- 12 Oct, 2021 24 commits
-
-
Bob Pearson authored
The pd field in struct rxe_ah is redundant with the pd field in the rdma-core's ib_ah. Eliminate the pd field in rxe_ah and add an inline to extract the pd from the ibah field. Link: https://lore.kernel.org/r/20211007204051.10086-5-rpearsonhpe@gmail.comSigned-off-by:
Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by:
Jason Gunthorpe <jgg@nvidia.com>
-
Bob Pearson authored
Make changes to rdma_user_rxe.h to allow indexing AH objects, passing the index in UD send WRs to the driver and returning the index to the rxe provider. Modify rxe_create_ah() to add an index to AH when created and if called from a new user provider return it to user space. If called from an old provider mark the AH as not having a useful index. Modify rxe_destroy_ah to drop the index before deleting the object. Link: https://lore.kernel.org/r/20211007204051.10086-4-rpearsonhpe@gmail.comSigned-off-by:
-
Bob Pearson authored
Make changes to rxe_param.h and rxe_pool.c to allow indexing of AH objects. Valid indices are non-zero so older providers can be detected. Link: https://lore.kernel.org/r/20211007204051.10086-3-rpearsonhpe@gmail.comSigned-off-by:
-
Bob Pearson authored
Move the struct rxe_av av from struct rxe_send_wqe to struct rxe_send_wr placing it in wr.ud at the same offset as it was previously. This has the effect of increasing the size of struct rxe_send_wr while keeping the size of struct rxe_send_wqe the same. This better reflects the use of this field which is only used for UD sends. This change has no effect on ABI compatibility so the modified rxe driver will operate with older versions of rdma-core. Link: https://lore.kernel.org/r/20211007204051.10086-2-rpearsonhpe@gmail.comSigned-off-by:
-
Leon Romanovsky authored
The error flow fixed in this patch is not possible because all kernel users of create QP interface check that device supports steering before set IB_QP_CREATE_NETIF_QP flag. Fixes: c1c98501 ("IB/mlx4: Add support for steerable IB UD QPs") Link: https://lore.kernel.org/r/91c61f6e60eb0240f8bbc321fda7a1d2986dd03c.1634023677.git.leonro@nvidia.comReported-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
Leon Romanovsky <leonro@nvidia.com> Signed-off-by:
-
Zhu Yanjun authored
The function irdma_cqp_up_map_cmd() is not used. So remove it. Link: https://lore.kernel.org/r/20211011110128.4057-5-yanjun.zhu@linux.devSigned-off-by:
Zhu Yanjun <yanjun.zhu@linux.dev> Acked-by:
Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by:
-
Zhu Yanjun authored
The function irdma_get_hw_addr() is not used. So remove it. Link: https://lore.kernel.org/r/20211011110128.4057-4-yanjun.zhu@linux.devSigned-off-by:
-
Zhu Yanjun authored
The function irdma_sc_send_lsmm_nostag is not used. So remove it. Link: https://lore.kernel.org/r/20211011110128.4057-3-yanjun.zhu@linux.devSigned-off-by:
-
Zhu Yanjun authored
The function irdma_uk_mw_bind is not used. So remove it. Link: https://lore.kernel.org/r/20211011110128.4057-2-yanjun.zhu@linux.devSigned-off-by:
-
Christophe JAILLET authored
'destroy_workqueue()' already drains the queue before destroying it, so there is no need to flush it explicitly. Remove the redundant 'flush_workqueue()' calls. This was generated with coccinelle: @@ expression E; @@ - flush_workqueue(E); destroy_workqueue(E); Link: https://lore.kernel.org/r/ca7bac6e6c9c5cc8d04eec3944edb13de0e381a3.1633874776.git.christophe.jaillet@wanadoo.frSigned-off-by:
Christophe JAILLET <christophe.jaillet@wanadoo.fr> Reviewed-by:
-
Colin Ian King authored
The pointer err_str is being initialized with a value that is never read, it is being updated later on. The assignment is redundant and can be removed. Link: https://lore.kernel.org/r/20211007173942.21933-1-colin.king@canonical.com Addresses-Coverity: ("Unused value") Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
Cai Huoqing authored
Replacing kmalloc/kfree/dma_map_single/dma_unmap_single() with dma_alloc_coherent/dma_free_coherent() helps to reduce code size, and simplify the code, and coherent DMA will not clear the cache every time. The SOC that this driver supports does not have incoherent DMA, so this makes the code follow the DMA API properly with no performance impact. Currently there are missing dma sync calls around the DMA transfers. Link: https://lore.kernel.org/r/20210926061116.282-1-caihuoqing@baidu.comSigned-off-by:
Cai Huoqing <caihuoqing@baidu.com> Reviewed-by:
Wenpeng Liang <liangwenpeng@huawei.com> Tested-by:
-
Aharon Landau authored
When get_hw_stats is called, query and return the optional counter statistic as well. Link: https://lore.kernel.org/r/20211008122439.166063-14-markzhang@nvidia.comSigned-off-by:
Aharon Landau <aharonl@nvidia.com> Signed-off-by:
Mark Zhang <markzhang@nvidia.com> Signed-off-by:
-
Aharon Landau authored
Add support for ib callback modify_op_stat() to add or remove an optional counter. When adding, a steering flow table is created with a rule that catches and counts all the matching packets. When removing, the table and flow counter are destroyed. Link: https://lore.kernel.org/r/20211008122439.166063-13-markzhang@nvidia.comSigned-off-by:
-
Aharon Landau authored
Adding steering infrastructure for adding and removing optional counter. This allows to add and remove the counters dynamically in order not to hurt performance. Link: https://lore.kernel.org/r/20211008122439.166063-12-markzhang@nvidia.comSigned-off-by:
Maor Gottlieb <maorg@nvidia.com> Signed-off-by:
-
Aharon Landau authored
Add optional counter support when allocate and initialize hw_stats structure. Optional counters have IB_STAT_FLAG_OPTIONAL flag set and are disabled by default. Link: https://lore.kernel.org/r/20211008122439.166063-11-markzhang@nvidia.comSigned-off-by:
-
Aharon Landau authored
Provide an option to allow users to enable/disable optional counters through RDMA netlink. Limiting it to users with ADMIN capability only. Examples: 1. Enable optional counters cc_rx_ce_pkts and cc_rx_cnp_pkts (and disable all others): $ sudo rdma statistic set link rocep8s0f0/1 optional-counters \ cc_rx_ce_pkts,cc_rx_cnp_pkts 2. Remove all optional counters: $ sudo rdma statistic unset link rocep8s0f0/1 optional-counters Link: https://lore.kernel.org/r/20211008122439.166063-10-markzhang@nvidia.comSigned-off-by: -
Aharon Landau authored
In order to allow expansion of the set command with more set options, take the set mode out of the main set function. Link: https://lore.kernel.org/r/20211008122439.166063-9-markzhang@nvidia.comSigned-off-by:
-
Aharon Landau authored
This patch adds the ability to get the name, index and status of all counters for each link through RDMA netlink. This can be used for user-space to get the current optional-counter mode. Examples: $ rdma statistic mode link rocep8s0f0/1 optional-counters cc_rx_ce_pkts $ rdma statistic mode supported link rocep8s0f0/1 supported optional-counters cc_rx_ce_pkts,cc_rx_cnp_pkts,cc_tx_cnp_pkts link rocep8s0f1/1 supported optional-counters cc_rx_ce_pkts,cc_rx_cnp_pkts,cc_tx_cnp_pkts Link: https://lore.kernel.org/r/20211008122439.166063-8-markzhang@nvidia.comSigned-off-by:
Neta Ostrovsky <netao@nvidia.com> Signed-off-by:
-
Aharon Landau authored
An optional counter is a driver-specific counter that may be dynamically enabled/disabled. This enhancement allows drivers to expose counters which are, for example, mutually exclusive and cannot be enabled at the same time, counters that might degrades performance, optional debug counters, etc. Optional counters are marked with IB_STAT_FLAG_OPTIONAL flag. They are not exported in sysfs, and must be at the end of all stats, otherwise the attr->show() in sysfs would get wrong indexes for hwcounters that are behind optional counters. Link: https://lore.kernel.org/r/20211008122439.166063-7-markzhang@nvidia.comSigned-off-by:
-
Aharon Landau authored
Add a bitmap in rdma_hw_stat structure, with each bit indicates whether the corresponding counter is currently disabled or not. By default hwcounters are enabled. Link: https://lore.kernel.org/r/20211008122439.166063-6-markzhang@nvidia.comSigned-off-by:
-
Mark Zhang authored
Add a new API rdma_free_hw_stats_struct to pair with rdma_alloc_hw_stats_struct (which is also de-inlined). This will be useful when there are more alloc/free works in following patches. Link: https://lore.kernel.org/r/20211008122439.166063-5-markzhang@nvidia.comSigned-off-by:
-
Aharon Landau authored
Add a counter statistic descriptor structure in rdma_hw_stats. In addition to the counter name, more meta-information will be added. This code extension is needed for optional-counter support in the following patches. Link: https://lore.kernel.org/r/20211008122439.166063-4-markzhang@nvidia.comSigned-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linuxJason Gunthorpe authored
For dependencies in the following patches. * mellanox/mlx5-next: net/mlx5: Add priorities for counters in RDMA namespaces net/mlx5: Add ifc bits to support optional counters Signed-off-by:
-
- 09 Oct, 2021 2 commits
-
-
Aharon Landau authored
Add additional flow steering priorities in the RDMA namespace. This allows adding flow counters to count filtered RDMA traffic and then continue processing in the regular RDMA steering flow. Signed-off-by:
-
Aharon Landau authored
Adding bth_opcode field and the relevant bits. This field will be used to capture and count congestion notification packets (CNP). Adding source_vhca_port support bit. This field will be used to check the capability to use the source_vhca_port as a match criteria in cases of dual port. Signed-off-by:
-
- 06 Oct, 2021 6 commits
-
-
Gal Pressman authored
This patch adds support for CQ notifications through the standard verbs api. In order to achieve that, a new event queue (EQ) object is introduced, which is in charge of reporting completion events to the driver. On driver load, EQs are allocated and their affinity is set to a single cpu. When a user app creates a CQ with a completion channel, the completion vector number is converted to a EQ number, which is in charge of reporting the CQ events. In addition, the CQ creation admin command now returns an offset for the CQ doorbell, which is mapped to the userspace provider and is used to arm the CQ when requested by the user. The EQs use a single doorbell (located on the registers BAR), which encodes the EQ number and arm as part of the doorbell value. The EQs are polled by the driver on each new EQE, and arm it when the poll is completed. Link: https://lore.kernel.org/r/20211003105605.29222-1-galpress@amazon.comReviewed-by:
Firas JahJah <firasj@amazon.com> Reviewed-by:
Yossi Leybovich <sleybo@amazon.com> Signed-off-by:
Gal Pressman <galpress@amazon.com> Signed-off-by:
-
Xiao Yang authored
Remove duplicate settings for vendor_err and qp_num. Link: https://lore.kernel.org/r/20210930094813.226888-5-yangx.jy@fujitsu.comSigned-off-by:
Xiao Yang <yangx.jy@fujitsu.com> Signed-off-by:
-
Xiao Yang authored
As ibv_poll_cq()'s manual said, only partial attributes are valid when completion status != IBV_WC_SUCCESS. Link: https://lore.kernel.org/r/20210930094813.226888-4-yangx.jy@fujitsu.comSigned-off-by:
-
Xiao Yang authored
Make the is_user members of struct rxe_qp/rxe_cq has the same type. Link: https://lore.kernel.org/r/20210930094813.226888-3-yangx.jy@fujitsu.comSigned-off-by:
-
Xiao Yang authored
The is_user members of struct rxe_sq/rxe_rq/rxe_srq are unsed since commit ae6e843f ("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.comSigned-off-by:
-
Zhu Yanjun authored
The struct irdma_bth is not used, so remove it. Link: https://lore.kernel.org/r/20211006201531.469650-1-yanjun.zhu@linux.devSigned-off-by:
-
- 05 Oct, 2021 2 commits
-
-
Andy Shevchenko authored
Use string_upper() from the string helper module instead of an open coded variant. Link: https://lore.kernel.org/r/20211001123153.67379-1-andriy.shevchenko@linux.intel.comSigned-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by:
-
Logan Gunthorpe authored
There are a couple of subtle error path bugs related to mapping the sgls: - In rdma_rw_ctx_init(), dma_unmap would be called with an sg that could have been incremented from the original call, as well as an nents that is the dma mapped entries not the original number of nents called when mapped. - Similarly in rdma_rw_ctx_signature_init, both sg and prot_sg were unmapped with the incorrect number of nents. To fix this, switch to the sgtable interface for mapping which conveniently stores the original nents for unmapping. This will get cleaned up further once the dma mapping interface supports P2PDMA and pci_p2pdma_map_sg() can be removed. Fixes: 0e353e34 ("IB/core: add RW API support for signature MRs") Fixes: a060b562 ("IB/core: generic RDMA READ/WRITE API") Link: https://lore.kernel.org/r/20211001213215.3761-1-logang@deltatee.comSigned-off-by:
Logan Gunthorpe <logang@deltatee.com> Signed-off-by:
-
- 04 Oct, 2021 6 commits
-
-
Aharon Landau authored
Currently, if a cache entry is empty, the driver will try to take MRs from larger cache entries. This behavior consumes a lot of memory. In addition, when searching for an mkey in an entry, the entry is locked. When using a multithreaded application with the old behavior, the threads will block each other more often, which can hurt performance as can be seen in the table below. Therefore, avoid it by creating a new mkey when the requested cache entry is empty. The test was performed on a machine with Intel(R) Xeon(R) CPU E5-2699 v4 @ 2.20GHz 44 cores. Here are the time measures for allocating MRs of 2^6 pages. The search in the cache started from entry 6. +------------+---------------------+---------------------+ | | Old behavior | New behavior | | +----------+----------+----------+----------+ | | 1 thread | 5 thread | 1 thread | 5 thread | +============+==========+==========+==========+==========+ | 1,000 MRs | 14 ms | 30 ms | 14 ms | 80 ms | +------------+----------+----------+----------+----------+ | 10,000 MRs | 135 ms | 6 sec | 173 ms | 880 ms | +------------+----------+----------+----------+----------+ |100,000 MRs | 11.2 sec | 57 sec | 1.74 sec | 8.8 sec | +------------+----------+----------+----------+----------+ Link: https://lore.kernel.org/r/71af2770c737b936f7b10f457f0ef303ffcf7ad7.1632644527.git.leonro@nvidia.comSigned-off-by:
-
Md Haris Iqbal authored
This commit divides the sysfs entry cpu_migration into 2 different entries One for "from cpus" and the other for "to cpus". Link: https://lore.kernel.org/r/20210922125333.351454-8-haris.iqbal@ionos.comSigned-off-by:
Md Haris Iqbal <haris.iqbal@ionos.com> Reviewed-by:
Gioh Kim <gi-oh.kim@ionos.com> Reviewed-by:
Aleksei Marov <aleksei.marov@ionos.com> Signed-off-by:
-
Md Haris Iqbal authored
Allowing these characters in sessname can lead to unexpected results, particularly because / is used as a separator between files in a path, and . points to the current directory. Link: https://lore.kernel.org/r/20210922125333.351454-7-haris.iqbal@ionos.comSigned-off-by:
-
Md Haris Iqbal authored
The same code snip used twice, to avoid duplicate, replace it with a destroy_cq helper. Link: https://lore.kernel.org/r/20210922125333.351454-6-haris.iqbal@ionos.comSigned-off-by:
Jack Wang <jinpu.wang@ionos.com> Signed-off-by:
-
Jack Wang authored
if (con->cid >= con->sess->irq_con_num) check can be replaced with a is_pollqueue helper. Link: https://lore.kernel.org/r/20210922125333.351454-5-haris.iqbal@ionos.comSigned-off-by:
-
Jack Wang authored
When testing with poll mode, it will fail and lead to warning below on client side: $ echo "sessname=bla path=gid:fe80::2:c903:4e:d0b3@gid:fe80::2:c903:8:ca17 device_path=/dev/nullb2 nr_poll_queues=-1" | \ sudo tee /sys/devices/virtual/rnbd-client/ctl/map_device rnbd_client L597: Mapping device /dev/nullb2 on session bla, (access_mode: rw, nr_poll_queues: 8) WARNING: CPU: 3 PID: 9886 at drivers/infiniband/core/cq.c:447 ib_cq_pool_get+0x26f/0x2a0 [ib_core] The problem is in case of poll queue, we need to still call ib_alloc_cq/ib_free_cq, we can't use cq_poll api for poll queue. As both client and server use shared function from rtrs, set irq_con_num to con_num on server side, which is number of total connection of the session, this way we can differ if the rtrs_con requires pollqueue. Following up patches will replace the duplicate code with helpers. Link: https://lore.kernel.org/r/20210922125333.351454-4-haris.iqbal@ionos.comSigned-off-by:
Md Haris Iqbal <haris.iqbal@cloud.ionos.com> Signed-off-by:
-
