summaryrefslogtreecommitdiff
path: root/drivers/infiniband
AgeCommit message (Collapse)AuthorFilesLines
2016-03-19Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-nextLinus Torvalds4-6/+13
Pull networking updates from David Miller: "Highlights: 1) Support more Realtek wireless chips, from Jes Sorenson. 2) New BPF types for per-cpu hash and arrap maps, from Alexei Starovoitov. 3) Make several TCP sysctls per-namespace, from Nikolay Borisov. 4) Allow the use of SO_REUSEPORT in order to do per-thread processing of incoming TCP/UDP connections. The muxing can be done using a BPF program which hashes the incoming packet. From Craig Gallek. 5) Add a multiplexer for TCP streams, to provide a messaged based interface. BPF programs can be used to determine the message boundaries. From Tom Herbert. 6) Add 802.1AE MACSEC support, from Sabrina Dubroca. 7) Avoid factorial complexity when taking down an inetdev interface with lots of configured addresses. We were doing things like traversing the entire address less for each address removed, and flushing the entire netfilter conntrack table for every address as well. 8) Add and use SKB bulk free infrastructure, from Jesper Brouer. 9) Allow offloading u32 classifiers to hardware, and implement for ixgbe, from John Fastabend. 10) Allow configuring IRQ coalescing parameters on a per-queue basis, from Kan Liang. 11) Extend ethtool so that larger link mode masks can be supported. From David Decotigny. 12) Introduce devlink, which can be used to configure port link types (ethernet vs Infiniband, etc.), port splitting, and switch device level attributes as a whole. From Jiri Pirko. 13) Hardware offload support for flower classifiers, from Amir Vadai. 14) Add "Local Checksum Offload". Basically, for a tunneled packet the checksum of the outer header is 'constant' (because with the checksum field filled into the inner protocol header, the payload of the outer frame checksums to 'zero'), and we can take advantage of that in various ways. From Edward Cree" * git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1548 commits) bonding: fix bond_get_stats() net: bcmgenet: fix dma api length mismatch net/mlx4_core: Fix backward compatibility on VFs phy: mdio-thunder: Fix some Kconfig typos lan78xx: add ndo_get_stats64 lan78xx: handle statistics counter rollover RDS: TCP: Remove unused constant RDS: TCP: Add sysctl tunables for sndbuf/rcvbuf on rds-tcp socket net: smc911x: convert pxa dma to dmaengine team: remove duplicate set of flag IFF_MULTICAST bonding: remove duplicate set of flag IFF_MULTICAST net: fix a comment typo ethernet: micrel: fix some error codes ip_tunnels, bpf: define IP_TUNNEL_OPTS_MAX and use it bpf, dst: add and use dst_tclassid helper bpf: make skb->tc_classid also readable net: mvneta: bm: clarify dependencies cls_bpf: reset class and reuse major in da ldmvsw: Checkpatch sunvnet.c and sunvnet_common.c ldmvsw: Add ldmvsw.c driver code ...
2016-03-18Merge tag 'for-linus' of ↵Linus Torvalds66-1838/+3116
git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma Pull rdma updates from Doug Ledford: "Initial roundup of 4.6 merge window patches. This is the first of two pull requests. It is the smaller request, but touches for more different things (this is everything but what is in or going into staging). The pull request for the code in staging/rdma is on hold until after we decide what to do on the write/writev API issue and may be partially deferred until 4.7 as a result. Summary: - cxgb4 updates - nes updates - unification of iwarp portmapper code to core - add drain_cq API - various ib_core updates - minor ipoib updates - minor mlx4 updates - more significant mlx5 updates (including a minor merge conflict with net-next tree...merge is simple to resolve and Stephen's resolution was confirmed by Mellanox) - trivial net/9p rdma conversion - ocrdma RoCEv2 update - srpt updates" * tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma: (85 commits) iwpm: crash fix for large connections test iw_cxgb3: support for iWARP port mapping iw_cxgb4: remove port mapper related code iw_nes: remove port mapper related code iwcm: common code for port mapper net/9p: convert to new CQ API IB/mlx5: Add support for don't trap rules net/mlx5_core: Introduce forward to next priority action net/mlx5_core: Create anchor of last flow table iser: Accept arbitrary sg lists mapping if the device supports it mlx5: Add arbitrary sg list support IB/core: Add arbitrary sg_list support IB/mlx5: Expose correct max_fast_reg_page_list_len IB/mlx5: Make coding style more consistent IB/mlx5: Convert UMR CQ to new CQ API IB/ocrdma: Skip using unneeded intermediate variable IB/ocrdma: Skip using unneeded intermediate variable IB/ocrdma: Delete unnecessary variable initialisations in 11 functions IB/core: Documentation fix in the MAD header file IB/core: trivial prink cleanup. ...
2016-03-18Merge branch 'for-linus' of ↵Linus Torvalds1-1/+1
git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial Pull trivial tree updates from Jiri Kosina. * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial: drivers/rtc: broken link fix drm/i915 Fix typos in i915_gem_fence.c Docs: fix missing word in REPORTING-BUGS lib+mm: fix few spelling mistakes MAINTAINERS: add git URL for APM driver treewide: Fix typo in printk
2016-03-18Merge tag 'configfs-for-linus' of git://git.infradead.org/users/hch/configfsLinus Torvalds1-22/+9
Pull configfs updates from Christoph Hellwig: - A large patch from me to simplify setting up the list of default groups by actually implementing it as a list instead of an array. - a small Y2083 prep patch from Deepa Dinamani. Probably doesn't matter on it's own, but it seems like he is trying to get rid of all CURRENT_TIME uses in file systems, which is a worthwhile goal. * tag 'configfs-for-linus' of git://git.infradead.org/users/hch/configfs: configfs: switch ->default groups to a linked list configfs: Replace CURRENT_TIME by current_fs_time()
2016-03-16Merge branches 'nes', 'cxgb4' and 'iwpm' into k.o/for-4.6Doug Ledford19-750/+432
2016-03-16iwpm: crash fix for large connections testFaisal Latif3-13/+13
During large connection test, there is a crash at wake_up() in the callback as waitq is not yet initialized. Callback can happen before iwpm_wait_complete_req() is called to initialize waitq. To resolve, using signaling semaphore instead of waitq. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Reviewed-by: Tatyana E Nikolova <tatyana.e.nikolova@intel.com> Signed-off-by: Faisal Latif <faisal.latif@intel.com> Reviewed-by: Steve Wise <swise@opengridcomputing.com> Tested-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-16iw_cxgb3: support for iWARP port mappingSteve Wise1-8/+8
Now with the new iWARP port mapping service in the iwcm, it is trivial to add cxgb3 support. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-16iw_cxgb4: remove port mapper related codeSteve Wise3-264/+65
Now that most of the port mapper code been moved to iwcm, we can remove it from iw_cxgb4. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-16iw_nes: remove port mapper related codeFaisal Latif4-309/+80
Now that most of the port mapper code been moved to iwcm, we can remove it from port mapper service user drivers. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Tatyana E. Nikolova <tatyana.e.nikolova@intel.com> Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-16iwcm: common code for port mapperFaisal Latif2-27/+165
moved port mapper related code from drivers into common code Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Tatyana E. Nikolova <tatyana.e.nikolova@intel.com> Signed-off-by: Faisal Latif <faisal.latif@intel.com> Reviewed-by: Steve Wise <swise@opengridcomputing.com> Tested-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-16Merge branches 'mlx4', 'mlx5' and 'ocrdma' into k.o/for-4.6Doug Ledford31-281/+1998
2016-03-15Merge branches 'ib_core', 'ib_ipoib', 'srpt', 'drain-cq-v4' and 'net/9p' ↵Doug Ledford15-681/+574
into k.o/for-4.6
2016-03-10IB/mlx5: Add support for don't trap rulesMaor Gottlieb2-7/+49
Each bypass flow steering priority will be split into two priorities: 1. Priority for don't trap rules. 2. Priority for normal rules. When user creates a flow using IB_FLOW_ATTR_FLAGS_DONT_TRAP flag, the driver creates two flow rules, one used for receiving the traffic and the other one for forwarding the packet to continue matching in lower or equal priorities. Signed-off-by: Maor Gottlieb <maorg@mellanox.com> Reviewed-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-08Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/netDavid S. Miller4-23/+30
Several cases of overlapping changes, as well as one instance (vxlan) of a bug fix in 'net' overlapping with code movement in 'net-next'. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-03-06configfs: switch ->default groups to a linked listChristoph Hellwig1-22/+9
Replace the current NULL-terminated array of default groups with a linked list. This gets rid of lots of nasty code to size and/or dynamically allocate the array. While we're at it also provide a conveniant helper to remove the default groups. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Felipe Balbi <balbi@kernel.org> [drivers/usb/gadget] Acked-by: Joel Becker <jlbec@evilplan.org> Acked-by: Nicholas Bellinger <nab@linux-iscsi.org> Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
2016-03-05Merge tag 'for-linus' of ↵Linus Torvalds4-23/+30
git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma Pull rdma fixes from Doug Ledford: "Additional 4.5-rc6 fixes. I have four patches today. I had previously thought I had submitted two of them last week, but they were accidentally skipped :-(. - One fix to an error path in the core - One fix for RoCE in the core - Two related fixes for the core/mlx5" * tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma: IB/core: Use GRH when the path hop-limit > 0 IB/{core, mlx5}: Fix input len in vendor part of create_qp/srq IB/mlx5: Avoid using user-index for SRQs IB/core: Fix missed clean call in registration path
2016-03-04iser: Accept arbitrary sg lists mapping if the device supports itSagi Grimberg2-9/+25
If the device support arbitrary sg list mapping (device cap IB_DEVICE_SG_GAPS_REG set) we allocate the memory regions with IB_MR_TYPE_SG_GAPS and allow the block layer to pass us gaps by skip setting the queue virt_boundary. Signed-off-by: Sagi Grimberg <sagig@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-04mlx5: Add arbitrary sg list supportSagi Grimberg4-8/+60
Allocate proper context for arbitrary scatterlist registration If ib_alloc_mr is called with IB_MR_MAP_ARB_SG, the driver allocate a private klm list instead of a private page list. Set the UMR wqe correctly when posting the fast registration. Also, expose device cap IB_DEVICE_MAP_ARB_SG according to the device id (until we have a FW bit that correctly exposes it). Signed-off-by: Sagi Grimberg <sagig@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-04IB/core: Add arbitrary sg_list supportSagi Grimberg1-0/+2
Devices that are capable in registering SG lists with gaps can now expose it in the core to ULPs using a new device capability IB_DEVICE_SG_GAPS_REG (in a new field device_cap_flags_ex in the device attributes as we ran out of bits), and a new mr_type IB_MR_TYPE_SG_GAPS_REG which allocates a memory region which is capable of handling SG lists with gaps. Signed-off-by: Sagi Grimberg <sagig@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-04IB/mlx5: Expose correct max_fast_reg_page_list_lenSagi Grimberg1-1/+2
While documentation indicates that the number of translation entries per memory key is unlimited, in practice, we can only fit a finite amount of translation entries in a single registration wqe (which is log_max_klm_list_size). Signed-off-by: Sagi Grimberg <sagig@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-04IB/mlx5: Make coding style more consistentDoug Ledford1-4/+2
These three related functions can't agree whether to put the umrwr on the stack dirty and then memset it, or to initialize it on the stack. Make them all agree. Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-04IB/mlx5: Convert UMR CQ to new CQ APIChristoph Hellwig3-40/+27
Simplifies the code, and makes it more fair vs other users by using a softirq for polling. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Haggai Eran <haggaie@mellanox.com> Reviewed-by: Sagi Grimberg <sagig@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-04net: mellanox: add DEVLINK dependenciesArnd Bergmann1-0/+1
The new NET_DEVLINK infrastructure can be a loadable module, but the drivers using it might be built-in, which causes link errors like: drivers/net/built-in.o: In function `mlx4_load_one': :(.text+0x2fbfda): undefined reference to `devlink_port_register' :(.text+0x2fc084): undefined reference to `devlink_port_unregister' drivers/net/built-in.o: In function `mlxsw_sx_port_remove': :(.text+0x33a03a): undefined reference to `devlink_port_type_clear' :(.text+0x33a04e): undefined reference to `devlink_port_unregister' There are multiple ways to avoid this: a) add 'depends on NET_DEVLINK || !NET_DEVLINK' dependencies for each user b) use 'select NET_DEVLINK' from each driver that uses it and hide the symbol in Kconfig. c) make NET_DEVLINK a 'bool' option so we don't have to list it as a dependency, and rely on the APIs to be stubbed out when it is disabled d) use IS_REACHABLE() rather than IS_ENABLED() to check for NET_DEVLINK in include/net/devlink.h This implements a variation of approach a) by adding an intermediate symbol that drivers can depend on, and changes the three drivers using it. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Fixes: 09d4d087cd48 ("mlx4: Implement devlink interface") Fixes: c4745500e988 ("mlxsw: Implement devlink interface") Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-03-03IB/ocrdma: Skip using unneeded intermediate variableMarkus Elfring1-3/+1
Return the value from a call of the ocrdma_mbx_modify_qp() function without using an extra assignment for the local variable "status". Signed-off-by: Markus Elfring <elfring@users.sourceforge.net> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-03IB/ocrdma: Skip using unneeded intermediate variableMarkus Elfring1-2/+1
Return zero at the end without using the local variable "status". Signed-off-by: Markus Elfring <elfring@users.sourceforge.net> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-03IB/ocrdma: Delete unnecessary variable initialisations in 11 functionsMarkus Elfring4-11/+11
The variable "status" will be set to an appropriate value a bit later. Thus let us omit the explicit initialisation at the beginning. Signed-off-by: Markus Elfring <elfring@users.sourceforge.net> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-03IB/core: Use GRH when the path hop-limit > 0Or Gerlitz1-1/+1
According to IBTA spec v1.3 section 12.7.19, QPs should use GRH when the path returned by the SA has hop-limit > 0. Currently, we do that only for the > 1 case, fix that. Fixes: 6d969a471ba1 ('IB/sa: Add ib_init_ah_from_path()') Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Reviewed-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-03IB/core: trivial prink cleanup.Parav Pandit10-89/+77
1. Replaced printk with appropriate pr_warn, pr_err, pr_info. 2. Removed unnecessary prints around memory allocation failure which are not required, as reported by the checkpatch script. Signed-off-by: Parav Pandit <pandit.parav@gmail.com> Reviewed-by: Haggai Eran <haggaie@mellanox.com> Reviewed-by: Sagi Grimberg <sagig@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-03IB/core: Replace memset with eth_zero_addrAmitoj Kaur Chawla1-1/+1
Use eth_zero_addr to assign the zero address to the given address array instead of memset when second argument is address of zero. The Coccinelle semantic patch used to make this change is as follows: // <smpl> @eth_zero_addr@ expression e; @@ -memset(e,0x00,ETH_ALEN); +eth_zero_addr(e); // </smpl> Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-03IB/core: Modify conditional on ucontext existenceEli Cohen1-6/+6
Since we allow to call legacy verbs using their extended counterpart, the check on ucontext has to move up to a common area in case this verb is ever extended. Signed-off-by: Eli Cohen <eli@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-03IB/core: IB/core: Allow legacy verbs through extended interfacesEli Cohen1-10/+19
When an extended verb is an extension to a legacy verb, the original functionality is preserved. Hence we do not require each hardware driver to set the extended capability. This will allow the use of the extended verb in its simple form with drivers that do not support the extended capability. Signed-off-by: Eli Cohen <eli@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-03IB/core: Avoid duplicate codeEli Cohen1-20/+9
Move the check on the validity of the command to a common area. Signed-off-by: Eli Cohen <eli@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-03Merge branch 'k.o/for-4.5-rc' into HEADDoug Ledford3-22/+29
2016-03-03IB/{core, mlx5}: Fix input len in vendor part of create_qp/srqMajd Dibbiny2-11/+9
Currently, the inlen field of the vendor's part of the command doesn't match the command buffer. This happens because the inlen accommodates ib_uverbs_cmd_hdr which is deducted from the in buffer. This is problematic since the vendor function could be called either from the legacy verb (where the input length mismatches the actual length) or by the extended verb (where the length matches). The vendor has no idea which function calls it and therefore has no way to know how the length variable should be treated. Fixing this by aligning the inlen to the correct length. All vendor drivers either assumed that inlen >= sizeof(vendor_uhw_cmd) or just failed wrongly (mlx5) and fixed in this patch. Fixes: cfb5e088e26a ('IB/mlx5: Add CQE version 1 support to user QPs and SRQs') Signed-off-by: Majd Dibbiny <majd@mellanox.com> Reviewed-by: Matan Barak <matanb@mellanox.com> Reviewed-by: Haggai Eran <haggaie@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-03IB/mlx5: Avoid using user-index for SRQsMajd Dibbiny1-11/+19
Normal SRQs, unlike XRC SRQs, don't have user-index, therefore avoid verifying it and using it. Fixes: cfb5e088e26a ('IB/mlx5: Add CQE version 1 support to user QPs and SRQs') Signed-off-by: Majd Dibbiny <majd@mellanox.com> Reviewed-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-03IB/ipoib: Add handling for sending of skb with many fragsHans Westgaard Ry4-3/+45
IPoIB converts skb-fragments to sge adding 1 extra sge when SG is enabled. Current codepath assumes that the max number of sge a device support is at least MAX_SKB_FRAGS+1, there is no interaction with upper layers to limit number of fragments in an skb if a device suports fewer sges. The assumptions also lead to requesting a fixed number of sge when IPoIB creates queue-pairs with SG enabled. A fallback/slowpath is implemented using skb_linearize to handle cases where the conversion would result in more sges than supported. Signed-off-by: Hans Westgaard Ry <hans.westgaard.ry@oracle.com> Reviewed-by: Håkon Bugge <haakon.bugge@oracle.com> Reviewed-by: Wei Lin Guay <wei.lin.guay@oracle.com> Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-02mlx4: Implement devlink interfaceJiri Pirko1-0/+7
Implement newly introduced devlink interface. Add devlink port instances for every port and set the port types accordingly. Signed-off-by: Jiri Pirko <jiri@mellanox.com> v2->v3: -add dev param to devlink_register (api change) Signed-off-by: David S. Miller <davem@davemloft.net>
2016-03-01IB/mlx5: Add memory windows allocation supportMatan Barak4-0/+117
This patch adds user-space support for memory windows allocation and deallocation. It also exposes the supported types via query_device_caps verb. Signed-off-by: Matan Barak <matanb@mellanox.com> Reviewed-by: Yishai Hadas <yishaih@mellanox.com> Tested-by: Max Gurtovoy <maxg@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-01IB/core: Add vendor's specific data to alloc mwMatan Barak7-7/+21
Passing udata to the vendor's driver in order to pass data from the user-space driver to the kernel-space driver. This data will be used in downstream patches. Signed-off-by: Matan Barak <matanb@mellanox.com> Reviewed-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-01net/mlx5: Refactor mlx5_core_mr to mkeyMatan Barak4-45/+45
Mlx5's mkey mechanism is also used for memory windows. The current code base uses MR (memory region) naming, which is inaccurate. Changing MR to mkey in order to represent its different usages more accurately. Signed-off-by: Matan Barak <matanb@mellanox.com> Reviewed-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-01IB/mlx5: Added support for re-registration of MRsNoa Osherovich4-3/+227
This patch adds support for re-registration of memory regions in MLX5. The functionality is basically the same as deregister followed by register, but attempts to reuse the existing resources as much as possible. Original memory keys are kept if possible, saving the need to communicate new ones to remote peers. Signed-off-by: Noa Osherovich <noaos@mellanox.com> Reviewed-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-01IB/mlx5: Refactoring register MR codeNoa Osherovich1-75/+132
In order to add re-registration of memory region, some logic was extracted to separate functions: - ODP related logic. - Some of the UMR WQE preparation code. - DMA mapping. - Umem creation. - Creating MKey using FW interface. - MR fields assignments after successful creation. Signed-off-by: Noa Osherovich <noaos@mellanox.com> Reviewed-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-01IB/cma: Print warning on different inner and header P_KeysHaggai Eran1-0/+8
Commit 4c21b5bcef73 ("IB/cma: Add net_dev and private data checks to RDMA CM") added checks for incoming RDMA CM requests that they can be matched to a netdev based on the P_Key in the BTH of the request. This behavior was reverted in commit ab3964ad2acf ("IB/cma: Use inner P_Key to determine netdev"), since the mlx5 and ipath drivers didn't send the correct value in the BTH P_Key. Since the ipath driver was removed, and the mlx5 driver can now send GSI packets on different P_Keys, we could revert the patch to let the rdma_cm module look on the BTH P_Key when deciding to what netdev a packet belongs. However, that still breaks compatibility with the older drivers. Change the behavior to print a warning when receiving a request that has a different BTH P_Key and inner payload P_Key. In the future, after users have seen the warnings and upgraded their setups, remove the warning and block these requests. Signed-off-by: Haggai Eran <haggaie@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-01IB/mlx5: Eliminate GSI RX QP's send buffersHaggai Eran1-0/+5
Now that the transmission of GSI MADs is done with the special transmission QPs, eliminate the send buffers in the GSI receive QP. Reviewed-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Haggai Eran <haggaie@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-01IB/mlx5: Pick the right GSI transmission QP for sendingHaggai Eran1-1/+47
Pick the QP to use according to the wr.ud.pkey_index field in the work request. If the QP doesn't exist, it means the P_Key is zero and the packet would have been dropped, so just generate a completion and move on. Reviewed-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Haggai Eran <haggaie@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-01IB/mlx5: Reorder GSI completionsHaggai Eran1-6/+150
The emulated GSI QP's send completions are generated by multiple hardware QPs, so their completions could arrive out of order with respect to the order their work request were submitted. Reorder the completions by keeping a list of the posted work request and their completions. A newly received completion from the hardware updates the list and marks its work request as completed. However, the completions are only reported to the client according to the list order. In order to support that, create a new private CQ to handle the hardware completions. Signed-off-by: Haggai Eran <haggaie@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-01IB/mlx5: Generate completions in softwareHaggai Eran2-5/+88
The GSI QP emulation requires also emulating completions for transmitted MADs. The CQ on which these completions are generated can also be used by the hardware, and the MAD layer is free to use any CQ of the device for the GSI QP. Add a method for generating software completions to each mlx5 CQ. Software completions are polled first, and generate calls to the completion handler callback if necessary. Reviewed-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Haggai Eran <haggaie@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-01IB/mlx5: Create GSI transmission QPs when P_Key table is changedHaggai Eran3-0/+41
Whenever the P_Key table is changed, we create the required GSI transmission QPs on-demand. Reviewed-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Haggai Eran <haggaie@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-01IB/mlx5: Create multiple transmission GSI QPsHaggai Eran1-3/+167
In order to send GSI MADs on different P_Keys, mlx5 needs different QPs to be created, each with a different P_Key set when the QP is modified to the INIT state. Create QPs for each non-zero P_Key in the P_Key table. Signed-off-by: Haggai Eran <haggaie@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2016-03-01IB/mlx5: Add GSI QP wrapperHaggai Eran5-14/+249
mlx5 creates special GSI QPs that has limited ability to control the P_Key of transmitted packets. The sent P_Key is taken from the QP object, similarly to what happens with regular UD QPs. Create a software wrapper around GSI QPs that with the following patches will be able to emulate the functionality of a GSI QP including control of the P_Key per work request. Reviewed-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Haggai Eran <haggaie@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
generated by cgit v1.2.3 (git 2.39.0) at 2025-02-09 08:56:10 +0300