From: pengcheng chen Date: Thu, 1 Nov 2018 12:22:29 +0000 (+0800) Subject: osd: osd shift-workaround one line fixed [1/1] X-Git-Tag: khadas-vims-v0.9.6-release~1156 X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=e0063c417bd64d12ced78d4c7462d16d4bc92814;p=platform%2Fkernel%2Flinux-amlogic.git osd: osd shift-workaround one line fixed [1/1] PD#163001 Problem: current osd code have one line missing issue in meson-hwc Solution: osd blend used +1 line to fix shift issue, and used freescale to cut it. Verify: verified in u200 board Change-Id: I2354742d456327d9f1aa5713ee782648895034db Signed-off-by: pengcheng chen --- diff --git a/drivers/amlogic/media/osd/osd_hw.c b/drivers/amlogic/media/osd/osd_hw.c index a6903eb..762a151 100644 --- a/drivers/amlogic/media/osd/osd_hw.c +++ b/drivers/amlogic/media/osd/osd_hw.c @@ -4404,6 +4404,7 @@ static void osd_update_disp_freescale_enable(u32 index) int vf_bank_len = 4; struct hw_osd_reg_s *osd_reg = &hw_osd_reg_array[index]; u32 data32 = 0x0; + u32 shift_workaround = 0; if (osd_hw.osd_meson_dev.osd_ver != OSD_HIGH_ONE) osd_reg = &hw_osd_reg_array[0]; @@ -4413,6 +4414,13 @@ static void osd_update_disp_freescale_enable(u32 index) else vf_bank_len = 4; + if (osd_hw.hwc_enable && (index == OSD1) + && ((osd_hw.osd_meson_dev.cpu_id == + __MESON_CPU_MAJOR_ID_G12A) || + (osd_hw.osd_meson_dev.cpu_id == + __MESON_CPU_MAJOR_ID_G12B))) + shift_workaround = 1; + #ifndef NEW_PPS_PHASE if (osd_hw.bot_type == 1) { vsc_bot_rcv_num = 4; @@ -4462,7 +4470,10 @@ static void osd_update_disp_freescale_enable(u32 index) hf_phase_step = (src_w << 18) / dst_w; hf_phase_step = (hf_phase_step << 6); - vf_phase_step = (src_h << 20) / dst_h; + if (shift_workaround) + vf_phase_step = ((src_h - 1) << 20) / dst_h; + else + vf_phase_step = (src_h << 20) / dst_h; #ifdef NEW_PPS_PHASE if (osd_hw.field_out_en) { @@ -4506,6 +4517,13 @@ static void osd_update_disp_freescale_enable(u32 index) vf_phase_step = (vf_phase_step << 4); /* config osd scaler in/out hv size */ data32 = 0x0; + if (shift_workaround) { + vsc_ini_rcv_num++; + if (osd_hw.field_out_en) + vsc_bot_rcv_num++; + } + + if (osd_hw.free_scale_enable[index]) { data32 = (((src_h - 1) & 0x1fff) | ((src_w - 1) & 0x1fff) << 16); @@ -6041,8 +6059,13 @@ static void osd_setting_blend1(struct hw_osd_blending_s *blending) return; if (osd_hw.hdr_used) workaround_line = osd_hw.workaround_hdr; - else - workaround_line = osd_hw.workaround_not_hdr; + else { + if (blending->layer_cnt == 2) + workaround_line = 0; + else + workaround_line = osd_hw.workaround_not_hdr; + } + layer_blend = &(blending->layer_blend); blend_reg = &(blending->blend_reg); #ifdef OSD_BLEND_SHIFT_WORKAROUND @@ -6567,8 +6590,7 @@ static void osd_setting_blend0_input(u32 index, layer_blend->input1_data.w = osd_hw.src_data[index].w; layer_blend->input1_data.h = - osd_hw.src_data[index].h - - workaround_line; + osd_hw.src_data[index].h; } else { #ifdef OSD_BLEND_SHIFT_WORKAROUND layer_blend->input1_data.x = 0; @@ -6579,14 +6601,12 @@ static void osd_setting_blend0_input(u32 index, + workaround_line; #endif layer_blend->input1_data.w = osd_hw.src_data[index].w; - layer_blend->input1_data.h = osd_hw.src_data[index].h - - workaround_line; + layer_blend->input1_data.h = osd_hw.src_data[index].h; } background_w = (osd_hw.disp_info.position_w * blending->pic_w_ratio) >> OSD_CALC; background_h = (osd_hw.disp_info.position_h * - blending->pic_h_ratio >> OSD_CALC) - - workaround_line; + blending->pic_h_ratio >> OSD_CALC); } else { layer_blend->input1_data.x = osd_hw.free_dst_data[index].x_start; @@ -6598,11 +6618,11 @@ static void osd_setting_blend0_input(u32 index, osd_hw.free_dst_data[index].x_start + 1; layer_blend->input1_data.h = osd_hw.free_dst_data[index].y_end - - osd_hw.free_dst_data[index].y_start + - 1 - workaround_line; + osd_hw.free_dst_data[index].y_start + 1; background_w = layer_blend->input1_data.w; background_h = layer_blend->input1_data.h; } + layer_blend->background_w = background_w; layer_blend->background_h = background_h; osd_log_dbg2("index=%d,src_data: x=%d,y=%d,w=%d,h=%d\n", @@ -6624,9 +6644,6 @@ static void set_blend_path(struct hw_osd_blending_s *blending) struct dispdata_s output1_data; u32 index = 0; u8 input1 = 0, input2 = 0; -#ifdef OSD_BLEND_SHIFT_WORKAROUND - u32 workaround_line = 1; -#endif if (!blending) return; @@ -7190,8 +7207,6 @@ static void set_blend_path(struct hw_osd_blending_s *blending) layer_blend->input2 = BLEND1_DIN; memcpy(&layer_blend->input1_data, &output1_data, sizeof(struct dispdata_s)); - layer_blend->input2_data.y += workaround_line; - layer_blend->input2_data.h -= workaround_line; } else { layer_blend->input1 = BLEND1_DIN; layer_blend->input2 = BLEND2_DIN; @@ -7201,8 +7216,6 @@ static void set_blend_path(struct hw_osd_blending_s *blending) memcpy(&layer_blend->input2_data, &output1_data, sizeof(struct dispdata_s)); - layer_blend->input1_data.y += workaround_line; - layer_blend->input1_data.h -= workaround_line; } vpp_setting_blend(blending); @@ -7463,9 +7476,9 @@ static void osd_setting_default_hwc(void) /* Do later: different format select different dummy_data */ /* used default dummy data */ VSYNCOSD_WR_MPEG_REG(VIU_OSD_BLEND_DUMMY_DATA0, - 0x0 << 16 | - 0x0 << 8 | - 0x0); + 0x80 << 16 | + 0x80 << 8 | + 0x80); /* used default dummy alpha data */ VSYNCOSD_WR_MPEG_REG(VIU_OSD_BLEND_DUMMY_ALPHA, 0x0 << 20 |