[PATCH v11 7/9] drm/mediatek: enable dither function
Yongqiang Niu
yongqiang.niu at mediatek.com
Thu Jan 28 08:32:09 UTC 2021
On Thu, 2021-01-28 at 16:28 +0800, CK Hu wrote:
> On Thu, 2021-01-28 at 16:18 +0800, Hsin-Yi Wang wrote:
> > On Thu, Jan 28, 2021 at 4:10 PM Yongqiang Niu
> > <yongqiang.niu at mediatek.com> wrote:
> > >
> > > On Thu, 2021-01-28 at 16:07 +0800, CK Hu wrote:
> > > > On Thu, 2021-01-28 at 15:59 +0800, Yongqiang Niu wrote:
> > > > > On Thu, 2021-01-28 at 15:42 +0800, CK Hu wrote:
> > > > > > Hi, Hsin-Yi:
> > > > > >
> > > > > > On Thu, 2021-01-28 at 15:28 +0800, Hsin-Yi Wang wrote:
> > > > > > > From: Yongqiang Niu <yongqiang.niu at mediatek.com>
> > > > > > >
> > > > > > > for 5 or 6 bpc panel, we need enable dither function
> > > > > > > to improve the display quality
> > > > > > >
> > > > > > > Signed-off-by: Yongqiang Niu <yongqiang.niu at mediatek.com>
> > > > > > > Signed-off-by: Hsin-Yi Wang <hsinyi at chromium.org>
> > > > > > > ---
> > > > > > > drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c | 44 ++++++++++++++++++++-
> > > > > > > 1 file changed, 43 insertions(+), 1 deletion(-)
> > > > > > >
> > > > > > > diff --git a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c
> > > > > > > index 8173f709272be..e85625704d611 100644
> > > > > > > --- a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c
> > > > > > > +++ b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c
> > > > > > > @@ -53,7 +53,9 @@
> > > > > > > #define DITHER_EN BIT(0)
> > > > > > > #define DISP_DITHER_CFG 0x0020
> > > > > > > #define DITHER_RELAY_MODE BIT(0)
> > > > > > > +#define DITHER_ENGINE_EN BIT(1)
> > > > > > > #define DISP_DITHER_SIZE 0x0030
> > > > > > > +#define DITHER_REG(idx) (0x100 + (idx) * 4)
> > > > > > >
> > > > > > > #define LUT_10BIT_MASK 0x03ff
> > > > > > >
> > > > > > > @@ -313,8 +315,48 @@ static void mtk_dither_config(struct device *dev, unsigned int w,
> > > > > > > {
> > > > > > > struct mtk_ddp_comp_dev *priv = dev_get_drvdata(dev);
> > > > > > >
> > > > > > > + bool enable = false;
> > > > > > > +
> > > > > > > + /* default value for dither reg 5 to 14 */
> > > > > > > + const u32 dither_setting[] = {
> > > > > > > + 0x00000000, /* 5 */
> > > > > > > + 0x00003002, /* 6 */
> > > > > > > + 0x00000000, /* 7 */
> > > > > > > + 0x00000000, /* 8 */
> > > > > > > + 0x00000000, /* 9 */
> > > > > > > + 0x00000000, /* 10 */
> > > > > > > + 0x00000000, /* 11 */
> > > > > > > + 0x00000011, /* 12 */
> > > > > > > + 0x00000000, /* 13 */
> > > > > > > + 0x00000000, /* 14 */
> > > > > >
> > > > > > Could you explain what is this?
> > > > >
> > > > > this is dither 5 to dither 14 setting
> > > > > this will be useless, we just need set dither 5 and dither 7 like
> > > > > mtk_ddp_write(cmdq_pkt, 0, comp, DISP_DITHER_5);
> > > > > mtk_ddp_write(cmdq_pkt, 0, comp, DISP_DITHER_7);
> > > > > other value is same with hardware default value.
> > > > >
> > > > >
> > > > > >
> > > > > > > + };
> > > > > > > +
> > > > > > > + if (bpc == 5 || bpc == 6) {
> > > > > > > + enable = true;
> > > > > > > + mtk_ddp_write(cmdq_pkt,
> > > > > > > + DITHER_LSB_ERR_SHIFT_R(MTK_MAX_BPC - bpc) |
> > > > > > > + DITHER_ADD_LSHIFT_R(MTK_MAX_BPC - bpc) |
> > > > > > > + DITHER_NEW_BIT_MODE,
> > > > > > > + &priv->cmdq_reg, priv->regs, DITHER_REG(15));
> > > > > > > + mtk_ddp_write(cmdq_pkt,
> > > > > > > + DITHER_LSB_ERR_SHIFT_B(MTK_MAX_BPC - bpc) |
> > > > > > > + DITHER_ADD_LSHIFT_B(MTK_MAX_BPC - bpc) |
> > > > > > > + DITHER_LSB_ERR_SHIFT_G(MTK_MAX_BPC - bpc) |
> > > > > > > + DITHER_ADD_LSHIFT_G(MTK_MAX_BPC - bpc),
> > > > > >
> > > > > > This result in 0x50505050, but previous version is 0x50504040, so this
> > > > > > version is correct and previous version is incorrect?
> > > > >
> > > > > the new version set r g b 3 channel same, seams more reasonable
> > > > >
> > > > >
> > > >
> > > > So all the setting of DISP_DITHER_5, DISP_DITHER_7, DISP_DITHER_15,
> > > > DISP_DITHER_16 is identical to mtk_dither_set(), so call
> > > > mtk_dither_set() instead of duplication here.
> > > >
> > >
> > > dither enable set in mtk_dither_set is
> > > mtk_ddp_write(cmdq_pkt, DISP_DITHERING, comp, CFG);
> > >
> > > that is different 8183 and mt8192.
> > > mt8173 dither enable in gamma is bit2
> > > mt8183 and mt8192 dither engine enable is bit 1
> > >
> > >
> >
> > We can still call mtk_dither_set() for bpc is 5 or 6 here, though it
> > will be set to bit2,
> > but later in mtk_ddp_write(cmdq_pkt, enable ? DITHER_ENGINE_EN :
> > DITHER_RELAY_MODE, &priv->cmdq_reg, priv->regs, DISP_DITHER_CFG); it
> > will be correct back to bit 1.
> >
> > Is this reasonable?
>
> Looks weird. Maybe pass some information into mtk_dither_set() to set
> DISP_DITHERING correctly.
>
> I find one thing need to be fixed. CFG should be lower case.
we could modify this like this:
void mtk_dither_set(struct mtk_ddp_comp *comp, unsigned int bpc,
unsigned int cfg, u32 dither_enable, struct cmdq_pkt *cmdq_pkt)
mtk_ddp_write(cmdq_pkt, dither_enable, comp, cfg);
>
> Regards,
> CK
>
> >
> > > > Regards,
> > > > CK
> > > > > >
> > > > > > Regards,
> > > > > > CK
> > > > > >
> > > > > > > + &priv->cmdq_reg, priv->regs, DITHER_REG(16));
> > > > > > > + }
> > > > > > > +
> > > > > > > +
> > > > > > > + if (enable) {
> > > > > > > + u32 idx;
> > > > > > > +
> > > > > > > + for (idx = 0; idx < ARRAY_SIZE(dither_setting); idx++)
> > > > > > > + mtk_ddp_write(cmdq_pkt, dither_setting[idx], &priv->cmdq_reg, priv->regs,
> > > > > > > + DITHER_REG(idx + 5));
> > > > > > > + }
> > > > > > > +
> > > > > > > mtk_ddp_write(cmdq_pkt, h << 16 | w, &priv->cmdq_reg, priv->regs, DISP_DITHER_SIZE);
> > > > > > > - mtk_ddp_write(cmdq_pkt, DITHER_RELAY_MODE, &priv->cmdq_reg, priv->regs, DISP_DITHER_CFG);
> > > > > > > + mtk_ddp_write(cmdq_pkt, enable ? DITHER_ENGINE_EN : DITHER_RELAY_MODE, &priv->cmdq_reg, priv->regs, DISP_DITHER_CFG);
> > > > > > > }
> > > > > > >
> > > > > > > static void mtk_dither_start(struct device *dev)
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > >
> > > >
> > >
>
>
More information about the dri-devel
mailing list