- 28 May, 2018 4 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdmaJason Gunthorpe authored
Update mlx4 to support user MR creation against read-only memory, previously it required the memory to be writable. Based on rdma for-rc due to dependencies. * mr_fix: (2 commits) IB/mlx4: Mark user MR as writable if actual virtual memory is writable IB/core: Make testing MR flags for writability a static inline function
-
Jack Morgenstein authored
To allow rereg_user_mr to modify the MR from read-only to writable without using get_user_pages again, we needed to define the initial MR as writable. However, this was originally done unconditionally, without taking into account the writability of the underlying virtual memory. As a result, any attempt to register a read-only MR over read-only virtual memory failed. To fix this, do not add the writable flag bit when the user virtual memory is not writable (e.g. const memory). However, when the underlying memory is NOT writable (and we therefore do not define the initial MR as writable), the IB core adds a "force writable" flag to its user-pages request. If this succeeds, the reg_user_mr caller gets a writable copy of the original pages. If the user-space caller then does a rereg_user_mr operation to enable writability, this will succeed. This should not be allowed, since the original virtual memory was not writable. Cc: <stable@vger.kernel.org> Fixes: 9376932d ("IB/mlx4_ib: Add support for user MR re-registration") Signed-off-by:
Jason Gunthorpe <jgg@mellanox.com> Signed-off-by:
Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by:
Leon Romanovsky <leonro@mellanox.com>
-
Jack Morgenstein authored
Make the MR writability flags check, which is performed in umem.c, a static inline function in file ib_verbs.h This allows the function to be used by low-level infiniband drivers. Cc: <stable@vger.kernel.org> Signed-off-by:
-
Zhu Yanjun authored
The function rxe_remove_all is only used in this modules. There is no other modules that call this function. So it is not necessary to export it. Signed-off-by:
Zhu Yanjun <yanjun.zhu@oracle.com> Reviewed-by:
-
- 24 May, 2018 21 commits
-
-
Wei Hu(Xavier) authored
This patch increases checking CMQ status timeout value and uses the same value with NIC driver to avoid deficiency of time. Signed-off-by:
Wei Hu (Xavier) <xavier.huwei@huawei.com> Reviewed-by:
-
Wei Hu(Xavier) authored
This patch modified uar allocation algorithm in hns_roce_uar_alloc function to avoid bitmap exhaust. Signed-off-by:
-
Jason Gunthorpe authored
Merge tag 'mlx5-updates-2018-05-17' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux into for-next mlx5-updates-2018-05-17 mlx5 core dirver updates for both net-next and rdma-next branches. From Christophe JAILLET, first three patche to use kvfree where needed. From: Or Gerlitz <ogerlitz@mellanox.com> Next six patches from Roi and Co adds support for merged sriov e-switch which comes to serve cases where both PFs, VFs set on them and both uplinks are to be used in single v-switch SW model. When merged e-switch is supported, the per-port e-switch is logically merged into one e-switch that spans both physical ports and all the VFs. This model allows to offload TC eswitch rules between VFs belonging to different PFs (and hence have different eswitch affinity), it also sets the some of the foundations needed for uplink LAG support. * tag 'mlx5-updates-2018-05-17' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux: net/mlx5e: Explicitly set source e-switch in offloaded TC rules net/mlx5: Add source e-switch owner net/mlx5e: Explicitly set destination e-switch in FDB rules net/mlx5: Add destination e-switch owner net/mlx5: Properly handle a vport destination when setting FTE net/mlx5: Add merged e-switch cap IB/mlx5: Use 'kvfree()' for memory allocated by 'kvzalloc()' net/mlx5: Eswitch, Use 'kvfree()' for memory allocated by 'kvzalloc()' net/mlx5: Vport, Use 'kvfree()' for memory allocated by 'kvzalloc()'
-
Parav Pandit authored
There are several places a gid table is accessed. Have a helper tiny function rdma_gid_table() to avoid code duplication at such places. Signed-off-by:
Parav Pandit <parav@mellanox.com> Signed-off-by:
-
Parav Pandit authored
Instead of open coding memcmp() to check whether a given GID is zero or not, use a helper function to do so, and replace instances of memcpy(z,&zgid) with memset. Signed-off-by:
-
Erez Shitrit authored
On fatal error the driver simulates CQE's for ULPs that rely on completion of all their posted work-request. For the GSI traffic, the mlx5 has its own mechanism that sends the completions via software CQE's directly to the relevant CQ. This should be kept in fatal error too, so the driver should simulate such CQE's with the specified error state in order to complete GSI QP work requests. Without the fix the next deadlock might appears: schedule_timeout+0x274/0x350 wait_for_common+0xec/0x240 mcast_remove_one+0xd0/0x120 [ib_core] ib_unregister_device+0x12c/0x230 [ib_core] mlx5_ib_remove+0xc4/0x270 [mlx5_ib] mlx5_detach_device+0x184/0x1a0 [mlx5_core] mlx5_unload_one+0x308/0x340 [mlx5_core] mlx5_pci_err_detected+0x74/0xe0 [mlx5_core] Cc: <stable@vger.kernel.org> # 4.7 Fixes: 89ea94a7 ("IB/mlx5: Reset flow support for IB kernel ULPs") Signed-off-by:Erez Shitrit <erezsh@mellanox.com> Signed-off-by:
-
Leon Romanovsky authored
In commit 357d23c811a7 ("Remove the obsolete libibcm library") in rdma-core [1], we removed obsolete library which used the /dev/infiniband/ucmX interface. Following multiple syzkaller reports about non-sanitized user input in the UCMA module, the short audit reveals the same issues in UCM module too. It is better to disable this interface in the kernel, before syzkaller team invests time and energy to harden this unused interface. [1] https://github.com/linux-rdma/rdma-core/pull/279Signed-off-by: -
Parav Pandit authored
Remove duplicate declaration of gid_cache_wq. Fixes: d4186194 ("IB/core: Add generic function to extract IB speed from netdev") Signed-off-by:
-
Leon Romanovsky authored
Remove various prints of VMA pointers. Reported-by:
Michal Kalderon <Michal.Kalderon@cavium.com> Signed-off-by:
-
oulijun authored
The lower 15 bit of paramter of db structure means different meanings when db type is sq, rq and srq. Signed-off-by:
Lijun Ou <oulijun@huawei.com> Signed-off-by:
-
Mike Marciniszyn authored
The following error occurs in a debug build when running MPI PSM: [ 307.415911] WARNING: CPU: 4 PID: 23867 at lib/dma-debug.c:1158 check_unmap+0x4ee/0xa20 [ 307.455661] ib_qib 0000:05:00.0: DMA-API: device driver failed to check map error[device address=0x00000000df82b000] [size=4096 bytes] [mapped as page] [ 307.517494] Modules linked in: [ 307.531584] ib_isert iscsi_target_mod ib_srpt target_core_mod rpcrdma sunrpc ib_srp scsi_transport_srp scsi_tgt ib_iser libiscsi ib_ipoib scsi_transport_iscsi rdma_ucm ib_ucm ib_uverbs ib_umad rdma_cm ib_cm iw_cm ib_qib intel_powerclamp coretemp rdmavt intel_rapl iosf_mbi kvm_intel kvm irqbypass crc32_pclmul ghash_clmulni_intel ipmi_ssif ib_core aesni_intel sg ipmi_si lrw gf128mul dca glue_helper ipmi_devintf iTCO_wdt gpio_ich hpwdt iTCO_vendor_support ablk_helper hpilo acpi_power_meter cryptd ipmi_msghandler ie31200_edac shpchp pcc_cpufreq lpc_ich pcspkr ip_tables xfs libcrc32c sd_mod crc_t10dif crct10dif_generic mgag200 i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm ahci crct10dif_pclmul crct10dif_common drm crc32c_intel libahci tg3 libata serio_raw ptp i2c_core [ 307.846113] pps_core dm_mirror dm_region_hash dm_log dm_mod [ 307.866505] CPU: 4 PID: 23867 Comm: mpitests-IMB-MP Kdump: loaded Not tainted 3.10.0-862.el7.x86_64.debug #1 [ 307.911178] Hardware name: HP ProLiant DL320e Gen8, BIOS J05 11/09/2013 [ 307.944206] Call Trace: [ 307.956973] [<ffffffffbd9e915b>] dump_stack+0x19/0x1b [ 307.982201] [<ffffffffbd2a2f58>] __warn+0xd8/0x100 [ 308.005999] [<ffffffffbd2a2fdf>] warn_slowpath_fmt+0x5f/0x80 [ 308.034260] [<ffffffffbd5f667e>] check_unmap+0x4ee/0xa20 [ 308.060801] [<ffffffffbd41acaa>] ? page_add_file_rmap+0x2a/0x1d0 [ 308.090689] [<ffffffffbd5f6c4d>] debug_dma_unmap_page+0x9d/0xb0 [ 308.120155] [<ffffffffbd4082e0>] ? might_fault+0xa0/0xb0 [ 308.146656] [<ffffffffc07761a5>] qib_tid_free.isra.14+0x215/0x2a0 [ib_qib] [ 308.180739] [<ffffffffc0776bf4>] qib_write+0x894/0x1280 [ib_qib] [ 308.210733] [<ffffffffbd540b00>] ? __inode_security_revalidate+0x70/0x80 [ 308.244837] [<ffffffffbd53c2b7>] ? security_file_permission+0x27/0xb0 [ 308.266025] qib_ib0.8006: multicast join failed for ff12:401b:8006:0000:0000:0000:ffff:ffff, status -22 [ 308.323421] [<ffffffffbd46f5d3>] vfs_write+0xc3/0x1f0 [ 308.347077] [<ffffffffbd492a5c>] ? fget_light+0xfc/0x510 [ 308.372533] [<ffffffffbd47045a>] SyS_write+0x8a/0x100 [ 308.396456] [<ffffffffbd9ff355>] system_call_fastpath+0x1c/0x21 The code calls a qib_map_page() which has never correctly tested for a mapping error. Fix by testing for pci_dma_mapping_error() in all cases and properly handling the failure in the caller. Additionally, streamline qib_map_page() arguments to satisfy just the single caller. Cc: <stable@vger.kernel.org> Reviewed-by:
Alex Estrin <alex.estrin@intel.com> Tested-by:
Don Dutile <ddutile@redhat.com> Reviewed-by:
Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by:
Dennis Dalessandro <dennis.dalessandro@intel.com> Signed-off-by:
Doug Ledford <dledford@redhat.com>
-
Alex Estrin authored
The following error message occurs on a target host in a debug build during session login: [ 3524.411874] WARNING: CPU: 5 PID: 12063 at lib/dma-debug.c:1207 check_sync+0x4ec/0x5b0 [ 3524.421057] infiniband hfi1_0: DMA-API: device driver tries to sync DMA memory it has not allocated [device address=0x0000000000000000] [size=76 bytes] ......snip ..... [ 3524.535846] CPU: 5 PID: 12063 Comm: iscsi_np Kdump: loaded Not tainted 3.10.0-862.el7.x86_64.debug #1 [ 3524.546764] Hardware name: Dell Inc. PowerEdge R430/03XKDV, BIOS 1.2.6 06/08/2015 [ 3524.555740] Call Trace: [ 3524.559102] [<ffffffffa5fe915b>] dump_stack+0x19/0x1b [ 3524.565477] [<ffffffffa58a2f58>] __warn+0xd8/0x100 [ 3524.571557] [<ffffffffa58a2fdf>] warn_slowpath_fmt+0x5f/0x80 [ 3524.578610] [<ffffffffa5bf5b8c>] check_sync+0x4ec/0x5b0 [ 3524.585177] [<ffffffffa58efc3f>] ? set_cpus_allowed_ptr+0x5f/0x1c0 [ 3524.592812] [<ffffffffa5bf5cd0>] debug_dma_sync_single_for_cpu+0x80/0x90 [ 3524.601029] [<ffffffffa586add3>] ? x2apic_send_IPI_mask+0x13/0x20 [ 3524.608574] [<ffffffffa585ee1b>] ? native_smp_send_reschedule+0x5b/0x80 [ 3524.616699] [<ffffffffa58e9b76>] ? resched_curr+0xf6/0x140 [ 3524.623567] [<ffffffffc0879af0>] isert_create_send_desc.isra.26+0xe0/0x110 [ib_isert] [ 3524.633060] [<ffffffffc087af95>] isert_put_login_tx+0x55/0x8b0 [ib_isert] [ 3524.641383] [<ffffffffa58ef114>] ? try_to_wake_up+0x1a4/0x430 [ 3524.648561] [<ffffffffc098cfed>] iscsi_target_do_tx_login_io+0xdd/0x230 [iscsi_target_mod] [ 3524.658557] [<ffffffffc098d827>] iscsi_target_do_login+0x1a7/0x600 [iscsi_target_mod] [ 3524.668084] [<ffffffffa59f9bc9>] ? kstrdup+0x49/0x60 [ 3524.674420] [<ffffffffc098e976>] iscsi_target_start_negotiation+0x56/0xc0 [iscsi_target_mod] [ 3524.684656] [<ffffffffc098c2ee>] __iscsi_target_login_thread+0x90e/0x1070 [iscsi_target_mod] [ 3524.694901] [<ffffffffc098ca50>] ? __iscsi_target_login_thread+0x1070/0x1070 [iscsi_target_mod] [ 3524.705446] [<ffffffffc098ca50>] ? __iscsi_target_login_thread+0x1070/0x1070 [iscsi_target_mod] [ 3524.715976] [<ffffffffc098ca78>] iscsi_target_login_thread+0x28/0x60 [iscsi_target_mod] [ 3524.725739] [<ffffffffa58d60ff>] kthread+0xef/0x100 [ 3524.732007] [<ffffffffa58d6010>] ? insert_kthread_work+0x80/0x80 [ 3524.739540] [<ffffffffa5fff1b7>] ret_from_fork_nospec_begin+0x21/0x21 [ 3524.747558] [<ffffffffa58d6010>] ? insert_kthread_work+0x80/0x80 [ 3524.755088] ---[ end trace 23f8bf9238bd1ed8 ]--- [ 3595.510822] iSCSI/iqn.1994-05.com.redhat:537fa56299: Unsupported SCSI Opcode 0xa3, sending CHECK_CONDITION. The code calls dma_sync on login_tx_desc->dma_addr prior to initializing it with dma-mapped address. login_tx_desc is a part of iser_conn structure and is used only once during login negotiation, so the issue is fixed by eliminating dma_sync call for this buffer using a special case routine. Cc: <stable@vger.kernel.org> Reviewed-by:
-
Mike Marciniszyn authored
Given we are dealing with nano-second level timers, when the timer pops, ensure it happens on the CPU which caused the timer to be set in the first place. This avoids excessive jitter from the desired expiration time by avoiding the cost of switching our context to another CPU that is cache cold for this given timer. Reviewed-by:
Kaike Wan <kaike.wan@intel.com> Signed-off-by:
-
Michael J. Ruhl authored
For errorinfo MAD requests, the response has a 0 port number left over from a memset. Instead we should always set the port number in the response. Reviewed-by:
Michael J. Ruhl <michael.j.ruhl@intel.com> Signed-off-by:
-
Mike Marciniszyn authored
The knowledge of the internal workings of the expect receive is too distributed. Fix by: - right size several rcd fields associated with expect receive - making an init entrance to init all the lists - consolidate all the allocations into an array anchored in the rcd Reviewed-by:
-
Don Hiatt authored
Add trace support for 16B Management Packets. Reviewed-by:
Ira Weiny <ira.weiny@intel.com> Signed-off-by:
Don Hiatt <don.hiatt@intel.com> Signed-off-by:
-
Don Hiatt authored
16B Management Packets (L4=0x08) replace the BTH and DETH of normal MAD packet packets with a header containing the the source and destination queue pair numbers; fields that were originally retrieved from the BTH/DETH are now populated from this header as well as from the 16B LRH (e.g. pkey). 16B Management Packets are used as an optimized management format on 16B fabrics. These management packets have an opcode of IB_OPCODE_UD_SEND_ONLY, a fixed 3Byte pad, and a header length of 24Bytes. The decision as to when we send a management packet is based upon either the source or destination queue pair number being 0 or 1. Reviewed-by:
-
Don Hiatt authored
Add 16B Management Packet definition. This optimized packet format replaces the ib_other_headers and BTH with a source and destination QP number. To support these packets we introduce struct opa_16b_mgmt into the struct hfi1_16b_header. This packet format is only used for MAD packets using the IB_OPCODE_UD_SEND_ONLY opcode on QP0/1. The original 16B implementation failed to use 16B management packets so now we add their definition. Reviewed-by:
-
Steve Wise authored
Add a table of important fields from the fw_ri_tpte structure to the mr resource tracking table. This is helpful in debugging. Signed-off-by:
Steve Wise <swise@opengridcomputing.com> Signed-off-by:
-
Steve Wise authored
Add a table of important fields from the c4iw_cq* structures to the cq resource tracking table. This is helpful in debugging. Signed-off-by:
-
Steve Wise authored
Add a table of important fields from the c4iw_ep* structures to the cm_id resource tracking table. This is helpful in debugging. Signed-off-by:
-
- 23 May, 2018 4 commits
-
-
oulijun authored
When posted work request, it need to compute the length of all sges of every wr and fill it into the msg_len field of send wqe. Thus, While posting multiple wr, tmp_len should be reinitialized to zero. Fixes: 8b9b8d14 ("RDMA/hns: Fix the endian problem for hns") Signed-off-by:
-
oulijun authored
When use cq record db for kernel, it needs to set the hr_cq->db_en to 1 and configure the dma address of record cq db of qp context. Fixes: 86188a88 ("RDMA/hns: Support cq record doorbell for kernel space") Signed-off-by:
-
Jason Gunthorpe authored
The err pointer comes from uverbs_attr_get, not from the uobject member, which does not store an ERR_PTR. Fixes: be934cca ("IB/uverbs: Add device memory registration ioctl support") Signed-off-by:
-
Kalderon, Michal authored
Each user_context receives a separate dpi value and thus a different address on the doorbell bar. The qedr_mmap function needs to validate the address and map the doorbell bar accordingly. The current implementation always checked against dpi=0 doorbell range leading to a wrong mapping for doorbell bar. (It entered an else case that mapped the address differently). qedr_mmap should only be used for doorbells, so the else was actually wrong in the first place. This only has an affect on arm architecture and not an issue on a x86 based architecture. This lead to doorbells not occurring on arm based systems and left applications that use more than one dpi (or several applications run simultaneously ) to hang. Fixes: ac1b36e5 ("qedr: Add support for user context verbs") Signed-off-by:
Ariel Elior <Ariel.Elior@cavium.com> Signed-off-by:
-
- 22 May, 2018 3 commits
-
-
Steve Wise authored
Add a helper function for iwarp drivers to be able to map an rdma_cm_id to an iw_cm_id. This is useful for dumping driver specific NLDEV/RESTRACK connection state. Add a helper to return the rdma_cm_id pointer from the rdma_restack pointer. This is needed for rdma drivers to map a res entry back to the public rdma_cm_id struct. Signed-off-by:
-
Steve Wise authored
In active side connections, the provider_data field is not getting set. This will be used in a subsequent patch to dump state, so always set it. Signed-off-by:
-
Doug Ledford authored
We do a light flush on CLIENT_REREG and SM_CHANGE events. This goes through and marks paths invalid. But we weren't always checking for this validity when we needed to, and so we could keep using a path marked invalid. What's more, once we establish a path with a valid ah, we put a pointer to the ah in the neigh struct directly, so even if we mark the path as invalid, as long as the neigh has a direct pointer to the ah, it keeps using the old, outdated ah. To fix this we do several things. 1) Put the valid flag in the ah instead of the path struct, so when we put the ah pointer directly in the neigh struct, we can easily check the validity of the ah on send events. 2) Check the neigh->ah and neigh->ah->valid elements in the needed places, and if we have an ah, but it's invalid, then invoke a refresh of the ah. 3) Fix the various places that check for path, but didn't check for path->valid (now path->ah && path->ah->valid). Reported-by:
Evgenii Smirnov <evgenii.smirnov@profitbricks.com> Fixes: ee1e2c82 ("IPoIB: Refresh paths instead of flushing them on SM change events") Signed-off-by:
-
- 17 May, 2018 8 commits
-
-
Shahar Klein authored
Set a specific source e-switch when setting a rule that matches on the ingress port. Signed-off-by:
Shahar Klein <shahark@mellanox.com> Reviewed-by:
Or Gerlitz <ogerlitz@mellanox.com> Reviewed-by:
Roi Dayan <roid@mellanox.com> Signed-off-by:
Saeed Mahameed <saeedm@mellanox.com>
-
Shahar Klein authored
The source e-switch owner allows a vport on one e-switch port be associated with a rule defined on the second port e-switch. The role of the source eswitch owner valid bit in the flow group is to allow the firmware fail driver attempts to wild card the source eswitch match field. If this bit is not set, the firmware ignores the source eswitch owner field totally. Signed-off-by:
-
Rabie Loulou authored
Set a specific destination e-switch when setting a destination vport. Signed-off-by:
Rabie Loulou <rabiel@mellanox.com> Reviewed-by:
-
Shahar Klein authored
The destination e-switch owner allows a rule in namespace of one e-switch owner to point to a vport that is natively associated with another e-switch owner. Signed-off-by:
-
Shahar Klein authored
When creating FTE, properly distinguish between destination being vport or tir. The previous code just worked accidentally b/c of both dest being in the same offset within a union. Signed-off-by:
-
Roi Dayan authored
When merged e-switch is supported, the per-port e-switch is logically merged into one e-switch that spans both physical ports and all the VFs. Under merged eswitch, both the matching on source vport and setting destination vport can have a 2nd attribute which is the vhca id of the eswitch owner. For example: esw0: {match: <src vport=1 owner=0> action: fwd to <dst vport=7, owner=1>} is a flow set on eswitch0 matching on source vport=1 from his eswitch and the action being fwd to dest vport=7 of eswitch1. Signed-off-by: -
Zhu Yanjun authored
In the exit branch, WARN_ON_ONCE is called to show stack. So it is not necessary to call WARN_ON_ONCE before going to exit. Signed-off-by:
-
Yixian Liu authored
This patch adds the support of 64KB page size for hip08 in kernel. Signed-off-by:
Yixian Liu <liuyixian@huawei.com> Signed-off-by:
-
