[PATCH v2 04/16] drm/msm/dpu: move csc matrices to dpu_hw_util
Abhinav Kumar
quic_abhinavk at quicinc.com
Fri Dec 8 16:35:22 UTC 2023
On 12/8/2023 8:27 AM, Dmitry Baryshkov wrote:
> On Fri, 8 Dec 2023 at 18:24, Abhinav Kumar <quic_abhinavk at quicinc.com> wrote:
>>
>>
>>
>> On 12/8/2023 3:12 AM, Dmitry Baryshkov wrote:
>>> On Fri, 8 Dec 2023 at 07:07, Abhinav Kumar <quic_abhinavk at quicinc.com> wrote:
>>>>
>>>> Since the type and usage of CSC matrices is spanning across DPU
>>>> lets introduce a helper to the dpu_hw_util to return the CSC
>>>> corresponding to the request type. This will help to add more
>>>> supported CSC types such as the RGB to YUV one which is used in
>>>> the case of CDM.
>>>>
>>>> Signed-off-by: Abhinav Kumar <quic_abhinavk at quicinc.com>
>>>> ---
>>>> drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c | 54 +++++++++++++++++++++
>>>> drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h | 7 +++
>>>> drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c | 39 ++-------------
>>>> 3 files changed, 64 insertions(+), 36 deletions(-)
>>>>
>>>> diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c
>>>> index 0b05061e3e62..59a153331194 100644
>>>> --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c
>>>> +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c
>>>> @@ -87,6 +87,60 @@ static u32 dpu_hw_util_log_mask = DPU_DBG_MASK_NONE;
>>>> #define QOS_QOS_CTRL_VBLANK_EN BIT(16)
>>>> #define QOS_QOS_CTRL_CREQ_VBLANK_MASK GENMASK(21, 20)
>>>>
>>>> +static const struct dpu_csc_cfg dpu_csc_YUV2RGB_601L = {
>>>> + {
>>>> + /* S15.16 format */
>>>> + 0x00012A00, 0x00000000, 0x00019880,
>>>> + 0x00012A00, 0xFFFF9B80, 0xFFFF3000,
>>>> + 0x00012A00, 0x00020480, 0x00000000,
>>>> + },
>>>> + /* signed bias */
>>>> + { 0xfff0, 0xff80, 0xff80,},
>>>> + { 0x0, 0x0, 0x0,},
>>>> + /* unsigned clamp */
>>>> + { 0x10, 0xeb, 0x10, 0xf0, 0x10, 0xf0,},
>>>> + { 0x00, 0xff, 0x00, 0xff, 0x00, 0xff,},
>>>> +};
>>>> +
>>>> +static const struct dpu_csc_cfg dpu_csc10_YUV2RGB_601L = {
>>>> + {
>>>> + /* S15.16 format */
>>>> + 0x00012A00, 0x00000000, 0x00019880,
>>>> + 0x00012A00, 0xFFFF9B80, 0xFFFF3000,
>>>> + 0x00012A00, 0x00020480, 0x00000000,
>>>> + },
>>>> + /* signed bias */
>>>> + { 0xffc0, 0xfe00, 0xfe00,},
>>>> + { 0x0, 0x0, 0x0,},
>>>> + /* unsigned clamp */
>>>> + { 0x40, 0x3ac, 0x40, 0x3c0, 0x40, 0x3c0,},
>>>> + { 0x00, 0x3ff, 0x00, 0x3ff, 0x00, 0x3ff,},
>>>> +};
>>>> +
>>>> +/**
>>>> + * dpu_hw_get_csc_cfg - get the CSC matrix based on the request type
>>>> + * @type: type of the requested CSC matrix from caller
>>>> + * Return: CSC matrix corresponding to the request type in DPU format
>>>> + */
>>>> +const struct dpu_csc_cfg *dpu_hw_get_csc_cfg(enum dpu_hw_csc_cfg_type type)
>>>> +{
>>>> + const struct dpu_csc_cfg *csc_cfg = NULL;
>>>> +
>>>> + switch (type) {
>>>> + case DPU_HW_YUV2RGB_601L:
>>>> + csc_cfg = &dpu_csc_YUV2RGB_601L;
>>>> + break;
>>>> + case DPU_HW_YUV2RGB_601L_10BIT:
>>>> + csc_cfg = &dpu_csc10_YUV2RGB_601L;
>>>> + break;
>>>> + default:
>>>> + DPU_ERROR("unknown csc_cfg type\n");
>>>> + break;
>>>> + }
>>>> +
>>>> + return csc_cfg;
>>>> +}
>>>> +
>>>> void dpu_reg_write(struct dpu_hw_blk_reg_map *c,
>>>> u32 reg_off,
>>>> u32 val,
>>>> diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h
>>>> index fe083b2e5696..49f2bcf6de15 100644
>>>> --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h
>>>> +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h
>>>> @@ -19,6 +19,11 @@
>>>> #define MISR_CTRL_STATUS_CLEAR BIT(10)
>>>> #define MISR_CTRL_FREE_RUN_MASK BIT(31)
>>>>
>>>> +enum dpu_hw_csc_cfg_type {
>>>> + DPU_HW_YUV2RGB_601L,
>>>> + DPU_HW_YUV2RGB_601L_10BIT,
>>>> +};
>>>> +
>>>> /*
>>>> * This is the common struct maintained by each sub block
>>>> * for mapping the register offsets in this block to the
>>>> @@ -368,4 +373,6 @@ bool dpu_hw_clk_force_ctrl(struct dpu_hw_blk_reg_map *c,
>>>> const struct dpu_clk_ctrl_reg *clk_ctrl_reg,
>>>> bool enable);
>>>>
>>>> +const struct dpu_csc_cfg *dpu_hw_get_csc_cfg(enum dpu_hw_csc_cfg_type type);
>>>
>>> I don't think we need extra enum and wrapper. Just export const data
>>> structures directly.
>>>
>>
>> I liked this approach because the blocks of DPU such as plane and CDM
>> are clients to the dpu_hw_util and just request the type and the util
>> handles their request of returning the correct csc matrix.
>>
>> Do you see any issue with this?
>
> Not an issue, but I don't see anything that requires an extra
> abstraction. We perfectly know which CSC config we would like to get.
>
Correct, so the clients know which "type" of matrix they need and not
the matrix itself. That was the idea behind this.
>>
>>>> +
>>>> #endif /* _DPU_HW_UTIL_H */
>>>> diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
>>>> index 3235ab132540..31641889b9f0 100644
>>>> --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
>>>> +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
>>>> @@ -21,6 +21,7 @@
>>>> #include "dpu_kms.h"
>>>> #include "dpu_formats.h"
>>>> #include "dpu_hw_sspp.h"
>>>> +#include "dpu_hw_util.h"
>>>> #include "dpu_trace.h"
>>>> #include "dpu_crtc.h"
>>>> #include "dpu_vbif.h"
>>>> @@ -508,50 +509,16 @@ static void _dpu_plane_setup_pixel_ext(struct dpu_hw_scaler3_cfg *scale_cfg,
>>>> }
>>>> }
>>>>
>>>> -static const struct dpu_csc_cfg dpu_csc_YUV2RGB_601L = {
>>>> - {
>>>> - /* S15.16 format */
>>>> - 0x00012A00, 0x00000000, 0x00019880,
>>>> - 0x00012A00, 0xFFFF9B80, 0xFFFF3000,
>>>> - 0x00012A00, 0x00020480, 0x00000000,
>>>> - },
>>>> - /* signed bias */
>>>> - { 0xfff0, 0xff80, 0xff80,},
>>>> - { 0x0, 0x0, 0x0,},
>>>> - /* unsigned clamp */
>>>> - { 0x10, 0xeb, 0x10, 0xf0, 0x10, 0xf0,},
>>>> - { 0x00, 0xff, 0x00, 0xff, 0x00, 0xff,},
>>>> -};
>>>> -
>>>> -static const struct dpu_csc_cfg dpu_csc10_YUV2RGB_601L = {
>>>> - {
>>>> - /* S15.16 format */
>>>> - 0x00012A00, 0x00000000, 0x00019880,
>>>> - 0x00012A00, 0xFFFF9B80, 0xFFFF3000,
>>>> - 0x00012A00, 0x00020480, 0x00000000,
>>>> - },
>>>> - /* signed bias */
>>>> - { 0xffc0, 0xfe00, 0xfe00,},
>>>> - { 0x0, 0x0, 0x0,},
>>>> - /* unsigned clamp */
>>>> - { 0x40, 0x3ac, 0x40, 0x3c0, 0x40, 0x3c0,},
>>>> - { 0x00, 0x3ff, 0x00, 0x3ff, 0x00, 0x3ff,},
>>>> -};
>>>> -
>>>> static const struct dpu_csc_cfg *_dpu_plane_get_csc(struct dpu_sw_pipe *pipe,
>>>> const struct dpu_format *fmt)
>>>> {
>>>> - const struct dpu_csc_cfg *csc_ptr;
>>>> -
>>>> if (!DPU_FORMAT_IS_YUV(fmt))
>>>> return NULL;
>>>>
>>>> if (BIT(DPU_SSPP_CSC_10BIT) & pipe->sspp->cap->features)
>>>> - csc_ptr = &dpu_csc10_YUV2RGB_601L;
>>>> + return dpu_hw_get_csc_cfg(DPU_HW_YUV2RGB_601L_10BIT);
>>>> else
>>>> - csc_ptr = &dpu_csc_YUV2RGB_601L;
>>>> -
>>>> - return csc_ptr;
>>>> + return dpu_hw_get_csc_cfg(DPU_HW_YUV2RGB_601L);
>>>> }
>>>>
>>>> static void _dpu_plane_setup_scaler(struct dpu_sw_pipe *pipe,
>>>> --
>>>> 2.40.1
>>>>
>>>
>>>
>
>
>
More information about the dri-devel
mailing list