ibmvscsi_set_request_limit(9)
π ibmvscsi_set_request_limit π
okay so bestie, welcome to the ibmvscsi_set_request_limit(9) page from the Kernel API Manual β¨ β this is the little function that sets the adapter's request_limit when the adapter's having a Moment (failure, reset, or SRP Login) π«Ά
β¨ NAME β¨
ibmvscsi_set_request_limit β sets the adapter request_limit in response to an adapter failure, a reset, or an SRP Login π and the whole thing runs under host lock, specifically so it doesn't race with SCSI command submission. no cap, that lock is the main character here π
π¦ SYNOPSIS π¦
void ibmvscsi_set_request_limit (struct ibmvscsi_host_data *hostdata ,
int limit );
it returns void btw β nothing comes back, it just does the thing π
π ARGUMENTS π
hostdataβ the adapter you're adjusting, like, the whole point π§limitβ the brand new request limit, the fresh number, the glow-up β
π SEE ALSO π
the kernel file lives at ./drivers/scsi/ibmvscsi/ibmvscsi.c (iykyk) and honestly the friend group here is huge, so buckle up buttercup, here's everyone π:
ibmvscsi_handle_event(9)ibmvscsi_release_crq_queue(9)crq_queue_next_crq(9)ibmvscsi_send_crq(9)ibmvscsi_task(9)ibmvscsi_reset_crq_queue(9)ibmvscsi_init_crq_queue(9)ibmvscsi_reenable_crq_queue(9)initialize_event_pool(9)release_event_pool(9)valid_event_struct(9)free_event_struct(9)get_event_struct(9)init_event_struct(9)unmap_cmd_data(9)map_sg_data(9)map_data_for_srp_cmd(9)purge_requests(9)ibmvscsi_reset_host(9)ibmvscsi_timeout(9)ibmvscsi_send_srp_event(9)handle_cmd_rsp(9)lun_from_dev(9)ibmvscsi_queuecommand_lck(9)map_persist_bufs(9)unmap_persist_bufs(9)login_rsp(9)send_srp_login(9)capabilities_rsp(9)send_mad_capabilities(9)fast_fail_rsp(9)enable_fast_fail(9)adapter_info_rsp(9)send_mad_adapter_info(9)ibmvscsi_eh_host_reset_handler(9)ibmvscsi_handle_crq(9)ibmvscsi_sdev_configure(9)ibmvscsi_change_queue_depth(9)ibmvscsi_get_desired_dma(9)ibmvscsi_resume(9)
like, go read all of them, they're literally all connected, it's giving one big happy driver family ππ
lowkey reminder: this thing touches request_limit and it does it under host lock β that's not decoration, that's the whole reason it's safe against SCSI command submission racing in. respect the lock π
and that's the page!! sourced from ./drivers/scsi/ibmvscsi πΈ dated October 2026 π mwah, slay, go forth and set those request limits responsibly β¨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