[PATCH 02/13] drm/i915: Clean up the cursor register defines

Ville Syrjälä ville.syrjala at linux.intel.com
Mon May 20 16:23:46 UTC 2024


On Mon, May 20, 2024 at 12:10:30PM +0300, Jani Nikula wrote:
> On Thu, 16 May 2024, Ville Syrjala <ville.syrjala at linux.intel.com> wrote:
> > From: Ville Syrjälä <ville.syrjala at linux.intel.com>
> >
> > Group the cursor register defines such that everything to
> > do with one register is in one place.
> >
> > Signed-off-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
> 
> Reviewed-by: Jani Nikula <jani.nikula at intel.com>
> 
> but a couple of nitpicks inline...
> 
> > ---
> >  .../gpu/drm/i915/display/intel_cursor_regs.h  | 52 +++++++++----------
> >  1 file changed, 26 insertions(+), 26 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/display/intel_cursor_regs.h b/drivers/gpu/drm/i915/display/intel_cursor_regs.h
> > index c2190af1e9f5..270c26c2e6df 100644
> > --- a/drivers/gpu/drm/i915/display/intel_cursor_regs.h
> > +++ b/drivers/gpu/drm/i915/display/intel_cursor_regs.h
> > @@ -9,6 +9,7 @@
> >  #include "intel_display_reg_defs.h"
> >  
> >  #define _CURACNTR		0x70080
> > +#define CURCNTR(dev_priv, pipe)	_MMIO_CURSOR2((dev_priv), (pipe), _CURACNTR)
> 
> In addition to code movement, these add braces around (dev_priv) and
> (pipe). While it makes review harder by breaking 'git show
> --color-moved',

Sorry. Forgot I snuck it in there.

> I also think it's kind of unnecessary when they're only
> passed on as parameters. Or is there some corner case where it matters?

I think cargo-culting is probably the best argument for protecting
each and every macro argument. If used universally then I think
it'll be a bit more likely that newly added macros, where it
might matter more, will inherit it as well.

And we've certainly had incidents with misplaced commas (the
i915_reg_t addition was a reaction to one such event), so I
wouldn't dare claim that there is zero chance of screwups
with these.

> Comma has the lowest precedence, and I don't think you could easily pass
> in a value with a comma operator.
> 
> No need to change for this, it's not wrong either.
> 
> >  /* Old style CUR*CNTR flags (desktop 8xx) */
> >  #define   CURSOR_ENABLE			REG_BIT(31)
> >  #define   CURSOR_PIPE_GAMMA_ENABLE	REG_BIT(30)
> > @@ -38,61 +39,60 @@
> >  #define   MCURSOR_MODE_128_ARGB_AX	(0x20 | MCURSOR_MODE_128_32B_AX)
> >  #define   MCURSOR_MODE_256_ARGB_AX	(0x20 | MCURSOR_MODE_256_32B_AX)
> >  #define   MCURSOR_MODE_64_ARGB_AX	(0x20 | MCURSOR_MODE_64_32B_AX)
> > +
> >  #define _CURABASE		0x70084
> > +#define CURBASE(dev_priv, pipe)	_MMIO_CURSOR2((dev_priv), (pipe), _CURABASE)
> > +
> >  #define _CURAPOS		0x70088
> > -#define _CURAPOS_ERLY_TPT	0x7008c
> > +#define CURPOS(dev_priv, pipe)	_MMIO_CURSOR2((dev_priv), (pipe), _CURAPOS)
> >  #define   CURSOR_POS_Y_SIGN		REG_BIT(31)
> >  #define   CURSOR_POS_Y_MASK		REG_GENMASK(30, 16)
> >  #define   CURSOR_POS_Y(y)		REG_FIELD_PREP(CURSOR_POS_Y_MASK, (y))
> >  #define   CURSOR_POS_X_SIGN		REG_BIT(15)
> >  #define   CURSOR_POS_X_MASK		REG_GENMASK(14, 0)
> >  #define   CURSOR_POS_X(x)		REG_FIELD_PREP(CURSOR_POS_X_MASK, (x))
> > +
> > +#define _CURAPOS_ERLY_TPT	0x7008c
> > +#define CURPOS_ERLY_TPT(dev_priv, pipe)	_MMIO_CURSOR2((dev_priv), (pipe), _CURAPOS_ERLY_TPT)
> > +
> >  #define _CURASIZE		0x700a0 /* 845/865 */
> > +#define CURSIZE(dev_priv, pipe)	_MMIO_CURSOR2((dev_priv), (pipe), _CURASIZE)
> >  #define   CURSOR_HEIGHT_MASK		REG_GENMASK(21, 12)
> >  #define   CURSOR_HEIGHT(h)		REG_FIELD_PREP(CURSOR_HEIGHT_MASK, (h))
> >  #define   CURSOR_WIDTH_MASK		REG_GENMASK(9, 0)
> >  #define   CURSOR_WIDTH(w)		REG_FIELD_PREP(CURSOR_WIDTH_MASK, (w))
> > +
> >  #define _CUR_FBC_CTL_A		0x700a0 /* ivb+ */
> > +#define CUR_FBC_CTL(dev_priv, pipe)	_MMIO_CURSOR2((dev_priv), (pipe), _CUR_FBC_CTL_A)
> >  #define   CUR_FBC_EN			REG_BIT(31)
> >  #define   CUR_FBC_HEIGHT_MASK		REG_GENMASK(7, 0)
> >  #define   CUR_FBC_HEIGHT(h)		REG_FIELD_PREP(CUR_FBC_HEIGHT_MASK, (h))
> > +
> >  #define _CUR_CHICKEN_A		0x700a4 /* mtl+ */
> > +#define CUR_CHICKEN(dev_priv, pipe)	_MMIO_CURSOR2((dev_priv), (pipe), _CUR_CHICKEN_A)
> > +
> >  #define _CURASURFLIVE		0x700ac /* g4x+ */
> > -#define _CURBCNTR		0x700c0
> > -#define _CURBBASE		0x700c4
> > -#define _CURBPOS		0x700c8
> > -
> > -#define _CURBCNTR_IVB		0x71080
> > -#define _CURBBASE_IVB		0x71084
> > -#define _CURBPOS_IVB		0x71088
> > -
> > -#define CURCNTR(dev_priv, pipe) _MMIO_CURSOR2(dev_priv, pipe, _CURACNTR)
> > -#define CURBASE(dev_priv, pipe) _MMIO_CURSOR2(dev_priv, pipe, _CURABASE)
> > -#define CURPOS(dev_priv, pipe) _MMIO_CURSOR2(dev_priv, pipe, _CURAPOS)
> > -#define CURPOS_ERLY_TPT(dev_priv, pipe) _MMIO_CURSOR2(dev_priv, pipe, _CURAPOS_ERLY_TPT)
> > -#define CURSIZE(dev_priv, pipe) _MMIO_CURSOR2(dev_priv, pipe, _CURASIZE)
> > -#define CUR_FBC_CTL(dev_priv, pipe) _MMIO_CURSOR2(dev_priv, pipe, _CUR_FBC_CTL_A)
> > -#define CUR_CHICKEN(dev_priv, pipe) _MMIO_CURSOR2(dev_priv, pipe, _CUR_CHICKEN_A)
> > -#define CURSURFLIVE(dev_priv, pipe) _MMIO_CURSOR2(dev_priv, pipe, _CURASURFLIVE)
> > +#define CURSURFLIVE(dev_priv, pipe)	_MMIO_CURSOR2((dev_priv), (pipe), _CURASURFLIVE)
> >  
> >  /* skl+ */
> >  #define _CUR_WM_A_0		0x70140
> >  #define _CUR_WM_B_0		0x71140
> > +#define CUR_WM(pipe, level)	_MMIO(_PIPE((pipe), _CUR_WM_A_0, _CUR_WM_B_0) + (level) * 4)
> > +
> >  #define _CUR_WM_SAGV_A		0x70158
> >  #define _CUR_WM_SAGV_B		0x71158
> > +#define CUR_WM_SAGV(pipe)	_MMIO_PIPE((pipe), _CUR_WM_SAGV_A, _CUR_WM_SAGV_B)
> > +
> >  #define _CUR_WM_SAGV_TRANS_A	0x7015C
> >  #define _CUR_WM_SAGV_TRANS_B	0x7115C
> > +#define CUR_WM_SAGV_TRANS(pipe)	_MMIO_PIPE((pipe), _CUR_WM_SAGV_TRANS_A, _CUR_WM_SAGV_TRANS_B)
> > +
> >  #define _CUR_WM_TRANS_A		0x70168
> >  #define _CUR_WM_TRANS_B		0x71168
> > -#define _CUR_WM_0(pipe) _PIPE(pipe, _CUR_WM_A_0, _CUR_WM_B_0)
> > -#define CUR_WM(pipe, level) _MMIO(_CUR_WM_0(pipe) + ((4) * (level)))
> 
> There's some unmentioned drive-by cleanup here too. No biggie, but I've
> found 'git show --color-moved' to be such a powerful aid in reviewing
> code movement patches that I'd prefer these to be separate. No need to
> change now, because I already reviewed it. :)

Agreed. I suppose I got a bit of tunnel vision here and didn't do
a proper job of reviewing my own stuff before sending it out.
I'll need to add --color-moved to my "this should be pure
code movement" checklist...

> 
> BR,
> Jani.
> 
> > -#define CUR_WM_SAGV(pipe) _MMIO_PIPE(pipe, _CUR_WM_SAGV_A, _CUR_WM_SAGV_B)
> > -#define CUR_WM_SAGV_TRANS(pipe) _MMIO_PIPE(pipe, _CUR_WM_SAGV_TRANS_A, _CUR_WM_SAGV_TRANS_B)
> > -#define CUR_WM_TRANS(pipe) _MMIO_PIPE(pipe, _CUR_WM_TRANS_A, _CUR_WM_TRANS_B)
> > +#define CUR_WM_TRANS(pipe)	_MMIO_PIPE((pipe), _CUR_WM_TRANS_A, _CUR_WM_TRANS_B)
> >  
> > -/* skl+ */
> > -#define _CUR_BUF_CFG_A				0x7017c
> > -#define _CUR_BUF_CFG_B				0x7117c
> > -#define CUR_BUF_CFG(pipe)	_MMIO_PIPE(pipe, _CUR_BUF_CFG_A, _CUR_BUF_CFG_B)
> > +#define _CUR_BUF_CFG_A		0x7017c
> > +#define _CUR_BUF_CFG_B		0x7117c
> > +#define CUR_BUF_CFG(pipe)	_MMIO_PIPE((pipe), _CUR_BUF_CFG_A, _CUR_BUF_CFG_B)
> >  
> >  #endif /* __INTEL_CURSOR_REGS_H__ */
> 
> -- 
> Jani Nikula, Intel

-- 
Ville Syrjälä
Intel


More information about the Intel-gfx mailing list