blk_rq_passthrough_stats(9)

๐Ÿ“ฆ version 7.3.1, in nixpkgs unstable

๐Ÿ’… blk_rq_passthrough_stats(9) โ€” the manual page, but make it make sense

okay so you found blk_rq_passthrough_stats in section 9 of the Kernel API Manual โ€” bestie, welcome, this one is short and sweet and honestly kind of a vibe. Let's get into it ๐Ÿซถ


โœจ NAME โœจ

blk_rq_passthrough_stats โ€” check if this request should account stats ๐Ÿงฎ

Like literally, it's a yes/no vibe check on whether a request gets to be counted.


๐ŸŒธ SYNOPSIS ๐ŸŒธ

bool blk_rq_passthrough_stats (struct request *rq , struct request_queue
*q );

That's it. Two args, one bool out. Clean. Iconic. No cap. ๐Ÿ’…


๐ŸŽ€ ARGUMENTS ๐ŸŽ€

  • rq โ€” the request you're checking ๐Ÿ‘€ (you know, the one you're wondering about)
  • q โ€” the queue that's accumulating the stats ๐Ÿ“Š (she's the one doing the counting)

๐Ÿฆ‹ DESCRIPTION ๐Ÿฆ‹

Okay so here's the one (1) thing you need to internalize, and it's lowkey important:

Note, q does not necessarily need to be the request_queue that provides rq. ๐Ÿซ 

Read that again. I'll wait. q and rq are not required to be each other's people โ€” the queue collecting the stats doesn't have to be the same queue that handed you the request. So don't go assuming they're a matched set, bestie. That's the whole trick. ๐Ÿ™ˆ


โญ RETURN โญ

Returns true if stats should be accounted. โœ…

That's the entire answer. true = count it. No ambiguity, no cliffhanger, we love a straightforward return. ๐Ÿ’–


๐ŸŒˆ SEE ALSO ๐ŸŒˆ

Kernel file ./include/linux/blk-mq.h, plus the whole friend group:

Go read those if you're nosy. I support it. ๐Ÿ‘‘


and that's the page!! ๐ŸŽ‰ sourced from ./include/linux, dated October 2026 โ€” mwah, go forth and count your stats correctly, bestie ๐Ÿงš๐Ÿ’œ

๐Ÿ“ฆ where this page comes from