Re: [PATCH 2/2] tpm_crb: Raise timeouts for Amazon NitroTPM

[Jarkko Sakkinen <jarkko@kernel.org>]

On Mon, Sep 28, 2026 at 09:09:53PM +0000, Surendran Kanagaraj wrote:
> On Fri, Sep 25, 2026, Jarkko Sakkinen wrote:
> > On Wed, Sep 23, 2026 at 02:28:34PM +0000, Surendran Kanagaraj wrote:
> > > +/* Longest unavailability seen from NitroTPM */
> > > +#define CRB_AMZN_BUSY_TIMEOUT_MS     30000
> >
> > Use TPM_CRB prefix.
> 
> Thanks for taking time to review my patches. Will update this in v2.
> 
> > > +	crb_apply_vendor_quirks(priv, chip);
> >
> > I.e., all this complexity equals to:
> >
> >         if (crb_vendor_id(priv) == CRB_INTF_VID_AMZN)
> >                 chip->busy_timeout_ms = CRB_AMZN_BUSY_TIMEOUT_MS;
> >
> > ?
> 
> Yes. I added the table to keep vendor specific quirks out of the probe
> and could be extended without touching the probe. tpm_tis_core does
> something similar with vendor_dur_overrides. what do you think? I am
> happy to switch to the plain check if you prefer.

See my suggestion for 1/2.

I.e. we need two patches:

1. One that makes all duration assignments indirect lookups to
   chip->duration in the main TPM driver.
2. This patch should then update that array.

> 
> Thanks,
> Surendran

Br, Jarkko