drm/i915: Adding dbuf support for skl nv12 format.
Subsystems (2)
PATCH 1/1 drm/i915: Adding dbuf support for skl nv12 format.
Skylake nv12 format requires dbuf (aka. ddb) calculations and programming for each of y and uv sub-planes. Made minor changes to reuse current dbuf calculations and programming for uv plane. i.e., with this change, existing computation is used for either packed format or uv portion of nv12 depending on incoming format. Added new code for dbuf computation and programming for y plane.
This patch is a pre-requisite for adding NV12 format support. Actual nv12 support is coming in later patches.
--- drivers/gpu/drm/i915/i915_drv.h | 3 +- drivers/gpu/drm/i915/i915_reg.h | 11 ++++++ drivers/gpu/drm/i915/intel_drv.h | 8 ++++ drivers/gpu/drm/i915/intel_pm.c | 79 ++++++++++++++++++++++++++++++++------ 4 files changed, 88 insertions(+), 13 deletions(-)
drivers/gpu/drm/i915/i915_drv.h +2 -1
@@ -1450,7 +1450,8 @@ static inline bool skl_ddb_entry_equal(const struct skl_ddb_entry *e1,struct skl_ddb_allocation {struct skl_ddb_entry pipe[I915_MAX_PIPES ]; -struct skl_ddb_entry plane[I915_MAX_PIPES ][I915_MAX_PLANES ]; +struct skl_ddb_entry plane[I915_MAX_PIPES ][I915_MAX_PLANES ];/* packed/uv */ +struct skl_ddb_entry y_plane[I915_MAX_PIPES ][I915_MAX_PLANES ];/* y-plane */ struct skl_ddb_entry cursor[I915_MAX_PIPES ]; };
drivers/gpu/drm/i915/i915_reg.h +11 -0
@@ -4880,6 +4880,8 @@ enum skl_disp_power_wells {#define _PLANE_KEYMAX_2_A 0x702a0#define _PLANE_BUF_CFG_1_A 0x7027c#define _PLANE_BUF_CFG_2_A 0x7037c +#define _PLANE_NV12_BUF_CFG_1_A 0x70278 +#define _PLANE_NV12_BUF_CFG_2_A 0x70378#define _PLANE_CTL_1_B 0x71180#define _PLANE_CTL_2_B 0x71280
@@ -4966,6 +4968,15 @@ enum skl_disp_power_wells {#define PLANE_BUF_CFG (pipe, plane) \ _PLANE(plane, _PLANE_BUF_CFG_1(pipe), _PLANE_BUF_CFG_2(pipe)) +#define _PLANE_NV12_BUF_CFG_1_B 0x71278 +#define _PLANE_NV12_BUF_CFG_2_B 0x71378 +#define _PLANE_NV12_BUF_CFG_1 (pipe) \ + _PIPE(pipe, _PLANE_NV12_BUF_CFG_1_A, _PLANE_NV12_BUF_CFG_1_B) +#define _PLANE_NV12_BUF_CFG_2 (pipe) \ + _PIPE(pipe, _PLANE_NV12_BUF_CFG_2_A, _PLANE_NV12_BUF_CFG_2_B) +#define PLANE_NV12_BUF_CFG (pipe, plane) \ + _PLANE(plane, _PLANE_NV12_BUF_CFG_1(pipe), _PLANE_NV12_BUF_CFG_2(pipe)) +/* SKL new cursor registers */ #define _CUR_BUF_CFG_A 0x7017c#define _CUR_BUF_CFG_B 0x7117c
drivers/gpu/drm/i915/intel_drv.h +8 -0
@@ -566,7 +566,15 @@ struct intel_crtc {struct intel_plane_wm_parameters {uint32_t horiz_pixels;uint32_t vert_pixels; +/* +* For packed pixel formats: +* bytes_per_pixel - holds bytes per pixel +* For planar pixel formats: +* bytes_per_pixel - holds bytes per pixel for uv-plane +* y_bytes_per_pixel - holds bytes per pixel for y-plane +*/ uint8_t bytes_per_pixel; +uint8_t y_bytes_per_pixel;bool enabled;bool scaled;u64 tiling;
drivers/gpu/drm/i915/intel_pm.c +67 -12
181 lines not on this page · raw message
Signature · 6 lines
1.7.9.5 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Task id: 6273 -------------------------------------Summary-------------------------------------
Platform Delta drm-intel-nightly Series Applied
PNV 276/276 276/276 ILK 302/302 302/302 SNB 318/318 318/318 IVB 341/341 341/341 BYT 287/287 287/287 BDW 318/318 318/318
-------------------------------------Detailed-------------------------------------
Platform Test drm-intel-nightly Series Applied
Note: You need to pay more attention to line start with '*'
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoted 8 lines · Chandra Konduru
Skylake nv12 format requires dbuf (aka. ddb) calculations and programming for each of y and uv sub-planes. Made minor changes to reuse current dbuf calculations and programming for uv plane. i.e., with this change, existing computation is used for either packed format or uv portion of nv12 depending on incoming format. Added new code for dbuf computation and programming for y plane.
Quoted 4 lines
This patch is a pre-requisite for adding NV12 format support. Actual nv12 support is coming in later patches.
Signed-off-by Chandra Konduru <[email protected]>
Just aside (and not a blocker for this patch really). Do we have nv12 subtest in all the relevant plane tests (rotation, scaling, ...) to exercise this code? Not sure whether we can redo the yuv->rgb conversion correctly in software to match hw, might need to pick special colors that convert accurately.
Anyway given how often we've broken the sprite code I really think we need full functional tests (i.e. including crc checks).
-Daniel
Signature · 7 lines
Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoted 23 lines
-----Original Message-----
From: Daniel Vetter [mailto:[email protected]] On Behalf Of Daniel Vetter
Sent: Monday, May 04, 2015 8:01 AM
To: Konduru, Chandra
Cc: [email protected]; Vetter, Daniel; Syrjala, Ville
Subject: Re: [Intel-gfx] [PATCH] drm/i915: Adding dbuf support for skl nv12 format.Quoted 11 lines · Chandra Konduru
Skylake nv12 format requires dbuf (aka. ddb) calculations and programming for each of y and uv sub-planes. Made minor changes to reuse current dbuf calculations and programming for uv plane. i.e., with this change, existing computation is used for either packed format or uv portion of nv12 depending on incoming format. Added new code for dbuf computation and programming for y plane.
This patch is a pre-requisite for adding NV12 format support. Actual nv12 support is coming in later patches.
Signed-off-by Chandra Konduru <[email protected]>
Quoted 7 lines
Just aside (and not a blocker for this patch really). Do we have nv12 subtest in all the relevant plane tests (rotation, scaling, ...) to exercise this code? Not sure whether we can redo the
yuv->rgbconversion correctly in software to match hw, might need to pick special colors that convert accurately.Anyway given how often we've broken the sprite code I really think we need full functional tests (i.e. including crc checks).
I have written an i-g-t test kms_nv12 to test/exercise this patch and subsequent NV12 patch series which was submitted last week after
this patch. It tests nv12 along with rotation, scaling and does
some crc checks.
But because chrome upsampling is done using scaler which uses multiple taps, crcs aren't going to match with original reference flip with RGB format. I tried with simple solid color. Can you please take a look that?
Quoted 5 lines
-Daniel
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoted 31 lines · Chandra
Quoted 23 lines
-----Original Message-----
From: Daniel Vetter [mailto:[email protected]] On Behalf Of Daniel Vetter
Sent: Monday, May 04, 2015 8:01 AM
To: Konduru, Chandra
Cc: [email protected]; Vetter, Daniel; Syrjala, Ville
Subject: Re: [Intel-gfx] [PATCH] drm/i915: Adding dbuf support for skl nv12 format.Quoted 11 lines · Chandra Konduru
Skylake nv12 format requires dbuf (aka. ddb) calculations and programming for each of y and uv sub-planes. Made minor changes to reuse current dbuf calculations and programming for uv plane. i.e., with this change, existing computation is used for either packed format or uv portion of nv12 depending on incoming format. Added new code for dbuf computation and programming for y plane.
This patch is a pre-requisite for adding NV12 format support. Actual nv12 support is coming in later patches.
Signed-off-by Chandra Konduru <[email protected]>Quoted 7 lines
Just aside (and not a blocker for this patch really). Do we have nv12 subtest in all the relevant plane tests (rotation, scaling, ...) to exercise this code? Not sure whether we can redo the
yuv->rgbconversion correctly in software to match hw, might need to pick special colors that convert accurately.Anyway given how often we've broken the sprite code I really think we need full functional tests (i.e. including crc checks).
Quoted 7 lines
I have written an i-g-t test
kms_nv12to test/exercise this patch and subsequent NV12 patch series which was submitted last week after
this patch. It tests nv12 along with rotation, scaling and does
some crc checks.
But because chrome upsampling is done using scaler which uses multiple taps, crcs aren't going to match with original reference flip with RGB format. I tried with simple solid color. Can you please take a look that?
Oh right the upsampling will result in a smudge when we change visible colors, so the 4 color pattern used for checking rotation doesn't work. Just an idea (without yet looking at your code):
- Create a tall&narrow sprite view, and a big buffer which would fit that sprite view in any orentiation.
- Only draw a slice exactly matching the sprite view (horizontal/vertical) in one solid color, leave the remaining buffer black.
That way we can at least do some basic checks to make sure the rotation/alignment is correct. We won't be able to detect when the buffer is mirrored wrongly, but at least we'll have some coverage. -Daniel
Signature · 7 lines
Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Hi Chandra,
Quoted 8 lines · Chandra Konduru
Skylake nv12 format requires dbuf (aka. ddb) calculations and programming for each of y and uv sub-planes. Made minor changes to reuse current dbuf calculations and programming for uv plane. i.e., with this change, existing computation is used for either packed format or uv portion of nv12 depending on incoming format. Added new code for dbuf computation and programming for y plane.
Quoted 4 lines
This patch is a pre-requisite for adding NV12 format support. Actual nv12 support is coming in later patches.
Signed-off-by Chandra Konduru <[email protected]>
I found a few things to talk about. Nothing major, except maybe that the cof for Y tiled buffers minimum allocation that is not there and would need to take the UV plane size into account.
A more general remark though, is that I believe we should make the code more generic and talk about fb planes 0 and 1 instead of having the
if (y_bytes_per_pixel)
everywhere. It'd be a good fit for the WM code as we're always interested in fb plane 0 (either RGB or Y plane). Concretely it means having per-fb plane variables in the structures, ie bytes_per_pixel[2], indexed by the fb plane number.
Otherwise the rest looks good to me. I'd love to see the more generic version (especially the remark about deriving the sub-sampling factors from the format), but this can be done in tree if you want.
Quoted 78 lines
--- drivers/gpu/drm/i915/i915_drv./h | 3 +- drivers/gpu/drm/i915/i915_reg.h | 11 ++++++ drivers/gpu/drm/i915/intel_drv.h | 8 ++++ drivers/gpu/drm/i915/intel_pm.c | 79 ++++++++++++++++++++++++++++++++------ 4 files changed, 88 insertions(+), 13 deletions(-)drivers/gpu/drm/i915/i915_drv.h +2
-1@@ -1450,7 +1450,8 @@ static inline bool skl_ddb_entry_equal(const struct skl_ddb_entry *e1,struct skl_ddb_allocation {struct skl_ddb_entry pipe[I915_MAX_PIPES ]; -struct skl_ddb_entry plane[I915_MAX_PIPES ][I915_MAX_PLANES ]; +struct skl_ddb_entry plane[I915_MAX_PIPES ][I915_MAX_PLANES ];/* packed/uv */ +struct skl_ddb_entry y_plane[I915_MAX_PIPES ][I915_MAX_PLANES ];/* y-plane */ struct skl_ddb_entry cursor[I915_MAX_PIPES ]; };drivers/gpu/drm/i915/i915_reg.h +11
-0@@ -4880,6 +4880,8 @@ enum skl_disp_power_wells {#define _PLANE_KEYMAX_2_A 0x702a0#define _PLANE_BUF_CFG_1_A 0x7027c#define _PLANE_BUF_CFG_2_A 0x7037c +#define _PLANE_NV12_BUF_CFG_1_A 0x70278 +#define _PLANE_NV12_BUF_CFG_2_A 0x70378#define _PLANE_CTL_1_B 0x71180#define _PLANE_CTL_2_B 0x71280@@ -4966,6 +4968,15 @@ enum skl_disp_power_wells {#define PLANE_BUF_CFG (pipe, plane) \ _PLANE(plane, _PLANE_BUF_CFG_1(pipe), _PLANE_BUF_CFG_2(pipe)) +#define _PLANE_NV12_BUF_CFG_1_B 0x71278 +#define _PLANE_NV12_BUF_CFG_2_B 0x71378 +#define _PLANE_NV12_BUF_CFG_1 (pipe) \ + _PIPE(pipe, _PLANE_NV12_BUF_CFG_1_A, _PLANE_NV12_BUF_CFG_1_B) +#define _PLANE_NV12_BUF_CFG_2 (pipe) \ + _PIPE(pipe, _PLANE_NV12_BUF_CFG_2_A, _PLANE_NV12_BUF_CFG_2_B) +#define PLANE_NV12_BUF_CFG (pipe, plane) \ + _PLANE(plane, _PLANE_NV12_BUF_CFG_1(pipe), _PLANE_NV12_BUF_CFG_2(pipe)) +/* SKL new cursor registers */ #define _CUR_BUF_CFG_A 0x7017c#define _CUR_BUF_CFG_B 0x7117cdrivers/gpu/drm/i915/intel_drv.h +8
-0@@ -566,7 +566,15 @@ struct intel_crtc {struct intel_plane_wm_parameters {uint32_t horiz_pixels;uint32_t vert_pixels; +/* +* For packed pixel formats: +* bytes_per_pixel - holds bytes per pixel +* For planar pixel formats: +* bytes_per_pixel - holds bytes per pixel for uv-plane +* y_bytes_per_pixel - holds bytes per pixel for y-plane +*/ uint8_t bytes_per_pixel; +uint8_t y_bytes_per_pixel;bool enabled;bool scaled;u64 tiling;drivers/gpu/drm/i915/intel_pm.c +0
-0@@ -2616,8 +2616,18 @@ void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv, } static unsigned int
drivers/gpu/drm/i915/intel_pm.c void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv, · line 2627
Quoted 10 lines
drivers/gpu/drm/i915/intel_pm.c +8
-1-skl_plane_relative_data_rate(const struct intel_plane_wm_parameters *p) +skl_plane_relative_data_rate(const struct intel_plane_wm_parameters *p, int y) { + +/* for planar format */ +if (p->y_bytes_per_pixel) { +if (y)/* y-plane data rate */ +return p->horiz_pixels * p->vert_pixels * p->y_bytes_per_pixel; +else /* uv-plane data rate */ +return (p->horiz_pixels/2 ) * (p->vert_pixels/2 ) * p->bytes_per_pixel;
The knowledge that the UV plane is sub-sampled by 2 in both directions is hidden there, we may want to add a new drm_ function to give the sub-sample factors depending on the format and the plane number. This way, the code would more generic if the hw ever adds support for more planar formats. If we still want to cache the parameters in intel_plane_wm_parameters, I think we should have per-plane (ie the surfaces in the fb) bytes_per_pixel, width, height, ...
I'd also try use something a bit more general rather than the 'y' parameter, maybe the plane number (0->y, 1->UV in the NV12 case 0->RGB otherwise).
drivers/gpu/drm/i915/intel_pm.c void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv, · line 2632
Quoted 5 lines
drivers/gpu/drm/i915/intel_pm.c +3
-0+ } + +/* for packed formats */ return p->horiz_pixels * p->vert_pixels * p->bytes_per_pixel; }
Oh, and something else I noticed, we apparently need to compute a scaling factor in there. As you've looked at the scalers for SKL, you may be interested in looking into it.
Quoted 22 lines
drivers/gpu/drm/i915/intel_pm.c +5
-1@@ -2640,7 +2650,10 @@ skl_get_total_relative_data_rate(struct intel_crtc *intel_crtc,if (!p->enabled)continue ; - total_data_rate +=skl_plane_relative_data_rate (p); + total_data_rate +=skl_plane_relative_data_rate (p,0 );/* packed/uv */ +if (p->y_bytes_per_pixel) { + total_data_rate +=skl_plane_relative_data_rate (p,1 );/* y-plane */ + } }return total_data_rate;@@ -2659,6 +2672,7 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc,struct skl_ddb_entry *alloc = &ddb->pipe[pipe];uint16_t alloc_size, start, cursor_blocks;uint16_t minimum[I915_MAX_PLANES ]; +uint16_t y_minimum[I915_MAX_PLANES ];unsigned int total_data_rate;int plane;
drivers/gpu/drm/i915/intel_pm.c skl_allocate_pipe_ddb(struct drm_crtc *crtc, · line 2706
Quoted 6 lines
drivers/gpu/drm/i915/intel_pm.c +2
-0@@ -2687,6 +2701,8 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc, minimum[plane] =8 ; alloc_size -= minimum[plane]; + y_minimum[plane] = p->y_bytes_per_pixel ?8 :0 ; + alloc_size -= y_minimum[plane]; }
Hum, this made me look at the DDB allocation page and it seems we're missing the code to compute the minimum allocate for Y tiled FBs. It's somewhat important here as that code is using the width of the plane, width that would be halved for a Y tiled UV plane.
Quoted 48 lines
drivers/gpu/drm/i915/intel_pm.c +20
-3/* @@ -2700,16 +2716,17 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc, start = alloc->start;for (plane =0 ; plane <intel_num_planes (intel_crtc); plane++) {const struct intel_plane_wm_parameters *p; -unsigned int data_rate; -uint16_t plane_blocks; +unsigned int data_rate, y_data_rate; +uint16_t plane_blocks, y_plane_blocks =0 ; p = ¶ms->plane[plane];if (!p->enabled)continue ; - data_rate =skl_plane_relative_data_rate (p); + data_rate =skl_plane_relative_data_rate (p,0 );/* +* allocation for (packed formats) or (uv-plane part of planar format): * promote the expression to 64 bits to avoid overflowing, the * result is < available as data_rate / total_data_rate < 1 */ @@ -2721,6 +2738,22 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc, ddb->plane[pipe][plane].end = start + plane_blocks; start += plane_blocks; + +/* +* allocation for y_plane part of planar format: +*/ +if (p->y_bytes_per_pixel) { + y_data_rate =skl_plane_relative_data_rate (p,1 ); + y_plane_blocks = y_minimum[plane]; + y_plane_blocks +=div_u64 ((uint64_t )alloc_size * y_data_rate, + total_data_rate); + + ddb->y_plane[pipe][plane].start = start; + ddb->y_plane[pipe][plane].end = start + y_plane_blocks; + + start += y_plane_blocks; + } + } }@@ -2833,13 +2866,18 @@ static void skl_compute_wm_pipe_parameters(struct drm_crtc *crtc, p->pixel_rate =skl_pipe_pixel_rate (intel_crtc->config);
drivers/gpu/drm/i915/intel_pm.c static void skl_compute_wm_pipe_parameters(struct drm_crtc *crtc, · line 2875
Quoted 10 lines
drivers/gpu/drm/i915/intel_pm.c +5
-1fb = crtc->primary->state->fb; +/* For planar: Bpp is for uv plane, y_Bpp is for y plane */ if (fb) { p->plane[0 ].enabled = true; - p->plane[0 ].bytes_per_pixel = fb->bits_per_pixel /8 ; + p->plane[0 ].bytes_per_pixel = fb->pixel_format ==DRM_FORMAT_NV12 ? +drm_format_plane_cpp (fb->pixel_format,1 ) : fb->bits_per_pixel /8 ; + p->plane[0 ].y_bytes_per_pixel = fb->pixel_format ==DRM_FORMAT_NV12 ? +drm_format_plane_cpp (fb->pixel_format,0 ) :0 ;
drm_format_plane_cpp() seems general enough to always work?
Quoted 75 lines
drivers/gpu/drm/i915/intel_pm.c +15
-6p->plane[0 ].tiling = fb->modifier[0 ]; }else { p->plane[0 ].enabled = false; p->plane[0 ].bytes_per_pixel =0 ; + p->plane[0 ].y_bytes_per_pixel =0 ; p->plane[0 ].tiling =DRM_FORMAT_MOD_NONE ; } p->plane[0 ].horiz_pixels = intel_crtc->config->pipe_src_w;@@ -2847,6 +2885,7 @@ static void skl_compute_wm_pipe_parameters(struct drm_crtc *crtc, p->plane[0 ].rotation = crtc->primary->state->rotation; fb = crtc->cursor->state->fb; + p->cursor.y_bytes_per_pixel =0 ;if (fb) { p->cursor.enabled = true; p->cursor.bytes_per_pixel = fb->bits_per_pixel /8 ;@@ -2882,22 +2921,25 @@ static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,uint32_t plane_bytes_per_line, plane_blocks_per_line;uint32_t res_blocks, res_lines;uint32_t selected_result; +uint8_t bytes_per_pixel;if (latency ==0 || !p->active || !p_params->enabled)return false; + bytes_per_pixel = p_params->y_bytes_per_pixel ? + p_params->y_bytes_per_pixel : + p_params->bytes_per_pixel; method1 =skl_wm_method1 (p->pixel_rate, - p_params->bytes_per_pixel, + bytes_per_pixel, latency); method2 =skl_wm_method2 (p->pixel_rate, p->pipe_htotal, p_params->horiz_pixels, - p_params->bytes_per_pixel, + bytes_per_pixel, p_params->tiling, latency); - plane_bytes_per_line = p_params->horiz_pixels * - p_params->bytes_per_pixel; + plane_bytes_per_line = p_params->horiz_pixels * bytes_per_pixel; plane_blocks_per_line =DIV_ROUND_UP (plane_bytes_per_line,512 ); if (p_params->tiling == I915_FORMAT_MOD_Y_TILED ||@@ -3114,10 +3156,14 @@ static void skl_write_wm_values(struct drm_i915_private *dev_priv, new->plane_trans[pipe][i]);I915_WRITE (CUR_WM_TRANS (pipe), new->cursor_trans[pipe]); -for (i =0 ; i <intel_num_planes (crtc); i++) +for (i =0 ; i <intel_num_planes (crtc); i++) {skl_ddb_entry_write (dev_priv,PLANE_BUF_CFG (pipe, i), &new->ddb.plane[pipe][i]); +skl_ddb_entry_write (dev_priv, +PLANE_NV12_BUF_CFG (pipe, i), + &new->ddb.y_plane[pipe][i]); + }skl_ddb_entry_write (dev_priv,CUR_BUF_CFG (pipe), &new->ddb.cursor[pipe]);@@ -3275,6 +3321,7 @@ static bool skl_update_pipe_wm(struct drm_crtc *crtc,return false; intel_crtc->wm.skl_active = *pipe_wm; +return true; }@@ -3368,8 +3415,16 @@ skl_update_sprite_wm(struct drm_plane *plane, struct drm_crtc *crtc, intel_plane->wm.scaled = scaled; intel_plane->wm.horiz_pixels = sprite_width; intel_plane->wm.vert_pixels = sprite_height; - intel_plane->wm.bytes_per_pixel = pixel_size;
drivers/gpu/drm/i915/intel_pm.c skl_update_sprite_wm(struct drm_plane *plane, struct drm_crtc *crtc, · line 3427
Quoted 10 lines
drivers/gpu/drm/i915/intel_pm.c +9
-0intel_plane->wm.tiling =DRM_FORMAT_MOD_NONE ; + +/* For planar: Bpp is for UV plane, y_Bpp is for Y plane */ + intel_plane->wm.bytes_per_pixel = + (fb && fb->pixel_format ==DRM_FORMAT_NV12 ) ? +drm_format_plane_cpp (plane->state->fb->pixel_format,1 ) : pixel_size; + intel_plane->wm.y_bytes_per_pixel = + (fb && fb->pixel_format ==DRM_FORMAT_NV12 ) ? +drm_format_plane_cpp (plane->state->fb->pixel_format,0 ) :0 ; +
drm_format_plane_cpp() seems general enough to always work?
Quoted 6 lines
drivers/gpu/drm/i915/intel_pm.c +0
-0/* * Framebuffer can be NULL on plane disable, but it does not * matter for watermarks if we assume no tiling in that case. Signature · 1 line
1.7.9.5
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoted 8 lines · Chandra Konduru
Skylake nv12 format requires dbuf (aka. ddb) calculations and programming for each of y and uv sub-planes. Made minor changes to reuse current dbuf calculations and programming for uv plane. i.e., with this change, existing computation is used for either packed format or uv portion of nv12 depending on incoming format. Added new code for dbuf computation and programming for y plane.
Quoted 4 lines
This patch is a pre-requisite for adding NV12 format support. Actual nv12 support is coming in later patches.
Signed-off-by Chandra Konduru <[email protected]>
Talked with Chandra about the previous comments. The Y tiling problem I was mentioning isn't quite part of NV12 and can be fixed on top of this. I'll probaby try my suggestions and create a series on top of this patch.
Signature · 1 line
Damien
Quoted 270 lines
--- drivers/gpu/drm/i915/i915_drv.h | 3 +- drivers/gpu/drm/i915/i915_reg.h | 11 ++++++ drivers/gpu/drm/i915/intel_drv.h | 8 ++++ drivers/gpu/drm/i915/intel_pm.c | 79 ++++++++++++++++++++++++++++++++------ 4 files changed, 88 insertions(+), 13 deletions(-)drivers/gpu/drm/i915/i915_drv.h +2
-1@@ -1450,7 +1450,8 @@ static inline bool skl_ddb_entry_equal(const struct skl_ddb_entry *e1,struct skl_ddb_allocation {struct skl_ddb_entry pipe[I915_MAX_PIPES ]; -struct skl_ddb_entry plane[I915_MAX_PIPES ][I915_MAX_PLANES ]; +struct skl_ddb_entry plane[I915_MAX_PIPES ][I915_MAX_PLANES ];/* packed/uv */ +struct skl_ddb_entry y_plane[I915_MAX_PIPES ][I915_MAX_PLANES ];/* y-plane */ struct skl_ddb_entry cursor[I915_MAX_PIPES ]; };drivers/gpu/drm/i915/i915_reg.h +11
-0@@ -4880,6 +4880,8 @@ enum skl_disp_power_wells {#define _PLANE_KEYMAX_2_A 0x702a0#define _PLANE_BUF_CFG_1_A 0x7027c#define _PLANE_BUF_CFG_2_A 0x7037c +#define _PLANE_NV12_BUF_CFG_1_A 0x70278 +#define _PLANE_NV12_BUF_CFG_2_A 0x70378#define _PLANE_CTL_1_B 0x71180#define _PLANE_CTL_2_B 0x71280@@ -4966,6 +4968,15 @@ enum skl_disp_power_wells {#define PLANE_BUF_CFG (pipe, plane) \ _PLANE(plane, _PLANE_BUF_CFG_1(pipe), _PLANE_BUF_CFG_2(pipe)) +#define _PLANE_NV12_BUF_CFG_1_B 0x71278 +#define _PLANE_NV12_BUF_CFG_2_B 0x71378 +#define _PLANE_NV12_BUF_CFG_1 (pipe) \ + _PIPE(pipe, _PLANE_NV12_BUF_CFG_1_A, _PLANE_NV12_BUF_CFG_1_B) +#define _PLANE_NV12_BUF_CFG_2 (pipe) \ + _PIPE(pipe, _PLANE_NV12_BUF_CFG_2_A, _PLANE_NV12_BUF_CFG_2_B) +#define PLANE_NV12_BUF_CFG (pipe, plane) \ + _PLANE(plane, _PLANE_NV12_BUF_CFG_1(pipe), _PLANE_NV12_BUF_CFG_2(pipe)) +/* SKL new cursor registers */ #define _CUR_BUF_CFG_A 0x7017c#define _CUR_BUF_CFG_B 0x7117cdrivers/gpu/drm/i915/intel_drv.h +8
-0@@ -566,7 +566,15 @@ struct intel_crtc {struct intel_plane_wm_parameters {uint32_t horiz_pixels;uint32_t vert_pixels; +/* +* For packed pixel formats: +* bytes_per_pixel - holds bytes per pixel +* For planar pixel formats: +* bytes_per_pixel - holds bytes per pixel for uv-plane +* y_bytes_per_pixel - holds bytes per pixel for y-plane +*/ uint8_t bytes_per_pixel; +uint8_t y_bytes_per_pixel;bool enabled;bool scaled;u64 tiling;drivers/gpu/drm/i915/intel_pm.c +67
-12@@ -2616,8 +2616,18 @@ void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv, }static unsigned int -skl_plane_relative_data_rate (const struct intel_plane_wm_parameters *p) +skl_plane_relative_data_rate (const struct intel_plane_wm_parameters *p,int y) { + +/* for planar format */ +if (p->y_bytes_per_pixel) { +if (y)/* y-plane data rate */ +return p->horiz_pixels * p->vert_pixels * p->y_bytes_per_pixel; +else /* uv-plane data rate */ +return (p->horiz_pixels/2 ) * (p->vert_pixels/2 ) * p->bytes_per_pixel; + } + +/* for packed formats */ return p->horiz_pixels * p->vert_pixels * p->bytes_per_pixel; }@@ -2640,7 +2650,10 @@ skl_get_total_relative_data_rate(struct intel_crtc *intel_crtc,if (!p->enabled)continue ; - total_data_rate +=skl_plane_relative_data_rate (p); + total_data_rate +=skl_plane_relative_data_rate (p,0 );/* packed/uv */ +if (p->y_bytes_per_pixel) { + total_data_rate +=skl_plane_relative_data_rate (p,1 );/* y-plane */ + } }return total_data_rate;@@ -2659,6 +2672,7 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc,struct skl_ddb_entry *alloc = &ddb->pipe[pipe];uint16_t alloc_size, start, cursor_blocks;uint16_t minimum[I915_MAX_PLANES ]; +uint16_t y_minimum[I915_MAX_PLANES ];unsigned int total_data_rate;int plane;@@ -2687,6 +2701,8 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc, minimum[plane] =8 ; alloc_size -= minimum[plane]; + y_minimum[plane] = p->y_bytes_per_pixel ?8 :0 ; + alloc_size -= y_minimum[plane]; }/* @@ -2700,16 +2716,17 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc, start = alloc->start;for (plane =0 ; plane <intel_num_planes (intel_crtc); plane++) {const struct intel_plane_wm_parameters *p; -unsigned int data_rate; -uint16_t plane_blocks; +unsigned int data_rate, y_data_rate; +uint16_t plane_blocks, y_plane_blocks =0 ; p = ¶ms->plane[plane];if (!p->enabled)continue ; - data_rate =skl_plane_relative_data_rate (p); + data_rate =skl_plane_relative_data_rate (p,0 );/* +* allocation for (packed formats) or (uv-plane part of planar format): * promote the expression to 64 bits to avoid overflowing, the * result is < available as data_rate / total_data_rate < 1 */ @@ -2721,6 +2738,22 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc, ddb->plane[pipe][plane].end = start + plane_blocks; start += plane_blocks; + +/* +* allocation for y_plane part of planar format: +*/ +if (p->y_bytes_per_pixel) { + y_data_rate =skl_plane_relative_data_rate (p,1 ); + y_plane_blocks = y_minimum[plane]; + y_plane_blocks +=div_u64 ((uint64_t )alloc_size * y_data_rate, + total_data_rate); + + ddb->y_plane[pipe][plane].start = start; + ddb->y_plane[pipe][plane].end = start + y_plane_blocks; + + start += y_plane_blocks; + } + } }@@ -2833,13 +2866,18 @@ static void skl_compute_wm_pipe_parameters(struct drm_crtc *crtc, p->pixel_rate =skl_pipe_pixel_rate (intel_crtc->config); fb = crtc->primary->state->fb; +/* For planar: Bpp is for uv plane, y_Bpp is for y plane */ if (fb) { p->plane[0 ].enabled = true; - p->plane[0 ].bytes_per_pixel = fb->bits_per_pixel /8 ; + p->plane[0 ].bytes_per_pixel = fb->pixel_format ==DRM_FORMAT_NV12 ? +drm_format_plane_cpp (fb->pixel_format,1 ) : fb->bits_per_pixel /8 ; + p->plane[0 ].y_bytes_per_pixel = fb->pixel_format ==DRM_FORMAT_NV12 ? +drm_format_plane_cpp (fb->pixel_format,0 ) :0 ; p->plane[0 ].tiling = fb->modifier[0 ]; }else { p->plane[0 ].enabled = false; p->plane[0 ].bytes_per_pixel =0 ; + p->plane[0 ].y_bytes_per_pixel =0 ; p->plane[0 ].tiling =DRM_FORMAT_MOD_NONE ; } p->plane[0 ].horiz_pixels = intel_crtc->config->pipe_src_w;@@ -2847,6 +2885,7 @@ static void skl_compute_wm_pipe_parameters(struct drm_crtc *crtc, p->plane[0 ].rotation = crtc->primary->state->rotation; fb = crtc->cursor->state->fb; + p->cursor.y_bytes_per_pixel =0 ;if (fb) { p->cursor.enabled = true; p->cursor.bytes_per_pixel = fb->bits_per_pixel /8 ;@@ -2882,22 +2921,25 @@ static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,uint32_t plane_bytes_per_line, plane_blocks_per_line;uint32_t res_blocks, res_lines;uint32_t selected_result; +uint8_t bytes_per_pixel;if (latency ==0 || !p->active || !p_params->enabled)return false; + bytes_per_pixel = p_params->y_bytes_per_pixel ? + p_params->y_bytes_per_pixel : + p_params->bytes_per_pixel; method1 =skl_wm_method1 (p->pixel_rate, - p_params->bytes_per_pixel, + bytes_per_pixel, latency); method2 =skl_wm_method2 (p->pixel_rate, p->pipe_htotal, p_params->horiz_pixels, - p_params->bytes_per_pixel, + bytes_per_pixel, p_params->tiling, latency); - plane_bytes_per_line = p_params->horiz_pixels * - p_params->bytes_per_pixel; + plane_bytes_per_line = p_params->horiz_pixels * bytes_per_pixel; plane_blocks_per_line =DIV_ROUND_UP (plane_bytes_per_line,512 ); if (p_params->tiling == I915_FORMAT_MOD_Y_TILED ||@@ -3114,10 +3156,14 @@ static void skl_write_wm_values(struct drm_i915_private *dev_priv, new->plane_trans[pipe][i]);I915_WRITE (CUR_WM_TRANS (pipe), new->cursor_trans[pipe]); -for (i =0 ; i <intel_num_planes (crtc); i++) +for (i =0 ; i <intel_num_planes (crtc); i++) {skl_ddb_entry_write (dev_priv,PLANE_BUF_CFG (pipe, i), &new->ddb.plane[pipe][i]); +skl_ddb_entry_write (dev_priv, +PLANE_NV12_BUF_CFG (pipe, i), + &new->ddb.y_plane[pipe][i]); + }skl_ddb_entry_write (dev_priv,CUR_BUF_CFG (pipe), &new->ddb.cursor[pipe]);@@ -3275,6 +3321,7 @@ static bool skl_update_pipe_wm(struct drm_crtc *crtc,return false; intel_crtc->wm.skl_active = *pipe_wm; +return true; }@@ -3368,8 +3415,16 @@ skl_update_sprite_wm(struct drm_plane *plane, struct drm_crtc *crtc, intel_plane->wm.scaled = scaled; intel_plane->wm.horiz_pixels = sprite_width; intel_plane->wm.vert_pixels = sprite_height; - intel_plane->wm.bytes_per_pixel = pixel_size; intel_plane->wm.tiling =DRM_FORMAT_MOD_NONE ; + +/* For planar: Bpp is for UV plane, y_Bpp is for Y plane */ + intel_plane->wm.bytes_per_pixel = + (fb && fb->pixel_format ==DRM_FORMAT_NV12 ) ? +drm_format_plane_cpp (plane->state->fb->pixel_format,1 ) : pixel_size; + intel_plane->wm.y_bytes_per_pixel = + (fb && fb->pixel_format ==DRM_FORMAT_NV12 ) ? +drm_format_plane_cpp (plane->state->fb->pixel_format,0 ) :0 ; +/* * Framebuffer can be NULL on plane disable, but it does not * matter for watermarks if we assume no tiling in that case. Signature · 1 line
1.7.9.5
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoted 12 lines · Damien Lespiau
Quoted 11 lines · Chandra Konduru
Skylake nv12 format requires dbuf (aka. ddb) calculations and programming for each of y and uv sub-planes. Made minor changes to reuse current dbuf calculations and programming for uv plane. i.e., with this change, existing computation is used for either packed format or uv portion of nv12 depending on incoming format. Added new code for dbuf computation and programming for y plane.
This patch is a pre-requisite for adding NV12 format support. Actual nv12 support is coming in later patches.
Quoted 6 lines
Quoted 1 line
Signed-off-by Chandra Konduru <[email protected]>Talked with Chandra about the previous comments. The Y tiling problem I was mentioning isn't quite part of NV12 and can be fixed on top of this. I'll probaby try my suggestions and create a series on top of this patch.
Wrt unifying the wm code, I honestly wonder whether we need our current indirection. I kinda made sense before universal planes and atomic. But know we have drm_plane_state subclassed. So imo the better approach would be to directly fish things out of there, or if _really_ need add something to intel_plane_state. But on a quick look a few weeks ago most of the wm structs seemed redundant with drm_plane_state or plane_state->fb.
Cheers, Daniel
Signature · 7 lines
Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoted 12 lines · Damien Lespiau
Quoted 11 lines · Chandra Konduru
Skylake nv12 format requires dbuf (aka. ddb) calculations and programming for each of y and uv sub-planes. Made minor changes to reuse current dbuf calculations and programming for uv plane. i.e., with this change, existing computation is used for either packed format or uv portion of nv12 depending on incoming format. Added new code for dbuf computation and programming for y plane.
This patch is a pre-requisite for adding NV12 format support. Actual nv12 support is coming in later patches.
Quoted 8 lines
Quoted 1 line
Signed-off-by Chandra Konduru <[email protected]>Talked with Chandra about the previous comments. The Y tiling problem I was mentioning isn't quite part of NV12 and can be fixed on top of this. I'll probaby try my suggestions and create a series on top of this patch.
Reviewed-by Damien Lespiau <[email protected]>
Oh and ofc: Queued for -next, thanks for the patch. -Daniel
Signature · 7 lines
Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx