- 12 Jul, 2019 1 commit
-
-
Jason Gunthorpe authored
The new siw driver fails to build on i386 with drivers/infiniband/sw/siw/siw_qp.c:1025:3: error: invalid output size for constraint '+q' smp_store_mb(*cq->notify, SIW_NOTIFY_NOT); As it is using 64 bit values with the smp_store_mb. Since the entire scheme here seems questionable, and we are in the merge window, fix the compile failures by disabling 32 bit support on this driver. A proper fix will be reviewed post merge window. Fixes: c0cf5bdd ("rdma/siw: addition to kernel build environment") Reported-by:Arnd Bergmann <arnd@arndb.de> Signed-off-by:
Jason Gunthorpe <jgg@mellanox.com>
-
- 11 Jul, 2019 5 commits
-
-
Gustavo A. R. Silva authored
In preparation to enabling -Wimplicit-fallthrough, mark switch cases where we are expecting to fall through. This patch fixes the following warnings: drivers/infiniband/sw/siw/siw_qp_rx.c: In function ‘siw_rdmap_complete’: drivers/infiniband/sw/siw/siw_qp_rx.c:1214:18: warning: this statement may fall through [-Wimplicit-fallthrough=] wqe->rqe.flags |= SIW_WQE_SOLICITED; ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~ drivers/infiniband/sw/siw/siw_qp_rx.c:1215:2: note: here case RDMAP_SEND: ^~~~ drivers/infiniband/sw/siw/siw_qp_tx.c: In function ‘siw_qp_sq_process’: drivers/infiniband/sw/siw/siw_qp_tx.c:1044:4: warning: this statement may fall through [-Wimplicit-fallthrough=] siw_wqe_put_mem(wqe, tx_type); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ drivers/infiniband/sw/siw/siw_qp_tx.c:1045:3: note: here case SIW_OP_INVAL_STAG: ^~~~ drivers/infiniband/sw/siw/siw_qp_tx.c:1128:4: warning: this statement may fall through [-Wimplicit-fallthrough=] siw_wqe_put_mem(wqe, tx_type); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ drivers/infiniband/sw/siw/siw_qp_tx.c:1129:3: note: here case SIW_OP_INVAL_STAG: ^~~~ Warning level 3 was used: -Wimplicit-fallthrough=3 This patch is part of the ongoing efforts to enable -Wimplicit-fallthrough. Signed-off-by:Gustavo A. R. Silva <gustavo@embeddedor.com> Reviewed-by:
Bernard Metzler <bmt@zurich.ibm.com> Signed-off-by:
-
Qian Cai authored
The commit below introduced a few compilation warnings. In file included from ./include/rdma/ib_verbs.h:64, from ./include/linux/mlx5/device.h:37, from ./include/linux/mlx5/driver.h:51, from drivers/net/ethernet/mellanox/mlx5/core/uar.c:36: ./include/linux/dim.h:378:1: warning: 'rdma_dim_prof' defined but not used [-Wunused-const-variable=] rdma_dim_prof[RDMA_DIM_PARAMS_NUM_PROFILES] = { ^~~~~~~~~~~~~ In file included from ./include/rdma/ib_verbs.h:64, from ./include/linux/mlx5/device.h:37, from ./include/linux/mlx5/driver.h:51, from drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c:37: ./include/linux/dim.h:378:1: warning: 'rdma_dim_prof' defined but not used [-Wunused-const-variable=] rdma_dim_prof[RDMA_DIM_PARAMS_NUM_PROFILES] = { ^~~~~~~~~~~~~ Since only ib_cq_rdma_dim_work() in drivers/infiniband/core/cq.c uses it, just move the definition over there. Fixes: f4915455 ("linux/dim: Implement RDMA adaptive moderation (DIM)") Signed-off-by:Qian Cai <cai@lca.pw> Reviewed-by:
-
YueHaibing authored
Fixes gcc '-Wunused-but-set-variable' warning: drivers/infiniband/sw/siw/siw_cm.c: In function siw_cm_llp_state_change: drivers/infiniband/sw/siw/siw_cm.c:1278:17: warning: variable s set but not used [-Wunused-but-set-variable] Fixes: 6c52fdc2 ("rdma/siw: connection management") Reported-by:
Hulk Robot <hulkci@huawei.com> Signed-off-by:
YueHaibing <yuehaibing@huawei.com> Reviewed-by:
-
Geert Uytterhoeven authored
If LIBCRC32C and DMA_VIRT_OPS are not enabled: drivers/infiniband/sw/siw/siw_main.o: In function `siw_newlink': siw_main.c:(.text+0x35c): undefined reference to `dma_virt_ops' drivers/infiniband/sw/siw/siw_qp_rx.o: In function `siw_csum_update': siw_qp_rx.c:(.text+0x16): undefined reference to `crc32c' Fix the first issue by adding a select of DMA_VIRT_OPS. Fix the second issue by replacing the unneeded dependency on CRYPTO_CRC32 by a dependency on LIBCRC32C. Reported-by: noreply@ellerman.id.au (first issue) Fixes: c0cf5bdd ("rdma/siw: addition to kernel build environment") Signed-off-by:Geert Uytterhoeven <geert@linux-m68k.org> Reviewed-by:
-
Jason Gunthorpe authored
ifa is protected by rcu or rtnl, add the missing locking. In this case we have to use rtnl since siw_listen_address() is sleeping. Fixes: 6c52fdc2 ("rdma/siw: connection management") Reviewed-by:
-
- 10 Jul, 2019 3 commits
-
-
Nathan Chancellor authored
clang warns several times: drivers/infiniband/sw/siw/siw_cq.c:31:4: warning: implicit conversion from enumeration type 'enum siw_wc_status' to different enumeration type 'enum siw_opcode' [-Wenum-conversion] { SIW_WC_SUCCESS, IB_WC_SUCCESS }, ~ ^~~~~~~~~~~~~~ Fixes: b0fff731 ("rdma/siw: completion queue methods") Link: https://github.com/ClangBuiltLinux/linux/issues/596Signed-off-by:Nathan Chancellor <natechancellor@gmail.com> Reviewed-by:
-
Bernard Metzler authored
There is already a warning if we cannot start any thread, and stopping those threads is not worth spamming the console. This also corrects a warning from gcc: drivers/infiniband/sw/siw/siw_main.c: In function 'siw_create_tx_threads': drivers/infiniband/sw/siw/siw_main.c:91:11: warning: variable 'rv' set but not used [-Wunused-but-set-variable] Reported-by:
-
Nathan Chancellor authored
clang warns: drivers/infiniband/sw/rdmavt/cq.c:260:7: warning: variable 'err' is used uninitialized whenever 'if' condition is true [-Wsometimes-uninitialized] if (err) ^~~ drivers/infiniband/sw/rdmavt/cq.c:310:9: note: uninitialized use occurs here return err; ^~~ drivers/infiniband/sw/rdmavt/cq.c:260:3: note: remove the 'if' if its condition is always false if (err) ^~~~~~~~ drivers/infiniband/sw/rdmavt/cq.c:253:7: warning: variable 'err' is used uninitialized whenever 'if' condition is true [-Wsometimes-uninitialized] if (!cq->ip) { ^~~~~~~ drivers/infiniband/sw/rdmavt/cq.c:310:9: note: uninitialized use occurs here return err; ^~~ drivers/infiniband/sw/rdmavt/cq.c:253:3: note: remove the 'if' if its condition is always false if (!cq->ip) { ^~~~~~~~~~~~~~ drivers/infiniband/sw/rdmavt/cq.c:211:9: note: initialize the variable 'err' to silence this warning int err; ^ = 0 2 warnings generated. The function scoped err variable is uninitialized when the flow jumps into the if statement. The if scoped err variable shadows the function scoped err variable, preventing the err assignments within the if statement to be reflected at the function level, which will cause uninitialized use when the goto statements are taken. Just remove the if scoped err declaration so that there is only one copy of the err variable for this function. Fixes: 239b0e52 ("IB/hfi1: Move rvt_cq_wc struct into uapi directory") Link: https://github.com/ClangBuiltLinux/linux/issues/594Reviewed-by:Nick Desaulniers <ndesaulniers@google.com> Signed-off-by:
Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by:
-
- 09 Jul, 2019 2 commits
-
-
Dag Moxnes authored
Use the neighbour lock when copying the MAC address from the neighbour data struct in dst_fetch_ha. When not using the lock, it is possible for the function to race with neigh_update(), causing it to copy an torn MAC address: rdma_resolve_addr() rdma_resolve_ip() addr_resolve() addr_resolve_neigh() fetch_ha() dst_fetch_ha() memcpy(dev_addr->dst_dev_addr, n->ha, MAX_ADDR_LEN) and net_ioctl() arp_ioctl() arp_rec_delete() arp_invalidate() neigh_update() __neigh_update() memcpy(&neigh->ha, lladdr, dev->addr_len) It is possible to provoke this error by calling rdma_resolve_addr() in a tight loop, while deleting the corresponding ARP entry in another tight loop. Fixes: 51d45974 ("infiniband: addr: Consolidate code to fetch neighbour hardware address from dst.") Signed-off-by:Dag Moxnes <dag.moxnes@oracle.com> Signed-off-by:
Håkon Bugge <haakon.bugge@oracle.com> Reviewed-by:
-
Jason Gunthorpe authored
5.4-rc1 will have new compile time debugging to test that headers can be compiled stand alone. Many rdma headers are already broken and excluded from the mechanism, however to avoid compile failures during the merge window fix enough so that the newly added header compiles clean. Fixes: 413d3347 ("RDMA/counter: Add set/clear per-port auto mode support") Reported-by:
Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by:
Mark Zhang <markz@mellanox.com>
-
- 08 Jul, 2019 13 commits
-
-
Parav Pandit authored
While creating new RDMA devices based on netdevice name, consider the net namespace of the caller skb's socket similar to rest of the doit() callbacks and nldev_dellink() which deletes the RDMA device created using nldev_newlink(). Fixes: 3856ec4b ("RDMA/core: Add RDMA_NLDEV_CMD_NEWLINK/DELLINK support") Signed-off-by:
Parav Pandit <parav@mellanox.com> Signed-off-by:
Leon Romanovsky <leonro@mellanox.com> Reviewed-by:
-
Konstantin Taranov authored
Calculate the correct byte_len on the receiving side when a work completion is generated with IB_WC_RECV_RDMA_WITH_IMM opcode. According to the IBA byte_len must indicate the number of written bytes, whereas it was always equal to zero for the IB_WC_RECV_RDMA_WITH_IMM opcode, even though data was transferred. Fixes: 8700e3e7 ("Soft RoCE driver") Signed-off-by:
Konstantin Taranov <konstantin.taranov@inf.ethz.ch> Signed-off-by:
-
Leon Romanovsky authored
Enable RDMA DIM by default for better user experience. Signed-off-by:
-
Yamin Friedman authored
Added parameter in ib_device for enabling dynamic interrupt moderation so that it can be configured in userspace using rdma tool. In order to set adaptive-moderation for an ib device the command is: rdma dev set [DEV] adaptive-moderation [on|off] Please set on/off. rdma dev show 0: mlx5_0: node_type ca fw 16.26.0055 node_guid 248a:0703:00a5:29d0 sys_image_guid 248a:0703:00a5:29d0 adaptive-moderation on rdma resource show cq dev mlx5_0 cqn 0 cqe 1023 users 4 poll-ctx UNBOUND_WORKQUEUE adaptive-moderation off comm [ib_core] Signed-off-by:
Yamin Friedman <yaminf@mellanox.com> Reviewed-by:
Sagi Grimberg <sagi@grimberg.me> Signed-off-by:
-
Yamin Friedman authored
Added the interface in the infiniband driver that applies the rdma_dim adaptive moderation. There is now a special function for allocating an ib_cq that uses rdma_dim. Performance improvement (ConnectX-5 100GbE, x86) running FIO benchmark over NVMf between two equal end-hosts with 56 cores across a Mellanox switch using null_blk device: READS without DIM: blk size | BW | IOPS | 99th percentile latency | 99.99th latency 512B | 3.8GiB/s | 7.7M | 1401 usec | 2442 usec 4k | 7.0GiB/s | 1.8M | 4817 usec | 6587 usec 64k | 10.7GiB/s| 175k | 9896 usec | 10028 usec IO WRITES without DIM: blk size | BW | IOPS | 99th percentile latency | 99.99th latency 512B | 3.6GiB/s | 7.5M | 1434 usec | 2474 usec 4k | 6.3GiB/s | 1.6M | 938 usec | 1221 usec 64k | 10.7GiB/s| 175k | 8979 usec | 12780 usec IO READS with DIM: blk size | BW | IOPS | 99th percentile latency | 99.99th latency 512B | 4GiB/s | 8.2M | 816 usec | 889 usec 4k | 10.1GiB/s| 2.65M| 3359 usec | 5080 usec 64k | 10.7GiB/s| 175k | 9896 usec | 10028 usec IO WRITES with DIM: blk size | BW | IOPS | 99th percentile latency | 99.99th latency 512B | 3.9GiB/s | 8.1M | 799 usec | 922 usec 4k | 9.6GiB/s | 2.5M | 717 usec | 1004 usec 64k | 10.7GiB/s| 176k | 8586 usec | 12256 usec The rdma_dim algorithm was designed to measure the effectiveness of moderation on the flow in a general way and thus should be appropriate for all RDMA storage protocols. rdma_dim is configured to be the default option based on performance improvement seen after extensive tests. Signed-off-by:
Max Gurtovoy <maxg@mellanox.com> Reviewed-by:
Saeed Mahameed <saeedm@mellanox.com> Signed-off-by:
-
Yamin Friedman authored
RDMA DIM implements a different algorithm from net DIM and is based on completions which is how we can implement interrupt moderation in RDMA. The algorithm optimizes for number of completions and ratio between completions and events. In order to avoid long latencies, the implementation performs fast reduction of moderation level when the traffic changes. Signed-off-by:
-
Jason Gunthorpe authored
Generic DIM From: Tal Gilboa and Yamin Fridman Implement net DIM over a generic DIM library, add RDMA DIM dim.h lib exposes an implementation of the DIM algorithm for dynamically-tuned interrupt moderation for networking interfaces. We want a similar functionality for other protocols, which might need to optimize interrupts differently. Main motivation here is DIM for NVMf storage protocol. Current DIM implementation prioritizes reducing interrupt overhead over latency. Also, in order to reduce DIM's own overhead, the algorithm might take some time to identify it needs to change profiles. While this is acceptable for networking, it might not work well on other scenarios. Here we propose a new structure to DIM. The idea is to allow a slightly modified functionality without the risk of breaking Net DIM behavior for netdev. We verified there are no degradations in current DIM behavior with the modified solution. Suggested solution: - Common logic is implemented in lib/dim/dim.c - Net DIM (existing) logic is implemented in lib/dim/net_dim.c, which uses the common logic in dim.c - Any new DIM logic will be implemented in "lib/dim/new_dim.c". This new implementation will expose modified versions of profiles, dim_step() and dim_decision(). - DIM API is declared in include/linux/dim.h for all implementations. Pros for this solution are: - Zero impact on existing net_dim implementation and usage - Relatively more code reuse (compared to two separate solutions) - Increased extensibility Required for dependencies in the next series. Signed-off-by: -
Danit Goldberg authored
Userspace expects the IB_TM_CAP_RC bit to indicate that the device supports RC transport tag matching with rendezvous offload. However the firmware splits this into two capabilities for eager and rendezvous tag matching. Only if the FW supports both modes should userspace be told the tag matching capability is available. Cc: <stable@vger.kernel.org> # 4.13 Fixes: eb761894 ("IB/mlx5: Fill XRQ capabilities") Signed-off-by:
Danit Goldberg <danitg@mellanox.com> Reviewed-by:
Yishai Hadas <yishaih@mellanox.com> Reviewed-by:
Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by:
-
Mauro Carvalho Chehab authored
While this contains some uAPI stuff, it was intended to be read by a kernel doc. So, let's not move it to a different dir, but, instead, just add it to the driver-api bookset. Signed-off-by:
Mauro Carvalho Chehab <mchehab+samsung@kernel.org> Signed-off-by:
-
Jason Gunthorpe authored
Max Gurtovoy says: ==================== Those two patches introduce VHCA tunnel mechanism to DEVX interface needed for Bluefield SOC. See extensive commit messages for more information. ==================== Based on the mlx5-next branch from git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux for dependencies * branch 'vcha-tunnel': IB/mlx5: Implement VHCA tunnel mechanism in DEVX net/mlx5: Introduce VHCA tunnel device capability
-
Max Gurtovoy authored
This mechanism will allow function-A to perform operations "on behalf" of function-B via tunnel object. Function-A will have privileges for creating and using this tunnel object. For example, in the device emulation feature presented in Bluefield-1 SoC, using device emulation capability, one can present NVMe function to the host OS. Since the NVMe function doesn't have a normal command interface to the HCA HW, here is a need to create a channel that will be able to issue commands "on behalf" of this function. This channel is the VHCA_TUNNEL general object. The emulation software will create this tunnel for every managed function and issue commands via devx general cmd interface using the appropriate tunnel ID. When devX context will receive a command with non-zero vhca_tunnel_id, it will pass the command as-is down to the HCA. All the validation, security and resource tracking of the commands and the created tunneled objects is in the responsibility of the HCA FW. When a VHCA_TUNNEL object destroyed, the device will issue an internal FLR (function level reset) to the emulated function associated with this tunnel. This will destroy all the created resources using the tunnel mechanism. Signed-off-by:
-
Jason Gunthorpe authored
rvt was using ib_sge as part of it's ABI, which is not allowed. Introduce a new struct with the same layout and use it instead. Fixes: dabac6e4 ("IB/hfi1: Move receive work queue struct into uapi directory") Reported-by:
-
Jason Gunthorpe authored
The initializer for the variable cannot be inside the macro (and zero initialization isn't needed anyhow). include/linux/percpu-defs.h:92:33: warning: '__pcpu_unique_use_cnt' initialized and declared 'extern' extern __PCPU_DUMMY_ATTRS char __pcpu_unique_##name; \ ^~~~~~~~~~~~~~ include/linux/percpu-defs.h:115:2: note: in expansion of macro 'DEFINE_PER_CPU_SECTION' DEFINE_PER_CPU_SECTION(type, name, "") ^~~~~~~~~~~~~~~~~~~~~~ drivers/infiniband/sw/siw/siw_main.c:129:8: note: in expansion of macro 'DEFINE_PER_CPU' static DEFINE_PER_CPU(atomic_t, use_cnt = ATOMIC_INIT(0)); ^~~~~~~~~~~~~~ Also the rules for PER_CPU require the variable names to be globally unique, so prefix them with siw_ Fixes: b9be6f18 ("rdma/siw: transmit path") Fixes: bdcf26bf ("rdma/siw: network and RDMA core interface") Reported-by:
-
- 07 Jul, 2019 6 commits
-
-
Maksym Planeta authored
In some cases (not in this particular one) variable self-initialization can lead to undefined behavior. In this case, it is just obscure code. Signed-off-by:
Maksym Planeta <mplaneta@os.inf.tu-dresden.de> Signed-off-by:
-
Lang Cheng authored
Here Clean up unnecessary initial value for some variable. Signed-off-by:
Lang Cheng <chenglang@huawei.com> Signed-off-by:
-
Xi Wang authored
When smmu is enable, if execute the perftest command and then use 'kill -9' to exit, follow this operation repeatedly, the kernel will have a high probability to print the following smmu event: arm-smmu-v3 arm-smmu-v3.1.auto: event 0x10 received: arm-smmu-v3 arm-smmu-v3.1.auto: 0x00007d0000000010 arm-smmu-v3 arm-smmu-v3.1.auto: 0x0000020900000080 arm-smmu-v3 arm-smmu-v3.1.auto: 0x00000000f47cf000 arm-smmu-v3 arm-smmu-v3.1.auto: 0x00000000f47cf000 This is because the hw will periodically refresh the qpc cache until the next reset. This patch fixed it by removing the action that release qpc memory in the 'hns_roce_qp_free' function. Fixes: 9a443537 ("IB/hns: Add driver files for hns RoCE driver") Signed-off-by:
Xi Wang <wangxi11@huawei.com> Signed-off-by:
-
Lang Cheng authored
The format specifier \"%p\" can leak kernel addresses. Use \"%pK\" instead. Signed-off-by:
-
Lijun Ou authored
The buffer size of qp which used to allocate qp buffer space for storing sqwqe and rqwqe will be the length of buffer space. The kernel driver will use the buffer address and the same size to get the user memory. The same size named buff_size of qp. According the algorithm of calculating, The size of the two is not equal when users set the max sge of sq. Fixes: b28ca7cc ("RDMA/hns: Limit extend sq sge num") Signed-off-by:
Lijun Ou <oulijun@huawei.com> Signed-off-by:
-
Max Gurtovoy authored
When using the device emulation feature (introduced in Bluefield-1 SOC), a privileged function (the device emulation manager) will be able to create a channel to execute commands on behalf of the emulated function. This channel will be a general object of type VHCA_TUNNEL that will have a unique ID for each emulated function. This ID will be passed in each cmd that will be issued by the emulation SW in a well known offset in the command header. This channel is needed since the emulated function doesn't have a normal command interface to the HCA HW, but some basic configuration for that function is needed (e.g. initialize and enable the HCA). For that matter, a specific command-set was defined and only those commands will be issued by the HCA. Signed-off-by:
-
- 05 Jul, 2019 10 commits
-
-
Lang Cheng authored
When hw resetting, there is no response from hw when driver sending cmdq. If driver still send cmdq to hw, the reset process may be blocked. So reset flag should be set to intercept the cmdq command when driver receiving "notify down" signal. Signed-off-by:
-
Yangyang Li authored
Currently, the depth of cq only supports 64K. According to the UM, the depth of cq is up to 4M, Therefore the ba page size of cqe was modified to support the maximum specification of cq depth. Signed-off-by:
Yangyang Li <liyangyang20@huawei.com> Signed-off-by:
-
chenglang authored
Hip06 reserve 12 qps, Hip08 reserve 8 qps. When the QP is released, the chip model is not judged, and the Hip08 cannot release the qpn 8~12 Signed-off-by:
-
Lijun Ou authored
It uses hns_roce_mtr_init in hns_roce_create_qp_common function. As a result, it should use hns_roce_mtr_cleanup function for cleaning mtr when destroying qp. Fixes: 8d18ad83 ("RDMA/hns: Fix bug when wqe num is larger than 16K") Signed-off-by:
-
Mark Zhang authored
This patch adds the ability to return the hwstats of per-port default counters (which can also be queried through sysfs nodes). Signed-off-by:
-
Mark Zhang authored
Provide an option to get current counter mode through RDMA netlink. Signed-off-by:
Majd Dibbiny <majd@mellanox.com> Signed-off-by:
-
Mark Zhang authored
Provide an option to allow users to manually bind a qp with a counter through RDMA netlink. Limit it to users with ADMIN capability only. Signed-off-by:
-
Mark Zhang authored
In manual mode a QP is bound to a counter manually. If counter is not specified then a new one will be allocated. Manual mode is enabled when user binds a QP, and disabled when the last manually bound QP is unbound. When auto-mode is turned off and there are counters left, manual mode is enabled so that the user is able to access these counters. Signed-off-by:
-
Mark Zhang authored
Since a QP can only be bound to one counter, then if it is bound to a separate counter, for backward compatibility purpose, the statistic value must be: * stat of default counter + stat of all running allocated counters + stat of all deallocated counters (history stats) Signed-off-by:
-
Mark Zhang authored
Add support for ib callback counter_alloc_stats() and counter_update_stats(). Signed-off-by:
-
