irdma_close_...objects_type(9)
π irdma_close_...objects_type(9)
okay so this is a page from the Kernel API Manual, section 9 β yes bestie, the big-kid kernel section, we're literally in the driver guts rn π«Ά
β¨ NAME β¨
irdma_close_hmc_objects_type β deletes hmc objects of a given type. that's it, that's the whole vibe. cleanup era π
πΈ SYNOPSIS πΈ
void irdma_close_hmc_objects_type (struct irdma_sc_dev *dev , enum
irdma_hmc_rsrc_type obj_type , struct irdma_hmc_info *hmc_info , bool
privileged , bool reset );
no return value, it's a void β she does the work and leaves, no notes π
π ARGUMENTS π
devβ iwarp device. the main character, the thing we're operating on π¦obj_typeβ the hmc object type to be deleted. like, which flavor of object is getting the β¨ goodbye treatment β¨hmc_infoβ host memory info struct. all the host-memory lore lives in here πprivilegedβ permission to close HMC objects. no permission? no closing. simple as that πresetβtrueif called before reset. not during, not after β before. clock it β
π₯ SEE ALSO π₯
first up, the source file so you can go read the actual receipts:
Kernel file ./drivers/infiniband/hw/irdma/hw.c
and now the extended friend group β every single one of these, no cap, i'm not skipping a single mutual π«Ά
irdma_iwarp_ce_handler(9), irdma_puda_ce_handler(9), irdma_process_normal_ceqe(9), irdma_process_reserved_ceqe(9), irdma_process_ceq(9), irdma_complete_cqp_request(9), irdma_process_ae_def_cmpl(9), irdma_process_aeq(9), irdma_ena_intr(9), irdma_dpc(9), irdma_ceq_dpc(9), irdma_save_msix_info(9), irdma_irq_handler(9), irdma_ceq_handler(9), irdma_destroy_irq(9), irdma_destroy_cqp(9), irdma_destroy_aeq(9), irdma_destroy_ceq(9), irdma_del_ceq_0(9), irdma_del_ceqs(9), irdma_destroy_ccq(9), irdma_del_hmc_objects(9), irdma_create_hmc_obj_type(9), irdma_create_hmc_objs(9), irdma_obj_aligned_mem(9), irdma_create_cqp(9), irdma_create_ccq(9), irdma_alloc_set_mac(9), irdma_cfg_ceq_vector(9), irdma_cfg_aeq_vector(9), irdma_create_ceq(9), irdma_setup_ceq_0(9), irdma_setup_ceqs(9), irdma_create_aeq(9), irdma_setup_aeq(9), irdma_initialize_ilq(9), irdma_initialize_ieq(9), irdma_reinitialize_ieq(9), irdma_hmc_setup(9), irdma_del_init_mem(9), irdma_initialize_dev(9), irdma_rt_deinit_hw(9), irdma_get_used_rsrc(9), irdma_rt_init_hw(9), irdma_ctrl_init_hw(9), irdma_set_hw_rsrc(9), irdma_calc_mem_rsrc_size(9), irdma_initialize_hw_rsrc(9), irdma_cqp_ce_handler(9), cqp_compl_worker(9), irdma_lookup_apbvt_entry(9), irdma_next_iw_state(9), irdma_del_local_mac_entry(9), irdma_add_local_mac_entry(9), irdma_alloc_local_mac_entry(9), irdma_cqp_manage_apbvt_cmd(9), irdma_add_apbvt(9), irdma_del_apbvt(9), irdma_manage_arp_cache(9), irdma_send_syn_cqp_callback(9), irdma_manage_qhash(9), irdma_hw_flush_wqes_callback(9), irdma_hw_flush_wqes(9), irdma_gen_ae(9)
ngl bestie, this is kernel HMC cleanup territory β you're deleting objects that the hardware expects to exist. know what you're doing before you call her, iykyk π
okay that's the whole page, love u, go touch some HMC objects responsibly π§π
and for the β¨creditsβ¨: this one lives under ./drivers/infiniband/hw/irdma, timestamped October 2026, and it's irdma_close_...objects_type(9) β iconic, no notes, periodt ππ
π¦ where this page comes from
- nixpkgs unstable: linux-manual 7.3.1
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms