sdma_txadd_daddr(9)
π NAME π
okay so bestie, this is sdma_txadd_daddr β from the Kernel API Manual, no less β and it does exactly one iconic thing: it adds a dma address to the sdma_txreq. that's the whole vibe. π
β¨ SYNOPSIS β¨
int sdma_txadd_daddr (struct hfi1_devdata *dd , struct sdma_txreq *tx ,
dma_addr_t addr , u16 len );
π ARGUMENTS π
ddβ the device to use for mapping. πtxβ thesdma_txreqto which the page is added. π¦addrβ the dma address mapped by caller. literally right there. πlenβ length in bytes. π
πΈ DESCRIPTION πΈ
This is used to add a descriptor for memory that is already dma mapped. no cap, that's the move. β¨
In this case, there is no unmapping as part of the progress processing for this memory location. so don't go looking for an unmap arc, because this memory location is not giving that. π«Ά
β RETURN β
0β success. slay. π-ENOMEMβ couldn't extend descriptor array. π that's the whole mood.
π¦ SEE ALSO π¦
The kernel file is ./drivers/infiniband/hw/hfi1/sdma.h. also go peek at these:
sdma exported routines(9)sdma PSM/verbs API(9)Infrastructure calls(9)struct sdma_engine(9)sdma_empty(9)sdma_running(9)sdma_txinit_ahg(9)sdma_txinit(9)sdma_txadd_page(9)sdma_txadd_kvaddr(9)sdma_progress(9)struct sdma_map_elem(9)sdma_engine_progress_schedule(9)
mwah π this little page lives in ./drivers/infiniband/hw/hfi1, dated October 2026 β periodt. π
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