On Thu, Oct 29, 2026 at 11:30:55AM +0200, Laurentiu Palcu wrote:
> On Mon, Oct 26, 2026 at 06:05:12PM +0800, Liu Ying wrote:
[...]
> > diff --git a/drivers/gpu/drm/imx/dc/dc-prg.c
> > b/drivers/gpu/drm/imx/dc/dc-prg.c
> > new file mode 100644
> > index 000000000000..43df5683b24d
> > --- /dev/null
> > +++ b/drivers/gpu/drm/imx/dc/dc-prg.c
> > @@ -0,0 +1,314 @@
> > +// SPDX-License-Identifier: GPL-2.0+
> > +/*
> > + * Copyright 2025 NXP
> > + */
> > +
> > +#include <linux/bitfield.h>
> > +#include <linux/bitops.h>
>
> This header can be removed.
Done.
>
> > +#include <linux/bits.h>
> > +#include <linux/cleanup.h>
> > +#include <linux/clk.h>
> > +#include <linux/delay.h>
> > +#include <linux/device.h>
> > +#include <linux/list.h>
> > +#include <linux/module.h>
>
> This one too.
Done.
>
> [...]
>
> > +
> > +void dc_prg_disable(struct dc_prg *prg)
> > +{
> > + regmap_write(prg->reg, PRG_CTRL, BYPASS);
> > +
> > + pm_runtime_put(prg->dev);
>
> I have a question about this. I see Sashiko also flagged an issue regarding
> the
> unconditional put when the pm_runtime_resume_and_get() fails in
> dc_prg_configure(). The resume is, probably, unlikely to fail.
Yes, dc_prg_configure() is only called at atomic commit stage which is supposed
to run successfully without any error. If pm_runtime_resume_and_get() fails
there, something should be fixed, so I don't bother to handle the failing case.
> But, I see a
> situation where the dc_prg_configure() is never called: when the plane is
> invisible. The dc_prg_disable() would still be called in flush() when the
> plane
> is disabled, would it not?
Yes, patch 15 doesn't handle the invisible plane properly with the prefetch
engine enabled. I'll fix this in next version in patch 15.
[...]
> > +bool dc_prg_stride_supported(struct dc_prg *prg,
>
> It looks like you're not using prg. Can it be removed?
dc_dprc_stride_supported() in patch 8 uses it, so it cannot be removed.
--
Regards,
Liu Ying
