Re: [PATCH RFC 2/2] tpm: tpm_tis: Provide error messages in tpm_tis_core_init

[Stefan Wahren <wahrenst@xxxxxxx>]

Am 06.09.24 um 11:50 schrieb Jarkko Sakkinen:
> On Fri Sep 6, 2024 at 9:09 AM EEST, Stefan Wahren wrote:
> > In case of SPI wiring issues (e.g. MISO pin permanent high) the
> > core init won't provide any error message on failure. Add some
> > helpful error messages to tpm_tis_core_init(), which helps
> > to better understand why the driver won't probe.
> >
> > Signed-off-by: Stefan Wahren <wahrenst@xxxxxxx>
> > ---
> >   drivers/char/tpm/tpm_tis_core.c | 15 ++++++++++++---
> >   1 file changed, 12 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/char/tpm/tpm_tis_core.c b/drivers/char/tpm/tpm_tis_core.c
> > index fdef214b9f6b..830e331fcebe 100644
> > --- a/drivers/char/tpm/tpm_tis_core.c
> > +++ b/drivers/char/tpm/tpm_tis_core.c
> > @@ -1133,8 +1133,10 @@ int tpm_tis_core_init(struct device *dev, struct tpm_tis_data *priv, int irq,
> >   	dev_set_drvdata(&chip->dev, priv);
> >
> >   	rc = tpm_tis_read32(priv, TPM_DID_VID(0), &vendor);
> > -	if (rc < 0)
> > +	if (rc < 0) {
> > +		dev_err(dev, "Failed to read TPM_DID_VID register: %d\n", rc);
> >   		return rc;
> > +	}
> >
> >   	priv->manufacturer_id = vendor;
> >
> > @@ -1162,19 +1164,25 @@ int tpm_tis_core_init(struct device *dev, struct tpm_tis_data *priv, int irq,
> >   		chip->ops->clk_enable(chip, true);
> >
> >   	if (wait_startup(chip, 0) != 0) {
> > +		dev_err(dev, "TPM device not accessible (0x%0x)\n",
> > +			vendor);
> >   		rc = -ENODEV;
> >   		goto out_err;
> >   	}
> >
> >   	/* Take control of the TPM's interrupt hardware and shut it off */
> >   	rc = tpm_tis_read32(priv, TPM_INT_ENABLE(priv->locality), &intmask);
> > -	if (rc < 0)
> > +	if (rc < 0) {
> > +		dev_err(dev, "Failed to read TPM_INT_ENABLE register: %d\n", rc);
> >   		goto out_err;
> > +	}
> >
> >   	/* Figure out the capabilities */
> >   	rc = tpm_tis_read32(priv, TPM_INTF_CAPS(priv->locality), &intfcaps);
> > -	if (rc < 0)
> > +	if (rc < 0) {
> > +		dev_err(dev, "Failed to read TPM_INTF_CAPS register: %d\n", rc);
> >   		goto out_err;
> > +	}
> >
> >   	dev_dbg(dev, "TPM interface capabilities (0x%x):\n",
> >   		intfcaps);
> > @@ -1209,6 +1217,7 @@ int tpm_tis_core_init(struct device *dev, struct tpm_tis_data *priv, int irq,
> >
> >   	rc = tpm_tis_request_locality(chip, 0);
> >   	if (rc < 0) {
> > +		dev_err(dev, "Failed to request locality (0x%0x)\n", vendor);
> >   		rc = -ENODEV;
> >   		goto out_err;
> >   	}
> > --
> > 2.34.1
> No thanks. It not only adds helpful messages but has potential to add
> unwanted merge conflicts.
So this is a general NAK?

So with every setup where the driver exits without an error message, the
developer should add its own debug?
> BR, Jarkko