reclaim_completed_tx_imm(9)

📦 version 7.3.1, in nixpkgs unstable

💅 reclaim_completed_tx_imm(9) 💅

okay so real talk — this is a page from the Kernel API Manual, and the star of the show is reclaim_completed_tx_imm. she's niche. she's specific. she's lowkey carrying the control queues. let's get into it 🫶

✨ NAME ✨

reclaim_completed_tx_imm — reclaims the completed control-queue Tx descriptors, bestie. that's the whole job. 💖

🌸 SYNOPSIS 🌸

void reclaim_completed_tx_imm (struct sge_txq *q );

literally one function, one argument, returns void. she's not complicated, she's just focused. ✨

🎀 ARGUMENTS 🎀

  • q — the SGE control Tx queue 🌈 — this is the queue our girl is gonna clean up after.

💜 DESCRIPTION 💜

okay so here's the tea, and it's genuinely important: reclaim_completed_tx_imm is a variant of cxgb4_reclaim_completed_tx. she's the cousin who does a slightly different job. 👑

she's specifically for Tx queues that send only immediate data — and yes, presently that means just the control queues, no cap. 🙈

and because those queues only ever deal in immediate data, they do not have any sk_buffs to release. that's the whole reason she exists as her own separate function! no sk_buff cleanup, no drama, no mess. she just reclaims and moves on. periodt. 💅

🦋 SEE ALSO 🦋

okay this list is long but every single one of these is iconic in her orbit, so we're listing them all, no skipping, no cutting corners:

like… sixty-ish entries. she keeps her circle big and her circle documented. 🔥

ngl this one is short but she's doing exactly what she says on the tin — immediate-data queues only, no sk_buffs, no wasted cycles. iykyk. 💀

and that's a wrap, bestie! 🧚 this lil page came from ./drivers/net/ethernet/chelsio... and it's dated October 2026, so tuck that away for later — reclaim_completed_tx_imm(9), signed, sealed, slayed. 🍓💖⭐

📦 where this page comes from