Hi,
On 2024/4/24 16:39, Dmitry Baryshkov wrote:
Ok, what is the_bug_ that is being fixed by this patch?
I didn't have a way to use that driver under non DT environment,
this is the bug. Note: I demanding full features.
Both st7735r.c and repaper.c requires additional device property.
- For st7735r.c: device_property_read_u32(dev, "rotation", &rotation);
- For repaper.c: device_property_read_string(dev, "pervasive,thermal-zone", &thermal_zone)
Under non DT environment, those device properties can not be read.
Software node backend provided a way. Otherwise, the net results
of the patch doesn't meet the description in the commit message.
If you check
the 'submitting-patches.rst', you'll find this phrase as a description
of the Fixes: tag.
I have readthe 'submitting-patches.rst', the first sentence tell us that
"A Fixes: tag indicates that the patch fixes an issue in a previous commit."
So what's the problem?
Hence, before my patch is applied, the two "Make driver OF-independent" patch
have no effect. Using device_get_match_data() itself is exactly*same* with
using of_device_get_match_data() as long as the .device_get_match_data hook is
not implemented.
As far as I can see, repaper correctly handles the case by falling
back to the spi_id. So does st7735r.c.
Yeah, this explicitly is the issue.
Falling back to other means is robust design, but it explicitly says
that the freshly addeddevice_get_match_data() don't works at all
under non DT environment. This is the bug, so what's you concern?
According to the commit message, the purpose of the introduction of
thedevice_get_match_data() is to achieve OF independent. But as you said,
it will fall back, then how does the goal "Make driver
OF-independent" can be achieved by the patch itself?
--
Best regards,
Sui
