Re: [PATCH v3 16/16] arm_mpam: detect and enable MPAM-Fb PCC support
Andre Przywara <[email protected]>
| Newsgroups | org.kernel.vger.linux-acpi,org.infradead.lists.linux-arm-kernel,org.kernel.vger.linux-kernel |
|---|---|
| Message-ID | <[email protected]> |
Hi Niyas, On 7/14/26 14:32, Niyas Sait wrote: > On 10/07/2026 15:45, Andre Przywara wrote: >> + cur->pcc_cl.dev = dev; >> + cur->pcc_cl.rx_callback = mpam_pcc_rx_callback; >> + cur->pcc_cl.tx_block = true; >> + cur->pcc_cl.tx_tout = 1000; /* 1s */ > > Should this timeout be based on nominal latency reported in PCC table > instead of being hard coded to 1s here? May be some multiple of nominal > latency reported I originally thought that this would be handled by the PCC code, but that just parses the value, and puts it in the pcc_mbox_chan struct. Which we only get access to *after* we request the channel in the next line here. But it's only mbox_send_message() that uses the value, so I think we can populate that value later on still. And yeah, some multiple of the nominal latency sounds reasonable. Cheers, Andre