summaryrefslogtreecommitdiff
path: root/drivers/infiniband/ulp
AgeCommit message (Collapse)Author
2021-04-13RDMA/rtrs-clt: Simplify error messageGioh Kim
Two error messages are only different message but have common code to generate the path string. Link: https://lore.kernel.org/r/20210406123639.202899-4-gi-oh.kim@ionos.com Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-13RDMA/rtrs-srv: More debugging info when fail to send replyGioh Kim
It does not help to debug if it only print error message without any debugging information which session and connection the error happened. Link: https://lore.kernel.org/r/20210406123639.202899-3-gi-oh.kim@ionos.com Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-13RDMA/rtrs-clt: Print more info when an error happensGioh Kim
Client prints only error value and it is not enough for debugging. 1. When client receives an error from server: the client does not only print the error value but also more information of server connection. 2. When client failes to send IO: the client gets an error from RDMA layer. It also print more information of server connection. Link: https://lore.kernel.org/r/20210406123639.202899-2-gi-oh.kim@ionos.com Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-13RDMA/rtrs-clt: New sysfs attribute to print the latency of each pathGioh Kim
It shows the latest latency that the client checked when sending the heart-beat. Link: https://lore.kernel.org/r/20210407113444.150961-3-gi-oh.kim@ionos.com Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-13RDMA/rtrs-clt: Add a minimum latency multipath policyGioh Kim
This patch adds new multipath policy: min-latency. Client checks the latency of each path when it sends the heart-beat. And it sends IO to the path with the minimum latency. Link: https://lore.kernel.org/r/20210407113444.150961-2-gi-oh.kim@ionos.com Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-12RDMA/rtrs-clt: destroy sysfs after removing session from active listGioh Kim
A session can be removed dynamically by sysfs interface "remove_path" that eventually calls rtrs_clt_remove_path_from_sysfs function. The current rtrs_clt_remove_path_from_sysfs first removes the sysfs interfaces and frees sess->stats object. Second it removes the session from the active list. Therefore some functions could access non-connected session and access the freed sess->stats object even-if they check the session status before accessing the session. For instance rtrs_clt_request and get_next_path_min_inflight check the session status and try to send IO to the session. The session status could be changed when they are trying to send IO but they could not catch the change and update the statistics information in sess->stats object, and generate use-after-free problem. (see: "RDMA/rtrs-clt: Check state of the rtrs_clt_sess before reading its stats") This patch changes the rtrs_clt_remove_path_from_sysfs to remove the session from the active session list and then destroy the sysfs interfaces. Each function still should check the session status because closing or error recovery paths can change the status. Fixes: 6a98d71daea1 ("RDMA/rtrs: client: main functionality") Link: https://lore.kernel.org/r/20210412084002.33582-1-gi-oh.kim@ionos.com Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Reviewed-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-12RDMA/srpt: Fix error return code in srpt_cm_req_recv()Wang Wensheng
Fix to return a negative error code from the error handling case instead of 0, as done elsewhere in this function. Fixes: db7683d7deb2 ("IB/srpt: Fix login-related race conditions") Link: https://lore.kernel.org/r/20210408113132.87250-1-wangwensheng4@huawei.com Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Wang Wensheng <wangwensheng4@huawei.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-07IB/{ipoib,hfi1}: Add a timeout handler for rdma_netdevMike Marciniszyn
The current rdma_netdev handling in ipoib hooks the tx_timeout handler, but prints out a totally useless message that prevents effective debugging especially when multiple transmit queues are being used. Add a tx_timeout rdma_netdev hook and implement the callback in the hfi1 to print additional information. The existing non-helpful message is avoided when the driver has presented a callback. Link: https://lore.kernel.org/r/1617026056-50483-3-git-send-email-dennis.dalessandro@cornelisnetworks.com Reviewed-by: Kaike Wan <kaike.wan@intel.com> Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-01RDMA/rtrs-clt: Cap max_io_sizeJack Wang
Max io size is limited by both remote buffer size and the max fr pages per mr. Link: https://lore.kernel.org/r/20210325153308.1214057-20-gi-oh.kim@ionos.com Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Reviewed-by: Md Haris Iqbal <haris.iqbal@ionos.com> Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-01RDMA/rtrs: Cleanup unused 's' variable in __alloc_sessJack Wang
Link: https://lore.kernel.org/r/20210325153308.1214057-18-gi-oh.kim@ionos.com Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-01RDMA/rtrs-srv: Report temporary sessname for error messageGioh Kim
Before receiving the session name, the error message cannot include any information about which connection generates the error. This patch stores the addresses of source and target in the sessname field to show which generates the error. That field will be over-written when receiving the session name from client. Link: https://lore.kernel.org/r/20210325153308.1214057-17-gi-oh.kim@ionos.com Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-01RDMA/rtrs: New function converting rtrs_addr to stringGioh Kim
There is common code converting addresses of source machine and destination machine to a string. We already have a struct rtrs_addr to store two addresses. This patch introduces a new function that converts two addresses into one string with struct rtrs_addr. Link: https://lore.kernel.org/r/20210325153308.1214057-14-gi-oh.kim@ionos.com Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-01RDMA/rtrs-clt: Close rtrs client conn before destroying rtrs clt session filesMd Haris Iqbal
KASAN detected the following BUG: BUG: KASAN: use-after-free in rtrs_clt_update_wc_stats+0x41/0x100 [rtrs_client] Read of size 8 at addr ffff88bf2fb4adc0 by task swapper/0/0 CPU: 0 PID: 0 Comm: swapper/0 Tainted: G O 5.4.84-pserver #5.4.84-1+feature+linux+5.4.y+dbg+20201216.1319+b6b887b~deb10 Hardware name: Supermicro H8QG6/H8QG6, BIOS 3.00 09/04/2012 Call Trace: <IRQ> dump_stack+0x96/0xe0 print_address_description.constprop.4+0x1f/0x300 ? irq_work_claim+0x2e/0x50 __kasan_report.cold.8+0x78/0x92 ? rtrs_clt_update_wc_stats+0x41/0x100 [rtrs_client] kasan_report+0x10/0x20 rtrs_clt_update_wc_stats+0x41/0x100 [rtrs_client] rtrs_clt_rdma_done+0xb1/0x760 [rtrs_client] ? lockdep_hardirqs_on+0x1a8/0x290 ? process_io_rsp+0xb0/0xb0 [rtrs_client] ? mlx4_ib_destroy_cq+0x100/0x100 [mlx4_ib] ? add_interrupt_randomness+0x1a2/0x340 __ib_process_cq+0x97/0x100 [ib_core] ib_poll_handler+0x41/0xb0 [ib_core] irq_poll_softirq+0xe0/0x260 __do_softirq+0x127/0x672 irq_exit+0xd1/0xe0 do_IRQ+0xa3/0x1d0 common_interrupt+0xf/0xf </IRQ> RIP: 0010:cpuidle_enter_state+0xea/0x780 Code: 31 ff e8 99 48 47 ff 80 7c 24 08 00 74 12 9c 58 f6 c4 02 0f 85 53 05 00 00 31 ff e8 b0 6f 53 ff e8 ab 4f 5e ff fb 8b 44 24 04 <85> c0 0f 89 f3 01 00 00 48 8d 7b 14 e8 65 1e 77 ff c7 43 14 00 00 RSP: 0018:ffffffffab007d58 EFLAGS: 00000246 ORIG_RAX: ffffffffffffffca RAX: 0000000000000002 RBX: ffff88b803d69800 RCX: ffffffffa91a8298 RDX: 0000000000000007 RSI: dffffc0000000000 RDI: ffffffffab021414 RBP: ffffffffab6329e0 R08: 0000000000000002 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000002 R13: 000000bf39d82466 R14: ffffffffab632aa0 R15: ffffffffab632ae0 ? lockdep_hardirqs_on+0x1a8/0x290 ? cpuidle_enter_state+0xe5/0x780 cpuidle_enter+0x3c/0x60 do_idle+0x2fb/0x390 ? arch_cpu_idle_exit+0x40/0x40 ? schedule+0x94/0x120 cpu_startup_entry+0x19/0x1b start_kernel+0x5da/0x61b ? thread_stack_cache_init+0x6/0x6 ? load_ucode_amd_bsp+0x6f/0xc4 ? init_amd_microcode+0xa6/0xa6 ? x86_family+0x5/0x20 ? load_ucode_bsp+0x182/0x1fd secondary_startup_64+0xa4/0xb0 Allocated by task 5730: save_stack+0x19/0x80 __kasan_kmalloc.constprop.9+0xc1/0xd0 kmem_cache_alloc_trace+0x15b/0x350 alloc_sess+0xf4/0x570 [rtrs_client] rtrs_clt_open+0x3b4/0x780 [rtrs_client] find_and_get_or_create_sess+0x649/0x9d0 [rnbd_client] rnbd_clt_map_device+0xd7/0xf50 [rnbd_client] rnbd_clt_map_device_store+0x4ee/0x970 [rnbd_client] kernfs_fop_write+0x141/0x240 vfs_write+0xf3/0x280 ksys_write+0xba/0x150 do_syscall_64+0x68/0x270 entry_SYSCALL_64_after_hwframe+0x49/0xbe Freed by task 5822: save_stack+0x19/0x80 __kasan_slab_free+0x125/0x170 kfree+0xe7/0x3f0 kobject_put+0xd3/0x240 rtrs_clt_destroy_sess_files+0x3f/0x60 [rtrs_client] rtrs_clt_close+0x3c/0x80 [rtrs_client] close_rtrs+0x45/0x80 [rnbd_client] rnbd_client_exit+0x10f/0x2bd [rnbd_client] __x64_sys_delete_module+0x27b/0x340 do_syscall_64+0x68/0x270 entry_SYSCALL_64_after_hwframe+0x49/0xbe When rtrs_clt_close is triggered, it iterates over all the present rtrs_clt_sess and triggers close on them. However, the call to rtrs_clt_destroy_sess_files is done before the rtrs_clt_close_conns. This is incorrect since during the initialization phase we allocate rtrs_clt_sess first, and then we go ahead and create rtrs_clt_con for it. If we free the rtrs_clt_sess structure before closing the rtrs_clt_con, it may so happen that an inflight IO completion would trigger the function rtrs_clt_rdma_done, which would lead to the above UAF case. Hence close the rtrs_clt_con connections first, and then trigger the destruction of session files. Fixes: 6a98d71daea1 ("RDMA/rtrs: client: main functionality") Link: https://lore.kernel.org/r/20210325153308.1214057-12-gi-oh.kim@ionos.com Signed-off-by: Md Haris Iqbal <haris.iqbal@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-01RDMA/rtrs: Cleanup the code in rtrs_srv_rdma_cm_handlerGuoqing Jiang
Let these cases share the same path since all of them need to close session. Link: https://lore.kernel.org/r/20210325153308.1214057-11-gi-oh.kim@ionos.com Signed-off-by: Guoqing Jiang <guoqing.jiang@ionos.com> Reviewed-by: Danil Kipnis <danil.kipnis@ionos.com> Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-01RDMA/rtrs: Remove sessname and sess_kobj from rtrs_attrsGuoqing Jiang
The two members are not used in the code, so remove them. Link: https://lore.kernel.org/r/20210325153308.1214057-10-gi-oh.kim@ionos.com Signed-off-by: Guoqing Jiang <guoqing.jiang@ionos.com> Reviewed-by: Danil Kipnis <danil.kipnis@ionos.com> Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-01RDMA/rtrs: Kill the put label in rtrs_srv_create_once_sysfs_root_foldersGuoqing Jiang
We can remove the label after move put_device to the right place. Link: https://lore.kernel.org/r/20210325153308.1214057-9-gi-oh.kim@ionos.com Signed-off-by: Guoqing Jiang <guoqing.jiang@ionos.com> Reviewed-by: Danil Kipnis <danil.kipnis@ionos.com> Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-04-01RDMA/rtrs-clt: Remove redundant code from rtrs_clt_read_reqGuoqing Jiang
There is no need to dereference 's' from 'sess', since we have "sess = to_clt_sess(s)" before. And we can deference 'dev' from 's' earlier. Link: https://lore.kernel.org/r/20210325153308.1214057-8-gi-oh.kim@ionos.com Signed-off-by: Guoqing Jiang <guoqing.jiang@ionos.com> Reviewed-by: Danil Kipnis <danil.kipnis@ionos.com> Signed-off-by: Gioh Kim <gi-oh.kim@ionos.com> Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-03-30RDMA/iser: struct iscsi_iser_task is declared twiceWan Jiabing
struct iscsi_iser_task has been declared at 201st line. Remove the duplicate. Link: https://lore.kernel.org/r/20210326113347.903976-1-wanjiabing@vivo.com Signed-off-by: Wan Jiabing <wanjiabing@vivo.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-03-26IB/isert: Fix a use after free in isert_connect_requestLv Yunlong
The device is got by isert_device_get() with refcount is 1, and is assigned to isert_conn by isert_conn->device = device. When isert_create_qp() failed, device will be freed with isert_device_put(). Later, the device is used in isert_free_login_buf(isert_conn) by the isert_conn->device->ib_device statement. Free the device in the correct order. Fixes: ae9ea9ed38c9 ("iser-target: Split some logic in isert_connect_request to routines") Link: https://lore.kernel.org/r/20210322161325.7491-1-lyl2019@mail.ustc.edu.cn Signed-off-by: Lv Yunlong <lyl2019@mail.ustc.edu.cn> Acked-by: Sagi Grimberg <sagi@grimberg.me> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Reviewed-by: Max Gurtovoy <mgurtovoy@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-03-26RDMA: Support more than 255 rdma portsMark Bloch
Current code uses many different types when dealing with a port of a RDMA device: u8, unsigned int and u32. Switch to u32 to clean up the logic. This allows us to make (at least) the core view consistent and use the same type. Unfortunately not all places can be converted. Many uverbs functions expect port to be u8 so keep those places in order not to break UAPIs. HW/Spec defined values must also not be changed. With the switch to u32 we now can support devices with more than 255 ports. U32_MAX is reserved to make control logic a bit easier to deal with. As a device with U32_MAX ports probably isn't going to happen any time soon this seems like a non issue. When a device with more than 255 ports is created uverbs will report the RDMA device as having 255 ports as this is the max currently supported. The verbs interface is not changed yet because the IBTA spec limits the port size in too many places to be u8 and all applications that relies in verbs won't be able to cope with this change. At this stage, we are extending the interfaces that are using vendor channel solely Once the limitation is lifted mlx5 in switchdev mode will be able to have thousands of SFs created by the device. As the only instance of an RDMA device that reports more than 255 ports will be a representor device and it exposes itself as a RAW Ethernet only device CM/MAD/IPoIB and other ULPs aren't effected by this change and their sysfs/interfaces that are exposes to userspace can remain unchanged. While here cleanup some alignment issues and remove unneeded sanity checks (mainly in rdmavt), Link: https://lore.kernel.org/r/20210301070420.439400-1-leon@kernel.org Signed-off-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-03-11RDMA/rtrs-clt: Use rdma_event_msg in logJack Wang
It's easier to understand a string instead of enum. Link: https://lore.kernel.org/r/20210222141551.54345-2-jinpu.wang@cloud.ionos.com Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-03-11RDMA/rtrs: Use new shared CQ mechanismJack Wang
Have the driver use shared CQs which provids a ~10%-20% improvement during test. Instead of opening a CQ for each QP per connection, a CQ for each QP will be provided by the RDMA core driver that will be shared between the QPs on that core reducing interrupt overhead. Link: https://lore.kernel.org/r/20210222141551.54345-1-jinpu.wang@cloud.ionos.com Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-03-04scsi: target: core: Add gfp_t arg to target_cmd_init_cdb()Mike Christie
tcm_loop could be used like a normal block device, so we can't use GFP_KERNEL and should use GFP_NOIO. This adds a gfp_t arg to target_cmd_init_cdb() and converts the users. For every driver but loop GFP_KERNEL is kept. This will also be useful in subsequent patches where loop needs to do target_submit_prep() from interrupt context to get a ref to the se_device, and so it will need to use GFP_ATOMIC. Link: https://lore.kernel.org/r/20210227170006.5077-16-michael.christie@oracle.com Tested-by: Laurence Oberman <loberman@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Mike Christie <michael.christie@oracle.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2021-03-04scsi: target: srpt: Convert to new submission APIMike Christie
target_submit_cmd_map_sgls() is being removed, so convert srpt to the new submission API. srpt uses target_stop_session() to sync session shutdown with LIO core, so we use target_init_cmd()/target_submit_prep()/target_submit(), because target_init_cmd() will detect the target_stop_session() call and return an error. Link: https://lore.kernel.org/r/20210227170006.5077-6-michael.christie@oracle.com Cc: Bart Van Assche <bvanassche@acm.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Mike Christie <michael.christie@oracle.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2021-02-16RDMA/rtrs-srv: Do not pass a valid pointer to PTR_ERR()Jack Wang
smatch gives the warning: drivers/infiniband/ulp/rtrs/rtrs-srv.c:1805 rtrs_rdma_connect() warn: passing zero to 'PTR_ERR' Which is trying to say smatch has shown that srv is not an error pointer and thus cannot be passed to PTR_ERR. The solution is to move the list_add() down after full initilization of rtrs_srv. To avoid holding the srv_mutex too long, only hold it during the list operation as suggested by Leon. Fixes: 03e9b33a0fd6 ("RDMA/rtrs: Only allow addition of path to an already established session") Link: https://lore.kernel.org/r/20210216143807.65923-1-jinpu.wang@cloud.ionos.com Reported-by: kernel test robot <lkp@intel.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-02-16RDMA/srp: Fix support for unpopulated and unbalanced NUMA nodesNicolas Morey-Chaisemartin
The current code computes a number of channels per SRP target and spreads them equally across all online NUMA nodes. Each channel is then assigned a CPU within this node. In the case of unbalanced, or even unpopulated nodes, some channels do not get a CPU associated and thus do not get connected. This causes the SRP connection to fail. This patch solves the issue by rewriting channel computation and allocation: - Drop channel to node/CPU association as it had no real effect on locality but added unnecessary complexity. - Tweak the number of channels allocated to reduce CPU contention when possible: - Up to one channel per CPU (instead of up to 4 by node) - At least 4 channels per node, unless ch_count module parameter is used. Link: https://lore.kernel.org/r/9cb4d9d3-30ad-2276-7eff-e85f7ddfb411@suse.com Signed-off-by: Nicolas Morey-Chaisemartin <nmoreychaisemartin@suse.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-02-16RDMA/rtrs-srv-sysfs: fix missing put_deviceGioh Kim
put_device() decreases the ref-count and then the device will be cleaned-up, while at is also add missing put_device in rtrs_srv_create_once_sysfs_root_folders This patch solves a kmemleak error as below: unreferenced object 0xffff88809a7a0710 (size 8): comm "kworker/4:1H", pid 113, jiffies 4295833049 (age 6212.380s) hex dump (first 8 bytes): 62 6c 61 00 6b 6b 6b a5 bla.kkk. backtrace: [<0000000054413611>] kstrdup+0x2e/0x60 [<0000000078e3120a>] kobject_set_name_vargs+0x2f/0xb0 [<00000000f1a17a6b>] dev_set_name+0xab/0xe0 [<00000000d5502e32>] rtrs_srv_create_sess_files+0x2fb/0x314 [rtrs_server] [<00000000ed11a1ef>] rtrs_srv_info_req_done+0x631/0x800 [rtrs_server] [<000000008fc5aa8f>] __ib_process_cq+0x94/0x100 [ib_core] [<00000000a9599cb4>] ib_cq_poll_work+0x32/0xc0 [ib_core] [<00000000cfc376be>] process_one_work+0x4bc/0x980 [<0000000016e5c96a>] worker_thread+0x78/0x5c0 [<00000000c20b8be0>] kthread+0x191/0x1e0 [<000000006c9c0003>] ret_from_fork+0x3a/0x50 Fixes: baa5b28b7a47 ("RDMA/rtrs-srv: Replace device_register with device_initialize and device_add") Link: https://lore.kernel.org/r/20210212134525.103456-5-jinpu.wang@cloud.ionos.com Signed-off-by: Gioh Kim <gi-oh.kim@cloud.ionos.com> Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-02-16RDMA/rtrs-srv: fix memory leak by missing kobject freeGioh Kim
kmemleak reported an error as below: unreferenced object 0xffff8880674b7640 (size 64): comm "kworker/4:1H", pid 113, jiffies 4296403507 (age 507.840s) hex dump (first 32 bytes): 69 70 3a 31 39 32 2e 31 36 38 2e 31 32 32 2e 31 ip:192.168.122.1 31 30 40 69 70 3a 31 39 32 2e 31 36 38 2e 31 32 10@ip:192.168.12 backtrace: [<0000000054413611>] kstrdup+0x2e/0x60 [<0000000078e3120a>] kobject_set_name_vargs+0x2f/0xb0 [<00000000ca2be3ee>] kobject_init_and_add+0xb0/0x120 [<0000000062ba5e78>] rtrs_srv_create_sess_files+0x14c/0x314 [rtrs_server] [<00000000b45b7217>] rtrs_srv_info_req_done+0x5b1/0x800 [rtrs_server] [<000000008fc5aa8f>] __ib_process_cq+0x94/0x100 [ib_core] [<00000000a9599cb4>] ib_cq_poll_work+0x32/0xc0 [ib_core] [<00000000cfc376be>] process_one_work+0x4bc/0x980 [<0000000016e5c96a>] worker_thread+0x78/0x5c0 [<00000000c20b8be0>] kthread+0x191/0x1e0 [<000000006c9c0003>] ret_from_fork+0x3a/0x50 It is caused by the not-freed kobject of rtrs_srv_sess. The kobject embedded in rtrs_srv_sess has ref-counter 2 after calling process_info_req(). Therefore it must call kobject_put twice. Currently it calls kobject_put only once at rtrs_srv_destroy_sess_files because kobject_del removes the state_in_sysfs flag and then kobject_put in free_sess() is not called. This patch moves kobject_del() into free_sess() so that the kobject of rtrs_srv_sess can be freed. And also this patch adds the missing call of sysfs_remove_group() to clean-up the sysfs directory. Fixes: 9cb837480424 ("RDMA/rtrs: server: main functionality") Link: https://lore.kernel.org/r/20210212134525.103456-4-jinpu.wang@cloud.ionos.com Signed-off-by: Gioh Kim <gi-oh.kim@cloud.ionos.com> Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-02-16RDMA/rtrs: Only allow addition of path to an already established sessionMd Haris Iqbal
While adding a path from the client side to an already established session, it was possible to provide the destination IP to a different server. This is dangerous. This commit adds an extra member to the rtrs_msg_conn_req structure, named first_conn; which is supposed to notify if the connection request is the first for that session or not. On the server side, if a session does not exist but the first_conn received inside the rtrs_msg_conn_req structure is 1, the connection request is failed. This signifies that the connection request is for an already existing session, and since the server did not find one, it is an wrong connection request. Fixes: 6a98d71daea1 ("RDMA/rtrs: client: main functionality") Fixes: 9cb837480424 ("RDMA/rtrs: server: main functionality") Link: https://lore.kernel.org/r/20210212134525.103456-3-jinpu.wang@cloud.ionos.com Signed-off-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com> Reviewed-by: Lutz Pogrell <lutz.pogrell@cloud.ionos.com> Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-02-16RDMA/rtrs-srv: Fix stack-out-of-boundsJack Wang
BUG: KASAN: stack-out-of-bounds in _mlx4_ib_post_send+0x1bd2/0x2770 [mlx4_ib] Read of size 4 at addr ffff8880d5a7f980 by task kworker/0:1H/565 CPU: 0 PID: 565 Comm: kworker/0:1H Tainted: G O 5.4.84-storage #5.4.84-1+feature+linux+5.4.y+dbg+20201216.1319+b6b887b~deb10 Hardware name: Supermicro H8QG6/H8QG6, BIOS 3.00 09/04/2012 Workqueue: ib-comp-wq ib_cq_poll_work [ib_core] Call Trace: dump_stack+0x96/0xe0 print_address_description.constprop.4+0x1f/0x300 ? irq_work_claim+0x2e/0x50 __kasan_report.cold.8+0x78/0x92 ? _mlx4_ib_post_send+0x1bd2/0x2770 [mlx4_ib] kasan_report+0x10/0x20 _mlx4_ib_post_send+0x1bd2/0x2770 [mlx4_ib] ? check_chain_key+0x1d7/0x2e0 ? _mlx4_ib_post_recv+0x630/0x630 [mlx4_ib] ? lockdep_hardirqs_on+0x1a8/0x290 ? stack_depot_save+0x218/0x56e ? do_profile_hits.isra.6.cold.13+0x1d/0x1d ? check_chain_key+0x1d7/0x2e0 ? save_stack+0x4d/0x80 ? save_stack+0x19/0x80 ? __kasan_slab_free+0x125/0x170 ? kfree+0xe7/0x3b0 rdma_write_sg+0x5b0/0x950 [rtrs_server] The problem is when we send imm_wr, the type should be ib_rdma_wr, so hw driver like mlx4 can do rdma_wr(wr), so fix it by use the ib_rdma_wr as type for imm_wr. Fixes: 9cb837480424 ("RDMA/rtrs: server: main functionality") Link: https://lore.kernel.org/r/20210212134525.103456-2-jinpu.wang@cloud.ionos.com Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Reviewed-by: Gioh Kim <gi-oh.kim@cloud.ionos.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-02-16RDMA/ipoib: Remove racy Subnet Manager sendonly join checksChristoph Lameter
When a system receives a REREG event from the SM, then the SM information in the kernel is marked as invalid and a request is sent to the SM to update the information. The SM information is invalid in that time period. However, receiving a REREG also occurs simultaneously in user space applications that are now trying to rejoin the multicast groups. Some of those may be sendonly multicast groups which are then failing. If the SM information is invalid then ib_sa_sendonly_fullmem_support() returns false. That is wrong because it just means that we do not know yet if the potentially new SM supports sendonly joins. Sendonly join was introduced in 2015 and all the Subnet managers have supported it ever since. So there is no point in checking if a subnet manager supports it. Should an old opensm get a request for a sendonly join then the request will fail. The code that is removed here accomodated that situation and fell back to a full join. Falling back to a full join is problematic in itself. The reason to use the sendonly join was to reduce the traffic on the Infiniband fabric otherwise one could have just stayed with the regular join. So this patch may cause users of very old opensms to discover that lots of traffic needlessly crosses their IB fabrics. Link: https://lore.kernel.org/r/alpine.DEB.2.22.394.2101281845160.13303@www.lameter.com Signed-off-by: Christoph Lameter <cl@linux.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-19IB/iser: Simplify prot_caps settingMax Gurtovoy
Reduce the number of instructions made for setting protection caps. No need to do bitwise OR with 0 since we can zero the return value in the beginning of the function. Link: https://lore.kernel.org/r/20210111145754.56727-5-mgurtovoy@nvidia.com Reviewed-by: Israel Rukshin <israelr@nvidia.com> Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-19IB/iser: Enforce iser_max_sectors to be greater than 0Max Gurtovoy
A value of 0 will casue the driver to fail establishing a valid connection to remote target. The following can be seen in the log in this case: iser: iser_connect: connecting to: 1.1.1.88:3260 iser: iser_cma_handler: address resolved (0): status 0 conn 00000000090aa4de id 00000000167d3b5a iser: iser_cma_handler: route resolved (2): status 0 conn 00000000090aa4de id 00000000167d3b5a iser: iscsi_iser_ep_poll: iser conn 00000000090aa4de rc = 0 iser: iser_create_ib_conn_res: setting conn 00000000090aa4de cma_id 00000000167d3b5a qp 00000000efa80660 max_send_wr 4619 iser_cma_handler: established (9): status 0 conn 00000000090aa4de id 00000000167d3b5a iser: iser_connected_handler: remote qpn:1c7 my qpn:1c6 iser: iser_connected_handler: conn 00000000090aa4de: negotiated remote invalidation iser: iscsi_iser_ep_poll: iser conn 00000000090aa4de rc = 1 scsi host10: iSCSI Initiator over iSER mlx5_core 0000:07:00.0: mlx5_cmd_check:769:(pid 616473): CREATE_MKEY(0x200) op_mod(0x0) failed, status bad parameter(0x3), syndrome (0x3bf6f) iser: iser_create_fastreg_desc: Failed to allocate ib_fast_reg_mr err=-22 iser: iser_alloc_rx_descriptors: failed allocating rx descriptors / data buffers iser: iscsi_iser_ep_disconnect: ep 00000000d2040785 iser conn 00000000090aa4de iser: iser_conn_terminate: iser_conn 00000000090aa4de state 3 iser: iser_free_ib_conn_res: freeing conn 00000000090aa4de cma_id 00000000167d3b5a qp 00000000efa80660 iser: iser_device_try_release: device 00000000dc871b1b refcount 0 Link: https://lore.kernel.org/r/20210111145754.56727-4-mgurtovoy@nvidia.com Reviewed-by: Israel Rukshin <israelr@nvidia.com> Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-19IB/iser: Protect iscsi_max_lun module param using callbackMax Gurtovoy
Remove the check from the module_init function. Link: https://lore.kernel.org/r/20210111145754.56727-3-mgurtovoy@nvidia.com Reviewed-by: Israel Rukshin <israelr@nvidia.com> Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-19IB/iser: Remove unneeded semicolonsMax Gurtovoy
No need to add semicolon after closing bracket. Link: https://lore.kernel.org/r/20210111145754.56727-2-mgurtovoy@nvidia.com Reviewed-by: Israel Rukshin <israelr@nvidia.com> Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-18IB/isert: Simplify signature cap checkMax Gurtovoy
Use if/else clause instead of "condition ? val1 : val2" to make the code cleaner and simpler. Link: https://lore.kernel.org/r/20210110111903.486681-3-mgurtovoy@nvidia.com Reviewed-by: Israel Rukshin <israelr@nvidia.com> Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com> Acked-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-18IB/isert: Remove unneeded semicolonMax Gurtovoy
No need to add semicolon after closing bracket. Link: https://lore.kernel.org/r/20210110111903.486681-2-mgurtovoy@nvidia.com Reviewed-by: Israel Rukshin <israelr@nvidia.com> Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-18IB/isert: Remove unneeded new linesMax Gurtovoy
The Linux convention is to have only 1 new line between functions. Link: https://lore.kernel.org/r/20210110111903.486681-1-mgurtovoy@nvidia.com Reviewed-by: Israel Rukshin <israelr@nvidia.com> Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-15RDMA/rtrs: Fix KASAN: stack-out-of-bounds bugJack Wang
When KASAN is enabled, we notice warning below: [ 483.436975] ================================================================== [ 483.437234] BUG: KASAN: stack-out-of-bounds in _mlx5_ib_post_send+0x188a/0x2560 [mlx5_ib] [ 483.437430] Read of size 4 at addr ffff88a195fd7d30 by task kworker/1:3/6954 [ 483.437731] CPU: 1 PID: 6954 Comm: kworker/1:3 Kdump: loaded Tainted: G O 5.4.82-pserver #5.4.82-1+feature+linux+5.4.y+dbg+20201210.1532+987e7a6~deb10 [ 483.437976] Hardware name: Supermicro Super Server/X11DDW-L, BIOS 3.3 02/21/2020 [ 483.438168] Workqueue: rtrs_server_wq hb_work [rtrs_core] [ 483.438323] Call Trace: [ 483.438486] dump_stack+0x96/0xe0 [ 483.438646] ? _mlx5_ib_post_send+0x188a/0x2560 [mlx5_ib] [ 483.438802] print_address_description.constprop.6+0x1b/0x220 [ 483.438966] ? _mlx5_ib_post_send+0x188a/0x2560 [mlx5_ib] [ 483.439133] ? _mlx5_ib_post_send+0x188a/0x2560 [mlx5_ib] [ 483.439285] __kasan_report.cold.9+0x1a/0x32 [ 483.439444] ? _mlx5_ib_post_send+0x188a/0x2560 [mlx5_ib] [ 483.439597] kasan_report+0x10/0x20 [ 483.439752] _mlx5_ib_post_send+0x188a/0x2560 [mlx5_ib] [ 483.439910] ? update_sd_lb_stats+0xfb1/0xfc0 [ 483.440073] ? set_reg_wr+0x520/0x520 [mlx5_ib] [ 483.440222] ? update_group_capacity+0x340/0x340 [ 483.440377] ? find_busiest_group+0x314/0x870 [ 483.440526] ? update_sd_lb_stats+0xfc0/0xfc0 [ 483.440683] ? __bitmap_and+0x6f/0x100 [ 483.440832] ? __lock_acquire+0xa2/0x2150 [ 483.440979] ? __lock_acquire+0xa2/0x2150 [ 483.441128] ? __lock_acquire+0xa2/0x2150 [ 483.441279] ? debug_lockdep_rcu_enabled+0x23/0x60 [ 483.441430] ? lock_downgrade+0x390/0x390 [ 483.441582] ? __lock_acquire+0xa2/0x2150 [ 483.441729] ? __lock_acquire+0xa2/0x2150 [ 483.441876] ? newidle_balance+0x425/0x8f0 [ 483.442024] ? __lock_acquire+0xa2/0x2150 [ 483.442172] ? debug_lockdep_rcu_enabled+0x23/0x60 [ 483.442330] hb_work+0x15d/0x1d0 [rtrs_core] [ 483.442479] ? schedule_hb+0x50/0x50 [rtrs_core] [ 483.442627] ? lock_downgrade+0x390/0x390 [ 483.442781] ? process_one_work+0x40d/0xa50 [ 483.442931] process_one_work+0x4ee/0xa50 [ 483.443082] ? pwq_dec_nr_in_flight+0x110/0x110 [ 483.443231] ? do_raw_spin_lock+0x119/0x1d0 [ 483.443383] worker_thread+0x65/0x5c0 [ 483.443532] ? process_one_work+0xa50/0xa50 [ 483.451839] kthread+0x1e2/0x200 [ 483.451983] ? kthread_create_on_node+0xc0/0xc0 [ 483.452139] ret_from_fork+0x3a/0x50 The problem is we use wrong type when send wr, hw driver expect the type of IB_WR_RDMA_WRITE_WITH_IMM wr should be ib_rdma_wr, and doing container_of to access member. The fix is simple use ib_rdma_wr instread of ib_send_wr. Fixes: c0894b3ea69d ("RDMA/rtrs: core: lib functions shared between client and server modules") Link: https://lore.kernel.org/r/20201217141915.56989-20-jinpu.wang@cloud.ionos.com Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Reviewed-by: Gioh Kim <gi-oh.kim@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-15RDMA/rtrs-srv: Init wr_cnt as 1Jack Wang
Fix up wr_avail accounting. if wr_cnt is 0, then we do SIGNAL for first wr, in completion we add queue_depth back, which is not right in the sense of tracking for available wr. So fix it by init wr_cnt to 1. Fixes: 9cb837480424 ("RDMA/rtrs: server: main functionality") Link: https://lore.kernel.org/r/20201217141915.56989-19-jinpu.wang@cloud.ionos.com Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Gioh Kim <gi-oh.kim@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-15RDMA/rtrs-srv: Do not signal REG_MRJack Wang
We do not need to wait for REG_MR completion, so remove the SIGNAL flag. Fixes: 9cb837480424 ("RDMA/rtrs: server: main functionality") Link: https://lore.kernel.org/r/20201217141915.56989-18-jinpu.wang@cloud.ionos.com Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Gioh Kim <gi-oh.kim@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-15RDMA/rtrs-clt: Use bitmask to check sess->flagsJack Wang
We may want to add new flags, so it's better to use bitmask to check flags. Fixes: 6a98d71daea1 ("RDMA/rtrs: client: main functionality") Link: https://lore.kernel.org/r/20201217141915.56989-17-jinpu.wang@cloud.ionos.com Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Gioh Kim <gi-oh.kim@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-15RDMA/rtrs: Do not signal for heatbeatJack Wang
For HB, there is no need to generate signal for completion. Also remove a comment accordingly. Fixes: c0894b3ea69d ("RDMA/rtrs: core: lib functions shared between client and server modules") Link: https://lore.kernel.org/r/20201217141915.56989-16-jinpu.wang@cloud.ionos.com Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Reported-by: Gioh Kim <gi-oh.kim@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-15RDMA/rtrs-clt: Refactor the failure cases in alloc_cltGuoqing Jiang
Make all failure cases go to the common path to avoid duplicate code. And some issued existed before. 1. clt need to be freed to avoid memory leak. 2. return ERR_PTR(-ENOMEM) if kobject_create_and_add fails, because rtrs_clt_open checks the return value of by call "IS_ERR(clt)". Fixes: 6a98d71daea1 ("RDMA/rtrs: client: main functionality") Link: https://lore.kernel.org/r/20201217141915.56989-15-jinpu.wang@cloud.ionos.com Signed-off-by: Guoqing Jiang <guoqing.jiang@cloud.ionos.com> Reviewed-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com> Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-15RDMA/rtrs-srv: Fix missing wr_cqeJack Wang
We had a few places wr_cqe is not set, which could lead to NULL pointer deref or GPF in error case. Fixes: 9cb837480424 ("RDMA/rtrs: server: main functionality") Link: https://lore.kernel.org/r/20201217141915.56989-14-jinpu.wang@cloud.ionos.com Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Reviewed-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com> Signed-off-by: Guoqing Jiang <guoqing.jiang@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-15RDMA/rtrs-clt: Rename __rtrs_clt_change_state to rtrs_clt_change_stateGuoqing Jiang
Let's rename it to rtrs_clt_change_state since the previous one is killed. Also update the comment to make it more clear. Link: https://lore.kernel.org/r/20201217141915.56989-13-jinpu.wang@cloud.ionos.com Signed-off-by: Guoqing Jiang <guoqing.jiang@cloud.ionos.com> Reviewed-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com> Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-15RDMA/rtrs-clt: Kill rtrs_clt_change_stateGuoqing Jiang
It is just a wrapper of rtrs_clt_change_state_get_old, and we can reuse rtrs_clt_change_state_get_old with add the checking of 'old_state' is valid or not. Link: https://lore.kernel.org/r/20201217141915.56989-12-jinpu.wang@cloud.ionos.com Signed-off-by: Guoqing Jiang <guoqing.jiang@cloud.ionos.com> Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-15RDMA/rtrs-clt: Remove unnecessary 'goto out'Guoqing Jiang
This is not needed since the label is just after the place. Link: https://lore.kernel.org/r/20201217141915.56989-11-jinpu.wang@cloud.ionos.com Signed-off-by: Guoqing Jiang <guoqing.jiang@cloud.ionos.com> Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-15RDMA/rtrs-clt: Kill wait_for_inflight_permitsGuoqing Jiang
Let's wait the inflight permits before free it. Link: https://lore.kernel.org/r/20201217141915.56989-10-jinpu.wang@cloud.ionos.com Signed-off-by: Guoqing Jiang <guoqing.jiang@cloud.ionos.com> Reviewed-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com> Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-15RDMA/rtrs-clt: Consolidate rtrs_clt_destroy_sysfs_root_{folder,files}Guoqing Jiang
Since the two functions are called together, let's consolidate them in a new function rtrs_clt_destroy_sysfs_root. Link: https://lore.kernel.org/r/20201217141915.56989-9-jinpu.wang@cloud.ionos.com Signed-off-by: Guoqing Jiang <guoqing.jiang@cloud.ionos.com> Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
generated by cgit (git 2.34.1) at 2025-07-18 02:43:20 +0000