moved to rdmavt we need to provide access to the reset function. This is only temporary and will be reverted back to a static, non exported function in the end. Reviewed-by:
-
Dennis Dalessandro authored
Add create queue pair verbs call as well as supporting functions. Reviewed-by:
-
Dennis Dalessandro authored
Use the flags originally provided for hfi1 in the rdmavt driver. These will be made available to drivers in the qp header file. Reviewed-by:
-
Harish Chegondi authored
Adding IB user context alloc and dealloc functions to rdmavt so that the drivers that use rdmavt can use these functions instead of defining their own functions. Reviewed-by:
-
Dennis Dalessandro authored
Remove srq from qib now that it has been moved into rdmavt. Reviewed-by:
-
Dennis Dalessandro authored
Original patch from Kamal Heib <kamalh@mellanox.com>, split apart from original. Remove AH from qib and use rdmavt version. Reviewed-by:
Kamal Heib <kamalh@mellanox.com> Signed-off-by:
-
