- 15 Dec, 2021 2 commits
-
-
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 <avihaih@nvidia.com> Reviewed-by:
Mark Zhang <markzhang@nvidia.com> Signed-off-by:
Leon Romanovsky <leonro@nvidia.com> Signed-off-by:
Jason Gunthorpe <jgg@nvidia.com>
-
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 1 commit
-
-
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:
Wenpeng Liang <liangwenpeng@huawei.com> Signed-off-by:
-
- 07 Dec, 2021 2 commits
-
-
Kamal Heib authored
Fix the wrongly reported max_send_wr and max_recv_wr attributes for user QP by making sure to save their valuse on QP creation, so when query QP is called the attributes will be reported correctly. Fixes: cecbcddf ("qedr: Add support for QP verbs") Link: https://lore.kernel.org/r/20211206201314.124947-1-kamalheib1@gmail.comSigned-off-by:
Kamal Heib <kamalheib1@gmail.com> Acked-by: Michal Kalderon <michal.kalderon@marvell.com> Signed-off-by:
-
Zhu Yanjun authored
The variable pkey is assigned from a macro. Then this variable is passed to a function bth_init directly, and pkey is not used again. So remove it and use the macro directly. Fixes: 76251e15 ("RDMA/rxe: Remove pkey table") Link: https://lore.kernel.org/r/20211207194057.713289-1-yanjun.zhu@linux.devSigned-off-by:
Zhu Yanjun <yanjun.zhu@linux.dev> Signed-off-by:
-
- 06 Dec, 2021 6 commits
-
-
Kamal Heib authored
Fix the following sparse warning: drivers/infiniband/hw/bnxt_re/qplib_fp.c:1260:26: sparse: warning: incorrect type in assignment (different base types) Fixes: 0e938533 ("RDMA/bnxt_re: Remove dynamic pkey table") Link: https://lore.kernel.org/r/20211205204537.14184-1-kamalheib1@gmail.comReported-by:
kernel test robot <lkp@intel.com> Signed-off-by:
Selvin Xavier <selvin.xavier@broadcom.com> Signed-off-by:
-
Christophe JAILLET authored
'bitmapbuf' is really used as a bitmap, so it should be defined as a 'unsigned long *' to be more consistent with the bitmap API. Link: https://lore.kernel.org/r/574b773fe7ced0cc87f1e1832350b38374815bd4.1638647428.git.christophe.jaillet@wanadoo.frSigned-off-by:
Christophe JAILLET <christophe.jaillet@wanadoo.fr> Signed-off-by:
-
Colin Ian King authored
The pointer mm is assigned a value but it is never used. The pointer is redundant and can be removed. Link: https://lore.kernel.org/r/20211204234904.105026-1-colin.i.king@gmail.comSigned-off-by:
Colin Ian King <colin.i.king@gmail.com> Signed-off-by:
-
Zhu Yanjun authored
The struct member variable create_flags is assigned twice. Remove the unnecessary assignment. Fixes: ece9ca97 ("RDMA/uverbs: Do not check the input length on create_cq/qp paths") Link: https://lore.kernel.org/r/20211207064607.541695-1-yanjun.zhu@linux.devSigned-off-by:
Håkon Bugge <haakon.bugge@oracle.com> Signed-off-by:
-
Yixing Liu authored
It is more general for ARM device drivers to use the device attribute to map PCI BAR spaces. Fixes: 9a443537 ("IB/hns: Add driver files for hns RoCE driver") Link: https://lore.kernel.org/r/20211206133652.27476-1-liangwenpeng@huawei.comSigned-off-by:
-
Jiapeng Chong authored
Fix following coccicheck warning: ./drivers/infiniband/sw/siw/siw_verbs.c:665:28-29: WARNING opportunity for max(). Link: https://lore.kernel.org/r/1638439679-114250-1-git-send-email-jiapeng.chong@linux.alibaba.comReported-by:
Abaci Robot <abaci@linux.alibaba.com> Signed-off-by:
Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Reviewed-by:
Bernard Metzler <bmt@zurich.ibm.com> Signed-off-by:
-
- 29 Nov, 2021 7 commits
-
-
Christophe JAILLET authored
Use 'bitmap_zalloc()' to simplify code, improve the semantic and avoid some open-coded arithmetic in allocator arguments. Also change the corresponding 'kfree()' into 'bitmap_free()' to keep consistency. Link: https://lore.kernel.org/r/5c029daf43b92fdc27926fe8a98084843437c498.1637872888.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
Christophe JAILLET authored
In 'pvrdma_uar_table_init()', the 'tbl->table' bitmap has just been allocated, so no concurrent accesses can occur. The other accesses to the 'tbl->table' bitmap are protected by the 'tbl->lock' spinlock, so no concurrent accesses can happen. So prefer the non-atomic '__[set|clear]_bit()' functions to save a few cycles. Link: https://lore.kernel.org/r/271b0e2c316e2b4cf34ac6fbca0701edd2d882ec.1637870667.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
Christophe JAILLET authored
Use 'bitmap_zalloc()' to simplify code, improve the semantic and avoid some open-coded arithmetic in allocator arguments. Also change the corresponding 'kfree()' into 'bitmap_free()' to keep consistency. Link: https://lore.kernel.org/r/33e8b993bfa6b7164e9bee95e3c27fb2c53949ce.1637870667.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
Christophe JAILLET authored
Use 'bitmap_zalloc()' to simplify code, improve the semantic and avoid some open-coded arithmetic in allocator arguments. Also change the corresponding 'kfree()' into 'bitmap_free()' to keep consistency. Link: https://lore.kernel.org/r/d46c6bc1869b8869244fa71943d2cad4104b3668.1637869925.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
Christophe JAILLET authored
Use 'bitmap_alloc()' to simplify code, improve the semantic and avoid some open-coded arithmetic in allocator arguments. Also change the corresponding 'kfree()' into 'bitmap_free()' to keep consistency. Link: https://lore.kernel.org/r/4c93b4e02f5d784ddfd3efd4af9e673b9117d641.1637869328.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
Christophe JAILLET authored
'pd_bitmap' does not need to be const. Without it, it is possible to use this variable when calling '__set_bit()'. This is less verbose and more logical. Link: https://lore.kernel.org/r/ec5cab9611ba062adea4cf8c98a63406ed510a71.1637868728.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
Christophe JAILLET authored
Use 'bitmap_zalloc()' to simplify code, improve the semantic and avoid some open-coded arithmetic in allocator arguments. Also change the corresponding 'kfree()' into 'bitmap_free()' to keep consistency. Link: https://lore.kernel.org/r/b157f9e1586fb4d1083cb4058d7ac81b10bb86d7.1637868728.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
- 25 Nov, 2021 11 commits
-
-
Kamal Heib authored
The RoCE spec requires RoCE devices to support only the default pkey. However the bnxt_re driver maintains a 0xFFFF entry pkey table and uses only the first entry. Remove the pkey table and hard code a table of length one hard wired with the default pkey. Link: https://lore.kernel.org/r/20211125033615.483750-1-kamalheib1@gmail.comSigned-off-by:
Devesh Sharma <devesh.s.sharma@oracle.com> Signed-off-by:
-
Kamal Heib authored
Use the addrconf_addr_eui48() helper function to set the sys_image_guid, Also make sure the GUID is valid EUI-64 identifier. Link: https://lore.kernel.org/r/20211124102336.427637-1-kamalheib1@gmail.comSigned-off-by:
-
Jack Wang authored
The type of min_latency is ktime_t, so use KTIME_MAX to initialize the initial value. Fixes: dc3b66a0 ("RDMA/rtrs-clt: Add a minimum latency multipath policy") Link: https://lore.kernel.org/r/20211124081040.19533-1-jinpu.wang@ionos.comSigned-off-by:
Jack Wang <jinpu.wang@ionos.com> Reviewed-by:
Guoqing Jiang <Guoqing.Jiang@linux.dev> Signed-off-by:
-
Håkon Bugge authored
The existing tests are a little hard to comprehend. Use check_add_overflow() instead. Fixes: 04ded167 ("RDMA/cma: Verify private data length") Link: https://lore.kernel.org/r/1637661978-18770-1-git-send-email-haakon.bugge@oracle.comSigned-off-by:
-
Christophe JAILLET authored
The accesses to the 'alloc->table' bitmap are protected by the 'alloc->lock' spinlock, so no concurrent accesses can happen. So prefer the non-atomic '__[set|clear]_bit()' functions to save a few cycles. Link: https://lore.kernel.org/r/0c1c4505ca32f5ba4126e3e324041da191513ef2.1637789139.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
Christophe JAILLET authored
The 'alloc->table' bitmap has just been allocated, so this is safe to use the faster and non-atomic 'bitmap_set()' function. There is no need to hand-write it. Link: https://lore.kernel.org/r/fd978b837935ed04863ffecfd495c4601a986df6.1637789139.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
Christophe JAILLET authored
Use 'bitmap_zalloc()' to simplify code, improve the semantic and avoid some open-coded arithmetic in allocator arguments. Using the 'zalloc' version of the allocator also saves a now useless 'bitmap_zero()' call. Also change the corresponding 'kfree()' into 'bitmap_free()' to keep consistency. While at it, remove an extra space in a statement just a few lines above. Link: https://lore.kernel.org/r/e396c4aa16cd8945d43877570a8f6d926cea555a.1637789139.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
Christophe JAILLET authored
In 'mthca_buddy_init()', the 'buddy->bits[n]' bitmap has just been allocated, so no concurrent accesses can occur. The other accesses to the 'buddy->bits[n]' bitmap are protected by the 'buddy->lock' spinlock, so no concurrent accesses can occur. So prefer the non-atomic '__[set|clear]_bit()' functions to save a few cycles. Link: https://lore.kernel.org/r/a19b88ccdbc03972fd97306b998731814283041f.1637785902.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
Christophe JAILLET authored
The accesses to the 'alloc->table' bitmap are protected by the 'alloc->lock' spinlock, so no concurrent accesses can happen. So prefer the non-atomic '__[set|clear]_bit()' functions to save a few cycles. Link: https://lore.kernel.org/r/5f909ca1284fa4d2cf13952b08b9e303b656c968.1637785902.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
Christophe JAILLET authored
The 'alloc->table' bitmap has just been allocated, so this is safe to use the faster and non-atomic 'bitmap_set()' function. There is no need to hand-write it. Link: https://lore.kernel.org/r/f1bd33f6ea6c8ad519a222db6e9aa17c55610557.1637785902.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
Christophe JAILLET authored
Use 'bitmap_zalloc()' to simplify code, improve the semantic and avoid some open-coded arithmetic in allocator arguments. Using the 'zalloc' version of the allocator also saves a now useless 'bitmap_zero()' call. Also change the corresponding 'kfree()' into 'bitmap_free()' to keep consistency. Link: https://lore.kernel.org/r/ea9031e28f453bc179033740f66f0c19293fcf0b.1637785902.git.christophe.jaillet@wanadoo.frSigned-off-by:
-
- 19 Nov, 2021 11 commits
-
-
Xinhao Liu authored
Don't use unintelligible constants. Link: https://lore.kernel.org/r/20211119140208.40416-10-liangwenpeng@huawei.comSigned-off-by:
Xinhao Liu <liuxinhao5@hisilicon.com> Signed-off-by:
-
Yixing Liu authored
These macros are no longer used, so remove them. Link: https://lore.kernel.org/r/20211119140208.40416-9-liangwenpeng@huawei.comSigned-off-by:
-
Xinhao Liu authored
Each member of Array[][] should be initialized on a separate line. Link: https://lore.kernel.org/r/20211119140208.40416-7-liangwenpeng@huawei.comSigned-off-by:
Xinhao Liu <liuxinhao@huawei.com> Signed-off-by:
-
Xinhao Liu authored
The type of the variable participating in the shift operation should be an unsigned type instead of a signed type. Link: https://lore.kernel.org/r/20211119140208.40416-5-liangwenpeng@huawei.comSigned-off-by:
-
Xinhao Liu authored
There should be a space between the code and the comment on the right. Link: https://lore.kernel.org/r/20211119140208.40416-4-liangwenpeng@huawei.comSigned-off-by:
-
Xinhao Liu authored
The print format should be consistent with the variable type. Link: https://lore.kernel.org/r/20211119140208.40416-3-liangwenpeng@huawei.comSigned-off-by:
-
Xinhao Liu authored
The hex printf format should be "0xff" instead of "ff". Link: https://lore.kernel.org/r/20211119140208.40416-2-liangwenpeng@huawei.comSigned-off-by:
-
Bob Pearson authored
RXE_POOL_ALIGN is only used in rxe_pool.c so move RXE_POOL_ALIGN to rxe_pool.c from rxe_pool.h. RXE_POOL_CACHE_FLAGS is never used so it is deleted from rxe_pool.h Link: https://lore.kernel.org/r/20211103050241.61293-8-rpearsonhpe@gmail.comSigned-off-by:
Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by:
-
Bob Pearson authored
rxe_loc.h is already included in rxe.h so do not include it in rxe_pool.c Link: https://lore.kernel.org/r/20211103050241.61293-7-rpearsonhpe@gmail.comSigned-off-by:
-
Bob Pearson authored
In rxe_pool.c currently there are many cases where it is necessary to compute the offset from a pool element struct to the object containing it in a type independent way where the offset is different for each type. By saving a pointer to the object when they are created extra work can be saved. Link: https://lore.kernel.org/r/20211103050241.61293-5-rpearsonhpe@gmail.comSigned-off-by:
-
Bob Pearson authored
In rxe_pool.c copy remaining pool setup parameters from rxe_pool_info into rxe_pool. This saves looking up rxe_pool_info in the performance path. Link: https://lore.kernel.org/r/20211103050241.61293-4-rpearsonhpe@gmail.comSigned-off-by:
-
