aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/infiniband/sw/rxe/rxe_qp.c (follow)
AgeCommit message (Collapse)AuthorFilesLines
2022-09-08RDMA/rxe: convert pr_warn to pr_debugLi Zhijian1-23/+22
They could be triggered by user APIs with invalid parameters. Signed-off-by: Li Zhijian <lizhijian@fujitsu.com> Link: https://lore.kernel.org/r/1662518901-2-2-git-send-email-lizhijian@fujitsu.com Signed-off-by: Leon Romanovsky <leon@kernel.org>
2022-09-08RDMA/rxe: use %u to print u32 variablesLi Zhijian1-5/+5
struct ib_qp_cap { u32 max_send_wr; u32 max_recv_wr; u32 max_send_sge; u32 max_recv_sge; u32 max_inline_data; ... To avoid getting a negative value from dmesg: [410580.579965] rdma_rxe: invalid send sge = 65535 > 32 [410580.583818] rdma_rxe: invalid send wr = -1 > 1048576 [410582.771323] rdma_rxe: invalid recv sge = 65535 > 32 [410582.775310] rdma_rxe: invalid recv wr = -1 > 1048576 Signed-off-by: Li Zhijian <lizhijian@fujitsu.com> Link: https://lore.kernel.org/r/1662518901-2-1-git-send-email-lizhijian@fujitsu.com Signed-off-by: Leon Romanovsky <leon@kernel.org>
2022-08-31RDMA/rxe: Remove the unused variable objZhu Yanjun1-3/+3
The member variable obj in struct rxe_task is not needed. So remove it to save memory. Link: https://lore.kernel.org/r/20220822011615.805603-4-yanjun.zhu@linux.dev Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev> Reviewed-by: Li Zhijian <lizhijian@fujitsu.com> Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Leon Romanovsky <leon@kernel.org>
2022-08-31RDMA/rxe: Fix the error caused by qp->skZhu Yanjun1-2/+4
When sock_create_kern in the function rxe_qp_init_req fails, qp->sk is set to NULL. Then the function rxe_create_qp will call rxe_qp_do_cleanup to handle allocated resource. Before handling qp->sk, this variable should be checked. Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://lore.kernel.org/r/20220822011615.805603-3-yanjun.zhu@linux.dev Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev> Reviewed-by: Li Zhijian <lizhijian@fujitsu.com> Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Leon Romanovsky <leon@kernel.org>
2022-08-31RDMA/rxe: Fix "kernel NULL pointer dereference" errorZhu Yanjun1-1/+3
When rxe_queue_init in the function rxe_qp_init_req fails, both qp->req.task.func and qp->req.task.arg are not initialized. Because of creation of qp fails, the function rxe_create_qp will call rxe_qp_do_cleanup to handle allocated resource. Before calling __rxe_do_task, both qp->req.task.func and qp->req.task.arg should be checked. Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://lore.kernel.org/r/20220822011615.805603-2-yanjun.zhu@linux.dev Reported-by: syzbot+ab99dc4c6e961eed8b8e@syzkaller.appspotmail.com Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev> Reviewed-by: Li Zhijian <lizhijian@fujitsu.com> Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Leon Romanovsky <leon@kernel.org>
2022-08-02RDMA/rxe: Fix error unwind in rxe_create_qp()Zhu Yanjun1-4/+8
In the function rxe_create_qp(), rxe_qp_from_init() is called to initialize qp, internally things like the spin locks are not setup until rxe_qp_init_req(). If an error occures before this point then the unwind will call rxe_cleanup() and eventually to rxe_qp_do_cleanup()/rxe_cleanup_task() which will oops when trying to access the uninitialized spinlock. Move the spinlock initializations earlier before any failures. Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://lore.kernel.org/r/20220731063621.298405-1-yanjun.zhu@linux.dev Reported-by: syzbot+833061116fa28df97f3b@syzkaller.appspotmail.com Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-08-02RDMA/rxe: Split qp state for requester and completerBob Pearson1-0/+5
Currently the requester can continue to process send wqes after an local qp operation error is detected because the setting of the qp state to the error state is deferred until later. This patch splits the qp state for the completer and requester into two separate states and sets qp->req.state = QP_STATE_ERROR as soon as the error is detected before another wqe can be executed. Link: https://lore.kernel.org/r/1658307368-1851-4-git-send-email-lizhijian@fujitsu.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-07-22RDMA/rxe: Fix rnr retry behaviorBob Pearson1-0/+1
Currently the completer tasklet when retransmit timer or the rnr timer fires the same flag (qp->req.need_retry) is set so that if either timer fires it will attempt to perform a retry flow on the send queue. This has the effect of responding to an RNR NAK at the first retransmit timer event which might not allow the requested rnr timeout. This patch adds a new flag (qp->req.wait_for_rnr_timer) which, if set, prevents a retry flow until the rnr nak timer fires. This patch fixes rnr retry errors which can be observed by running the pyverbs test_rdmacm_async_traffic_external_qp multiple times. With this patch applied they do not occur. Link: https://lore.kernel.org/linux-rdma/a8287823-1408-4273-bc22-99a0678db640@gmail.com/ Link: https://lore.kernel.org/linux-rdma/2bafda9e-2bb6-186d-12a1-179e8f6a2678@talpey.com/ Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://lore.kernel.org/r/20220630190425.2251-6-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-07-19RDMA/rxe: Remove unused qp parameterXiao Yang1-3/+3
The qp parameter in free_rd_atomic_resource() has become unused so remove it directly. Fixes: 15ae1375ea91 ("RDMA/rxe: Fix qp reference counting for atomic ops") Link: https://lore.kernel.org/all/20220708035547.6592-1-yangx.jy@fujitsu.com/ Signed-off-by: Xiao Yang <yangx.jy@fujitsu.com> Signed-off-by: Leon Romanovsky <leon@kernel.org>
2022-07-18RDMA/rxe: Fix BUG: KASAN: null-ptr-deref in rxe_qp_do_cleanupZhu Yanjun1-4/+6
The function rxe_create_qp calls rxe_qp_from_init. If some error occurs, the error handler of function rxe_qp_from_init will set both scq and rcq to NULL. Then rxe_create_qp calls rxe_put to handle qp. In the end, rxe_qp_do_cleanup is called by rxe_put. rxe_qp_do_cleanup directly accesses scq and rcq before checking them. This will cause null-ptr-deref error. The call graph is as below: rxe_create_qp { ... rxe_qp_from_init { ... err1: ... qp->rcq = NULL; <---rcq is set to NULL qp->scq = NULL; <---scq is set to NULL ... } qp_init: rxe_put{ ... rxe_qp_do_cleanup { ... atomic_dec(&qp->scq->num_wq); <--- scq is accessed ... atomic_dec(&qp->rcq->num_wq); <--- rcq is accessed } } Fixes: 4703b4f0d94a ("RDMA/rxe: Enforce IBA C11-17") Link: https://lore.kernel.org/r/20220705225414.315478-1-yanjun.zhu@linux.dev Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev> Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com> Reviewed-by: Md Haris Iqbal <haris.iqbal@ionos.com> Signed-off-by: Leon Romanovsky <leon@kernel.org>
2022-06-30RDMA/rxe: Merge normal and retry atomic flowsBob Pearson1-2/+0
Make the execution of the atomic operation in rxe_atomic_reply() conditional on res->replay and make duplicate_request() call into rxe_atomic_reply() to merge the two flows. This is modeled on the behavior of read reply. Delete the skb from the atomic responder resource since it is no longer used. Adjust the reference counting of the qp in send_atomic_ack() for this flow. Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://lore.kernel.org/r/20220606143836.3323-6-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-05-09RDMA/rxe: Enforce IBA C11-17Bob Pearson1-0/+10
Add a counter to keep track of the number of WQs connected to a CQ and return an error if destroy_cq() is called while the counter is non zero. Link: https://lore.kernel.org/r/20220421014042.26985-8-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-05-09RDMA/rxe: Move qp cleanup code to rxe_qp_do_cleanup()Bob Pearson1-8/+4
Move the code from rxe_qp_destroy() to rxe_qp_do_cleanup(). This allows flows holding references to qp to complete before the qp object is torn down. Link: https://lore.kernel.org/r/20220421014042.26985-5-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-04-08RDMA/rxe: Remove support for SMI QPs from rdma_rxeBob Pearson1-13/+1
Currently the rdma_rxe driver supports SMI type QPs in a few places which is incorrect. RoCE devices never should support SMI QPs. This commit removes SMI QP support from the driver. Link: https://lore.kernel.org/r/20220407185416.16372-1-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-03-16RDMA/rxe: Use standard names for ref countingBob Pearson1-14/+14
Rename rxe_add_ref() to rxe_get() and rxe_drop_ref() to rxe_put(). Significantly improves readability for new readers. Link: https://lore.kernel.org/r/20220304000808.225811-10-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-03-15RDMA/rxe: Replace mr by rkey in responder resourcesBob Pearson1-8/+2
Currently rxe saves a copy of MR in responder resources for RDMA reads. Since the responder resources are never freed just over written if more are needed this MR may not have a reference freed until the QP is destroyed. This patch uses the rkey instead of the MR and on subsequent packets of a multipacket read reply message it looks up the MR from the rkey for each packet. This makes it possible for a user to deregister an MR or unbind a MW on the fly and get correct behaviour. Link: https://lore.kernel.org/r/20220304000808.225811-3-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-01-28RDMA/rxe: Remove qp->grp_lock and qp->grp_listBob Pearson1-3/+0
Since it is no longer required to cleanup attachments to multicast groups when a QP is destroyed qp->grp_lock and qp->grp_list are no longer needed and are removed. Link: https://lore.kernel.org/r/20220127213755.31697-7-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-01-28RDMA/rxe: Remove rxe_drop_all_macst_groupsBob Pearson1-2/+0
With o10-2.2.3 enforced rxe_drop_all_mcast_groups is completely unnecessary. Remove it and references to it. Link: https://lore.kernel.org/r/20220127213755.31697-6-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-01-28RDMA/rxe: Enforce IBA o10-2.2.3Bob Pearson1-0/+14
Add code to check if a QP is attached to one or more multicast groups when destroy_qp is called and return an error if so. Link: https://lore.kernel.org/r/20220127213755.31697-5-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2022-01-20Merge branch 'akpm' (patches from Andrew)Linus Torvalds1-2/+1
Merge more updates from Andrew Morton: "55 patches. Subsystems affected by this patch series: percpu, procfs, sysctl, misc, core-kernel, get_maintainer, lib, checkpatch, binfmt, nilfs2, hfs, fat, adfs, panic, delayacct, kconfig, kcov, and ubsan" * emailed patches from Andrew Morton <akpm@linux-foundation.org>: (55 commits) lib: remove redundant assignment to variable ret ubsan: remove CONFIG_UBSAN_OBJECT_SIZE kcov: fix generic Kconfig dependencies if ARCH_WANTS_NO_INSTR lib/Kconfig.debug: make TEST_KMOD depend on PAGE_SIZE_LESS_THAN_256KB btrfs: use generic Kconfig option for 256kB page size limit arch/Kconfig: split PAGE_SIZE_LESS_THAN_256KB from PAGE_SIZE_LESS_THAN_64KB configs: introduce debug.config for CI-like setup delayacct: track delays from memory compact Documentation/accounting/delay-accounting.rst: add thrashing page cache and direct compact delayacct: cleanup flags in struct task_delay_info and functions use it delayacct: fix incomplete disable operation when switch enable to disable delayacct: support swapin delay accounting for swapping without blkio panic: remove oops_id panic: use error_report_end tracepoint on warnings fs/adfs: remove unneeded variable make code cleaner FAT: use io_schedule_timeout() instead of congestion_wait() hfsplus: use struct_group_attr() for memcpy() region nilfs2: remove redundant pointer sbufs fs/binfmt_elf: use PT_LOAD p_align values for static PIE const_structs.checkpatch: add frequently used ops structs ...
2022-01-20hash.h: remove unused define directiveIsabella Basso1-2/+1
Patch series "test_hash.c: refactor into KUnit", v3. We refactored the lib/test_hash.c file into KUnit as part of the student group LKCAMP [1] introductory hackathon for kernel development. This test was pointed to our group by Daniel Latypov [2], so its full conversion into a pure KUnit test was our goal in this patch series, but we ran into many problems relating to it not being split as unit tests, which complicated matters a bit, as the reasoning behind the original tests is quite cryptic for those unfamiliar with hash implementations. Some interesting developments we'd like to highlight are: - In patch 1/5 we noticed that there was an unused define directive that could be removed. - In patch 4/5 we noticed how stringhash and hash tests are all under the lib/test_hash.c file, which might cause some confusion, and we also broke those kernel config entries up. Overall KUnit developments have been made in the other patches in this series: In patches 2/5, 3/5 and 5/5 we refactored the lib/test_hash.c file so as to make it more compatible with the KUnit style, whilst preserving the original idea of the maintainer who designed it (i.e. George Spelvin), which might be undesirable for unit tests, but we assume it is enough for a first patch. This patch (of 5): Currently, there exist hash_32() and __hash_32() functions, which were introduced in a patch [1] targeting architecture specific optimizations. These functions can be overridden on a per-architecture basis to achieve such optimizations. They must set their corresponding define directive (HAVE_ARCH_HASH_32 and HAVE_ARCH__HASH_32, respectively) so that header files can deal with these overrides properly. As the supported 32-bit architectures that have their own hash function implementation (i.e. m68k, Microblaze, H8/300, pa-risc) have only been making use of the (more general) __hash_32() function (which only lacks a right shift operation when compared to the hash_32() function), remove the define directive corresponding to the arch-specific hash_32() implementation. [1] https://lore.kernel.org/lkml/20160525073311.5600.qmail@ns.sciencehorizons.net/ [akpm@linux-foundation.org: hash_32_generic() becomes hash_32()] Link: https://lkml.kernel.org/r/20211208183711.390454-1-isabbasso@riseup.net Link: https://lkml.kernel.org/r/20211208183711.390454-2-isabbasso@riseup.net Reviewed-by: David Gow <davidgow@google.com> Tested-by: David Gow <davidgow@google.com> Co-developed-by: Augusto Durães Camargo <augusto.duraes33@gmail.com> Signed-off-by: Augusto Durães Camargo <augusto.duraes33@gmail.com> Co-developed-by: Enzo Ferreira <ferreiraenzoa@gmail.com> Signed-off-by: Enzo Ferreira <ferreiraenzoa@gmail.com> Signed-off-by: Isabella Basso <isabbasso@riseup.net> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Brendan Higgins <brendanhiggins@google.com> Cc: Daniel Latypov <dlatypov@google.com> Cc: Shuah Khan <skhan@linuxfoundation.org> Cc: Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com> Cc: kernel test robot <lkp@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-12-14Merge tag 'v5.16-rc5' into rdma.git for-nextJason Gunthorpe1-0/+1
Required due to dependencies in following patches. Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-11-25RDMA: Fix use-after-free in rxe_queue_cleanupPavel Skripkin1-0/+1
On error handling path in rxe_qp_from_init() qp->sq.queue is freed and then rxe_create_qp() will drop last reference to this object. qp clean up function will try to free this queue one time and it causes UAF bug. Fix it by zeroing queue pointer after freeing queue in rxe_qp_from_init(). Fixes: 514aee660df4 ("RDMA: Globally allocate and release QP memory") Link: https://lore.kernel.org/r/20211121202239.3129-1-paskripkin@gmail.com Reported-by: syzbot+aab53008a5adf26abe91@syzkaller.appspotmail.com Signed-off-by: Pavel Skripkin <paskripkin@gmail.com> Reviewed-by: Zhu Yanjun <zyjzyj2000@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-11-19RDMA/rxe: Cleanup rxe_pool_entryBob Pearson1-3/+3
Currently three different names are used to describe rxe pool elements. They are referred to as entries, elems or pelems. This patch chooses one 'elem' and changes the other ones. Link: https://lore.kernel.org/r/20211103050241.61293-3-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-10-06RDMA/rxe: Remove the is_user members of struct rxe_sq/rxe_rq/rxe_srqXiao Yang1-2/+0
The is_user members of struct rxe_sq/rxe_rq/rxe_srq are unsed since commit ae6e843fe08d ("RDMA/rxe: Add memory barriers to kernel queues"). In this case, it is fine to remove them directly. Link: https://lore.kernel.org/r/20210930094813.226888-2-yangx.jy@fujitsu.com Signed-off-by: Xiao Yang <yangx.jy@fujitsu.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-24RDMA/rxe: Add memory barriers to kernel queuesBob Pearson1-8/+4
Earlier patches added memory barriers to protect user space to kernel space communications. The user space queues were previously shown to have occasional memory synchonization errors which were removed by adding smp_load_acquire, smp_store_release barriers. This patch extends that to the case where queues are used between kernel space threads. This patch also extends the queue types to include kernel ULP queues which access the other end of the queues in kernel verbs calls like poll_cq and post_send/recv. Link: https://lore.kernel.org/r/20210914164206.19768-2-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-14RDMA/rxe: remove the unnecessary variableZhu Yanjun1-2/+0
In the struct rxe_qp, the variable send_pkts is never used. So remove it. Link: https://lore.kernel.org/r/20210915075128.482919-1-yanjun.zhu@intel.com Signed-off-by: Zhu Yanjun <zyjzyj2000@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22Merge tag 'v5.13-rc7' into rdma.git for-nextJason Gunthorpe1-0/+7
Linux 5.13-rc7 Needed for dependencies in following patches. Merge conflict in rxe_cmop.c resolved by compining both patches. Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-16RDMA/rxe: Fix qp reference counting for atomic opsBob Pearson1-1/+0
Currently the rdma_rxe driver attempts to protect atomic responder resources by taking a reference to the qp which is only freed when the resource is recycled for a new read or atomic operation. This means that in normal circumstances there is almost always an extra qp reference once an atomic operation has been executed which prevents cleaning up the qp and associated pd and cqs when the qp is destroyed. This patch removes the call to rxe_add_ref() in send_atomic_ack() and the call to rxe_drop_ref() in free_rd_atomic_resource(). If the qp is destroyed while a peer is retrying an atomic op it will cause the operation to fail which is acceptable. Link: https://lore.kernel.org/r/20210604230558.4812-1-rpearsonhpe@gmail.com Reported-by: Zhu Yanjun <zyjzyj2000@gmail.com> Fixes: 86af61764151 ("IB/rxe: remove unnecessary skb_clone") Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-03RDMA/rxe: Protext kernel index from user spaceBob Pearson1-1/+9
In order to prevent user space from modifying the index that belongs to the kernel for shared queues let the kernel use a local copy of the index and copy any new values of that index to the shared rxe_queue_bus struct. This adds more switch statements which decreases the performance of the queue API. Move the type into the parameter list for these functions so that the compiler can optimize out the switch statements when the explicit type is known. Modify all the calls in the driver on performance paths to pass in the explicit queue type. Link: https://lore.kernel.org/r/20210527194748.662636-4-rpearsonhpe@gmail.com Link: https://lore.kernel.org/linux-rdma/20210526165239.GP1002214@@nvidia.com/ Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-03RDMA/rxe: Add a type flag to rxe_queue structsBob Pearson1-4/+8
To create optimal code only want to use smp_load_acquire() and smp_store_release() for user indices in rxe_queue APIs since kernel indices are protected by locks which also act as memory barriers. By adding a type to the queues we can determine which indices need to be protected. Link: https://lore.kernel.org/r/20210527194748.662636-2-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-05-11RDMA/rxe: Clear all QP fields if creation failedLeon Romanovsky1-0/+7
rxe_qp_do_cleanup() relies on valid pointer values in QP for the properly created ones, but in case rxe_qp_from_init() failed it was filled with garbage and caused tot the following error. refcount_t: underflow; use-after-free. WARNING: CPU: 1 PID: 12560 at lib/refcount.c:28 refcount_warn_saturate+0x1d1/0x1e0 lib/refcount.c:28 Modules linked in: CPU: 1 PID: 12560 Comm: syz-executor.4 Not tainted 5.12.0-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:refcount_warn_saturate+0x1d1/0x1e0 lib/refcount.c:28 Code: e9 db fe ff ff 48 89 df e8 2c c2 ea fd e9 8a fe ff ff e8 72 6a a7 fd 48 c7 c7 e0 b2 c1 89 c6 05 dc 3a e6 09 01 e8 ee 74 fb 04 <0f> 0b e9 af fe ff ff 0f 1f 84 00 00 00 00 00 41 56 41 55 41 54 55 RSP: 0018:ffffc900097ceba8 EFLAGS: 00010286 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 RDX: 0000000000040000 RSI: ffffffff815bb075 RDI: fffff520012f9d67 RBP: 0000000000000003 R08: 0000000000000000 R09: 0000000000000000 R10: ffffffff815b4eae R11: 0000000000000000 R12: ffff8880322a4800 R13: ffff8880322a4940 R14: ffff888033044e00 R15: 0000000000000000 FS: 00007f6eb2be3700(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007fdbe5d41000 CR3: 000000001d181000 CR4: 00000000001506e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: __refcount_sub_and_test include/linux/refcount.h:283 [inline] __refcount_dec_and_test include/linux/refcount.h:315 [inline] refcount_dec_and_test include/linux/refcount.h:333 [inline] kref_put include/linux/kref.h:64 [inline] rxe_qp_do_cleanup+0x96f/0xaf0 drivers/infiniband/sw/rxe/rxe_qp.c:805 execute_in_process_context+0x37/0x150 kernel/workqueue.c:3327 rxe_elem_release+0x9f/0x180 drivers/infiniband/sw/rxe/rxe_pool.c:391 kref_put include/linux/kref.h:65 [inline] rxe_create_qp+0x2cd/0x310 drivers/infiniband/sw/rxe/rxe_verbs.c:425 _ib_create_qp drivers/infiniband/core/core_priv.h:331 [inline] ib_create_named_qp+0x2ad/0x1370 drivers/infiniband/core/verbs.c:1231 ib_create_qp include/rdma/ib_verbs.h:3644 [inline] create_mad_qp+0x177/0x2d0 drivers/infiniband/core/mad.c:2920 ib_mad_port_open drivers/infiniband/core/mad.c:3001 [inline] ib_mad_init_device+0xd6f/0x1400 drivers/infiniband/core/mad.c:3092 add_client_context+0x405/0x5e0 drivers/infiniband/core/device.c:717 enable_device_and_get+0x1cd/0x3b0 drivers/infiniband/core/device.c:1331 ib_register_device drivers/infiniband/core/device.c:1413 [inline] ib_register_device+0x7c7/0xa50 drivers/infiniband/core/device.c:1365 rxe_register_device+0x3d5/0x4a0 drivers/infiniband/sw/rxe/rxe_verbs.c:1147 rxe_add+0x12fe/0x16d0 drivers/infiniband/sw/rxe/rxe.c:247 rxe_net_add+0x8c/0xe0 drivers/infiniband/sw/rxe/rxe_net.c:503 rxe_newlink drivers/infiniband/sw/rxe/rxe.c:269 [inline] rxe_newlink+0xb7/0xe0 drivers/infiniband/sw/rxe/rxe.c:250 nldev_newlink+0x30e/0x550 drivers/infiniband/core/nldev.c:1555 rdma_nl_rcv_msg+0x36d/0x690 drivers/infiniband/core/netlink.c:195 rdma_nl_rcv_skb drivers/infiniband/core/netlink.c:239 [inline] rdma_nl_rcv+0x2ee/0x430 drivers/infiniband/core/netlink.c:259 netlink_unicast_kernel net/netlink/af_netlink.c:1312 [inline] netlink_unicast+0x533/0x7d0 net/netlink/af_netlink.c:1338 netlink_sendmsg+0x856/0xd90 net/netlink/af_netlink.c:1927 sock_sendmsg_nosec net/socket.c:654 [inline] sock_sendmsg+0xcf/0x120 net/socket.c:674 ____sys_sendmsg+0x6e8/0x810 net/socket.c:2350 ___sys_sendmsg+0xf3/0x170 net/socket.c:2404 __sys_sendmsg+0xe5/0x1b0 net/socket.c:2433 do_syscall_64+0x3a/0xb0 arch/x86/entry/common.c:47 entry_SYSCALL_64_after_hwframe+0x44/0xae Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://lore.kernel.org/r/7bf8d548764d406dbbbaf4b574960ebfd5af8387.1620717918.git.leonro@nvidia.com Reported-by: syzbot+36a7f280de4e11c6f04e@syzkaller.appspotmail.com Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Reviewed-by: Zhu Yanjun <zyjzyj2000@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-02-08RDMA/rxe: Fix minor coding style issuesBob Pearson1-1/+1
checkpatch -f found 3 warnings in RDMA/rxe 1. a missing space following switch 2. return followed by else 3. use of strlcpy() instead of strscpy(). This patch fixes each of these. In ... } elseif (...) { ... return 0; } else ... The middle block can be safely moved since it is completely independent of the other code. Link: https://lore.kernel.org/r/20210205230525.49068-1-rpearson@hpe.com Signed-off-by: Bob Pearson <rpearson@hpe.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-01-12RDMA/rxe: Add check for supported QP typesXiao Yang1-0/+11
Current rdma_rxe only supports five QP types, attempting to create any others should return an error - the type check was missed. Link: https://lore.kernel.org/r/20201216071755.149449-2-yangx.jy@cn.fujitsu.com Signed-off-by: Xiao Yang <yangx.jy@cn.fujitsu.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2020-08-31RDMA/rxe: Add SPDX hdrs to rxe source filesBob Pearson1-28/+1
Add SPDX headers to all rxe .c and .h files. Link: https://lore.kernel.org/r/20200827145439.2273-1-rpearson@hpe.com Signed-off-by: Bob Pearson <rpearson@hpe.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2020-08-27RDMA/rxe: Fix style warningsBob Pearson1-2/+1
Fixed several minor checkpatch warnings in existing rxe source. Link: https://lore.kernel.org/r/20200820224638.3212-3-rpearson@hpe.com Signed-off-by: Bob Pearson <rpearson@hpe.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2020-02-19RDMA/rxe: Fix configuration of atomic queue pair attributesBart Van Assche1-3/+4
From the comment above the definition of the roundup_pow_of_two() macro: The result is undefined when n == 0. Hence only pass positive values to roundup_pow_of_two(). This patch fixes the following UBSAN complaint: UBSAN: Undefined behaviour in ./include/linux/log2.h:57:13 shift exponent 64 is too large for 64-bit type 'long unsigned int' Call Trace: dump_stack+0xa5/0xe6 ubsan_epilogue+0x9/0x26 __ubsan_handle_shift_out_of_bounds.cold+0x4c/0xf9 rxe_qp_from_attr.cold+0x37/0x5d [rdma_rxe] rxe_modify_qp+0x59/0x70 [rdma_rxe] _ib_modify_qp+0x5aa/0x7c0 [ib_core] ib_modify_qp+0x3b/0x50 [ib_core] cma_modify_qp_rtr+0x234/0x260 [rdma_cm] __rdma_accept+0x1a7/0x650 [rdma_cm] nvmet_rdma_cm_handler+0x1286/0x14cd [nvmet_rdma] cma_cm_event_handler+0x6b/0x330 [rdma_cm] cma_ib_req_handler+0xe60/0x22d0 [rdma_cm] cm_process_work+0x30/0x140 [ib_cm] cm_req_handler+0x11f4/0x1cd0 [ib_cm] cm_work_handler+0xb8/0x344e [ib_cm] process_one_work+0x569/0xb60 worker_thread+0x7a/0x5d0 kthread+0x1e6/0x210 ret_from_fork+0x24/0x30 Link: https://lore.kernel.org/r/20200217205714.26937-1-bvanassche@acm.org Fixes: 8700e3e7c485 ("Soft RoCE driver") Signed-off-by: Bart Van Assche <bvanassche@acm.org> Reviewed-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2020-01-15RDMA/rxe: Compute the maximum sges and inline size based on the WQE sizeRao Shoaib1-10/+8
The SGE buffer size and max_inline data should be derived from the size of the WQE. Each value individually sets the WQE size, so compute the actual sizes based on the actual WQE size and configure the QP with the maximums. Also fix the missing return of the actual maximum capability to the caller. Link: https://lore.kernel.org/r/1578962480-17814-3-git-send-email-rao.shoaib@oracle.com Signed-off-by: Rao Shoaib <rao.shoaib@oracle.com> Reviewed-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2019-04-01IB: Pass only ib_udata in function prototypesShamir Rabinovitch1-9/+6
Now when ib_udata is passed to all the driver's object create/destroy APIs the ib_udata will carry the ib_ucontext for every user command. There is no need to also pass the ib_ucontext via the functions prototypes. Make ib_udata the only argument psssed. Signed-off-by: Shamir Rabinovitch <shamir.rabinovitch@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2019-02-15IB/{hw,sw}: Remove 'uobject->context' dependency in object creation APIsShamir Rabinovitch1-3/+5
Now when we have the udata passed to all the ib_xxx object creation APIs and the additional macro 'rdma_udata_to_drv_context' to get the ib_ucontext from ib_udata stored in uverbs_attr_bundle, we can finally start to remove the dependency of the drivers in the ib_xxx->uobject->context. Signed-off-by: Shamir Rabinovitch <shamir.rabinovitch@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2019-02-04RDMA/rxe: Move rxe_init_av() to rxe_av.cKamal Heib1-5/+2
Move the function rxe_init_av() to rxe_av.c file and use it instead of calling rxe_av_from_attr() and rxe_av_fill_ip_info(), also remove the unused rxe_dev parameter from rxe_init_av(). Signed-off-by: Kamal Heib <kamalheib1@gmail.com> Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com> Reviewed-by: Zhu Yanjun <yanjun.zhu@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-12-18RDMA: Cleanup undesired pd->uobject usageShamir Rabinovitch1-2/+3
Drivers should be using udata to determine if a method is invoked from user space or kernel space. A pd does not necessarily say a different objects is kernel or user. Transforming the tests to use udata eliminates a large number of uobject references from the drivers. Signed-off-by: Shamir Rabinovitch <shamir.rabinovitch@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-12-06IB/rxe: Utilize generic function to validate port numberYuval Shaia1-3/+3
Utilize rdma_is_port_valid to validate the given port. Signed-off-by: Yuval Shaia <yuval.shaia@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-10-03RDMA: Remove unused parameter from ib_modify_qp_is_ok()Kamal Heib1-2/+1
The ll parameter is not used in ib_modify_qp_is_ok(), so remove it. Signed-off-by: Kamal Heib <kamalheib1@gmail.com> Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-09-30IB/rxe: replace kvfree with vfreeZhu Yanjun1-2/+3
The buf is allocated by vmalloc_user in the function rxe_queue_init. So it is better to free it by vfree. Fixes: 8700e3e7c485 ("Soft RoCE driver") Reviewed-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-08-30IB/rxe: vary the source udp port for receive scalingVijay Immanuel1-0/+10
Select the source udp port number for a QP based on the source QPN. This provides a better spread of traffic across NIC RX queues for RC/UC QPs. Signed-off-by: Vijay Immanuel <vijayi@attalasystems.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2018-06-18IB/core: add max_send_sge and max_recv_sge attributesSteve Wise1-4/+4
This patch replaces the ib_device_attr.max_sge with max_send_sge and max_recv_sge. It allows ulps to take advantage of devices that have very different send and recv sge depths. For example cxgb4 has a max_recv_sge of 4, yet a max_send_sge of 16. Splitting out these attributes allows much more efficient use of the SQ for cxgb4 with ulps that use the RDMA_RW API. Consider a large RDMA WRITE that has 16 scattergather entries. With max_sge of 4, the ulp would send 4 WRITE WRs, but with max_sge of 16, it can be done with 1 WRITE WR. Acked-by: Sagi Grimberg <sagi@grimberg.me> Acked-by: Christoph Hellwig <hch@lst.de> Acked-by: Selvin Xavier <selvin.xavier@broadcom.com> Acked-by: Shiraz Saleem <shiraz.saleem@intel.com> Acked-by: Dennis Dalessandro <dennis.dalessandro@intel.com> Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-06-18RDMA: Convert drivers to use sgid_attr instead of sgid_indexParav Pandit1-21/+2
The core code now ensures that all driver callbacks that receive an rdma_ah_attrs will have a sgid_attr's pointer if there is a GRH present. Drivers can use this pointer instead of calling a query function with sgid_index. This simplifies the drivers and also avoids races where a gid_index lookup may return different data if it is changed. Signed-off-by: Parav Pandit <parav@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
2018-03-15RDMA/rxe: Use structs to describe the uABI instead of opencodingJason Gunthorpe1-12/+14
Open coding pointer math is not acceptable for describing the uABI in RDMA. Provide structs for all the cases. The udata is casted to the struct as close to the verbs entry point as possible for maximum clarity. Function signatures and so forth are revised to allow for this. Tested-by: Yuval Shaia <yuval.shaia@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-02-28IB/rxe: Remove unused variable (char *rxe_qp_state_name[])Hernán Gonzalez1-9/+0
Note: This is compile only tested as I have no access to the hw. This variable was not used anywhere in the code. Removing it saves 24 bytes. add/remove: 0/1 grow/shrink: 0/0 up/down: 0/-24 (-24) Function old new delta rxe_qp_state_name 24 - -24 Total: Before=3348732, After=3348708, chg -0.00% Signed-off-by: Hernán Gonzalez <hernan@vanguardiasur.com.ar> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>