Re: [PATCH] ata: ahci: Disallow LPM for Asus B550-F motherboard

[Niklas Cassel <cassel@kernel.org>]

On Tue, Jun 17, 2025 at 03:32:17PM +0900, Damien Le Moal wrote:
> On 6/17/25 3:18 PM, Mikko Korhonen wrote:
> > Asus ROG STRIX B550-F GAMING (WI-FI) motherboard has problems on some
> > SATA ports with at least one hard drive model (WDC WD20EFAX-68FB5N0)
> > when LPM is enabled. Disabling LPM solves the issue.
> > 
> > Fixes: 7627a0edef54 ("ata: ahci: Drop low power policy board type")
> > Signed-off-by: Mikko Korhonen <mjkorhon@gmail.com>
> > ---
> > I think we might include the comment about the non working drive model, as
> > we have tested only one model and in principle this could also be an
> > interoperation issue. If one would need to enable LPM for the motherboard
> > and happens to have a working hard drive model, he/she would know
> > why we have made the patch in the first place.
> > 
> >  drivers/ata/ahci.c | 16 ++++++++++++++++
> >  1 file changed, 16 insertions(+)
> > 
> > diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
> > index c8ad8ace7496..ae8c4072a5a3 100644
> > --- a/drivers/ata/ahci.c
> > +++ b/drivers/ata/ahci.c
> > @@ -1454,6 +1454,22 @@ static bool ahci_broken_lpm(struct pci_dev *pdev)
> >  			},
> >  			/* 320 is broken, there is no known good version yet. */
> >  		},
> > +		{
> > +			/*
> > +			 * AMD 500 Series Chipset SATA Controller [1022:43eb]
> > +			 * on this motherboard timouts on ports 5 and 6

s/timouts/timeouts/

> > +			 * when LPM enabled at least with WDC WD20EFAX-68FB5N0
> > +			 * hard drives.
> > +			 */
> > +			.matches = {
> > +				DMI_MATCH(DMI_BOARD_VENDOR,
> > +					"ASUSTeK COMPUTER INC."),

align to open parenthesis


> > +				DMI_MATCH(DMI_BOARD_NAME,
> > +					"ROG STRIX B550-F GAMING (WI-FI)"),

align to open parenthesis


> > +			},
> > +			/* 3621 is broken, there is no known good version yet.
> > +			 */
> 
> Please use the correct comment format:
> 
> 			/*
> 			 * 3621 is broken and there is no known good version.
> 			 */

I think having it a single line as:
	/* 3621 is broken and there is no known good version. */

is fine, as it will be less than 80 columns, and matches the
existing single line comments in the same table.


Anyway, I can fix these nits up when applying, no need to send a V2.

Mikko, thank you for your efforts in debugging and helping out with
reviews.


Kind regards,
Niklas