- 22 Oct, 2015 19 commits
-
-
Matan Barak authored
The GID cache accompanies every GID with attributes. The GID attributes link the GID with its netdevice, which could be resolved to smac and vlan id easily. Since we've added the netdevice (ifindex and net) to the path record, storing the L2 attributes is duplicated data and hence these attributes are removed. Signed-off-by:
Matan Barak <matanb@mellanox.com> Reviewed-By:
Devesh Sharma <devesh.sharma@avagotech.com> Signed-off-by:
Doug Ledford <dledford@redhat.com>
-
Matan Barak authored
Smac and vlan id could be resolved from the GID attribute, and thus these attributes aren't needed anymore. Removing them. Signed-off-by:
-
Matan Barak authored
The cm and cma don't need to explicitly handle vlan and smac, as they are resolved from the GID index now. Removing this portion of code. Signed-off-by:
-
Matan Barak authored
Previously, vlan id and source MAC were used from QP attributes. Since the net device is now stored in the GID attributes, they could be used instead of getting this information from the QP attributes. IB_QP_SMAC, IB_QP_ALT_SMAC, IB_QP_VID and IB_QP_ALT_VID were removed because there is no known libibverbs that uses them. This commit also modifies the vendors (mlx4, ocrdma) drivers in order to use the new approach. ocrdma driver changes were done by Somnath Kotur <Somnath.Kotur@Avagotech.Com> Signed-off-by:
-
Matan Barak authored
GID cache API users might want to search for GIDs with specific attributes rather than just specifying GID, net device and port. This is used in a later patch, where we find the sgid index by L2 Ethernet attributes. Signed-off-by:
-
Matan Barak authored
Previously, cma_validate_port searched for GIDs in IB cache and then tried to verify the found port. This could fail when there are identical GIDs on both ports. In addition, netdevice should be taken into account when searching the GID table. Fixing cma_validate_port to search only the relevant port's cache and netdevice. Signed-off-by:
-
Matan Barak authored
Previously, the CM has searched the cache for any sgid_index whose GID matches the path's GID. Since the path record stores the net device, the CM should now search only for GIDs which originated from this net device. Signed-off-by:
-
Matan Barak authored
In order to find the sgid_index, one could just query the IB cache with the correct GID and netdevice. Therefore, instead of storing the L2 attributes directly in the path, we only store the ifindex and net and use them later to get the sgid_index. The vlan_id and smac L2 attributes are removed in a later patch. Signed-off-by:
-
Matan Barak authored
Sometime consumers might want to search for a GID in a specific port. For example, when a WC arrives and we want to search the GID that matches that port - it's better to search only the relevant port. Exposing and renaming ib_cache_gid_find_by_port in order to match the naming convention of the module. Signed-off-by:
-
Matan Barak authored
Adding an ability to query the IB cache by a netdev and get the attributes of a GID. These parameters are necessary in order to successfully resolve the required GID (when the netdevice is known) and get the Ethernet L2 attributes from a GID. Signed-off-by:
-
Eran Ben Elisha authored
MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK is now supported downstream. In addition, this flag was supported only for IB_QPT_UD, now, with the new implementation it is supported for all QP types. Support IB_USER_VERBS_EX_CMD_CREATE_QP in order to get the flag from user space using the extension create qp command. Signed-off-by:
Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by:
-
Eran Ben Elisha authored
Current implementation for MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK is not supported when link layer is Ethernet. This patch will add counter based implementation for multicast loopback prevention. HW can drop multicast loopback packets if sender QP counter index is equal to receiver QP counter index. If qp flag MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK is set and link layer is Ethernet, create a new counter and attach it to the QP so it will continue receiving multicast loopback traffic but it's own. The decision if to create a new counter is being made at the qp modification to RTR after the QP's port is set. When QP is destroyed or moved back to reset state, delete the counter. Signed-off-by:
-
Eran Ben Elisha authored
This is an infrastructure step for allocating and attaching more than one counter to QPs on the same port. Allocate a counters table and manage the insertion and removals of the counters in load and unload of mlx4 IB. Signed-off-by:
-
Maor Gottlieb authored
Set the mcast loopback prevention bit in the QPC for ETH MLX QPs (not RSS QPs), when the firmware supports this feature. In addition, all rx ring QPs need to be updated in order not to enforce loopback checks. This prevents getting packets we sent both from the network stack and the HCA. Loopback prevention is done by comparing the counter indices of the sent and receiving QPs. If they're equal, packets aren't loopback-ed. Signed-off-by:
Maor Gottlieb <maorg@mellanox.com> Signed-off-by:
-
Maor Gottlieb authored
Update device capabilities regarding HW filtering multicast loopback support. Add MLX4_UPDATE_QP_ETH_SRC_CHECK_MC_LB attribute to mlx4_update_qp to enable changing QP context to support filtering incoming multicast loopback traffic according the sender's counter index. Set the corresponding bits in QP context to force the loopback source checks if attribute is given and HW supports it. Signed-off-by:
-
Eran Ben Elisha authored
Allow setting IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK at create_flags in ib_uverbs_create_qp_ex. Signed-off-by:
Haggai Eran <haggaie@mellanox.com> Signed-off-by:
-
Eran Ben Elisha authored
ib_uverbs_ex_create_qp follows the extension verbs mechanism. New features (for example, QP creation flags field which is added in a downstream patch) could used via user-space libraries without breaking the ABI. Signed-off-by:
-
Hariprasad S authored
Signed-off-by:
Hariprasad Shenai <hariprasad@chelsio.com> Signed-off-by:
-
Hariprasad S authored
Signed-off-by:
-
- 21 Oct, 2015 18 commits
-
-
Selvin Xavier authored
Updating the version number to 11.0.0.0 Signed-off-by:
Selvin Xavier <selvin.xavier@avagotech.com> Signed-off-by:
-
Devesh Sharma authored
Changing CQ-Doorbell(DB) logic to prevent DB floods, it is supposed to be pressed only if any hw CQE is polled. If cq-arm was requested previously then don't bother about number of hw CQEs polled and arm the CQ. Signed-off-by:
-
Naga Irrinki authored
Some versions of the FW sends wrong QP or CQ IDs in the Async CQE. Adding a check to see whether qp or cq structures associated with the CQE is valid. Signed-off-by:
-
Selvin Xavier authored
debugfs_remove should be called before freeing the driver stats resources to avoid any crash during ocrdma_remove. Signed-off-by:
-
Selvin Xavier authored
ocrdma_dev_list is not used by the driver. So removing the references of this variable. dev->rcu was introduced for the ipv6 notifier for GID management. This is no longer required as the GID management is outside the HW driver. Signed-off-by:
Padmanabh Ratnakar <padmanabh.ratnakar@avagotech.com> Signed-off-by:
-
Hariprasad S authored
The ESTABLISHED event should have the peer's ord/ird so swap the values in the event before the upcall. Signed-off-by:
-
Hariprasad S authored
When calculating the minimum ird in c4iw_accept_cr(), we need to always have a value of at least 1 if the RTR message is a 0B read. The code was incorrectly using ep->ord for this logic which was incorrectly adjusting the ird and causing incorrect ord/ird negotiation when using MPAv2 to negotiate these values. Signed-off-by:
-
Hariprasad S authored
This allows client ULPs to get the negotiated ord/ird which is useful to avoid stalling the SQ due to exceeding the ORD. Signed-off-by:
-
Hariprasad S authored
In c4iw_create_listen(), if we're using listen filters, then bail out of the busy loop if the device becomes fatally dead Signed-off-by:
-
Arnd Bergmann authored
The INIT_UDATA() macro requires a pointer or unsigned long argument for both input and output buffer, and all callers had a cast from when the code was merged until a recent restructuring, so now we get core/uverbs_cmd.c: In function 'ib_uverbs_create_cq': core/uverbs_cmd.c:1481:66: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast] This makes the code behave as before by adding back the cast to unsigned long. Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Fixes: 565197dd ("IB/core: Extend ib_uverbs_create_cq") Reviewed-by:
Yann Droneaud <ydroneaud@opteya.com> Signed-off-by:
-
Arnd Bergmann authored
Casting a pointer to __be64 produces a warning on 32-bit architectures: drivers/infiniband/hw/cxgb4/mem.c:147:20: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast] req->wr.wr_lo = (__force __be64)&wr_wait; This was fixed at least twice for this driver in different places, and accidentally reverted once more. This puts the correct version back in place. Signed-off-by: -
Hal Rosenstock authored
IBA spec is now 1.3 not 3.1 and vol 1 should be mentioned as there is also vol 2. Signed-off-by:
Hal Rosenstock <hal@mellanox.com> Signed-off-by:
-
Geliang Tang authored
Just fix a typo in the code comment. Signed-off-by:
Geliang Tang <geliangtang@163.com> Acked-by:
Sagi Grimberg <sagig@mellanox.com> Signed-off-by:
-
Insu Yun authored
Since kzalloc returns memory address, not error code, it should be checked whether it is null or not. Signed-off-by:
Insu Yun <wuninsu@gmail.com> Reviewed-by:
Dave Goodell <dgoodell@cisco.com> Signed-off-by:
-
Dave Goodell \(dgoodell\) authored
Upinder hasn't worked for Cisco for a little while now, updating to more active maintainers. Signed-off-by:
-
Insu Yun authored
Since ib_alloc_device returns allocated memory address, not error, it should be checked as IS_NULL, not IS_ERR_OR_NULL. Signed-off-by:
-
Doug Ledford authored
Pick up the late fixes from the 4.3 cycle so we have them in our next branch.
-
Doron Tsur authored
ib_send_cm_sidr_rep could sometimes erase the node from the sidr (depending on errors in the process). Since ib_send_cm_sidr_rep is called both from cm_sidr_req_handler and cm_destroy_id, cm_id_priv could be either erased from the rb_tree twice or not erased at all. Fixing that by making sure it's erased only once before freeing cm_id_priv. Fixes: a977049d ('[PATCH] IB: Add the kernel CM implementation') Signed-off-by:
Doron Tsur <doront@mellanox.com> Signed-off-by:
-
- 20 Oct, 2015 3 commits
-
-
Haggai Eran authored
When discussing the patches to demux ids in rdma_cm instead of ib_cm, it was decided that it is best to use the P_Key value in the packet headers. However, the mlx5 and ipath drivers are currently unable to send correct P_Key values in GMP headers. They always send using a single P_Key that is set during the GSI QP initialization. Change the rdma_cm code to look at the P_Key value that is part of the packet payload as a workaround. Once the drivers are fixed this patch can be reverted. Fixes: 4c21b5bc ("IB/cma: Add net_dev and private data checks to RDMA CM") Signed-off-by:
-
Sasha Levin authored
Allocating a workqueue might fail, which wasn't checked so far and would lead to NULL ptr derefs when an attempt to use it was made. Signed-off-by:
Sasha Levin <sasha.levin@oracle.com> Signed-off-by:
-
Haggai Eran authored
If the lookup of a listening ID failed for an AF_IB request, the code would try to call dev_put() on a NULL net_dev. Fixes: be688195 ("IB/cma: Fix net_dev reference leak with failed requests") Reported-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
-
