cppc_get_transition_latency(9)

πŸ“¦ version 7.3.1, in nixpkgs unstable

πŸ’… cppc_get_transition_latency(9) πŸ’…

okay so bestie, welcome to the Kernel API Manual β€” yes we're in the kernel docs today, put your hair up, we're getting into it ✨


🌸 NAME 🌸

cppc_get_transition_latency β€” literally just returns the frequency transition latency, in ns, no cap πŸŽ€


✨ SYNOPSIS ✨

int cppc_get_transition_latency (int cpu_num );

πŸ¦‹ ARGUMENTS πŸ¦‹

  • cpu_num β€” the CPU number for per_cpu. that's it, that's the whole argument, she's simple and iconic πŸ’…

πŸ’– DESCRIPTION πŸ’–

so here's the tea πŸ“ β€” ACPI CPPC does not explicitly specify how a platform can specify the transition latency for performance change requests. like, it just… doesn't say. rude, but okay.

the closest we have is the timing information from the PCCT tables, which gives us the info on the number and frequency of PCC commands the platform can handle. that's our bestie in this situation 🫢

and if desired_reg is in the SystemMemory or SystemIo ACPI address space, then assume there is no latency. zero. nothing. blessed. ✨


πŸ‘‘ SEE ALSO πŸ‘‘

the kernel file is at ./drivers/acpi/cppc_acpi.c β€” go read her, she ate πŸ”₯

and then the whole friend group of related calls, all of them, not skipping a single one:

like that's the whole squad, iykyk πŸ’€


and that's the page, bestie!! 🧚 this little beauty lives in ./drivers/acpi, dated October 2026, and she's filed under cppc_get_transition_latency(9) β€” mwah πŸ’‹ go forth and measure your latency, periodt πŸ’œ

πŸ“¦ where this page comes from