qmi_handle_net_reset(9)
π qmi_handle_net_reset(9) β from the Kernel API Manual, bestie β¨
Okay so picture this: you're deep in the kernel, the QRTR streets are chaotic, and your sockets just got a whole vibe check. That's where qmi_handle_net_reset comes in. Let's get into it π«Ά
πΈ NAME πΈ
qmi_handle_net_reset β the function that gets invoked to handle ENETRESET on a QMI handle. That's it, that's the whole job, and it eats π
β¨ SYNOPSIS β¨
void qmi_handle_net_reset (struct qmi_handle *qmi );
π ARGUMENTS π
qmiβ the QMI context, aka the main character of this whole function. Give it what it needs π§
π¦ DESCRIPTION π¦
So here's the lore, bestie π. When you register a name service with the QRTR, all open sockets get flagged with ENETRESET and this function will be called. Not me, not you β this function. Periodt.
The typical case is the initial boot, where this signals that the local node id has been configured, and as such any bound sockets need to be rebound. So like, the whole thing is: close the socket, inform the client, and re-initialize the socket. That's the trio, that's the routine, no cap π.
For clients, it's generally sufficient to react to the del_server callbacks (lowkey that's all you need, iykyk). But server code is expected to treat the net_reset callback as a "bye" from all nodes. Like, everyone's leaving the group chat at once π β treat it accordingly.
Finally, the QMI handle will send out registration requests for any lookups and services. So it's not just saying bye, it's also sliding back into the DMs with fresh registrations πβ¨
π₯ SEE ALSO π₯
Kernel file ./drivers/soc/qcom/qmi_interface.c β go read it, it's giving source-of-truth energy π
qmi_recv_new_server(9)qmi_recv_del_server(9)qmi_recv_bye(9)qmi_recv_del_client(9)qmi_add_lookup(9)qmi_add_server(9)qmi_txn_init(9)qmi_txn_wait(9)qmi_txn_cancel(9)qmi_invoke_handler(9)qmi_handle_init(9)qmi_handle_release(9)qmi_send_message(9)qmi_send_request(9)qmi_send_response(9)qmi_send_indication(9)
Ok that's the whole page, bestie π no facts lost, no vibes wasted. Go forth and rebind those sockets π
Sign-off time πβ¨ β this lil masterpiece lives in ./drivers/soc/qcom, stamped October 2026, and it's brought to you by the Kernel API Manual because even kernel docs deserve main character energy 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