Re: [PATCH] ata: libata-sata: retry hardreset when device detected but PHY not established

[yangxingui]

On 2026/5/28 15:09, Niklas Cassel wrote:
> Hello Xingui,
>
> On Thu, May 28, 2026 at 03:01:39PM +0800, yangxingui wrote:
> > Latest update: The issue we are currently facing is that the SATA host side
> > is continuously waiting for the disk to return the COMWAKE primitive. From
> > the SATA analyzer, it can be seen that the disk has already returned the
> > COMWAKE primitive. This may be related to poor COMWAKE signal quality,
> > causing the host to fail to recognize it. Performing a hard reset might be
> > able to fix the issue.
>
> I'm a little confused. In your previous update you said that the issue had
> been resolved.
>
> Now you say that an additional hard reset (COMRESET) "might be able" to
> fix the issue.
> That does not sound like you have actually verified if it does so or not.
>
> Perhaps you can try implementing my suggestion in:
> https://lore.kernel.org/linux-ide/afMgoFBFPWR0f4gK@ryzen/
>
> which does an additional hard reset (COMRESET).
>
> But please only send out a patch if you have actually verified that the
> suggestion actually resolves your issue.

Hi, Niklas

Thank you for your suggestions.
Okay, yes, initially the disk actively initiated a COMINIT to 
re-establish the link, which was found to resolve the issue. However, 
the root cause is unknown. Recently, it was discovered that the disk 
fails to establish a link because the host did not receive the COMWAKE.

Let's check whether the customer still needs a software fix for this 
problem and try to test.

Thanks
Xingui