On Mon 04 Jun 05:58 PDT 2018, Tomasz Figa wrote:
> Hi Vikash,
>
> On Sat, Jun 2, 2018 at 5:27 AM Vikash Garodia <vgarodia@xxxxxxxxxxxxxx> wrote:
> [snip]
> > +int venus_boot(struct venus_core *core)
> > +{
> > + phys_addr_t mem_phys;
> > + size_t mem_size;
> > + int ret;
> > + struct device *dev;
> > +
> > + if (!IS_ENABLED(CONFIG_QCOM_MDT_LOADER))
> > + return -EPROBE_DEFER;
>
> Why are we deferring probe here? The option will not magically become
> enabled after probe is retried.
>
The original code should have read:
if (IS_ENABLED(CONFIG_QCOM_MDT_LOADER) && !qcom_scm_is_available())
return -EPROBE_DEFER;
The code does depend on CONFIG_QCOM_MDT_LOADER regardless of it using
scm for firmware verification.
Regards,
Bjorn
