Undefined This issue is carved out of https://issues.redhat.com/browse/RHEL-30146?focusedId=25700259&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-25700259
What were you trying to do that didn't work?
When running on an emulated ConnectX-8 device (800G):
$ ibv_devinfo -v -d mlx5_0 hca_id: mlx5_0 transport: InfiniBand (0) fw_ver: 24.10.1037 node_guid: 2442:53c0:fe21:5243 sys_image_guid: 2442:53c1:fe21:5243 vendor_id: 0x02c9 vendor_part_id: 4131 hw_ver: 0x0 board_id: MT_QEMU000000 phys_port_cnt: 1 max_mr_size: 0xffffffffffffffff page_size_cap: 0xfffffffffffff000 max_qp: 262144 max_qp_wr: 32768 device_cap_flags: 0x21361c36 BAD_PKEY_CNTR BAD_QKEY_CNTR AUTO_PATH_MIG CHANGE_PHY_PORT PORT_ACTIVE_EVENT SYS_IMAGE_GUID RC_RNR_NAK_GEN MEM_WINDOW UD_IP_CSUM XRC MEM_MGT_EXTENSIONS MEM_WINDOW_TYPE_2B MANAGED_FLOW_STEERING max_sge: 30 max_sge_rd: 30 max_cq: 16777216 max_cqe: 4194303 max_mr: 16777216 max_pd: 16777216 max_qp_rd_atom: 16 max_ee_rd_atom: 0 max_res_rd_atom: 4194304 max_qp_init_rd_atom: 16 max_ee_init_rd_atom: 0 atomic_cap: ATOMIC_HCA (1) max_ee: 0 max_rdd: 0 max_mw: 16777216 max_raw_ipv6_qp: 0 max_raw_ethy_qp: 0 max_mcast_grp: 2097152 max_mcast_qp_attach: 48 max_total_mcast_qp_attach: 100663296 max_ah: 2147483647 max_fmr: 0 max_srq: 8388608 max_srq_wr: 32767 max_srq_sge: 31 max_pkeys: 128 local_ca_ack_delay: 16 general_odp_caps: ODP_SUPPORT ODP_SUPPORT_IMPLICIT rc_odp_caps: SUPPORT_SEND SUPPORT_RECV SUPPORT_WRITE SUPPORT_READ SUPPORT_SRQ uc_odp_caps: NO SUPPORT ud_odp_caps: NO SUPPORT xrc_odp_caps: NO SUPPORT completion timestamp_mask: 0x7fffffffffffffff hca_core_clock: 78125kHZ device_cap_flags_ex: 0x1021361C36 PCI_WRITE_END_PADDING tso_caps: max_tso: 0 rss_caps: max_rwq_indirection_tables: 0 max_rwq_indirection_table_size: 0 rx_hash_function: 0x0 rx_hash_fields_mask: 0x0 max_wq_type_rq: 0 packet_pacing_caps: qp_rate_limit_min: 0kbps qp_rate_limit_max: 0kbps tag matching not supported cq moderation caps: max_cq_count: 65535 max_cq_period: 4095 us maximum available device memory: 131072Bytes num_comp_vectors: 4 port: 1 state: PORT_ACTIVE (4) max_mtu: 4096 (5) active_mtu: 4096 (5) sm_lid: 49151 port_lid: 21059 port_lmc: 0x00 link_layer: InfiniBand max_msg_sz: 0x40000000 port_cap_flags: 0x8209c868 port_cap_flags2: 0x1d63 max_vl_num: 4 (3) bad_pkey_cntr: 0x0 qkey_viol_cntr: 0x0 sm_sl: 0 pkey_tbl_len: 128 gid_tbl_len: 8 subnet_timeout: 0 init_type_reply: 0 active_width: 4X (2) active_speed: 100.0 Gbps (128) phys_state: LINK_UP (5) GID[ 0]: fe80:0000:0000:0000:2442:53c0:fe21:5243
Notice that the output of ibv_devinfo reads "active_speed: 100.0 Gbps (128)". I think it should be 200G.
What is the impact of this issue to you?
normal
Please provide the package NVR for which the bug is seen:
$ dnf info rdma-core Updating Subscription Management repositories. Unable to read consumer identity This system is not registered with an entitlement server. You can use "rhc" or "subscription-manager" to register. Last metadata expiration check: 2:27:25 ago on Tue 20 Aug 2024 03:20:20 PM IDT. Installed Packages Name : rdma-core Version : 51.0 Release : 1.el9 Architecture : x86_64 Size : 118 k Source : rdma-core-51.0-1.el9.src.rpm Repository : @System From repo : 1-BaseOS Summary : RDMA core userspace libraries and daemons URL : https://github.com/linux-rdma/rdma-core License : GPLv2 or BSD Description : RDMA core userspace infrastructure and documentation, including initialization : scripts, kernel driver-specific modprobe override configs, IPoIB network : scripts, dracut rules, and the rdma-ndd utility.
How reproducible is this bug?:
Every time
Steps to reproduce
- `ibv_devinfo -v -d mlx5_0`
Expected results
As mentioned in https://issues.redhat.com/browse/RHEL-30146?focusedId=25391263&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-25391263, when using the upstream rdma-core, the result is
active_speed: 200.0 Gbps (256)
Actual results
active_speed: 100.0 Gbps (128)