adm_dma_xlate(9)

✨ adm_dma_xlate(9) ✨

the little kernel-API page from the Kernel API Manual, bestie — yes the whole thing lives in there

---

🌸 NAME 🌸

adm_dma_xlate — okay so the original page literally leaves this one blank, no description, nothing, just the name and then silence 😭 so I'm not gonna invent a summary for you. Spoiler though: it's the DMA xlate hook. Keep reading, it all makes sense in a sec 💅

---

🦋 SYNOPSIS 🦋

struct dma_chan * adm_dma_xlate (struct of_phandle_args *dma_spec ,
struct of_dma *ofdma );

That's the whole signature, copied character for character, no notes ✨

---

🎀 ARGUMENTS 🎀

• dma_spec — pointer to the DMA specifier, exactly as you found it in the device tree 🌈
• ofdma — pointer to the DMA controller data 👑

---

💖 DESCRIPTION 💖

So this one can roll with either the 1-cell or the 2-cell format — flexible queen behavior, we love to see it 🫶

• the first cell identifies the slave device
• the optional second cell carries the crci value

And the return vibe: it gives you back a pointer to the appropriate dma channel on success, or NULL on error 💀 (yes, always check for the NULL, that's just basic self-care)

---

🔥 SEE ALSO 🔥

Kernel file ./drivers/dma/qcom/qcom_adm.c, plus the whole friend group:

• adm_free_chan(9)
• adm_get_blksize(9)
• adm_process_fc_descriptors(9)
• adm_process_non_fc_descriptors(9)
• adm_prep_slave_sg(9)
• adm_terminate_all(9)
• adm_start_dma(9)
• adm_dma_irq(9)
• adm_tx_status(9)
• adm_issue_pending(9)
• adm_dma_free_desc(9)

---

and that's the page!! shoutout to ./drivers/dma/qcom for being the source, and to October 2026 for the date — timestamped and iconic, mwah 💋✨