_scsih_block...tached_to_ex(9)

📦 version 7.3.1, in nixpkgs unstable

💅 _scsih_block...tached_to_ex(9) 💅

okay bestie, welcome to the Kernel API Manual page for _scsih_block...tached_to_ex(9) — yes the name is long, yes it's the main character, and yes we're gonna get through it together 🫶

🌸 NAME 🌸

_scsih_block_io_to_children_attached_to_ex — and that's literally it, the NAME section gives us the function and then just... vibes. No tagline. Iconic minimalist behavior, honestly ✨

🦋 SYNOPSIS 🦋

void _scsih_block_io_to_children_attached_to_ex (struct MPT3SAS_ADAPTER
*ioc , struct _sas_node *sas_expander );

it returns void (nothing, nada, she's done after doing the thing) and takes two pointers. copy it exactly, do not "fix" the spacing, we don't do that here 💅

🎀 ARGUMENTS 🎀

  • ioc — the per adapter object 🫶
  • sas_expander — the sas_device object. (yes, the parameter's named sas_expander and the docs call it the sas_device object — that's just the lore, we move 😭)

🔥 DESCRIPTION 🔥

okay so here's the tea: this routine sets the sdev state to SDEV_BLOCK for all devices attached to this expander. every single child, no cap, all of them get blocked 👑

and when does it fire? when the expander is pulled. that's the moment. expander gone → all her children get blocked. it's giving "the way that one removal cascades" 💀

🌈 SEE ALSO 🌈

Kernel file ./drivers/scsi/mpt3sas/mpt3sas_scsih.c, enum device_responsive_state(9), struct sense_info(9), struct fw_event_work(9), struct _scsi_io_transfer(9), plus the entire extended fam — buckle up, this list is a whole group chat 🔥

like, that's the whole friend group right there 💀 love that for them.

fr fr reminder: this page lives under ./drivers/scsi/mpt3sas and it's stamped October 2026 — so screenshot it, cite it, whatever, just remember where she came from 🎀

k bye bestie, go block some IO 😘✨💖

📦 where this page comes from