- 07 Jan, 2022 8 commits
-
-
Zhu Yanjun authored
Calculate and set UDP source port based on the flow label. If flow label is not defined in GRH then calculate it based on lqpn/rqpn. Link: https://lore.kernel.org/r/20220106180359.2915060-2-yanjun.zhu@linux.devSigned-off-by:
Zhu Yanjun <yanjun.zhu@linux.dev> Reviewed-by:
Leon Romanovsky <leonro@nvidia.com> Signed-off-by:
Jason Gunthorpe <jgg@nvidia.com>
-
Qinghua Jin authored
Change 'postion' to 'position'. Link: https://lore.kernel.org/r/20220106082722.354680-1-qhjin.dev@gmail.comSigned-off-by:
Qinghua Jin <qhjin.dev@gmail.com> Acked-by:
Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com> Signed-off-by:
-
Vaishali Thakkar authored
Structure rtrs_clt is used for sessions. So to avoid confusions rename it to rtrs_clt_sess. Transformations are done with the help of following coccinelle script. @@ @@ struct - rtrs_clt + rtrs_clt_sess Link: https://lore.kernel.org/r/20220105180708.7774-6-jinpu.wang@ionos.comSigned-off-by:
Vaishali Thakkar <vaishali.thakkar@ionos.com> Signed-off-by:
Jack Wang <jinpu.wang@ionos.com> Signed-off-by:
-
Vaishali Thakkar authored
Structure rtrs_srv is used for sessions so in order to avoid confusions rename it to rtrs_srv_sess. All changes were done with the help of following Coccinelle script: @@ @@ struct - rtrs_srv + rtrs_srv_sess Link: https://lore.kernel.org/r/20220105180708.7774-5-jinpu.wang@ionos.comSigned-off-by:
-
Vaishali Thakkar authored
rtrs_clt_sess is used for paths and not sessions on the client side. This creates confusion so let's rename it to rtrs_clt_path. Also, rename related variables and functions. Coccinelle is used to do the transformations for most of the occurrences and remaining ones were handled manually. Link: https://lore.kernel.org/r/20220105180708.7774-4-jinpu.wang@ionos.comSigned-off-by:
-
Vaishali Thakkar authored
rtrs_srv_sess is used for paths and not sessions on the server side. This creates confusion so let's rename it to rtrs_srv_path. Also, rename related variables and functions. Coccinelle is used to do the transformations for most of the occurrences and remaining ones were handled manually. Link: https://lore.kernel.org/r/20220105180708.7774-3-jinpu.wang@ionos.comSigned-off-by:
-
Vaishali Thakkar authored
rtrs_sess is in fact a path. This makes it confusing and difficult to get into the code. So let's rename the structure and related use cases of it. Coccinelle was used to do the transformation for most of the occurrences and remaining ones were handled manually. Link: https://lore.kernel.org/r/20220105180708.7774-2-jinpu.wang@ionos.comSigned-off-by:
-
Wenpeng Liang authored
HIP09 EQ does not support level 2 addressing. Link: https://lore.kernel.org/r/20211231101341.45759-3-liangwenpeng@huawei.comSigned-off-by:
Wenpeng Liang <liangwenpeng@huawei.com> Signed-off-by:
-
- 06 Jan, 2022 1 commit
-
-
Max Gurtovoy authored
The following changes were made: 1. Align function signatures to 80 characters per line. 2. Remove tabs for variable assignment and use 1 space instead. 3. Don't compare to NULL in "if" clause. 4. Remove strange indentations. This will ease on the maintenance of the driver for the future. Link: https://lore.kernel.org/r/20211215135721.3662-7-mgurtovoy@nvidia.comSigned-off-by:
Max Gurtovoy <mgurtovoy@nvidia.com> Signed-off-by:
-
- 05 Jan, 2022 17 commits
-
-
Max Gurtovoy authored
The void pointer can be typecasted to/from any type. Link: https://lore.kernel.org/r/20211215135721.3662-6-mgurtovoy@nvidia.comSigned-off-by:
-
Max Gurtovoy authored
In order to complete a scsi command and guarantee that the HCA will never perform an access violation when retrying a send operation we must complete a scsi request only when both send and receive completions has arrived. This is a preparation commit that remove the send completions suppression. Next step will be taking care of the local invalidation mechanism and adding a reference counter for commands. Currently, we don't do anything upon getting the send completion and just "consume" it. Link: https://lore.kernel.org/r/20211215135721.3662-5-mgurtovoy@nvidia.comSigned-off-by:
Israel Rukshin <israelr@nvidia.com> Signed-off-by:
-
Max Gurtovoy authored
Use more common name for return values ("ret"). This commit doesn't change any logic. Link: https://lore.kernel.org/r/20211215135721.3662-4-mgurtovoy@nvidia.comSigned-off-by: -
Sergey Gorenko authored
Some users complain about RNR errors on the target, when heavy high-priority tasks run on the initiator. After the investigation, we found out that the receive WRs were exhausted, because the initiator could not post them on time. Receive work reqeusts are posted in chunks to reduce the number of hits to the HCA. The WRs are posted in the receive completion handler when the number of free receive buffers reaches the threshold. But on a high-loaded host, receive CQEs processing can be delayed and all receive WRs will be exhausted. In this case, the target will get an RNR error. To avoid this, we post receive WR, as soon as possible and not in a batch. This increases the number of hits to the HCA, but also the common implementation in most of Linux ULPs (e.g. NVMe-oF/RDMA). As a rule of thumb, performance improvements and heuristics are being added to the RDMA core layer or vendors low level drivers and it's about time to align iSER as well. Link: https://lore.kernel.org/r/20211215135721.3662-3-mgurtovoy@nvidia.comSigned-off-by:
Sergey Gorenko <sergeygo@nvidia.com> Signed-off-by:
-
Max Gurtovoy authored
No need for this dead code. This commit doesn't change any functionality since one can still run "modprobe ib_iser pi_guard=<type>". Link: https://lore.kernel.org/r/20211215135721.3662-2-mgurtovoy@nvidia.comSigned-off-by:
-
Maher Sanalla authored
Under MAD query port, Report NDR speed when NDR is supported in the port capability mask. Link: https://lore.kernel.org/r/a2ab630d2a634547db9b581faa9d65da2edb9d05.1639554831.git.leonro@nvidia.comSigned-off-by:
Maher Sanalla <msanalla@nvidia.com> Reviewed-by:
Michael Guralnik <michaelgur@nvidia.com> Signed-off-by:
-
Kamal Heib authored
The API for ib_query_qp requires the driver to set cur_qp_state on return, add the missing set. Fixes: 67bbc055 ("RDMA/cxgb4: Add query_qp support") Link: https://lore.kernel.org/r/20211220152530.60399-1-kamalheib1@gmail.comSigned-off-by:
Kamal Heib <kamalheib1@gmail.com> Reviewed-by:
-
Chengchang Tang authored
HIP06 is no longer supported. In order to reduce unnecessary maintenance, the code of HIP06 is removed. Link: https://lore.kernel.org/r/20211220130558.61585-1-liangwenpeng@huawei.comSigned-off-by:
Chengchang Tang <tangchengchang@huawei.com> Signed-off-by:
-
Leon Romanovsky authored
Starting from the commit 66920e1b ("rdma_rxe: Use netlink messages to add/delete links") from the 2019, the RXE modules parameters are marked as deprecated in favour of rdmatool. So remove the kernel code too. Link: https://lore.kernel.org/r/c8376d7517aebe7cc851f0baaeef7b13707cf767.1641372460.git.leonro@nvidia.comSigned-off-by:
Zhu Yanjun <zyjzyj2000@gmail.com> Signed-off-by:
-
Leon Romanovsky authored
Several drivers used same function to initialize query MAD, so move that function to global header file. Link: https://lore.kernel.org/r/af6f35c590ff5ef56d0137351b8b295af0f7c13c.1641369858.git.leonro@nvidia.comSigned-off-by:
Håkon Bugge <haakon.bugge@oracle.com> Signed-off-by:
-
Li Zhijian authored
* Fix these up to always have the '+', and '|' on the continuing line which is the normal kernel style. * Fix indentations correspondingly NOTE: this patch also remove the 2 redundant plus in IB_OPCODE_RD_FETCH_ADD and IB_OPCODE_RD_COMPARE_SWAP Link: https://lore.kernel.org/r/20220105042605.14343-1-lizhijian@fujitsu.comSigned-off-by:
Li Zhijian <lizhijian@fujitsu.com> Reviewed-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Greg Kroah-Hartman authored
There are currently 2 ways to create a set of sysfs files for a kobj_type, through the default_attrs field, and the default_groups field. Move the IB code to use default_groups field which has been the preferred way since commit aa30f47c ("kobject: Add support for default attribute groups to kobj_type") so that we can soon get rid of the obsolete default_attrs field. Link: https://lore.kernel.org/r/20220103152259.531034-1-gregkh@linuxfoundation.orgSigned-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Reviewed-by:
-
Dust Li authored
mlx5_handle_error_cqe() only dump the content of the CQE which is raw hex data, and not straighforward for debug. Print WC status message when we got CQE error and dump is need. Here is an example of how the dmesg log looks like with this: infiniband mlx5_0: mlx5_handle_error_cqe:333:(pid 0): WC error: 10, message: remote access error infiniband mlx5_0: dump_cqe:272:(pid 0): dump error cqe 00000000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00000020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00000030: 00 00 00 00 00 00 88 13 08 03 61 b3 1e a1 42 d3 Link: https://lore.kernel.org/r/20211227123806.47530-1-dust.li@linux.alibaba.comSigned-off-by:
Dust Li <dust.li@linux.alibaba.com> Acked-by:
-
Chengguang Xu authored
There is a redundant ']' in the name of opcode IB_OPCODE_RC_SEND_MIDDLE, so just fix it. Fixes: 8700e3e7 ("Soft RoCE driver") Link: https://lore.kernel.org/r/20211218112320.3558770-1-cgxu519@mykernel.netSigned-off-by:
Chengguang Xu <cgxu519@mykernel.net> Acked-by:
Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by:
-
Zhu Yanjun authored
The member variable xmit_errors can be replaced with rxe_counter_inc(rxe, RXE_CNT_SEND_ERR) Link: https://lore.kernel.org/r/20211216054842.1099428-1-yanjun.zhu@linux.devSigned-off-by:
-
Minghao Chi authored
Return value directly instead of taking this in another redundant variable. Link: https://lore.kernel.org/20211215075258.442930-1-chi.minghao@zte.com.cnReported-by:
Zeal Robot <zealci@zte.com.cn> Signed-off-by:
Minghao Chi <chi.minghao@zte.com.cn> Acked-by:
Devesh Sharma <Devesh.s.sharma@oracle.com> Signed-off-by:
-
Minghao Chi authored
Return status directly from function called. Link: https://lore.kernel.org/r/20211215055421.441375-1-chi.minghao@zte.com.cnReported-by:
-
- 15 Dec, 2021 8 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linuxJason Gunthorpe authored
Saeed Mahameed says: ==================== Currently, the driver ignores the user's priority for flow steering rules in FDB namespace. Change it and create the rule in the right priority. It will allow to create FDB steering rules in up to 16 different priorities. ==================== Signed-off-by: -
Kees Cook authored
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields. Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct. Link: https://lore.kernel.org/r/20211213223331.135412-15-keescook@chromium.orgSigned-off-by:
Kees Cook <keescook@chromium.org> Signed-off-by:
-
Kees Cook authored
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields. Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct. Additionally, since everything appears to perform a roundup (including allocation), just change the size of the struct itself and add a build-time check to validate the expected size. Link: https://lore.kernel.org/r/20211213223331.135412-13-keescook@chromium.orgSigned-off-by:
-
Kees Cook authored
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields. Use memset_after() to zero the end of struct mlx5_ib_mr that should be initialized. Link: https://lore.kernel.org/r/20211213223331.135412-10-keescook@chromium.orgAcked-by:
-
Jason Gunthorpe authored
Required due to dependencies in following patches. Signed-off-by: -
Avihai Horon authored
Currently, when cma_resolve_ib_dev() searches for a matching GID it will stop searching after encountering the first empty GID table entry. This behavior is wrong since neither IB nor RoCE spec enforce tightly packed GID tables. For example, when the matching valid GID entry exists at index N, and if a GID entry is empty at index N-1, cma_resolve_ib_dev() will fail to find the matching valid entry. Fix it by making cma_resolve_ib_dev() continue searching even after encountering missing entries. Fixes: f17df3b0 ("RDMA/cma: Add support for AF_IB to rdma_resolve_addr()") Link: https://lore.kernel.org/r/b7346307e3bb396c43d67d924348c6c496493991.1639055490.git.leonro@nvidia.comSigned-off-by:
Avihai Horon <avihaih@nvidia.com> Reviewed-by:
Mark Zhang <markzhang@nvidia.com> Signed-off-by:
-
Avihai Horon authored
Currently, ib_find_gid() will stop searching after encountering the first empty GID table entry. This behavior is wrong since neither IB nor RoCE spec enforce tightly packed GID tables. For example, when a valid GID entry exists at index N, and if a GID entry is empty at index N-1, ib_find_gid() will fail to find the valid entry. Fix it by making ib_find_gid() continue searching even after encountering missing entries. Fixes: 5eb620c8 ("IB/core: Add helpers for uncached GID and P_Key searches") Link: https://lore.kernel.org/r/e55d331b96cecfc2cf19803d16e7109ea966882d.1639055490.git.leonro@nvidia.comSigned-off-by:
-
Avihai Horon authored
Modify rdma_query_gid() to return -ENOENT for empty entries. This will make error reporting more accurate and will be used in next patches. Link: https://lore.kernel.org/r/1f2b65dfb4d995e74b621e3e21e7c7445d187956.1639055490.git.leonro@nvidia.comSigned-off-by:
-
- 14 Dec, 2021 5 commits
-
-
Yixing Liu authored
The current write wqe mechanism is to write to DDR first, and then notify the hardware through doorbell to read the data. Direct wqe is a mechanism to fill wqe directly into the hardware. In the case of light load, the wqe will be filled into pcie bar space of the hardware, this will reduce one memory access operation and therefore reduce the latency. SIMD instructions allows cpu to write the 512 bits at one time to device memory, thus it can be used for posting direct wqe. Add direct wqe enable switch and address mapping. Link: https://lore.kernel.org/r/20211207124901.42123-2-liangwenpeng@huawei.comSigned-off-by:
Yixing Liu <liuyixing1@huawei.com> Signed-off-by:
-
Maor Gottlieb authored
Currently, the driver ignores the user's priority for flow steering rules in FDB namespace. Change it and create the rule in the right priority. It will allow to create FDB steering rules in up to 16 different priorities. Signed-off-by:
Maor Gottlieb <maorg@nvidia.com> Reviewed-by:
Mark Bloch <mbloch@nvidia.com> Signed-off-by:
Saeed Mahameed <saeedm@nvidia.com> Acked-by:
-
Maor Gottlieb authored
Create 16 flow steering priorities for FDB bypass users. Signed-off-by:
-
Maor Gottlieb authored
Have all the namespace type check in the same switch case. Signed-off-by:
-
Maor Gottlieb authored
This patch doesn't add an additional namespaces, but just separates the naming to be used by each FDB user, bypass and kernel. Downstream patches will actually split this up and allow to have more than single priority for the bypass users. Signed-off-by:
-
- 12 Dec, 2021 1 commit
-
-
Linus Torvalds authored
-
