- 19 1月, 2018 2 次提交
-
-
由 Devesh Sharma 提交于
Shared receive queue (SRQ) is defined as a pool of receive buffers shared among multiple QPs which belong to same protection domain in a given process context. Use of SRQ reduces the memory foot print of IB applications. Broadcom adapters support SRQ, adding code-changes to enable shared receive queue. Signed-off-by: NDevesh Sharma <devesh.sharma@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Somnath Kotur 提交于
Depending on the OS page-table configurations, applications may request MRs which has page size alignment other than 4K Underlying provider driver needs to adjust its PBL boundaries according to the incoming page boundaries in the PA list. Adding a capability to register MRs having pages-sizes other than 4K (Hugepages). Signed-off-by: NSomnath Kotur <somnath.kotur@broadcom.com> Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDevesh Sharma <devesh.sharma@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
- 17 1月, 2018 2 次提交
-
-
由 Selvin Xavier 提交于
The device now reports firmware version thus, removing the hard coded values of the FW version string and redundant fw_rev hook from sysfs. Adding code to query firmware version from underlying device and report it through the kernel verb to get firmware version string. Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDevesh Sharma <devesh.sharma@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Selvin Xavier 提交于
RoCE can be used by virtual functions (VFs) as well. Adding code changes to allow resource reservation, initialization and avail the resources to the RDMA applications running on those VFs. Currently, fifty percent of the total available resources are reserved for PF and remaining are equally divided among active VFs. Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDevesh Sharma <devesh.sharma@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
- 14 11月, 2017 5 次提交
-
-
由 Devesh Sharma 提交于
In a real RoCE v2 network it is possible to have two Sections of network have same IP hence same gid. However those may have different vlans. During connection resolution it is important to report the actual vlan on which the MAD packet was received instead of relying on other means to resolve vlan-id. ib_find_gid_index should not be used to resolve the vlan-id using sgid of the local system where the packet was received. Our device has the capability to report the actual VLAN-ID in the GSI qp completions. Since we have the capability our driver should move away from resolving the vlan-id with the help of SGID at the destination port. Signed-off-by: NDevesh Sharma <devesh.sharma@broadcom.com> Reported-by: NParav Pandit <parav@mellanox.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Selvin Xavier 提交于
Synchronize poll_cq and req_notify_cq verbs using cq_lock, instead of the lower level qplib->hwq.lock. Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Selvin Xavier 提交于
Destroy_qp shall wait for any outstanding CQ notification to be flushed out before proceeding with QP destroy. Flushing the WQ before destroying the QP. Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Sriharsha Basavapatna 提交于
While adding a new gid, the driver currently does not return the context back to the stack. A subsequent del_gid() (e.g, when ip address is changed) doesn't find the right context in the driver and it ends up dropping that request. This results in the HW caching a stale gid entry and traffic fails because of that. Fix by returning the proper context in bnxt_re_add_gid(). Signed-off-by: NSriharsha Basavapatna <sriharsha.basavapatna@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Leon Romanovsky 提交于
The Broadcom driver produces the following compilation warning drivers/infiniband/hw/bnxt_re/ib_verbs.c: In function ‘bnxt_re_create_ah’: drivers/infiniband/hw/bnxt_re/ib_verbs.c:668:6: warning: variable ‘vlan_tag’ set but not used [-Wunused-but-set-variable] u16 vlan_tag; Let's remove it till vlan_tag will be implemented properly. Cc: Selvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NLeon Romanovsky <leonro@mellanox.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
- 19 10月, 2017 1 次提交
-
-
由 Parav Pandit 提交于
Since IB/core resolves the destination mac address for user and kernel consumers, avoid resolving in multiple provider drivers. Only ib_core resolves DMAC now, therefore resolve_eth_dmac is removed as exported symbol. Signed-off-by: NParav Pandit <parav@mellanox.com> Signed-off-by: NLeon Romanovsky <leon@kernel.org> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
- 15 10月, 2017 1 次提交
-
-
由 Bart Van Assche 提交于
Signed-off-by: NBart Van Assche <bart.vanassche@wdc.com> Cc: Somnath Kotur <somnath.kotur@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
- 23 9月, 2017 6 次提交
-
-
由 Somnath Kotur 提交于
FW needs the 0th GID Entry in the Table to be preserved before it's corresponding QP1 is deleted, else it will fail the cmd. Check for the same and return to prevent error msg being logged for cmd failure. Signed-off-by: NSomnath Kotur <somnath.kotur@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Selvin Xavier 提交于
This patch fixes a memory leak issue when alloc_mr is used. mr->pages and mr->npages are used only in alloc_mr path. mr->pages is allocated when alloc_mr is called or in the case of FRMR, while creating the MR. mr->npages is updated only when the MR created is used i.e. after invoking map_mr_sg verb, before data transfer. In the dereg_mr path, if mr->npages is 0, driver ends up not freeing the memory created. Removing the npages check from the dereg_mr path for kernel consumers. Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NSomnath Kotur <somnath.kotur@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Somnath Kotur 提交于
When there is a NETDEV_UNREGISTER event, bnxt_re driver calls ib_unregister_device() (RTNL lock held). ib_unregister_device attempts to flush a worker queue scheduled by ib_core and that queue might have a pending ib_query_port(). ib_query_port in turn calls bnxt_re_query_port(), which while querying the link speed using ib_get_eth_speed(), tries to acquire the rtnl_lock() which was already held by NETDEV_UNREGISTER. Fixing the issue by removing the link speed query from bnxt_re_query_port() Now the speed is queried post a successful ib_register_device or whenever there is a NETDEV_CHANGE event. Signed-off-by: NSomnath Kotur <somnath.kotur@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Devesh Sharma 提交于
Driver must assign the user supplied compare/swap values in the wqe to successfully complete the atomic compare and swap operation. Signed-off-by: NDevesh Sharma <devesh.sharma@broadcom.com> Signed-off-by: NSomnath Kotur <somnath.kotur@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Devesh Sharma 提交于
The MTU value in the qplib_qp.mtu should be consistent with whatever mtu was set during INIT to RTR.The Next PSN and number of packets are calculated based on this member in the qplib_qp structure. Signed-off-by: NNarender Reddy <narender.reddy@broadcom.com> Signed-off-by: NDevesh Sharma <devesh.sharma@broadcom.com> Signed-off-by: NSomnath Kotur <somnath.kotur@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Leon Romanovsky 提交于
Reduce stack size by dynamically allocating memory instead of declaring large struct on the stack: drivers/infiniband/hw/bnxt_re/ib_verbs.c: In function ‘bnxt_re_query_qp’: drivers/infiniband/hw/bnxt_re/ib_verbs.c:1600:1: warning: the frame size of 1216 bytes is larger than 1024 bytes [-Wframe-larger-than=] } ^ Cc: Selvin Xavier <selvin.xavier@broadcom.com> Fixes: 1ac5a404 ("RDMA/bnxt_re: Add bnxt_re RoCE driver") Signed-off-by: NLeon Romanovsky <leon@kernel.org> Acked-by: NSelvin Xavier <selvin.xavier@broadcom.com> Reviewed-by: NJonathan Toppins <jtoppins@redhat.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
- 29 8月, 2017 1 次提交
-
-
由 Himanshu Jha 提交于
call to memset to assign 0 value immediately after allocating memory with kzalloc is unnecesaary as kzalloc allocates the memory filled with 0 value. Signed-off-by: NHimanshu Jha <himanshujha199640@gmail.com> Reviewed-by: NYuval Shaia <yuval.shaia@oracle.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
- 23 8月, 2017 1 次提交
-
-
由 Selvin Xavier 提交于
ib_get_eth_speed() return 0 on success. Fixing the condition checking and prevent reporting failure for query_port verb. Fixes: d4186194 ("Add generic function to extract IB speed from netdev") Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Reviewed-by: NLeon Romanovsky <leonro@mellanox.com> Reviewed-by: NYuval Shaia <yuval.shaia@oracle.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
- 19 8月, 2017 1 次提交
-
-
由 Selvin Xavier 提交于
Enables multiple Interrupt vectors. Driver is requesting the max MSIX vectors based on the number of online cpus and creates upto 9 MSIx vectors (1 for control path and 8 for data path). A tasklet is created for each of these vectors. NQs are assigned to CQs in round robin fashion. This patch also adds IRQ affinity hint for the MSIX vector of each NQ. Signed-off-by: NRay Jui <ray.jui@broadcom.com> Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
- 18 8月, 2017 1 次提交
-
-
由 Colin Ian King 提交于
Trivial fix to spelling mistake in dev_err error message Signed-off-by: NColin Ian King <colin.king@canonical.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
- 24 7月, 2017 3 次提交
-
-
由 Selvin Xavier 提交于
This patch allows driver to post send and receive requests on QPs which are in error state. Instead of flushing the QP in the context of polling error CQEs, the QPs will be added to a flush list maintained per CQ. QP state is moved to error. QP is added to flush list if the user moves it to error state using modify_qp also. After polling the HW CQ in poll_cq routine, this flush list is traversed and driver completes work requests on each QP in the flush list, till the budget expires. The QP is moved out of flush list during QP destroy or during modify_QP to RESET. When ULPs post Work Requests while QP is in error state, driver will store the ULP data and then increment the QP producer s/w index, without ringing doorbell. It then schedules a worker to invoke the CQ handler since the interrupts wont be generated from the HW for this request. Signed-off-by: NSriharsha Basavapatna <sriharsha.basavapatna@broadcom.com> Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Yuval Shaia 提交于
Logic of retrieving netdev speed from net_device and translating it to IB speed is implemented in rxe, in usnic and in bnxt drivers. Define new function which merges all. Signed-off-by: NYuval Shaia <yuval.shaia@oracle.com> Reviewed-by: NChristian Benvenuti <benve@cisco.com> Reviewed-by: NSelvin Xavier <selvin.xavier@broadcom.com> Reviewed-by: NMoni Shoua <monis@mellanox.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Leon Romanovsky 提交于
There is no need to return always zero for function which is not supported. The IB stack treats uninitialized ib_device->functions as not implemented. Signed-off-by: NLeon Romanovsky <leonro@mellanox.com> Reviewed-by: NDennis Dalessandro <dennis.dalessandro@intel.com> Reviewed-by: NYuval Shaia <yuval.shaia@oracle.com> Acked-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
- 20 7月, 2017 9 次提交
-
-
由 Selvin Xavier 提交于
Local ack delay exposed by the driver is 0 which means infinite QP timeout. Reporting the default value to 16 (approx 260ms) Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Selvin Xavier 提交于
While invoking the req_notify_cq hook, ULPs can request whether the CQs have any CQEs pending. If CQEs are pending, drivers can indicate it by returning 1 for req_notify_cq. The stack will poll CQ again till CQ is empty. This patch peeks the CQ for any valid entries and return accordingly. Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Devesh Sharma 提交于
Fix the incorrect reporting of number of polled entries by taking into account the max CQ depth in the driver. Signed-off-by: NDevesh Sharma <devesh.sharma@broadcom.com> Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Reviewed-by: NLeon Romanovsky <leonro@mellanox.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Devesh Sharma 提交于
Driver shall check if the host system bios has enabled Atomic operations capability in PCI Device Control 2 register of the pci-device. Expose the ATOMIC_HCA flag only if the Atomic operations capability is set. Signed-off-by: NDevesh Sharma <devesh.sharma@broadcom.com> Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Eddie Wai 提交于
There's a couple of bugs in the support of max_rd_atomic and max_dest_rd_atomic. In the modify_qp, if the requested max_rd_atomic, which is the ORRQ size, is greater than what the chip can support, then we have to cap the request to chip max as we can't have the HW overflow the ORRQ. Capping the max_rd_atomic support internally is okay to do as the remaining read/atomic WRs will still be sitting in the SQ. However, for the max_dest_rd_atomic, the driver has to error out as this dictates the IRRQ size and we can't control what the remote side sends. Signed-off-by: NEddie Wai <eddie.wai@broadcom.com> Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Selvin Xavier 提交于
- Report supported value for max_mr_size to IB stack in query_device. Also, check and log if MR size requested by application in reg_user_mr() is greater than value currently supported by driver. - Report only 4K page size support for now - Fix Max_QP value returned by ibv_devinfo -vv. In case of PF, FW reserves 129 QPs for creating QP1s of VFs and PF. So the max_qp value reported by FW for PF doesn'tt include the QP1. Fixing this issue by adding 1 with the value reported by FW. Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Selvin Xavier 提交于
This fix is added only to avoid system crash in some a specific scenario. When bnxt_re driver is loaded and if user tries to change interface mac address, delete GID fails because QP1 is still associated with existing MAC (default GID). If the above command fails GID tables are not modified in the h/w or driver, but the GID context memory is freed. Now, if the user changes the mac back to the original value, another add_gid comes to the driver where the driver reports that the GID is already present in its table and tries to access the context which was already freed. So, in this case, in order to avoid NULL pointer de-reference, this patch removes the context memory free if delete_gid fails and the same context memory is re-used in new add_gid. Memory cleanup will be taken care during driver unload, while deleting the GID table. Signed-off-by: NKalesh AP <kalesh-anakkur.purayil@broadcom.com> Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Devesh Sharma 提交于
The driver must free the DPI during the dealloc_ucontext instead of freeing it during dealloc_pd. However, the DPI allocation scheme remains unchanged. Signed-off-by: NDevesh Sharma <devesh.sharma@broadcom.com> Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Dan Carpenter 提交于
bnxt_re_alloc_mw() doesn't return NULL, it returns error pointers. Fixes: 9152e0b7 ("RDMA/bnxt_re: HW workarounds for handling specific conditions") Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com> Acked-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
- 15 6月, 2017 5 次提交
-
-
由 Selvin Xavier 提交于
Some issues observed with FMR implementation while running stress traffic. So removing the FMR verbs support for now. Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Acked-by: NChristoph Hellwig <hch@lst.de> Reviewed-by: NLeon Romanovsky <leonro@mellanox.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Devesh Sharma 提交于
This patch adds code to ring RQ Doorbell aggressively so that the adapter can DMA RQ buffers sooner, instead of DMA all WQEs in the post_recv WR list together at the end of the post_recv verb. Also use spinlock to serialize RQ posting Signed-off-by: NKalesh AP <kalesh-anakkur.purayil@broadcom.com> Signed-off-by: NDevesh Sharma <devesh.sharma@broadcom.com> Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Somnath Kotur 提交于
HW stalls out after 0x800000 WQEs are posted for UD QPs. To workaround this problem, driver will send a modify_qp cmd to the HW at around the halfway mark(0x400000) so that FW can accordingly modify the QP context in the HW to prevent this stall. This workaround needs to be done for UD, QP1 and Raw Ethertype packets. Added a counter to keep track of WQEs posted during post_send. Signed-off-by: NSomnath Kotur <somnath.kotur@broadcom.com> Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Selvin Xavier 提交于
If the host buffers are freed before destroying MR in HW, HW could try accessing these buffers. This could cause a host crash. Fixing the code to avoid this condition. Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
由 Eddie Wai 提交于
This patch implements the following HW workarounds 1. The SQ depth needs to be augmented by 128 + 1 to avoid running into an Out of order CQE issue 2. Workaround to handle the problem where the HW fast path engine continues to access DMA memory in retranmission mode even after the WQE has already been completed. If the HW reports this condition, driver detects it and posts a Fence WQE. The driver stops reporting the completions to stack until it receives completion for Fence WQE. Signed-off-by: NEddie Wai <eddie.wai@broadcom.com> Signed-off-by: NSriharsha Basavapatna <sriharsha.basavapatna@broadcom.com> Signed-off-by: NSelvin Xavier <selvin.xavier@broadcom.com> Reviewed-by: NLeon Romanovsky <leonro@mellanox.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
- 02 5月, 2017 2 次提交
-
-
rdma_ah_attr can now be either ib or roce allowing core components to use one type or the other and also to define attributes unique to a specific type. struct ib_ah is also initialized with the type when its first created. This ensures that calls such as modify_ah dont modify the type of the address handle attribute. Reviewed-by: NIra Weiny <ira.weiny@intel.com> Reviewed-by: NDon Hiatt <don.hiatt@intel.com> Reviewed-by: NSean Hefty <sean.hefty@intel.com> Reviewed-by: NNiranjana Vishwanathapura <niranjana.vishwanathapura@intel.com> Signed-off-by: NDasaratharaman Chandramouli <dasaratharaman.chandramouli@intel.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-
Modify core and driver components to use accessor functions introduced to access individual fields of rdma_ah_attr Reviewed-by: NIra Weiny <ira.weiny@intel.com> Reviewed-by: NDon Hiatt <don.hiatt@intel.com> Reviewed-by: NSean Hefty <sean.hefty@intel.com> Reviewed-by: NNiranjana Vishwanathapura <niranjana.vishwanathapura@intel.com> Signed-off-by: NDasaratharaman Chandramouli <dasaratharaman.chandramouli@intel.com> Signed-off-by: NDoug Ledford <dledford@redhat.com>
-