Re: [PATCH v2 0/4] sun8i: r40: add AHCI

[Icenowy Zheng]

ä 2018å7æ10æ GMT+08:00 äå9:05:17, LABBE Corentin <clabbe@xxxxxxxxxxxx> åå:
>On Mon, Jul 09, 2018 at 11:44:20PM +0800, Icenowy Zheng wrote:
>> 
>> 
>> ä 2018å7æ9æ GMT+08:00 äå11:20:54, Corentin Labbe
><clabbe@xxxxxxxxxxxx> åå:
>> >Hello
>> >
>> >With Moeicenowy's agreement, I have take leadership ot this
>patchset.
>> >
>> >There are no really changes appart renaming struct quirck to
>variant.
>> >
>> >Since the last serie is really old, I will answer comment here.
>> >The two regulator (1.2 and 2.5V) are not for the PHY since:
>> >- nothing in the schematic said that they are for the PHY, they
>seems
>> >  only for controller
>> >- all other AHCI driver use 5V for the target/PHY (vs 1.2/2.5 which
>> >  cannot be used for target)
>> 
>> Target is not equal to PHY. Target means the supply
>> of the disk, which can be usually 5v (for 2.5" HDD) or
>> 12v (3.5" HDD).
>> 
>> By reading Wikipedia articles  about SATA and LVDS, I assume
>> 2.5V is for PHY and 1.2V is for internal digital logic (VDD-SYS
>> is commonly 1.2V on 40nm Allwinner SoCs; 2.5V VDD can be
>> used to efficiently deliver ~1.2V LVDS.)
>> 
>> P.S. VDD-SATA and VDD25-SATA also exist on A20, and by checking
>> Banana Pi M1 (the original Banana Pi) schematics, VDD-SATA
>> is connected to common VDD-SYS (called INTVDD on the
>> schematics) and VDD25-SATA is connected to an always-on
>> fixed LDO, maybe due to the lack of power outputs on AXP209.
>> 
>
>So we still need to add a regulator on the controller and add an
>optionnal "phy" regulator for AHCI port.

No, phy is not a part of the port. From the perspective of
the users, phy is part of the controller.

I still suggest generic multiple regulator support, to
reduce sunxi-specified code.

>
>
>_______________________________________________
>linux-arm-kernel mailing list
>linux-arm-kernel@xxxxxxxxxxxxxxxxxxx
>http://lists.infradead.org/mailman/listinfo/linux-arm-kernel