微知-ib_write_bw的各种参数汇总(-d -q -s -R --run_infinitely)

背景

经常忘记使用ib_write_bw打流的一些参数,特此整理记录在这里方便快速查阅。尤其是run_infinitely这个参数容易写错。

最简洁

bash 复制代码
ib_write_bw -d mlx5_0 # server
ib_write_bw -d mlx5_0 1.1.1.1 # client

常用参数

非常常用

  • -d mlx5_0, --ib-dev= 指定ib设备,比如: -d mlx5_0表示用mlx5_0设备
  • -q 8 指定qp数量,比如8个qp, --qp=<num of qp's> Num of qp's(default 1)。服务器和client都需要指定
  • -R 使用rdma_cm建链, --rdma_cm 。服务器和client都需要指定
  • --run_infinitely 一直持续的运行,每间隔-D的参数秒打印
  • -s 1024指定size大小, --size=

一般常用

  • -D 10, --duration 指定打流时间 比如-D 10 指定10秒
  • -a 默认只会用65535的msgsize,这里会使用从2到2^23(8M)大小的size, --all
  • -c RC指定类型,默认RC。, --connection=<RC/XRC/UC/DC>
  • -i 1 指定IB的port。, --ib-port= Use port of IB device (default 1)
  • -m 4096 指定mtu, --mtu= MTU size : 256 - 4096 (default port mtu)
  • -p 18516 Listen on/connect to port (default 18515) 指定建链监听端口, --port=
  • -u 14 指定qp超时时间,默认, --qp-timeout= QP timeout, timeout value is 4 usec * 2 ^(timeout), default 14 ,约等于 65ms
  • --report_gbits使用Gbps的方式报告
  • --rate_limit=<rate> 设置最大限速Set the maximum rate of sent packages. default unit is [Gbps]. use --rate_units to change that.

其他 全量help信息

bash 复制代码
root@localhost:~# ib_write_bw --help
Usage:
  ib_write_bw            start a server and wait for connection
  ib_write_bw <host>     connect to server at <host>

Options:
  -a, --all  Run sizes from 2 till 2^23
  -b, --bidirectional  Measure bidirectional bandwidth (default unidirectional)
  -c, --connection=<RC/XRC/UC/DC>  Connection type RC/XRC/UC/DC (default RC)
      --log_dci_streams=<log_num_dci_stream_channels> (default 0)  Run DC initiator as DCS instead of DCI with <log_num dci_stream_channels>
      --log_active_dci_streams=<log_num_active_dci_stream_channels> (default log_num_dci_stream_channels)
          --aes_xts Runs traffic with AES_XTS feature (encryption)
          --encrypt_on_tx Runs traffic with encryption on tx (default decryption on tx)
          --sig_before Puts signature on data before encrypting it (default after)
      --aes_block_size=<512,520,4048,4096,4160> (default 512)
          --data_enc_keys_number=<number of data encryption keys> (default 1)
          --kek_path path to the key encryption key file
          --credentials_path path to the credentials file
          --data_enc_key_app_path path to the data encryption key app
  -d, --ib-dev=<dev>  Use IB device <dev> (default first device found)
  -D, --duration  Run test for a customized period of seconds.
  -f, --margin  measure results within margins. (default=2sec)
  -F, --CPU-freq  Do not show a warning even if cpufreq_ondemand module is loaded, and cpu-freq is not on max.
  -h, --help  Show this help screen.
  -i, --ib-port=<port>  Use port <port> of IB device (default 1)
  -I, --inline_size=<size>  Max size of message to be sent in inline
  -l, --post_list=<list size>
 Post list of send WQEs of <list size> size (instead of single post)
      --recv_post_list=<list size> Post list of receive WQEs of <list size> size (instead of single post)
  -L, --hop_limit=<hop_limit>  Set hop limit value (ttl for IPv4 RawEth QP). Values 0-255 (default 64)
  -m, --mtu=<mtu>  MTU size : 256 - 4096 (default port mtu)
  -n, --iters=<iters>  Number of exchanges (at least 5, default 5000)
  -N, --noPeak Cancel peak-bw calculation (default with peak up to iters=20000)
  -O, --dualport  Run test in dual-port mode.
  -p, --port=<port>  Listen on/connect to port <port> (default 18515)
  -q, --qp=<num of qp's>  Num of qp's(default 1)
  -Q, --cq-mod  Generate Cqe only after <--cq-mod> completion
  -R, --rdma_cm  Connect QPs with rdma_cm and run test on those QPs
  -s, --size=<size>  Size of message to exchange (default 65536)
  -S, --sl=<sl>  SL (default 0)
  -t, --tx-depth=<dep>  Size of tx queue (default 128)
  -T, --tos=<tos value>  Set <tos_value> to RDMA-CM QPs. available only with -R flag. values 0-256 (default off)
  -u, --qp-timeout=<timeout>  QP timeout, timeout value is 4 usec * 2 ^(timeout), default 14
  -V, --version  Display version number
  -w, --limit_bw=<value>  Set verifier limit for bandwidth
  -W, --report-counters=<list of counter names>  Report performance counter change (example: "counters/port_xmit_data,hw_counters/out_of_buffer")
  -x, --gid-index=<index>  Test uses GID with GID index
  -y, --limit_msgrate=<value>  Set verifier limit for Msg Rate
  -z, --comm_rdma_cm  Communicate with rdma_cm module to exchange data - use regular QPs

      --out_json  Save the report in a json file
      --out_json_file=<file>  Name of the report json file. (Default: perftest_out.json in the working directory) 
      --cpu_util  Show CPU Utilization in report, valid only in Duration mode 
      --dlid  Set a Destination LID instead of getting it from the other side.
      --dont_xchg_versions  Do not exchange versions and MTU with other side 
      --force-link=<value>  Force the link(s) to a specific type: IB or Ethernet.
      --ipv6  Use IPv6 GID. Default is IPv4
      --ipv6-addr=<IPv6>  Use IPv6 address for parameters negotiation. Default is IPv4
      --bind_source_ip  Source IP of the interface used for connection establishment. By default taken from routing table.
      --mmap=file  Use an mmap'd file as the buffer for testing P2P transfers.
      --mmap-offset=<offset>  Use an mmap'd file as the buffer for testing P2P transfers.
      --mr_per_qp  Create memory region for each qp.
      --odp  Use On Demand Paging instead of Memory Registration.
      --output=<units> Set verbosity output level: bandwidth , message_rate, latency 
      --payload_file_path=<payload_txt_file_path> Set the payload by passing a txt file containing a pattern in the next form(little endian): '0xaaaaaaaa, 0xbbbbbbbb, ...' .
 Latency measurement is Average calculation 
      --use_old_post_send Use old post send flow (ibv_post_send).
      --perform_warm_up Perform some iterations before start measuring in order to warming-up memory cache, valid in Atomic, Read and Write BW tests
      --pkey_index=<pkey index> PKey index to use for QP
      --report-both  Report RX & TX results separately on Bidirectional BW tests
      --report_gbits  Report Max/Average BW of test in Gbit/sec (instead of MiB/sec)
        Note: MiB=2^20 byte, while Gb=10^9 bits. Use these formulas for conversion:
        Factor=10^9/(2^20*8)=119.2; MiB=Gb_result * factor; Gb=MiB_result / factor
      --report-per-port  Report BW data on both ports when running Dualport and Duration mode
      --reversed  Reverse traffic direction - Server send to client
      --run_infinitely  Run test forever, print results every <duration> seconds
      --retry_count=<value>  Set retry count value in rdma_cm mode
      --tclass=<value>  Set the Traffic Class in GRH (if GRH is in use)
      --flow_label=<value>  Set the flow_label in GRH (if GRH is in use)
      --use_hugepages  Use Hugepages instead of contig, memalign allocations.
      --use-null-mr  Allocate a null memory region for the client with ibv_alloc_null_mr.
      --wait_destroy=<seconds>  Wait <seconds> before destroying allocated resources (QP/CQ/PD/MR..)
      --disable_pcie_relaxed Disable PCIe relaxed ordering

 Rate Limiter:
      --burst_size=<size> Set the amount of messages to send in a burst when using rate limiter
      --typical_pkt_size=<bytes> Set the size of packet to send in a burst. Only supports PP rate limiter
      --rate_limit=<rate> Set the maximum rate of sent packages. default unit is [Gbps]. use --rate_units to change that.
      --rate_units=<units> [Mgp] Set the units for rate limit to MiBps (M), Gbps (g) or pps (p). default is Gbps (g).
        Note (1): pps not supported with HW limit.
        Note (2): When using PP rate_units is forced to Kbps.
      --rate_limit_type=<type> [HW/SW/PP] Limit the QP's by HW, PP or by SW. Disabled by default. When rate_limit is not specified HW limit is Default.
        Note: in Latency under load test SW rate limit is forced
      --write_with_imm  use write-with-immediate verb instead of write
相关推荐
北冥有鱼被烹1 个月前
NVIDIA Bluefield DPU上的启动流程4个阶段分别是什么?作用是什么?
dpu·mellanox
北冥有鱼被烹1 个月前
微知-Bluefield DPU使用flint烧录固件报错MFE_NO_FLASH_DETECTED是什么?MFE是什么?
dpu·mellanox·bluefield
KIDGINBROOK3 个月前
RDMA驱动学习(一)- 用户态到内核态的过程
rdma
羌俊恩3 个月前
Linux 常见的冷知识集锦
linux·rdma·posix
我想学LINUX4 个月前
【RDMA项目】如何使用rdma-core进行调用开发一个实战项目
服务器·网络·github·客户端·rdma·rdma-core·内存访问
我想学LINUX4 个月前
RDMA 高性能架构基本原理与设计方案
linux·架构·嵌入式·rdma·系统开发·rdma设计思路·rdma基本元素架构
北冥有鱼被烹5 个月前
RDMA建链的3次握手和断链的4次挥手流程?
rdma·rdma_cm
妙BOOK言5 个月前
Rcmp: Reconstructing RDMA-Based Memory Disaggregation via CXL——论文阅读
论文阅读·cxl·rdma·内存分解
Pretend ^^6 个月前
3. 使用tcpdump抓取rdma数据包
网络·wireshark·tcpdump·rdma