Hi, Yongqiang:
On Thu, 2019-08-29 at 22:50 +0800, yongqiang.niu@xxxxxxxxxxxx wrote:
> From: Yongqiang Niu <yongqiang.niu@xxxxxxxxxxxx>
>
> This patch add mmsys private data for ddp path config
> all these register offset and value will be different in future SOC
> add these define into mmsys private data
> u32 ovl0_mout_en;
> u32 rdma1_sout_sel_in;
> u32 rdma1_sout_dsi0;
> u32 dpi0_sel_in;
> u32 dpi0_sel_in_rdma1;
> u32 dsi0_sel_in;
> u32 dsi0_sel_in_rdma1;
>
> Signed-off-by: Yongqiang Niu <yongqiang.niu@xxxxxxxxxxxx>
> ---
> drivers/gpu/drm/mediatek/mtk_drm_crtc.c | 4 ++
> drivers/gpu/drm/mediatek/mtk_drm_ddp.c | 86 +++++++++++++++++++++++----------
> drivers/gpu/drm/mediatek/mtk_drm_ddp.h | 5 ++
> drivers/gpu/drm/mediatek/mtk_drm_drv.c | 3 ++
> drivers/gpu/drm/mediatek/mtk_drm_drv.h | 3 ++
> 5 files changed, 76 insertions(+), 25 deletions(-)
>
[snip]
>
> void mtk_ddp_add_comp_to_path(void __iomem *config_regs,
> + const struct mtk_mmsys_reg_data *reg_data,
> enum mtk_ddp_comp_id cur,
> enum mtk_ddp_comp_id next)
> {
> unsigned int addr, value, reg;
>
> - value = mtk_ddp_mout_en(cur, next, &addr);
> + value = mtk_ddp_mout_en(reg_data, cur, next, &addr);
> if (value) {
> reg = readl_relaxed(config_regs + addr) | value;
> writel_relaxed(reg, config_regs + addr);
> }
>
> - mtk_ddp_sout_sel(config_regs, cur, next);
> + value = mtk_ddp_sout_sel(reg_data, cur, next, &addr);
> + if (value)
> + writel_relaxed(value, config_regs + addr);
I think the register could be written inside mtk_ddp_sout_sel(), why do
you move out of that function?
Regards,
CK
>
> - value = mtk_ddp_sel_in(cur, next, &addr);
> + value = mtk_ddp_sel_in(reg_data, cur, next, &addr);
> if (value) {
> reg = readl_relaxed(config_regs + addr) | value;
> writel_relaxed(reg, config_regs + addr);
> @@ -420,18 +455,19 @@ void mtk_ddp_add_comp_to_path(void __iomem *config_regs,
> }
>
>
>
_______________________________________________
dri-devel mailing list
dri-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/dri-devel
