Re: [PATCH v2] ata: libahci_platform: support non-consecutive port numbers

[Josua Mayer <josua@solid-run.com>]

Am 07.02.25 um 20:45 schrieb Klaus Kudielka:
> On Fri, 2025-02-07 at 11:22 +0000, Josua Mayer wrote:
>> Can you confirm the physical number of sata ports on your board?
>>
> The second port indeed seems not wired on Turris Omnia.
> If the "masking port_map 0x3 -> 0x1" kernel warning had not suddenly appeared, I would not have noticed this at all.
>
>> I would be curious whether in another board that has two ports physically,
>> whether both of them were functional before my patch.
> I don't have such a board, but to me it seems the existing code was made exactly for that case.

I have such a board and tested how it behaves with Linux 6.1.124 from Debian.

I modified the dtb removing the port subnodes from sata nodes and found
just one difference:

- [    3.225848] ahci-mvebu f10a8000.sata: masking port_map 0x3 -> 0x3
  [    3.225882] ahci-mvebu f10a8000.sata: AHCI 0001.0000 32 slots 2 ports 6 Gbps 0x3 impl platform mode
  [    3.225891] ahci-mvebu f10a8000.sata: flags: 64bit ncq sntf led only pmp fbs pio slum part sxs
...
- [    3.248678] ahci-mvebu f10e0000.sata: masking port_map 0x3 -> 0x3
  [    3.248714] ahci-mvebu f10e0000.sata: AHCI 0001.0000 32 slots 2 ports 6 Gbps 0x3 impl platform mode
  [    3.248723] ahci-mvebu f10e0000.sata: flags: 64bit ncq sntf led only pmp fbs pio slum part sxs

So, only the masking message goes away.

When connecting drives to each port, both ports per controller were functional
contrary to my intuition.

>
> For reference, my board later reports
>   ata2: SATA link down (SStatus 0 SControl 300)
>   ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
Thanks!