summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2016-12-14Merge branches 'chelsio', 'debug-cleanup', 'hns' and 'i40iw' into merge-testDoug Ledford
2016-12-12i40iw: Reorganize structures to align with HW capabilitiesHenry Orosco
Some resources are incorrectly organized and at odds with HW capabilities. Specifically, ILQ, IEQ, QPs, MSS, QOS and statistics belong in a VSI. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-12i40iw: Fix incorrect check for errorMustafa Ismail
In i40iw_ieq_handle_partial() the check for !status is incorrect. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-12i40iw: Assign MSS only when it is a new MTUMustafa Ismail
Currently we are changing the MSS regardless of whether there is a change or not in MTU. Fix to make the assignment of MSS dependent on an MTU change. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-12i40iw: Fix race condition in terminate timer's handlerShiraz Saleem
Add a QP reference when terminate timer is started to ensure the destroy QP doesn't race ahead to free the QP while it is being referenced in the terminate timer's handler. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-12i40iw: Fix memory leak in CQP destroy when in resetMustafa Ismail
On a device close, the control QP (CQP) is destroyed by calling cqp_destroy which destroys the CQP and frees its SD buffer memory. However, if the reset flag is true, cqp_destroy is never called and leads to a memory leak on SD buffer memory. Fix this by always calling cqp_destroy, on device close, regardless of reset. The exception to this when CQP create fails. In this case, the SD buffer memory is already freed on an error check and there is no need to call cqp_destroy. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-12i40iw: Fix QP flush to not hang on empty queues or failureShiraz Saleem
When flush QP and there are no pending work requests, signal completion to unblock i40iw_drain_sq and i40iw_drain_rq which are waiting on completion for iwqp->sq_drained and iwqp->sq_drained respectively. Also, signal completion if flush QP fails to prevent the drain SQ or RQ from being blocked indefintely. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-12i40iw: Fix double free of QPMustafa Ismail
A QP can be double freed if i40iw_cm_disconn() is called while it is currently being freed by i40iw_rem_ref(). The fix in i40iw_cm_disconn() will first check if the QP is already freed before making another request for the QP to be freed. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-12i40iw: Use correct src address in memcpy to rdma stats countersShiraz Saleem
hw_stats is a pointer to i40_iw_dev_stats struct in i40iw_get_hw_stats(). Use hw_stats and not &hw_stats in the memcpy to copy the i40iw device stats data into rdma_hw_stats counters. Fixes: b40f4757daa1 ("IB/core: Make device counter infrastructure dynamic") Cc: stable@vger.kernel.org # 4.7+ Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-12i40iw: Remove macros I40IW_STAG_KEY_FROM_STAG and I40IW_STAG_INDEX_FROM_STAGThomas Huth
The macros I40IW_STAG_KEY_FROM_STAG and I40IW_STAG_INDEX_FROM_STAG are apparently bad - they are using the logical "&&" operation which does not make sense here. It should have been a bitwise "&" instead. Since the macros seem to be completely unused, let's simply remove them so that nobody accidentially uses them in the future. And while we're at it, also remove the unused macro I40IW_CREATE_STAG. Signed-off-by: Thomas Huth <thuth@redhat.com> Reviewed-by: Leon Romanovsky <leonro@mellanox.com> Acked-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-08iw_cxgb4: Fix error return code in c4iw_rdev_open()Wei Yongjun
Fix to return error code -ENOMEM from the __get_free_page() error handling case instead of 0, as done elsewhere in this function. Fixes: 05eb23893c2c ("cxgb4/iw_cxgb4: Doorbell Drop Avoidance Bug Fixes") Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Acked-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Add request for reset on CQP timeoutHenry Orosco
When CQP times out, send a request to LAN driver for reset. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Code cleanup, remove check of PBLE pagesHenry Orosco
Remove check for zero 'pages' of unallocated pbles calculated in add_pble_pool(); as it can never be true. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Correctly fail loopback connection if no listenerShiraz Saleem
Fail the connect and return the proper error code if a client is started with local IP address and there is no corresponding loopback listener. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Fill in IRD value when on connect requestShiraz Saleem
IRD is not populated on connect request and application is getting 0 for the value. Fill in the correct value on connect request. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Set TOS field in IP headerShiraz Saleem
Set the TOS field in IP header with the value passed in from application. If there is mismatch between the remote client's TOS and listener, set the listener Tos to the higher of the two values. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Add NULL check for ibqp event handlerShiraz Saleem
Add NULL check for ibqp event handler before calling it to report QP events, as it might not initialized. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Replace list_for_each_entry macro with safe versionMustafa Ismail
Use list_for_each_entry_safe macro for the IPv6 addr list as IPv6 addresses can be deleted while going through the list. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Add IP addr handling on netdev eventsMustafa Ismail
Disable listeners and disconnect all connected QPs on a netdev interface down event. On an interface up event, the listeners are re-enabled. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Add missing cleanup on device closeMustafa Ismail
On i40iw device close, disconnect all connected QPs by moving them to error state; and block further QPs, PDs and CQs from being created. Additionally, make sure all resources have been freed before deallocating the ibdev as part of the device close. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Add 2MB page supportHenry Orosco
Add support to allow each independent memory region to be configured for 2MB page size in addition to 4KB page size. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Utilize physically mapped memory regionsHenry Orosco
Add support to use physically mapped WQ's and MR's if determined that the OS registered user-memory for the region is physically contiguous. This feature will eliminate the need for unnecessarily setting up and using PBL's when not required. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Fix incorrect assignment of SQ headHenry Orosco
The SQ head is incorrectly incremented when the number of WQEs required is greater than the number available. The fix is to use the I40IW_RING_MOV_HEAD_BY_COUNT macro. This checks for the SQ full condition first and only if SQ has room for the request, then we move the head appropriately. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Remove variable flush_code and check to set qp->sq_flushHenry Orosco
The flush_code variable in i40iw_bld_terminate_hdr() is obsolete and the check to set qp->sq_flush is unreachable. Currently flush code is populated in setup_term_hdr() and both SQ and RQ are flushed always as part of the tear down flow. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Remove check on return from device_init_pestat()Henry Orosco
Remove unnecessary check for return code from device_init_pestat() and change func to void. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Use runtime check for IS_ENABLED(CONFIG_IPV6)Henry Orosco
To be consistent, use the runtime check instead of conditional compile. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Use actual page sizeHenry Orosco
In i40iw_post_send, use the actual page size instead of encoded page size. This is to be consistent with the rest of the file. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Remove NULL check for cm_node->iwdevHenry Orosco
It is not necessary to check cm_node->iwdev in i40iw_rem_ref_cm_node() as it can never be NULL after a successful call out of i40iw_make_cm_node(). Signed-off-by: Chien Tin Tung <chien.tin.tung@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Remove checks for more than 48 bytes inline dataHenry Orosco
Remove dead code, which isn't executed because we return error if the data size is greater than 48 bytes. Inline data size greater than 48 bytes isn't supported and the maximum WQE size is 64 bytes. Signed-off-by: Tatyana Nikolova <tatyana.e.nikolova@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Query device accounts for internal rsrcHenry Orosco
Some resources are consumed internally and not available to the user. After hw is initialized, figure out how many resources are consumed and subtract those numbers from the initial max device capability in i40iw_query_device(). Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Optimize inline data copyHenry Orosco
Use memcpy for inline data copy in sends and writes instead of byte by byte copy. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Fix for LAN handler removalHenry Orosco
If i40iw_open() fails for any reason, the LAN handler is not being removed. Modify i40iw_deinit_device() to always remove the handler. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Correct values for max_recv_sge, max_send_sgeHenry Orosco
When creating QPs, ensure init_attr->cap.max_recv_sge is clipped to MAX_FRAG_COUNT. Expose MAX_FRAG_COUNT for max_recv_sge and max_send_sge in i40iw_query_qp(). Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Reviewed-By: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Use vector when creating CQsHenry Orosco
Assign each CEQ vector to a different CPU when possible, then when creating a CQ, use the vector for the CEQ id. This allows completion work to be distributed over multiple cores. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-05i40iw: Convert page_size to encoded valueHenry Orosco
Passed in page_size was used as encoded value for writing the WQE and passed in value was usually 4096. This was working out since bit 0 was 0 and implies 4KB pages, but would not work for other page sizes. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03i40iw: Set MAX IRD, MAX ORD size to max supported valueHenry Orosco
Set the MAX_IRD and MAX_ORD size negotiated to the maximum supported values. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03i40iw: Remove workaround for pre-production errataHenry Orosco
Pre-production silicon incorrectly truncates 4 bytes of the MPA packet in UDP loopback case. Remove the workaround as it is no longer necessary. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03i40iw: Enable message packingHenry Orosco
Remove the parameter to disable message packing and always enable it. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03i40iw: Add Quality of Service supportHenry Orosco
Add support for QoS on QPs. Upon device initialization, a map is created from user priority to queue set handles. On QP creation, use ToS to look up the queue set handle for use with the QP. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03IB/hns: Move HNS RoCE user vendor structuresLeon Romanovsky
This patch moves HNS vendor's specific structures to common UAPI folder which will be visible to all consumers. Signed-off-by: Leon Romanovsky <leon@kernel.org> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03IB/hns: Fix the IB device nameLijun Ou
This patch mainly fix the name for IB device in order to match with libhns. Signed-off-by: Lijun Ou <oulijun@huawei.com> Signed-off-by: Salil Mehta <salil.mehta@huawei.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03IB/hns: Fix the bug when free cqShaobo Xu
If the resources of cq are freed while executing the user case, hardware can not been notified in hip06 SoC. Then hardware will hold on when it writes the cq buffer which has been released. In order to slove this problem, RoCE driver checks the CQE counter, and ensure that the outstanding CQE have been written. Then the cq buffer can be released. Signed-off-by: Shaobo Xu <xushaobo2@huawei.com> Reviewed-by: Wei Hu (Xavier) <xavier.huwei@huawei.com> Signed-off-by: Salil Mehta <salil.mehta@huawei.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03IB/hns: Delete the redundant memset operationWei Hu (Xavier)
It deleted the redundant memset operation because the memory allocated by ib_alloc_device has been set zero. Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com> Signed-off-by: Salil Mehta <salil.mehta@huawei.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03IB/hns: Fix the bug of setting port mtuWei Hu (Xavier)
In hns_roce driver, we need not call iboe_get_mtu to reduce IB headers from effective IBoE MTU because hr_dev->caps.max_mtu has already been reduced. Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com> Signed-off-by: Salil Mehta <salil.mehta@huawei.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03IB/hns: Fix the bug when free mrShaobo Xu
If the resources of mr are freed while executing the user case, hardware can not been notified in hip06 SoC. Then hardware will hold on when it reads the payload by the PA which has been released. In order to slove this problem, RoCE driver creates 8 reserved loopback QPs to ensure zero wqe when free mr. When the mac address is reset, in order to avoid loopback failure, we need to release the reserved loopback QPs and recreate them. Signed-off-by: Shaobo Xu <xushaobo2@huawei.com> Reviewed-by: Wei Hu (Xavier) <xavier.huwei@huawei.com> Signed-off-by: Salil Mehta <salil.mehta@huawei.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03IB/hns: Fix the bug when destroy qpWei Hu (Xavier)
If send queue is still working when qp is in reset state by modify qp in destroy qp function, hardware will hold on and don't work in hip06 SoC. In current codes, RoCE driver check hardware pointer of sending and hardware pointer of processing to ensure that hardware has processed all the dbs of this qp. But while the environment of wire becomes not good, The checking time maybe too long. In order to solve this problem, RoCE driver created a workqueue at probe function. If there is a timeout when checking the status of qp, driver initialize work entry and push it into the workqueue, Work function will finish checking and release the related resources later. Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com> Signed-off-by: Lijun Ou <oulijun@huawei.com> Signed-off-by: Dongdong Huang(Donald) <hdd.huang@huawei.com> Signed-off-by: Salil Mehta <salil.mehta@huawei.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03IB/hns: Fix for Checkpatch.pl comment style errorsSalil
This patch correct the comment style errors caught by checkpatch.pl script Signed-off-by: Salil Mehta <salil.mehta@huawei.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03IB/hns: Implement the add_gid/del_gid and optimize the GIDs managementShaobo Xu
IB core has implemented the calculation of GIDs and the management of GID tables, and it is now responsible to supply query function for GIDs. So the calculation of GIDs and the management of GID tables in the RoCE driver is redundant. The patch is to implement the add_gid/del_gid to set the GIDs in the RoCE driver, remove the redundant calculation and management of GIDs in the notifier call of the net device and the inet, and update the query_gid. Signed-off-by: Shaobo Xu <xushaobo2@huawei.com> Reviewed-by: Wei Hu (Xavier) <xavier.huwei@huawei.com> Signed-off-by: Salil Mehta <salil.mehta@huawei.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03IB/hns: Change qpn allocation to round-robin mode.Wei Hu (Xavier)
When using CM to establish connections, qp number that was freed just now will be rejected by ib core. To fix these problem, We change qpn allocation to round-robin mode. We added the round-robin mode for allocating resources using bitmap. We use round-robin mode for qp number and non round-robing mode for other resources like cq number, pd number etc. Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com> Signed-off-by: Salil Mehta <salil.mehta@huawei.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-12-03IB/hns: Modify query info named port_num when querying RC QPWei Hu (Xavier)
This patch modified the output query info qp_attr->port_num to fix bug in hip06. Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com> Signed-off-by: Salil Mehta <salil.mehta@huawei.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
generated by cgit (git 2.34.1) at 2025-07-15 21:57:20 +0000