diff options
Diffstat (limited to 'drivers/gpu/drm/radeon')
61 files changed, 1577 insertions, 590 deletions
diff --git a/drivers/gpu/drm/radeon/atombios.h b/drivers/gpu/drm/radeon/atombios.h index af10f8571d87..92be50c39ffd 100644 --- a/drivers/gpu/drm/radeon/atombios.h +++ b/drivers/gpu/drm/radeon/atombios.h @@ -1711,7 +1711,9 @@ typedef struct _PIXEL_CLOCK_PARAMETERS_V6 #define PIXEL_CLOCK_V6_MISC_HDMI_BPP_MASK 0x0c #define PIXEL_CLOCK_V6_MISC_HDMI_24BPP 0x00 #define PIXEL_CLOCK_V6_MISC_HDMI_36BPP 0x04 +#define PIXEL_CLOCK_V6_MISC_HDMI_36BPP_V6 0x08 //for V6, the correct defintion for 36bpp should be 2 for 36bpp(2:1) #define PIXEL_CLOCK_V6_MISC_HDMI_30BPP 0x08 +#define PIXEL_CLOCK_V6_MISC_HDMI_30BPP_V6 0x04 //for V6, the correct defintion for 30bpp should be 1 for 36bpp(5:4) #define PIXEL_CLOCK_V6_MISC_HDMI_48BPP 0x0c #define PIXEL_CLOCK_V6_MISC_REF_DIV_SRC 0x10 #define PIXEL_CLOCK_V6_MISC_GEN_DPREFCLK 0x40 @@ -2223,7 +2225,7 @@ typedef struct _SET_VOLTAGE_PARAMETERS_V2 USHORT usVoltageLevel; // real voltage level }SET_VOLTAGE_PARAMETERS_V2; - +// used by both SetVoltageTable v1.3 and v1.4 typedef struct _SET_VOLTAGE_PARAMETERS_V1_3 { UCHAR ucVoltageType; // To tell which voltage to set up, VDDC/MVDDC/MVDDQ/VDDCI @@ -2290,15 +2292,36 @@ typedef struct _GET_LEAKAGE_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_1 #define ATOM_GET_VOLTAGE_VID 0x00 #define ATOM_GET_VOTLAGE_INIT_SEQ 0x03 #define ATOM_GET_VOLTTAGE_PHASE_PHASE_VID 0x04 -// for SI, this state map to 0xff02 voltage state in Power Play table, which is power boost state -#define ATOM_GET_VOLTAGE_STATE0_LEAKAGE_VID 0x10 +#define ATOM_GET_VOLTAGE_SVID2 0x07 //Get SVI2 Regulator Info +// for SI, this state map to 0xff02 voltage state in Power Play table, which is power boost state +#define ATOM_GET_VOLTAGE_STATE0_LEAKAGE_VID 0x10 // for SI, this state map to 0xff01 voltage state in Power Play table, which is performance state #define ATOM_GET_VOLTAGE_STATE1_LEAKAGE_VID 0x11 -// undefined power state + #define ATOM_GET_VOLTAGE_STATE2_LEAKAGE_VID 0x12 #define ATOM_GET_VOLTAGE_STATE3_LEAKAGE_VID 0x13 +// New Added from CI Hawaii for GetVoltageInfoTable, input parameter structure +typedef struct _GET_VOLTAGE_INFO_INPUT_PARAMETER_V1_2 +{ + UCHAR ucVoltageType; // Input: To tell which voltage to set up, VDDC/MVDDC/MVDDQ/VDDCI + UCHAR ucVoltageMode; // Input: Indicate action: Get voltage info + USHORT usVoltageLevel; // Input: real voltage level in unit of mv or Voltage Phase (0, 1, 2, .. ) or Leakage Id + ULONG ulSCLKFreq; // Input: when ucVoltageMode= ATOM_GET_VOLTAGE_EVV_VOLTAGE, DPM state SCLK frequency, Define in PPTable SCLK/Voltage dependence table +}GET_VOLTAGE_INFO_INPUT_PARAMETER_V1_2; + +// New in GetVoltageInfo v1.2 ucVoltageMode +#define ATOM_GET_VOLTAGE_EVV_VOLTAGE 0x09 + +// New Added from CI Hawaii for EVV feature +typedef struct _GET_EVV_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_2 +{ + USHORT usVoltageLevel; // real voltage level in unit of mv + USHORT usVoltageId; // Voltage Id programmed in Voltage Regulator + ULONG ulReseved; +}GET_EVV_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_2; + /****************************************************************************/ // Structures used by TVEncoderControlTable /****************************************************************************/ @@ -3864,6 +3887,8 @@ typedef struct _ATOM_GPIO_PIN_ASSIGNMENT #define PP_AC_DC_SWITCH_GPIO_PINID 60 //from SMU7.x, if ucGPIO_ID=VDDC_REGULATOR_VRHOT_GPIO_PINID in GPIO_LUTable, VRHot feature is enable #define VDDC_VRHOT_GPIO_PINID 61 +//if ucGPIO_ID=VDDC_PCC_GPIO_PINID in GPIO_LUTable, Peak Current Control feature is enabled +#define VDDC_PCC_GPIO_PINID 62 typedef struct _ATOM_GPIO_PIN_LUT { @@ -4169,10 +4194,10 @@ typedef struct _ATOM_COMMON_RECORD_HEADER #define ATOM_OBJECT_LINK_RECORD_TYPE 18 //Once this record is present under one object, it indicats the oobject is linked to another obj described by the record #define ATOM_CONNECTOR_REMOTE_CAP_RECORD_TYPE 19 #define ATOM_ENCODER_CAP_RECORD_TYPE 20 - +#define ATOM_BRACKET_LAYOUT_RECORD_TYPE 21 //Must be updated when new record type is added,equal to that record definition! -#define ATOM_MAX_OBJECT_RECORD_NUMBER ATOM_ENCODER_CAP_RECORD_TYPE +#define ATOM_MAX_OBJECT_RECORD_NUMBER ATOM_BRACKET_LAYOUT_RECORD_TYPE typedef struct _ATOM_I2C_RECORD { @@ -4397,6 +4422,31 @@ typedef struct _ATOM_CONNECTOR_REMOTE_CAP_RECORD USHORT usReserved; }ATOM_CONNECTOR_REMOTE_CAP_RECORD; +typedef struct _ATOM_CONNECTOR_LAYOUT_INFO +{ + USHORT usConnectorObjectId; + UCHAR ucConnectorType; + UCHAR ucPosition; +}ATOM_CONNECTOR_LAYOUT_INFO; + +// define ATOM_CONNECTOR_LAYOUT_INFO.ucConnectorType to describe the display connector size +#define CONNECTOR_TYPE_DVI_D 1 +#define CONNECTOR_TYPE_DVI_I 2 +#define CONNECTOR_TYPE_VGA 3 +#define CONNECTOR_TYPE_HDMI 4 +#define CONNECTOR_TYPE_DISPLAY_PORT 5 +#define CONNECTOR_TYPE_MINI_DISPLAY_PORT 6 + +typedef struct _ATOM_BRACKET_LAYOUT_RECORD +{ + ATOM_COMMON_RECORD_HEADER sheader; + UCHAR ucLength; + UCHAR ucWidth; + UCHAR ucConnNum; + UCHAR ucReserved; + ATOM_CONNECTOR_LAYOUT_INFO asConnInfo[1]; +}ATOM_BRACKET_LAYOUT_RECORD; + /****************************************************************************/ // ASIC voltage data table /****************************************************************************/ @@ -4524,8 +4574,9 @@ typedef struct _ATOM_VOLTAGE_OBJECT_HEADER_V3{ #define VOLTAGE_OBJ_VR_I2C_INIT_SEQ 3 //VOLTAGE REGULATOR INIT sequece through I2C -> ATOM_I2C_VOLTAGE_OBJECT_V3 #define VOLTAGE_OBJ_PHASE_LUT 4 //Set Vregulator Phase lookup table ->ATOM_GPIO_VOLTAGE_OBJECT_V3 #define VOLTAGE_OBJ_SVID2 7 //Indicate voltage control by SVID2 ->ATOM_SVID2_VOLTAGE_OBJECT_V3 -#define VOLTAGE_OBJ_PWRBOOST_LEAKAGE_LUT 0x10 //Powerboost Voltage and LeakageId lookup table->ATOM_LEAKAGE_VOLTAGE_OBJECT_V3 -#define VOLTAGE_OBJ_HIGH_STATE_LEAKAGE_LUT 0x11 //High voltage state Voltage and LeakageId lookup table->ATOM_LEAKAGE_VOLTAGE_OBJECT_V3 +#define VOLTAGE_OBJ_EVV 8 +#define VOLTAGE_OBJ_PWRBOOST_LEAKAGE_LUT 0x10 //Powerboost Voltage and LeakageId lookup table->ATOM_LEAKAGE_VOLTAGE_OBJECT_V3 +#define VOLTAGE_OBJ_HIGH_STATE_LEAKAGE_LUT 0x11 //High voltage state Voltage and LeakageId lookup table->ATOM_LEAKAGE_VOLTAGE_OBJECT_V3 #define VOLTAGE_OBJ_HIGH1_STATE_LEAKAGE_LUT 0x12 //High1 voltage state Voltage and LeakageId lookup table->ATOM_LEAKAGE_VOLTAGE_OBJECT_V3 typedef struct _VOLTAGE_LUT_ENTRY_V2 @@ -4552,6 +4603,10 @@ typedef struct _ATOM_I2C_VOLTAGE_OBJECT_V3 VOLTAGE_LUT_ENTRY asVolI2cLut[1]; // end with 0xff }ATOM_I2C_VOLTAGE_OBJECT_V3; +// ATOM_I2C_VOLTAGE_OBJECT_V3.ucVoltageControlFlag +#define VOLTAGE_DATA_ONE_BYTE 0 +#define VOLTAGE_DATA_TWO_BYTE 1 + typedef struct _ATOM_GPIO_VOLTAGE_OBJECT_V3 { ATOM_VOLTAGE_OBJECT_HEADER_V3 sHeader; // voltage mode = VOLTAGE_OBJ_GPIO_LUT or VOLTAGE_OBJ_PHASE_LUT @@ -4584,7 +4639,8 @@ typedef struct _ATOM_SVID2_VOLTAGE_OBJECT_V3 // 1:0 offset trim, USHORT usLoadLine_PSI; // GPU GPIO pin Id to SVID2 regulator VRHot pin. possible value 0~31. 0 means GPIO0, 31 means GPIO31 - UCHAR ucReserved[2]; + UCHAR ucSVDGpioId; //0~31 indicate GPIO0~31 + UCHAR ucSVCGpioId; //0~31 indicate GPIO0~31 ULONG ulReserved; }ATOM_SVID2_VOLTAGE_OBJECT_V3; @@ -4637,6 +4693,49 @@ typedef struct _ATOM_ASIC_PROFILING_INFO_V2_1 USHORT usElbVDDCI_LevelArrayOffset; // offset of 2 dimension voltage level USHORT array }ATOM_ASIC_PROFILING_INFO_V2_1; +typedef struct _ATOM_ASIC_PROFILING_INFO_V3_1 +{ + ATOM_COMMON_TABLE_HEADER asHeader; + ULONG ulEvvDerateTdp; + ULONG ulEvvDerateTdc; + ULONG ulBoardCoreTemp; + ULONG ulMaxVddc; + ULONG ulMinVddc; + ULONG ulLoadLineSlop; + ULONG ulLeakageTemp; + ULONG ulLeakageVoltage; + ULONG ulCACmEncodeRange; + ULONG ulCACmEncodeAverage; + ULONG ulCACbEncodeRange; + ULONG ulCACbEncodeAverage; + ULONG ulKt_bEncodeRange; + ULONG ulKt_bEncodeAverage; + ULONG ulKv_mEncodeRange; + ULONG ulKv_mEncodeAverage; + ULONG ulKv_bEncodeRange; + ULONG ulKv_bEncodeAverage; + ULONG ulLkgEncodeLn_MaxDivMin; + ULONG ulLkgEncodeMin; + ULONG ulEfuseLogisticAlpha; + USHORT usPowerDpm0; + USHORT usCurrentDpm0; + USHORT usPowerDpm1; + USHORT usCurrentDpm1; + USHORT usPowerDpm2; + USHORT usCurrentDpm2; + USHORT usPowerDpm3; + USHORT usCurrentDpm3; + USHORT usPowerDpm4; + USHORT usCurrentDpm4; + USHORT usPowerDpm5; + USHORT usCurrentDpm5; + USHORT usPowerDpm6; + USHORT usCurrentDpm6; + USHORT usPowerDpm7; + USHORT usCurrentDpm7; +}ATOM_ASIC_PROFILING_INFO_V3_1; + + typedef struct _ATOM_POWER_SOURCE_OBJECT { UCHAR ucPwrSrcId; // Power source @@ -5808,6 +5907,8 @@ typedef struct _ATOM_ASIC_INTERNAL_SS_INFO_V3 #define ATOM_S7_DOS_MODE_PIXEL_DEPTHb0 0x0C #define ATOM_S7_DOS_MODE_PIXEL_FORMATb0 0xF0 #define ATOM_S7_DOS_8BIT_DAC_ENb1 0x01 +#define ATOM_S7_ASIC_INIT_COMPLETEb1 0x02 +#define ATOM_S7_ASIC_INIT_COMPLETE_MASK 0x00000200 #define ATOM_S7_DOS_MODE_NUMBERw1 0x0FFFF #define ATOM_S7_DOS_8BIT_DAC_EN_SHIFT 8 @@ -6242,6 +6343,7 @@ typedef struct _ATOM_MC_INIT_PARAM_TABLE #define _128Mx32 0x53 #define _256Mx8 0x61 #define _256Mx16 0x62 +#define _512Mx8 0x71 #define SAMSUNG 0x1 #define INFINEON 0x2 @@ -6987,9 +7089,10 @@ typedef struct _ATOM_DISP_OUT_INFO_V3 UCHAR ucMaxDispEngineNum; UCHAR ucMaxActiveDispEngineNum; UCHAR ucMaxPPLLNum; - UCHAR ucCoreRefClkSource; // value of CORE_REF_CLK_SOURCE - UCHAR ucReserved[3]; - ASIC_TRANSMITTER_INFO_V2 asTransmitterInfo[1]; // for alligment only + UCHAR ucCoreRefClkSource; // value of CORE_REF_CLK_SOURCE + UCHAR ucDispCaps; + UCHAR ucReserved[2]; + ASIC_TRANSMITTER_INFO_V2 asTransmitterInfo[1]; // for alligment only }ATOM_DISP_OUT_INFO_V3; //ucDispCaps diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index 86d9ee08b13f..0b9621c9aeea 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -1143,31 +1143,53 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc, } if (tiling_flags & RADEON_TILING_MACRO) { - if (rdev->family >= CHIP_BONAIRE) - tmp = rdev->config.cik.tile_config; - else if (rdev->family >= CHIP_TAHITI) - tmp = rdev->config.si.tile_config; - else if (rdev->family >= CHIP_CAYMAN) - tmp = rdev->config.cayman.tile_config; - else - tmp = rdev->config.evergreen.tile_config; + evergreen_tiling_fields(tiling_flags, &bankw, &bankh, &mtaspect, &tile_split); - switch ((tmp & 0xf0) >> 4) { - case 0: /* 4 banks */ - fb_format |= EVERGREEN_GRPH_NUM_BANKS(EVERGREEN_ADDR_SURF_4_BANK); - break; - case 1: /* 8 banks */ - default: - fb_format |= EVERGREEN_GRPH_NUM_BANKS(EVERGREEN_ADDR_SURF_8_BANK); - break; - case 2: /* 16 banks */ - fb_format |= EVERGREEN_GRPH_NUM_BANKS(EVERGREEN_ADDR_SURF_16_BANK); - break; + /* Set NUM_BANKS. */ + if (rdev->family >= CHIP_BONAIRE) { + unsigned tileb, index, num_banks, tile_split_bytes; + + /* Calculate the macrotile mode index. */ + tile_split_bytes = 64 << tile_split; + tileb = 8 * 8 * target_fb->bits_per_pixel / 8; + tileb = min(tile_split_bytes, tileb); + + for (index = 0; tileb > 64; index++) { + tileb >>= 1; + } + + if (index >= 16) { + DRM_ERROR("Wrong screen bpp (%u) or tile split (%u)\n", + target_fb->bits_per_pixel, tile_split); + return -EINVAL; + } + + num_banks = (rdev->config.cik.macrotile_mode_array[index] >> 6) & 0x3; + fb_format |= EVERGREEN_GRPH_NUM_BANKS(num_banks); + } else { + /* SI and older. */ + if (rdev->family >= CHIP_TAHITI) + tmp = rdev->config.si.tile_config; + else if (rdev->family >= CHIP_CAYMAN) + tmp = rdev->config.cayman.tile_config; + else + tmp = rdev->config.evergreen.tile_config; + + switch ((tmp & 0xf0) >> 4) { + case 0: /* 4 banks */ + fb_format |= EVERGREEN_GRPH_NUM_BANKS(EVERGREEN_ADDR_SURF_4_BANK); + break; + case 1: /* 8 banks */ + default: + fb_format |= EVERGREEN_GRPH_NUM_BANKS(EVERGREEN_ADDR_SURF_8_BANK); + break; + case 2: /* 16 banks */ + fb_format |= EVERGREEN_GRPH_NUM_BANKS(EVERGREEN_ADDR_SURF_16_BANK); + break; + } } fb_format |= EVERGREEN_GRPH_ARRAY_MODE(EVERGREEN_GRPH_ARRAY_2D_TILED_THIN1); - - evergreen_tiling_fields(tiling_flags, &bankw, &bankh, &mtaspect, &tile_split); fb_format |= EVERGREEN_GRPH_TILE_SPLIT(tile_split); fb_format |= EVERGREEN_GRPH_BANK_WIDTH(bankw); fb_format |= EVERGREEN_GRPH_BANK_HEIGHT(bankh); @@ -1180,23 +1202,18 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc, fb_format |= EVERGREEN_GRPH_ARRAY_MODE(EVERGREEN_GRPH_ARRAY_1D_TILED_THIN1); if (rdev->family >= CHIP_BONAIRE) { - u32 num_pipe_configs = rdev->config.cik.max_tile_pipes; - u32 num_rb = rdev->config.cik.max_backends_per_se; - if (num_pipe_configs > 8) - num_pipe_configs = 8; - if (num_pipe_configs == 8) - fb_format |= CIK_GRPH_PIPE_CONFIG(CIK_ADDR_SURF_P8_32x32_16x16); - else if (num_pipe_configs == 4) { - if (num_rb == 4) - fb_format |= CIK_GRPH_PIPE_CONFIG(CIK_ADDR_SURF_P4_16x16); - else if (num_rb < 4) - fb_format |= CIK_GRPH_PIPE_CONFIG(CIK_ADDR_SURF_P4_8x16); - } else if (num_pipe_configs == 2) - fb_format |= CIK_GRPH_PIPE_CONFIG(CIK_ADDR_SURF_P2); + /* Read the pipe config from the 2D TILED SCANOUT mode. + * It should be the same for the other modes too, but not all + * modes set the pipe config field. */ + u32 pipe_config = (rdev->config.cik.tile_mode_array[10] >> 6) & 0x1f; + + fb_format |= CIK_GRPH_PIPE_CONFIG(pipe_config); } else if ((rdev->family == CHIP_TAHITI) || (rdev->family == CHIP_PITCAIRN)) fb_format |= SI_GRPH_PIPE_CONFIG(SI_ADDR_SURF_P8_32x32_8x16); - else if (rdev->family == CHIP_VERDE) + else if ((rdev->family == CHIP_VERDE) || + (rdev->family == CHIP_OLAND) || + (rdev->family == CHIP_HAINAN)) /* for completeness. HAINAN has no display hw */ fb_format |= SI_GRPH_PIPE_CONFIG(SI_ADDR_SURF_P4_8x16); switch (radeon_crtc->crtc_id) { @@ -1910,6 +1927,21 @@ static void atombios_crtc_disable(struct drm_crtc *crtc) int i; atombios_crtc_dpms(crtc, DRM_MODE_DPMS_OFF); + if (crtc->fb) { + int r; + struct radeon_framebuffer *radeon_fb; + struct radeon_bo *rbo; + + radeon_fb = to_radeon_framebuffer(crtc->fb); + rbo = gem_to_radeon_bo(radeon_fb->obj); + r = radeon_bo_reserve(rbo, false); + if (unlikely(r)) + DRM_ERROR("failed to reserve rbo before unpin\n"); + else { + radeon_bo_unpin(rbo); + radeon_bo_unreserve(rbo); + } + } /* disable the GRPH */ if (ASIC_IS_DCE4(rdev)) WREG32(EVERGREEN_GRPH_ENABLE + radeon_crtc->crtc_offset, 0); @@ -1940,7 +1972,9 @@ static void atombios_crtc_disable(struct drm_crtc *crtc) break; case ATOM_PPLL0: /* disable the ppll */ - if ((rdev->family == CHIP_ARUBA) || (rdev->family == CHIP_BONAIRE)) + if ((rdev->family == CHIP_ARUBA) || + (rdev->family == CHIP_BONAIRE) || + (rdev->family == CHIP_HAWAII)) atombios_crtc_program_pll(crtc, radeon_crtc->crtc_id, radeon_crtc->pll_id, 0, 0, ATOM_DISABLE, 0, 0, 0, 0, 0, false, &ss); break; diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c index eb3ba60a2e92..a42d61571f49 100644 --- a/drivers/gpu/drm/radeon/atombios_encoders.c +++ b/drivers/gpu/drm/radeon/atombios_encoders.c @@ -707,24 +707,37 @@ atombios_get_encoder_mode(struct drm_encoder *encoder) switch (connector->connector_type) { case DRM_MODE_CONNECTOR_DVII: case DRM_MODE_CONNECTOR_HDMIB: /* HDMI-B is basically DL-DVI; analog works fine */ - if ((radeon_connector->audio == RADEON_AUDIO_ENABLE) || - (drm_detect_hdmi_monitor(radeon_connector->edid) && - (radeon_connector->audio == RADEON_AUDIO_AUTO))) - return ATOM_ENCODER_MODE_HDMI; - else if (radeon_connector->use_digital) + if (radeon_audio != 0) { + if (radeon_connector->use_digital && + (radeon_connector->audio == RADEON_AUDIO_ENABLE)) + return ATOM_ENCODER_MODE_HDMI; + else if (drm_detect_hdmi_monitor(radeon_connector->edid) && + (radeon_connector->audio == RADEON_AUDIO_AUTO)) + return ATOM_ENCODER_MODE_HDMI; + else if (radeon_connector->use_digital) + return ATOM_ENCODER_MODE_DVI; + else + return ATOM_ENCODER_MODE_CRT; + } else if (radeon_connector->use_digital) { return ATOM_ENCODER_MODE_DVI; - else + } else { return ATOM_ENCODER_MODE_CRT; + } break; case DRM_MODE_CONNECTOR_DVID: case DRM_MODE_CONNECTOR_HDMIA: default: - if ((radeon_connector->audio == RADEON_AUDIO_ENABLE) || - (drm_detect_hdmi_monitor(radeon_connector->edid) && - (radeon_connector->audio == RADEON_AUDIO_AUTO))) - return ATOM_ENCODER_MODE_HDMI; - else + if (radeon_audio != 0) { + if (radeon_connector->audio == RADEON_AUDIO_ENABLE) + return ATOM_ENCODER_MODE_HDMI; + else if (drm_detect_hdmi_monitor(radeon_connector->edid) && + (radeon_connector->audio == RADEON_AUDIO_AUTO)) + return ATOM_ENCODER_MODE_HDMI; + else + return ATOM_ENCODER_MODE_DVI; + } else { return ATOM_ENCODER_MODE_DVI; + } break; case DRM_MODE_CONNECTOR_LVDS: return ATOM_ENCODER_MODE_LVDS; @@ -732,14 +745,19 @@ atombios_get_encoder_mode(struct drm_encoder *encoder) case DRM_MODE_CONNECTOR_DisplayPort: dig_connector = radeon_connector->con_priv; if ((dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_DISPLAYPORT) || - (dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_eDP)) + (dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_eDP)) { return ATOM_ENCODER_MODE_DP; - else if ((radeon_connector->audio == RADEON_AUDIO_ENABLE) || - (drm_detect_hdmi_monitor(radeon_connector->edid) && - (radeon_connector->audio == RADEON_AUDIO_AUTO))) - return ATOM_ENCODER_MODE_HDMI; - else + } else if (radeon_audio != 0) { + if (radeon_connector->audio == RADEON_AUDIO_ENABLE) + return ATOM_ENCODER_MODE_HDMI; + else if (drm_detect_hdmi_monitor(radeon_connector->edid) && + (radeon_connector->audio == RADEON_AUDIO_AUTO)) + return ATOM_ENCODER_MODE_HDMI; + else + return ATOM_ENCODER_MODE_DVI; + } else { return ATOM_ENCODER_MODE_DVI; + } break; case DRM_MODE_CONNECTOR_eDP: return ATOM_ENCODER_MODE_DP; diff --git a/drivers/gpu/drm/radeon/atombios_i2c.c b/drivers/gpu/drm/radeon/atombios_i2c.c index deaf98cdca3a..f685035dbe39 100644 --- a/drivers/gpu/drm/radeon/atombios_i2c.c +++ b/drivers/gpu/drm/radeon/atombios_i2c.c @@ -44,7 +44,7 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan, PROCESS_I2C_CHANNEL_TRANSACTION_PS_ALLOCATION args; int index = GetIndexIntoMasterTable(COMMAND, ProcessI2cChannelTransaction); unsigned char *base; - u16 out; + u16 out = cpu_to_le16(0); memset(&args, 0, sizeof(args)); @@ -55,9 +55,14 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan, DRM_ERROR("hw i2c: tried to write too many bytes (%d vs 3)\n", num); return -EINVAL; } - args.ucRegIndex = buf[0]; - if (num > 1) - memcpy(&out, &buf[1], num - 1); + if (buf == NULL) + args.ucRegIndex = 0; + else + args.ucRegIndex = buf[0]; + if (num) + num--; + if (num) + memcpy(&out, &buf[1], num); args.lpI2CDataOut = cpu_to_le16(out); } else { if (num > ATOM_MAX_HW_I2C_READ) { @@ -94,14 +99,14 @@ int radeon_atom_hw_i2c_xfer(struct i2c_adapter *i2c_adap, struct radeon_i2c_chan *i2c = i2c_get_adapdata(i2c_adap); struct i2c_msg *p; int i, remaining, current_count, buffer_offset, max_bytes, ret; - u8 buf = 0, flags; + u8 flags; /* check for bus probe */ p = &msgs[0]; if ((num == 1) && (p->len == 0)) { ret = radeon_process_i2c_ch(i2c, p->addr, HW_I2C_WRITE, - &buf, 1); + NULL, 0); if (ret) return ret; else diff --git a/drivers/gpu/drm/radeon/btc_dpm.c b/drivers/gpu/drm/radeon/btc_dpm.c index b162e98a2953..9b6950d9b3c0 100644 --- a/drivers/gpu/drm/radeon/btc_dpm.c +++ b/drivers/gpu/drm/radeon/btc_dpm.c @@ -1930,7 +1930,7 @@ static int btc_set_mc_special_registers(struct radeon_device *rdev, } j++; - if (j > SMC_EVERGREEN_MC_REGISTER_ARRAY_SIZE) + if (j >= SMC_EVERGREEN_MC_REGISTER_ARRAY_SIZE) return -EINVAL; tmp = RREG32(MC_PMG_CMD_MRS); @@ -1945,7 +1945,7 @@ static int btc_set_mc_special_registers(struct radeon_device *rdev, } j++; - if (j > SMC_EVERGREEN_MC_REGISTER_ARRAY_SIZE) + if (j >= SMC_EVERGREEN_MC_REGISTER_ARRAY_SIZE) return -EINVAL; break; case MC_SEQ_RESERVE_M >> 2: @@ -1959,7 +1959,7 @@ static int btc_set_mc_special_registers(struct radeon_device *rdev, } j++; - if (j > SMC_EVERGREEN_MC_REGISTER_ARRAY_SIZE) + if (j >= SMC_EVERGREEN_MC_REGISTER_ARRAY_SIZE) return -EINVAL; break; default: diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c index 51e947a97edf..1ed479976358 100644 --- a/drivers/gpu/drm/radeon/ci_dpm.c +++ b/drivers/gpu/drm/radeon/ci_dpm.c @@ -40,6 +40,20 @@ #define VOLTAGE_VID_OFFSET_SCALE1 625 #define VOLTAGE_VID_OFFSET_SCALE2 100 +static const struct ci_pt_defaults defaults_hawaii_xt = +{ + 1, 0xF, 0xFD, 0x19, 5, 0x14, 0, 0xB0000, + { 0x84, 0x0, 0x0, 0x7F, 0x0, 0x0, 0x5A, 0x60, 0x51, 0x8E, 0x79, 0x6B, 0x5F, 0x90, 0x79 }, + { 0x1EA, 0x1EA, 0x1EA, 0x224, 0x224, 0x224, 0x24F, 0x24F, 0x24F, 0x28E, 0x28E, 0x28E, 0x2BC, 0x2BC, 0x2BC } +}; + +static const struct ci_pt_defaults defaults_hawaii_pro = +{ + 1, 0xF, 0xFD, 0x19, 5, 0x14, 0, 0x65062, + { 0x93, 0x0, 0x0, 0x97, 0x0, 0x0, 0x6B, 0x60, 0x51, 0x95, 0x79, 0x6B, 0x5F, 0x90, 0x79 }, + { 0x1EA, 0x1EA, 0x1EA, 0x224, 0x224, 0x224, 0x24F, 0x24F, 0x24F, 0x28E, 0x28E, 0x28E, 0x2BC, 0x2BC, 0x2BC } +}; + static const struct ci_pt_defaults defaults_bonaire_xt = { 1, 0xF, 0xFD, 0x19, 5, 45, 0, 0xB0000, @@ -187,22 +201,38 @@ static void ci_initialize_powertune_defaults(struct radeon_device *rdev) struct ci_power_info *pi = ci_get_pi(rdev); switch (rdev->pdev->device) { - case 0x6650: - case 0x6658: - case 0x665C: - default: + case 0x6650: + case 0x6658: + case 0x665C: + default: pi->powertune_defaults = &defaults_bonaire_xt; break; - case 0x6651: - case 0x665D: + case 0x6651: + case 0x665D: pi->powertune_defaults = &defaults_bonaire_pro; break; - case 0x6640: + case 0x6640: pi->powertune_defaults = &defaults_saturn_xt; break; - case 0x6641: + case 0x6641: pi->powertune_defaults = &defaults_saturn_pro; break; + case 0x67B8: + case 0x67B0: + case 0x67A0: + case 0x67A1: + case 0x67A2: + case 0x67A8: + case 0x67A9: + case 0x67AA: + case 0x67B9: + case 0x67BE: + pi->powertune_defaults = &defaults_hawaii_xt; + break; + case 0x67BA: + case 0x67B1: + pi->powertune_defaults = &defaults_hawaii_pro; + break; } pi->dte_tj_offset = 0; @@ -5142,9 +5172,15 @@ int ci_dpm_init(struct radeon_device *rdev) rdev->pm.dpm.dyn_state.valid_mclk_values.count = 0; rdev->pm.dpm.dyn_state.valid_mclk_values.values = NULL; - pi->thermal_temp_setting.temperature_low = 99500; - pi->thermal_temp_setting.temperature_high = 100000; - pi->thermal_temp_setting.temperature_shutdown = 104000; + if (rdev->family == CHIP_HAWAII) { + pi->thermal_temp_setting.temperature_low = 94500; + pi->thermal_temp_setting.temperature_high = 95000; + pi->thermal_temp_setting.temperature_shutdown = 104000; + } else { + pi->thermal_temp_setting.temperature_low = 99500; + pi->thermal_temp_setting.temperature_high = 100000; + pi->thermal_temp_setting.temperature_shutdown = 104000; + } pi->uvd_enabled = false; diff --git a/drivers/gpu/drm/radeon/ci_smc.c b/drivers/gpu/drm/radeon/ci_smc.c index 252e10a41cf5..9c745dd22438 100644 --- a/drivers/gpu/drm/radeon/ci_smc.c +++ b/drivers/gpu/drm/radeon/ci_smc.c @@ -217,6 +217,10 @@ int ci_load_smc_ucode(struct radeon_device *rdev, u32 limit) ucode_start_address = BONAIRE_SMC_UCODE_START; ucode_size = BONAIRE_SMC_UCODE_SIZE; break; + case CHIP_HAWAII: + ucode_start_address = HAWAII_SMC_UCODE_START; + ucode_size = HAWAII_SMC_UCODE_SIZE; + break; default: DRM_ERROR("unknown asic in smc ucode loader\n"); BUG(); diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c index 2e48f902e3b5..e950fabd7f5e 100644 --- a/drivers/gpu/drm/radeon/cik.c +++ b/drivers/gpu/drm/radeon/cik.c @@ -41,6 +41,14 @@ MODULE_FIRMWARE("radeon/BONAIRE_mc.bin"); MODULE_FIRMWARE("radeon/BONAIRE_rlc.bin"); MODULE_FIRMWARE("radeon/BONAIRE_sdma.bin"); MODULE_FIRMWARE("radeon/BONAIRE_smc.bin"); +MODULE_FIRMWARE("radeon/HAWAII_pfp.bin"); +MODULE_FIRMWARE("radeon/HAWAII_me.bin"); +MODULE_FIRMWARE("radeon/HAWAII_ce.bin"); +MODULE_FIRMWARE("radeon/HAWAII_mec.bin"); +MODULE_FIRMWARE("radeon/HAWAII_mc.bin"); +MODULE_FIRMWARE("radeon/HAWAII_rlc.bin"); +MODULE_FIRMWARE("radeon/HAWAII_sdma.bin"); +MODULE_FIRMWARE("radeon/HAWAII_smc.bin"); MODULE_FIRMWARE("radeon/KAVERI_pfp.bin"); MODULE_FIRMWARE("radeon/KAVERI_me.bin"); MODULE_FIRMWARE("radeon/KAVERI_ce.bin"); @@ -72,6 +80,8 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev); static void cik_program_aspm(struct radeon_device *rdev); static void cik_init_pg(struct radeon_device *rdev); static void cik_init_cg(struct radeon_device *rdev); +static void cik_fini_pg(struct radeon_device *rdev); +static void cik_fini_cg(struct radeon_device *rdev); static void cik_enable_gui_idle_interrupt(struct radeon_device *rdev, bool enable); @@ -1295,6 +1305,171 @@ static const u32 kalindi_mgcg_cgcg_init[] = 0xd80c, 0xff000ff0, 0x00000100 }; +static const u32 hawaii_golden_spm_registers[] = +{ + 0x30800, 0xe0ffffff, 0xe0000000 +}; + +static const u32 hawaii_golden_common_registers[] = +{ + 0x30800, 0xffffffff, 0xe0000000, + 0x28350, 0xffffffff, 0x3a00161a, + 0x28354, 0xffffffff, 0x0000002e, + 0x9a10, 0xffffffff, 0x00018208, + 0x98f8, 0xffffffff, 0x12011003 +}; + +static const u32 hawaii_golden_registers[] = +{ + 0x3354, 0x00000333, 0x00000333, + 0x9a10, 0x00010000, 0x00058208, + 0x9830, 0xffffffff, 0x00000000, + 0x9834, 0xf00fffff, 0x00000400, + 0x9838, 0x0002021c, 0x00020200, + 0xc78, 0x00000080, 0x00000000, + 0x5bb0, 0x000000f0, 0x00000070, + 0x5bc0, 0xf0311fff, 0x80300000, + 0x350c, 0x00810000, 0x408af000, + 0x7030, 0x31000111, 0x00000011, + 0x2f48, 0x73773777, 0x12010001, + 0x2120, 0x0000007f, 0x0000001b, + 0x21dc, 0x00007fb6, 0x00002191, + 0x3628, 0x0000003f, 0x0000000a, + 0x362c, 0x0000003f, 0x0000000a, + 0x2ae4, 0x00073ffe, 0x000022a2, + 0x240c, 0x000007ff, 0x00000000, + 0x8bf0, 0x00002001, 0x00000001, + 0x8b24, 0xffffffff, 0x00ffffff, + 0x30a04, 0x0000ff0f, 0x00000000, + 0x28a4c, 0x07ffffff, 0x06000000, + 0x3e78, 0x00000001, 0x00000002, + 0xc768, 0x00000008, 0x00000008, + 0xc770, 0x00000f00, 0x00000800, + 0xc774, 0x00000f00, 0x00000800, + 0xc798, 0x00ffffff, 0x00ff7fbf, + 0xc79c, 0x00ffffff, 0x00ff7faf, + 0x8c00, 0x000000ff, 0x00000800, + 0xe40, 0x00001fff, 0x00001fff, + 0x9060, 0x0000007f, 0x00000020, + 0x9508, 0x00010000, 0x00010000, + 0xae00, 0x00100000, 0x000ff07c, + 0xac14, 0x000003ff, 0x0000000f, + 0xac10, 0xffffffff, 0x7564fdec, + 0xac0c, 0xffffffff, 0x3120b9a8, + 0xac08, 0x20000000, 0x0f9c0000 +}; + +static const u32 hawaii_mgcg_cgcg_init[] = +{ + 0xc420, 0xffffffff, 0xfffffffd, + 0x30800, 0xffffffff, 0xe0000000, + 0x3c2a0, 0xffffffff, 0x00000100, + 0x3c208, 0xffffffff, 0x00000100, + 0x3c2c0, 0xffffffff, 0x00000100, + 0x3c2c8, 0xffffffff, 0x00000100, + 0x3c2c4, 0xffffffff, 0x00000100, + 0x55e4, 0xffffffff, 0x00200100, + 0x3c280, 0xffffffff, 0x00000100, + 0x3c214, 0xffffffff, 0x06000100, + 0x3c220, 0xffffffff, 0x00000100, + 0x3c218, 0xffffffff, 0x06000100, + 0x3c204, 0xffffffff, 0x00000100, + 0x3c2e0, 0xffffffff, 0x00000100, + 0x3c224, 0xffffffff, 0x00000100, + 0x3c200, 0xffffffff, 0x00000100, + 0x3c230, 0xffffffff, 0x00000100, + 0x3c234, 0xffffffff, 0x00000100, + 0x3c250, 0xffffffff, 0x00000100, + 0x3c254, 0xffffffff, 0x00000100, + 0x3c258, 0xffffffff, 0x00000100, + 0x3c25c, 0xffffffff, 0x00000100, + 0x3c260, 0xffffffff, 0x00000100, + 0x3c27c, 0xffffffff, 0x00000100, + 0x3c278, 0xffffffff, 0x00000100, + 0x3c210, 0xffffffff, 0x06000100, + 0x3c290, 0xffffffff, 0x00000100, + 0x3c274, 0xffffffff, 0x00000100, + 0x3c2b4, 0xffffffff, 0x00000100, + 0x3c2b0, 0xffffffff, 0x00000100, + 0x3c270, 0xffffffff, 0x00000100, + 0x30800, 0xffffffff, 0xe0000000, + 0x3c020, 0xffffffff, 0x00010000, + 0x3c024, 0xffffffff, 0x00030002, + 0x3c028, 0xffffffff, 0x00040007, + 0x3c02c, 0xffffffff, 0x00060005, + 0x3c030, 0xffffffff, 0x00090008, + 0x3c034, 0xffffffff, 0x00010000, + 0x3c038, 0xffffffff, 0x00030002, + 0x3c03c, 0xffffffff, 0x00040007, + 0x3c040, 0xffffffff, 0x00060005, + 0x3c044, 0xffffffff, 0x00090008, + 0x3c048, 0xffffffff, 0x00010000, + 0x3c04c, 0xffffffff, 0x00030002, + 0x3c050, 0xffffffff, 0x00040007, + 0x3c054, 0xffffffff, 0x00060005, + 0x3c058, 0xffffffff, 0x00090008, + 0x3c05c, 0xffffffff, 0x00010000, + 0x3c060, 0xffffffff, 0x00030002, + 0x3c064, 0xffffffff, 0x00040007, + 0x3c068, 0xffffffff, 0x00060005, + 0x3c06c, 0xffffffff, 0x00090008, + 0x3c070, 0xffffffff, 0x00010000, + 0x3c074, 0xffffffff, 0x00030002, + 0x3c078, 0xffffffff, 0x00040007, + 0x3c07c, 0xffffffff, 0x00060005, + 0x3c080, 0xffffffff, 0x00090008, + 0x3c084, 0xffffffff, 0x00010000, + 0x3c088, 0xffffffff, 0x00030002, + 0x3c08c, 0xffffffff, 0x00040007, + 0x3c090, 0xffffffff, 0x00060005, + 0x3c094, 0xffffffff, 0x00090008, + 0x3c098, 0xffffffff, 0x00010000, + 0x3c09c, 0xffffffff, 0x00030002, + 0x3c0a0, 0xffffffff, 0x00040007, + 0x3c0a4, 0xffffffff, 0x00060005, + 0x3c0a8, 0xffffffff, 0x00090008, + 0x3c0ac, 0xffffffff, 0x00010000, + 0x3c0b0, 0xffffffff, 0x00030002, + 0x3c0b4, 0xffffffff, 0x00040007, + 0x3c0b8, 0xffffffff, 0x00060005, + 0x3c0bc, 0xffffffff, 0x00090008, + 0x3c0c0, 0xffffffff, 0x00010000, + 0x3c0c4, 0xffffffff, 0x00030002, + 0x3c0c8, 0xffffffff, 0x00040007, + 0x3c0cc, 0xffffffff, 0x00060005, + 0x3c0d0, 0xffffffff, 0x00090008, + 0x3c0d4, 0xffffffff, 0x00010000, + 0x3c0d8, 0xffffffff, 0x00030002, + 0x3c0dc, 0xffffffff, 0x00040007, + 0x3c0e0, 0xffffffff, 0x00060005, + 0x3c0e4, 0xffffffff, 0x00090008, + 0x3c0e8, 0xffffffff, 0x00010000, + 0x3c0ec, 0xffffffff, 0x00030002, + 0x3c0f0, 0xffffffff, 0x00040007, + 0x3c0f4, 0xffffffff, 0x00060005, + 0x3c0f8, 0xffffffff, 0x00090008, + 0xc318, 0xffffffff, 0x00020200, + 0x3350, 0xffffffff, 0x00000200, + 0x15c0, 0xffffffff, 0x00000400, + 0x55e8, 0xffffffff, 0x00000000, + 0x2f50, 0xffffffff, 0x00000902, + 0x3c000, 0xffffffff, 0x96940200, + 0x8708, 0xffffffff, 0x00900100, + 0xc424, 0xffffffff, 0x0020003f, + 0x38, 0xffffffff, 0x0140001c, + 0x3c, 0x000f0000, 0x000f0000, + 0x220, 0xffffffff, 0xc060000c, + 0x224, 0xc0000fff, 0x00000100, + 0xf90, 0xffffffff, 0x00000100, + 0xf98, 0x00000101, 0x00000000, + 0x20a8, 0xffffffff, 0x00000104, + 0x55e4, 0xff000fff, 0x00000100, + 0x30cc, 0xc0000fff, 0x00000104, + 0xc1e4, 0x00000001, 0x00000001, + 0xd00c, 0xff000ff0, 0x00000100, + 0xd80c, 0xff000ff0, 0x00000100 +}; + static void cik_init_golden_registers(struct radeon_device *rdev) { switch (rdev->family) { @@ -1340,6 +1515,20 @@ static void cik_init_golden_registers(struct radeon_device *rdev) spectre_golden_spm_registers, (const u32)ARRAY_SIZE(spectre_golden_spm_registers)); break; + case CHIP_HAWAII: + radeon_program_register_sequence(rdev, + hawaii_mgcg_cgcg_init, + (const u32)ARRAY_SIZE(hawaii_mgcg_cgcg_init)); + radeon_program_register_sequence(rdev, + hawaii_golden_registers, + (const u32)ARRAY_SIZE(hawaii_golden_registers)); + radeon_program_register_sequence(rdev, + hawaii_golden_common_registers, + (const u32)ARRAY_SIZE(hawaii_golden_common_registers)); + radeon_program_register_sequence(rdev, + hawaii_golden_spm_registers, + (const u32)ARRAY_SIZE(hawaii_golden_spm_registers)); + break; default: break; } @@ -1371,17 +1560,17 @@ u32 cik_get_xclk(struct radeon_device *rdev) * cik_mm_rdoorbell - read a doorbell dword * * @rdev: radeon_device pointer - * @offset: byte offset into the aperture + * @index: doorbell index * * Returns the value in the doorbell aperture at the - * requested offset (CIK). + * requested doorbell index (CIK). */ -u32 cik_mm_rdoorbell(struct radeon_device *rdev, u32 offset) +u32 cik_mm_rdoorbell(struct radeon_device *rdev, u32 index) { - if (offset < rdev->doorbell.size) { - return readl(((void __iomem *)rdev->doorbell.ptr) + offset); + if (index < rdev->doorbell.num_doorbells) { + return readl(rdev->doorbell.ptr + index); } else { - DRM_ERROR("reading beyond doorbell aperture: 0x%08x!\n", offset); + DRM_ERROR("reading beyond doorbell aperture: 0x%08x!\n", index); return 0; } } @@ -1390,18 +1579,18 @@ u32 cik_mm_rdoorbell(struct radeon_device *rdev, u32 offset) * cik_mm_wdoorbell - write a doorbell dword * * @rdev: radeon_device pointer - * @offset: byte offset into the aperture + * @index: doorbell index * @v: value to write * * Writes @v to the doorbell aperture at the - * requested offset (CIK). + * requested doorbell index (CIK). */ -void cik_mm_wdoorbell(struct radeon_device *rdev, u32 offset, u32 v) +void cik_mm_wdoorbell(struct radeon_device *rdev, u32 index, u32 v) { - if (offset < rdev->doorbell.size) { - writel(v, ((void __iomem *)rdev->doorbell.ptr) + offset); + if (index < rdev->doorbell.num_doorbells) { + writel(v, rdev->doorbell.ptr + index); } else { - DRM_ERROR("writing beyond doorbell aperture: 0x%08x!\n", offset); + DRM_ERROR("writing beyond doorbell aperture: 0x%08x!\n", index); } } @@ -1447,6 +1636,35 @@ static const u32 bonaire_io_mc_regs[BONAIRE_IO_MC_REGS_SIZE][2] = {0x0000009f, 0x00b48000} }; +#define HAWAII_IO_MC_REGS_SIZE 22 + +static const u32 hawaii_io_mc_regs[HAWAII_IO_MC_REGS_SIZE][2] = +{ + {0x0000007d, 0x40000000}, + {0x0000007e, 0x40180304}, + {0x0000007f, 0x0000ff00}, + {0x00000081, 0x00000000}, + {0x00000083, 0x00000800}, + {0x00000086, 0x00000000}, + {0x00000087, 0x00000100}, + {0x00000088, 0x00020100}, + {0x00000089, 0x00000000}, + {0x0000008b, 0x00040000}, + {0x0000008c, 0x00000100}, + {0x0000008e, 0xff010000}, + {0x00000090, 0xffffefff}, + {0x00000091, 0xfff3efff}, + {0x00000092, 0xfff3efbf}, + {0x00000093, 0xf7ffffff}, + {0x00000094, 0xffffff7f}, + {0x00000095, 0x00000fff}, + {0x00000096, 0x00116fff}, + {0x00000097, 0x60010000}, + {0x00000098, 0x10010000}, + {0x0000009f, 0x00c79000} +}; + + /** * cik_srbm_select - select specific register instances * @@ -1491,11 +1709,17 @@ static int ci_mc_load_microcode(struct radeon_device *rdev) switch (rdev->family) { case CHIP_BONAIRE: - default: io_mc_regs = (u32 *)&bonaire_io_mc_regs; ucode_size = CIK_MC_UCODE_SIZE; regs_size = BONAIRE_IO_MC_REGS_SIZE; break; + case CHIP_HAWAII: + io_mc_regs = (u32 *)&hawaii_io_mc_regs; + ucode_size = HAWAII_MC_UCODE_SIZE; + regs_size = HAWAII_IO_MC_REGS_SIZE; + break; + default: + return -EINVAL; } running = RREG32(MC_SEQ_SUP_CNTL) & RUN_MASK; @@ -1557,8 +1781,8 @@ static int cik_init_microcode(struct radeon_device *rdev) { const char *chip_name; size_t pfp_req_size, me_req_size, ce_req_size, - mec_req_size, rlc_req_size, mc_req_size, - sdma_req_size, smc_req_size; + mec_req_size, rlc_req_size, mc_req_size = 0, + sdma_req_size, smc_req_size = 0; char fw_name[30]; int err; @@ -1576,6 +1800,17 @@ static int cik_init_microcode(struct radeon_device *rdev) sdma_req_size = CIK_SDMA_UCODE_SIZE * 4; smc_req_size = ALIGN(BONAIRE_SMC_UCODE_SIZE, 4); break; + case CHIP_HAWAII: + chip_name = "HAWAII"; + pfp_req_size = CIK_PFP_UCODE_SIZE * 4; + me_req_size = CIK_ME_UCODE_SIZE * 4; + ce_req_size = CIK_CE_UCODE_SIZE * 4; + mec_req_size = CIK_MEC_UCODE_SIZE * 4; + rlc_req_size = BONAIRE_RLC_UCODE_SIZE * 4; + mc_req_size = HAWAII_MC_UCODE_SIZE * 4; + sdma_req_size = CIK_SDMA_UCODE_SIZE * 4; + smc_req_size = ALIGN(HAWAII_SMC_UCODE_SIZE, 4); + break; case CHIP_KAVERI: chip_name = "KAVERI"; pfp_req_size = CIK_PFP_UCODE_SIZE * 4; @@ -1687,6 +1922,7 @@ static int cik_init_microcode(struct radeon_device *rdev) fw_name); release_firmware(rdev->smc_fw); rdev->smc_fw = NULL; + err = 0; } else if (rdev->smc_fw->size != smc_req_size) { printk(KERN_ERR "cik_smc: Bogus length %zu in firmware \"%s\"\n", @@ -1755,9 +1991,227 @@ static void cik_tiling_mode_table_init(struct radeon_device *rdev) num_pipe_configs = rdev->config.cik.max_tile_pipes; if (num_pipe_configs > 8) - num_pipe_configs = 8; /* ??? */ + num_pipe_configs = 16; - if (num_pipe_configs == 8) { + if (num_pipe_configs == 16) { + for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) { + switch (reg_offset) { + case 0: + gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B)); + break; + case 1: + gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B)); + break; + case 2: + gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B)); + break; + case 3: + gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B)); + break; + case 4: + gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) | + TILE_SPLIT(split_equal_to_row_size)); + break; + case 5: + gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING)); + break; + case 6: + gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B)); + break; + case 7: + gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) | + TILE_SPLIT(split_equal_to_row_size)); + break; + case 8: + gb_tile_moden = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16)); + break; + case 9: + gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING)); + break; + case 10: + gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) | + SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2)); + break; + case 11: + gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_8x16) | + SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2)); + break; + case 12: + gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) | + SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2)); + break; + case 13: + gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING)); + break; + case 14: + gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) | + SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2)); + break; + case 16: + gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_8x16) | + SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2)); + break; + case 17: + gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) | + SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2)); + break; + case 27: + gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING)); + break; + case 28: + gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) | + SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2)); + break; + case 29: + gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_8x16) | + SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2)); + break; + case 30: + gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) | + MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) | + PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) | + SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2)); + break; + default: + gb_tile_moden = 0; + break; + } + rdev->config.cik.tile_mode_array[reg_offset] = gb_tile_moden; + WREG32(GB_TILE_MODE0 + (reg_offset * 4), gb_tile_moden); + } + for (reg_offset = 0; reg_offset < num_secondary_tile_mode_states; reg_offset++) { + switch (reg_offset) { + case 0: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) | + NUM_BANKS(ADDR_SURF_16_BANK)); + break; + case 1: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) | + NUM_BANKS(ADDR_SURF_16_BANK)); + break; + case 2: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) | + NUM_BANKS(ADDR_SURF_16_BANK)); + break; + case 3: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) | + NUM_BANKS(ADDR_SURF_16_BANK)); + break; + case 4: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) | + NUM_BANKS(ADDR_SURF_8_BANK)); + break; + case 5: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) | + NUM_BANKS(ADDR_SURF_4_BANK)); + break; + case 6: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) | + NUM_BANKS(ADDR_SURF_2_BANK)); + break; + case 8: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) | + NUM_BANKS(ADDR_SURF_16_BANK)); + break; + case 9: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) | + NUM_BANKS(ADDR_SURF_16_BANK)); + break; + case 10: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) | + NUM_BANKS(ADDR_SURF_16_BANK)); + break; + case 11: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) | + NUM_BANKS(ADDR_SURF_8_BANK)); + break; + case 12: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) | + NUM_BANKS(ADDR_SURF_4_BANK)); + break; + case 13: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) | + NUM_BANKS(ADDR_SURF_2_BANK)); + break; + case 14: + gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) | + NUM_BANKS(ADDR_SURF_2_BANK)); + break; + default: + gb_tile_moden = 0; + break; + } + WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), gb_tile_moden); + } + } else if (num_pipe_configs == 8) { for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) { switch (reg_offset) { case 0: @@ -1973,6 +2427,7 @@ static void cik_tiling_mode_table_init(struct radeon_device *rdev) gb_tile_moden = 0; break; } + rdev->config.cik.macrotile_mode_array[reg_offset] = gb_tile_moden; WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), gb_tile_moden); } } else if (num_pipe_configs == 4) { @@ -2319,6 +2774,7 @@ static void cik_tiling_mode_table_init(struct radeon_device *rdev) gb_tile_moden = 0; break; } + rdev->config.cik.macrotile_mode_array[reg_offset] = gb_tile_moden; WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), gb_tile_moden); } } else if (num_pipe_configs == 2) { @@ -2536,6 +2992,7 @@ static void cik_tiling_mode_table_init(struct radeon_device *rdev) gb_tile_moden = 0; break; } + rdev->config.cik.macrotile_mode_array[reg_offset] = gb_tile_moden; WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), gb_tile_moden); } } else @@ -2600,7 +3057,7 @@ static u32 cik_create_bitmask(u32 bit_width) * Returns the disabled RB bitmask. */ static u32 cik_get_rb_disabled(struct radeon_device *rdev, - u32 max_rb_num, u32 se_num, + u32 max_rb_num_per_se, u32 sh_per_se) { u32 data, mask; @@ -2614,7 +3071,7 @@ static u32 cik_get_rb_disabled(struct radeon_device *rdev, data >>= BACKEND_DISABLE_SHIFT; - mask = cik_create_bitmask(max_rb_num / se_num / sh_per_se); + mask = cik_create_bitmask(max_rb_num_per_se / sh_per_se); return data & mask; } @@ -2631,7 +3088,7 @@ static u32 cik_get_rb_disabled(struct radeon_device *rdev, */ static void cik_setup_rb(struct radeon_device *rdev, u32 se_num, u32 sh_per_se, - u32 max_rb_num) + u32 max_rb_num_per_se) { int i, j; u32 data, mask; @@ -2641,24 +3098,35 @@ static void cik_setup_rb(struct radeon_device *rdev, for (i = 0; i < se_num; i++) { for (j = 0; j < sh_per_se; j++) { cik_select_se_sh(rdev, i, j); - data = cik_get_rb_disabled(rdev, max_rb_num, se_num, sh_per_se); - disabled_rbs |= data << ((i * sh_per_se + j) * CIK_RB_BITMAP_WIDTH_PER_SH); + data = cik_get_rb_disabled(rdev, max_rb_num_per_se, sh_per_se); + if (rdev->family == CHIP_HAWAII) + disabled_rbs |= data << ((i * sh_per_se + j) * HAWAII_RB_BITMAP_WIDTH_PER_SH); + else + disabled_rbs |= data << ((i * sh_per_se + j) * CIK_RB_BITMAP_WIDTH_PER_SH); } } cik_select_se_sh(rdev, 0xffffffff, 0xffffffff); mask = 1; - for (i = 0; i < max_rb_num; i++) { + for (i = 0; i < max_rb_num_per_se * se_num; i++) { if (!(disabled_rbs & mask)) enabled_rbs |= mask; mask <<= 1; } + rdev->config.cik.backend_enable_mask = enabled_rbs; + for (i = 0; i < se_num; i++) { cik_select_se_sh(rdev, i, 0xffffffff); data = 0; for (j = 0; j < sh_per_se; j++) { switch (enabled_rbs & 3) { + case 0: + if (j == 0) + data |= PKR_MAP(RASTER_CONFIG_RB_MAP_3); + else + data |= PKR_MAP(RASTER_CONFIG_RB_MAP_0); + break; case 1: data |= (RASTER_CONFIG_RB_MAP_0 << (i * sh_per_se + j) * 2); break; @@ -2711,6 +3179,23 @@ static void cik_gpu_init(struct radeon_device *rdev) rdev->config.cik.sc_earlyz_tile_fifo_size = 0x130; gb_addr_config = BONAIRE_GB_ADDR_CONFIG_GOLDEN; break; + case CHIP_HAWAII: + rdev->config.cik.max_shader_engines = 4; + rdev->config.cik.max_tile_pipes = 16; + rdev->config.cik.max_cu_per_sh = 11; + rdev->config.cik.max_sh_per_se = 1; + rdev->config.cik.max_backends_per_se = 4; + rdev->config.cik.max_texture_channel_caches = 16; + rdev->config.cik.max_gprs = 256; + rdev->config.cik.max_gs_threads = 32; + rdev->config.cik.max_hw_contexts = 8; + + rdev->config.cik.sc_prim_fifo_size_frontend = 0x20; + rdev->config.cik.sc_prim_fifo_size_backend = 0x100; + rdev->config.cik.sc_hiz_tile_fifo_size = 0x30; + rdev->config.cik.sc_earlyz_tile_fifo_size = 0x130; + gb_addr_config = HAWAII_GB_ADDR_CONFIG_GOLDEN; + break; case CHIP_KAVERI: rdev->config.cik.max_shader_engines = 1; rdev->config.cik.max_tile_pipes = 4; @@ -3076,17 +3561,24 @@ void cik_fence_compute_ring_emit(struct radeon_device *rdev, radeon_ring_write(ring, 0); } -void cik_semaphore_ring_emit(struct radeon_device *rdev, +bool cik_semaphore_ring_emit(struct radeon_device *rdev, struct radeon_ring *ring, struct radeon_semaphore *semaphore, bool emit_wait) { +/* TODO: figure out why semaphore cause lockups */ +#if 0 uint64_t addr = semaphore->gpu_addr; unsigned sel = emit_wait ? PACKET3_SEM_SEL_WAIT : PACKET3_SEM_SEL_SIGNAL; radeon_ring_write(ring, PACKET3(PACKET3_MEM_SEMAPHORE, 1)); radeon_ring_write(ring, addr & 0xffffffff); radeon_ring_write(ring, (upper_32_bits(addr) & 0xffff) | sel); + + return true; +#else + return false; +#endif } /** @@ -3129,13 +3621,8 @@ int cik_copy_cpdma(struct radeon_device *rdev, return r; } - if (radeon_fence_need_sync(*fence, ring->idx)) { - radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring, - ring->idx); - radeon_fence_note_sync(*fence, ring->idx); - } else { - radeon_semaphore_free(rdev, &sem, NULL); - } + radeon_semaphore_sync_to(sem, *fence); + radeon_semaphore_sync_rings(rdev, sem, ring->idx); for (i = 0; i < num_loops; i++) { cur_size_in_bytes = size_in_bytes; @@ -3254,6 +3741,7 @@ int cik_ib_test(struct radeon_device *rdev, struct radeon_ring *ring) r = radeon_ib_get(rdev, ring->idx, &ib, NULL, 256); if (r) { DRM_ERROR("radeon: failed to get ib (%d).\n", r); + radeon_scratch_free(rdev, scratch); return r; } ib.ptr[0] = PACKET3(PACKET3_SET_UCONFIG_REG, 1); @@ -3270,6 +3758,8 @@ int cik_ib_test(struct radeon_device *rdev, struct radeon_ring *ring) r = radeon_fence_wait(ib.fence, false); if (r) { DRM_ERROR("radeon: fence wait failed (%d).\n", r); + radeon_scratch_free(rdev, scratch); + radeon_ib_free(rdev, &ib); return r; } for (i = 0; i < rdev->usec_timeout; i++) { @@ -3471,7 +3961,8 @@ static int cik_cp_gfx_resume(struct radeon_device *rdev) int r; WREG32(CP_SEM_WAIT_TIMER, 0x0); - WREG32(CP_SEM_INCOMPLETE_TIMER_CNTL, 0x0); + if (rdev->family != CHIP_HAWAII) + WREG32(CP_SEM_INCOMPLETE_TIMER_CNTL, 0x0); /* Set the write pointer delay */ WREG32(CP_RB_WPTR_DELAY, 0); @@ -3568,7 +4059,7 @@ void cik_compute_ring_set_wptr(struct radeon_device *rdev, struct radeon_ring *ring) { rdev->wb.wb[ring->wptr_offs/4] = cpu_to_le32(ring->wptr); - WDOORBELL32(ring->doorbell_offset, ring->wptr); + WDOORBELL32(ring->doorbell_index, ring->wptr); } /** @@ -3909,10 +4400,6 @@ static int cik_cp_compute_resume(struct radeon_device *rdev) return r; } - /* doorbell offset */ - rdev->ring[idx].doorbell_offset = - (rdev->ring[idx].doorbell_page_num * PAGE_SIZE) + 0; - /* init the mqd struct */ memset(buf, 0, sizeof(struct bonaire_mqd)); @@ -4024,7 +4511,7 @@ static int cik_cp_compute_resume(struct radeon_device *rdev) RREG32(CP_HQD_PQ_DOORBELL_CONTROL); mqd->queue_state.cp_hqd_pq_doorbell_control &= ~DOORBELL_OFFSET_MASK; mqd->queue_state.cp_hqd_pq_doorbell_control |= - DOORBELL_OFFSET(rdev->ring[idx].doorbell_offset / 4); + DOORBELL_OFFSET(rdev->ring[idx].doorbell_index); mqd->queue_state.cp_hqd_pq_doorbell_control |= DOORBELL_EN; mqd->queue_state.cp_hqd_pq_doorbell_control &= ~(DOORBELL_SOURCE | DOORBELL_HIT); @@ -4259,6 +4746,10 @@ static void cik_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask) dev_info(rdev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n", RREG32(VM_CONTEXT1_PROTECTION_FAULT_STATUS)); + /* disable CG/PG */ + cik_fini_pg(rdev); + cik_fini_cg(rdev); + /* stop the rlc */ cik_rlc_stop(rdev); @@ -4804,12 +5295,17 @@ void cik_vm_fini(struct radeon_device *rdev) static void cik_vm_decode_fault(struct radeon_device *rdev, u32 status, u32 addr, u32 mc_client) { - u32 mc_id = (status & MEMORY_CLIENT_ID_MASK) >> MEMORY_CLIENT_ID_SHIFT; + u32 mc_id; u32 vmid = (status & FAULT_VMID_MASK) >> FAULT_VMID_SHIFT; u32 protections = (status & PROTECTIONS_MASK) >> PROTECTIONS_SHIFT; char block[5] = { mc_client >> 24, (mc_client >> 16) & 0xff, (mc_client >> 8) & 0xff, mc_client & 0xff, 0 }; + if (rdev->family == CHIP_HAWAII) + mc_id = (status & HAWAII_MEMORY_CLIENT_ID_MASK) >> MEMORY_CLIENT_ID_SHIFT; + else + mc_id = (status & MEMORY_CLIENT_ID_MASK) >> MEMORY_CLIENT_ID_SHIFT; + printk("VM fault (0x%02x, vmid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, addr, (status & MEMORY_CLIENT_RW_MASK) ? "write" : "read", @@ -5066,6 +5562,7 @@ static int cik_rlc_resume(struct radeon_device *rdev) switch (rdev->family) { case CHIP_BONAIRE: + case CHIP_HAWAII: default: size = BONAIRE_RLC_UCODE_SIZE; break; @@ -5822,6 +6319,10 @@ void cik_get_csb_buffer(struct radeon_device *rdev, volatile u32 *buffer) buffer[count++] = cpu_to_le32(0x00000000); /* XXX */ buffer[count++] = cpu_to_le32(0x00000000); break; + case CHIP_HAWAII: + buffer[count++] = 0x3a00161a; + buffer[count++] = 0x0000002e; + break; default: buffer[count++] = cpu_to_le32(0x00000000); buffer[count++] = cpu_to_le32(0x00000000); @@ -7341,14 +7842,14 @@ int cik_init(struct radeon_device *rdev) ring = &rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX]; ring->ring_obj = NULL; r600_ring_init(rdev, ring, 1024 * 1024); - r = radeon_doorbell_get(rdev, &ring->doorbell_page_num); + r = radeon_doorbell_get(rdev, &ring->doorbell_index); if (r) return r; ring = &rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX]; ring->ring_obj = NULL; r600_ring_init(rdev, ring, 1024 * 1024); - r = radeon_doorbell_get(rdev, &ring->doorbell_page_num); + r = radeon_doorbell_get(rdev, &ring->doorbell_index); if (r) return r; diff --git a/drivers/gpu/drm/radeon/cik_sdma.c b/drivers/gpu/drm/radeon/cik_sdma.c index 8d84ebe2b6fa..d08b83c6267b 100644 --- a/drivers/gpu/drm/radeon/cik_sdma.c +++ b/drivers/gpu/drm/radeon/cik_sdma.c @@ -102,14 +102,6 @@ void cik_sdma_fence_ring_emit(struct radeon_device *rdev, { struct radeon_ring *ring = &rdev->ring[fence->ring]; u64 addr = rdev->fence_drv[fence->ring].gpu_addr; - u32 extra_bits = (SDMA_POLL_REG_MEM_EXTRA_OP(1) | - SDMA_POLL_REG_MEM_EXTRA_FUNC(3)); /* == */ - u32 ref_and_mask; - - if (fence->ring == R600_RING_TYPE_DMA_INDEX) - ref_and_mask = SDMA0; - else - ref_and_mask = SDMA1; /* write the fence */ radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_FENCE, 0, 0)); @@ -119,12 +111,12 @@ void cik_sdma_fence_ring_emit(struct radeon_device *rdev, /* generate an interrupt */ radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_TRAP, 0, 0)); /* flush HDP */ - radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_POLL_REG_MEM, 0, extra_bits)); - radeon_ring_write(ring, GPU_HDP_FLUSH_DONE); - radeon_ring_write(ring, GPU_HDP_FLUSH_REQ); - radeon_ring_write(ring, ref_and_mask); /* REFERENCE */ - radeon_ring_write(ring, ref_and_mask); /* MASK */ - radeon_ring_write(ring, (4 << 16) | 10); /* RETRY_COUNT, POLL_INTERVAL */ + /* We should be using the new POLL_REG_MEM special op packet here + * but it causes sDMA to hang sometimes + */ + radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SRBM_WRITE, 0, 0xf000)); + radeon_ring_write(ring, HDP_MEM_COHERENCY_FLUSH_CNTL >> 2); + radeon_ring_write(ring, 0); } /** @@ -138,7 +130,7 @@ void cik_sdma_fence_ring_emit(struct radeon_device *rdev, * Add a DMA semaphore packet to the ring wait on or signal * other rings (CIK). */ -void cik_sdma_semaphore_ring_emit(struct radeon_device *rdev, +bool cik_sdma_semaphore_ring_emit(struct radeon_device *rdev, struct radeon_ring *ring, struct radeon_semaphore *semaphore, bool emit_wait) @@ -149,6 +141,8 @@ void cik_sdma_semaphore_ring_emit(struct radeon_device *rdev, radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SEMAPHORE, 0, extra_bits)); radeon_ring_write(ring, addr & 0xfffffff8); radeon_ring_write(ring, upper_32_bits(addr) & 0xffffffff); + + return true; } /** @@ -451,13 +445,8 @@ int cik_copy_dma(struct radeon_device *rdev, return r; } - if (radeon_fence_need_sync(*fence, ring->idx)) { - radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring, - ring->idx); - radeon_fence_note_sync(*fence, ring->idx); - } else { - radeon_semaphore_free(rdev, &sem, NULL); - } + radeon_semaphore_sync_to(sem, *fence); + radeon_semaphore_sync_rings(rdev, sem, ring->idx); for (i = 0; i < num_loops; i++) { cur_size_in_bytes = size_in_bytes; @@ -469,7 +458,7 @@ int cik_copy_dma(struct radeon_device *rdev, radeon_ring_write(ring, 0); /* src/dst endian swap */ radeon_ring_write(ring, src_offset & 0xffffffff); radeon_ring_write(ring, upper_32_bits(src_offset) & 0xffffffff); - radeon_ring_write(ring, dst_offset & 0xfffffffc); + radeon_ring_write(ring, dst_offset & 0xffffffff); radeon_ring_write(ring, upper_32_bits(dst_offset) & 0xffffffff); src_offset += cur_size_in_bytes; dst_offset += cur_size_in_bytes; @@ -720,18 +709,10 @@ void cik_sdma_vm_set_page(struct radeon_device *rdev, void cik_dma_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm) { struct radeon_ring *ring = &rdev->ring[ridx]; - u32 extra_bits = (SDMA_POLL_REG_MEM_EXTRA_OP(1) | - SDMA_POLL_REG_MEM_EXTRA_FUNC(3)); /* == */ - u32 ref_and_mask; if (vm == NULL) return; - if (ridx == R600_RING_TYPE_DMA_INDEX) - ref_and_mask = SDMA0; - else - ref_and_mask = SDMA1; - radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SRBM_WRITE, 0, 0xf000)); if (vm->id < 8) { radeon_ring_write(ring, (VM_CONTEXT0_PAGE_TABLE_BASE_ADDR + (vm->id << 2)) >> 2); @@ -766,12 +747,12 @@ void cik_dma_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm radeon_ring_write(ring, VMID(0)); /* flush HDP */ - radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_POLL_REG_MEM, 0, extra_bits)); - radeon_ring_write(ring, GPU_HDP_FLUSH_DONE); - radeon_ring_write(ring, GPU_HDP_FLUSH_REQ); - radeon_ring_write(ring, ref_and_mask); /* REFERENCE */ - radeon_ring_write(ring, ref_and_mask); /* MASK */ - radeon_ring_write(ring, (4 << 16) | 10); /* RETRY_COUNT, POLL_INTERVAL */ + /* We should be using the new POLL_REG_MEM special op packet here + * but it causes sDMA to hang sometimes + */ + radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SRBM_WRITE, 0, 0xf000)); + radeon_ring_write(ring, HDP_MEM_COHERENCY_FLUSH_CNTL >> 2); + radeon_ring_write(ring, 0); /* flush TLB */ radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SRBM_WRITE, 0, 0xf000)); diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h index 380cea311a2b..5964af5e5b2d 100644 --- a/drivers/gpu/drm/radeon/cikd.h +++ b/drivers/gpu/drm/radeon/cikd.h @@ -25,8 +25,10 @@ #define CIK_H #define BONAIRE_GB_ADDR_CONFIG_GOLDEN 0x12010001 +#define HAWAII_GB_ADDR_CONFIG_GOLDEN 0x12011003 -#define CIK_RB_BITMAP_WIDTH_PER_SH 2 +#define CIK_RB_BITMAP_WIDTH_PER_SH 2 +#define HAWAII_RB_BITMAP_WIDTH_PER_SH 4 /* DIDT IND registers */ #define DIDT_SQ_CTRL0 0x0 @@ -499,6 +501,7 @@ * bit 4: write */ #define MEMORY_CLIENT_ID_MASK (0xff << 12) +#define HAWAII_MEMORY_CLIENT_ID_MASK (0x1ff << 12) #define MEMORY_CLIENT_ID_SHIFT 12 #define MEMORY_CLIENT_RW_MASK (1 << 24) #define MEMORY_CLIENT_RW_SHIFT 24 @@ -1162,6 +1165,8 @@ # define ADDR_SURF_P8_32x32_16x16 12 # define ADDR_SURF_P8_32x32_16x32 13 # define ADDR_SURF_P8_32x64_32x32 14 +# define ADDR_SURF_P16_32x32_8x16 16 +# define ADDR_SURF_P16_32x32_16x16 17 # define TILE_SPLIT(x) ((x) << 11) # define ADDR_SURF_TILE_SPLIT_64B 0 # define ADDR_SURF_TILE_SPLIT_128B 1 @@ -1455,6 +1460,7 @@ # define RASTER_CONFIG_RB_MAP_1 1 # define RASTER_CONFIG_RB_MAP_2 2 # define RASTER_CONFIG_RB_MAP_3 3 +#define PKR_MAP(x) ((x) << 8) #define VGT_EVENT_INITIATOR 0x28a90 # define SAMPLE_STREAMOUTSTATS1 (1 << 0) diff --git a/drivers/gpu/drm/radeon/cypress_dpm.c b/drivers/gpu/drm/radeon/cypress_dpm.c index 91bb470de0a3..920e1e4a52c5 100644 --- a/drivers/gpu/drm/radeon/cypress_dpm.c +++ b/drivers/gpu/drm/radeon/cypress_dpm.c @@ -299,7 +299,9 @@ void cypress_program_response_times(struct radeon_device *rdev) static int cypress_pcie_performance_request(struct radeon_device *rdev, u8 perf_req, bool advertise) { +#if defined(CONFIG_ACPI) struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev); +#endif u32 tmp; udelay(10); diff --git a/drivers/gpu/drm/radeon/dce6_afmt.c b/drivers/gpu/drm/radeon/dce6_afmt.c index 5929056beca6..713a5d359901 100644 --- a/drivers/gpu/drm/radeon/dce6_afmt.c +++ b/drivers/gpu/drm/radeon/dce6_afmt.c @@ -93,11 +93,13 @@ void dce6_afmt_select_pin(struct drm_encoder *encoder) struct radeon_device *rdev = encoder->dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; - u32 offset = dig->afmt->offset; + u32 offset; - if (!dig->afmt->pin) + if (!dig || !dig->afmt || !dig->afmt->pin) return; + offset = dig->afmt->offset; + WREG32(AFMT_AUDIO_SRC_CONTROL + offset, AFMT_AUDIO_SRC_SELECT(dig->afmt->pin->id)); } @@ -112,7 +114,7 @@ void dce6_afmt_write_latency_fields(struct drm_encoder *encoder, struct radeon_connector *radeon_connector = NULL; u32 tmp = 0, offset; - if (!dig->afmt->pin) + if (!dig || !dig->afmt || !dig->afmt->pin) return; offset = dig->afmt->pin->offset; @@ -156,7 +158,7 @@ void dce6_afmt_write_speaker_allocation(struct drm_encoder *encoder) u8 *sadb; int sad_count; - if (!dig->afmt->pin) + if (!dig || !dig->afmt || !dig->afmt->pin) return; offset = dig->afmt->pin->offset; @@ -172,7 +174,7 @@ void dce6_afmt_write_speaker_allocation(struct drm_encoder *encoder) } sad_count = drm_edid_to_speaker_allocation(radeon_connector->edid, &sadb); - if (sad_count < 0) { + if (sad_count <= 0) { DRM_ERROR("Couldn't read Speaker Allocation Data Block: %d\n", sad_count); return; } @@ -217,7 +219,7 @@ void dce6_afmt_write_sad_regs(struct drm_encoder *encoder) { AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR13, HDMI_AUDIO_CODING_TYPE_WMA_PRO }, }; - if (!dig->afmt->pin) + if (!dig || !dig->afmt || !dig->afmt->pin) return; offset = dig->afmt->pin->offset; @@ -233,7 +235,7 @@ void dce6_afmt_write_sad_regs(struct drm_encoder *encoder) } sad_count = drm_edid_to_sad(radeon_connector->edid, &sads); - if (sad_count < 0) { + if (sad_count <= 0) { DRM_ERROR("Couldn't read SADs: %d\n", sad_count); return; } @@ -241,20 +243,30 @@ void dce6_afmt_write_sad_regs(struct drm_encoder *encoder) for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) { u32 value = 0; + u8 stereo_freqs = 0; + int max_channels = -1; int j; for (j = 0; j < sad_count; j++) { struct cea_sad *sad = &sads[j]; if (sad->format == eld_reg_to_type[i][1]) { - value = MAX_CHANNELS(sad->channels) | - DESCRIPTOR_BYTE_2(sad->byte2) | - SUPPORTED_FREQUENCIES(sad->freq); + if (sad->channels > max_channels) { + value = MAX_CHANNELS(sad->channels) | + DESCRIPTOR_BYTE_2(sad->byte2) | + SUPPORTED_FREQUENCIES(sad->freq); + max_channels = sad->channels; + } + if (sad->format == HDMI_AUDIO_CODING_TYPE_PCM) - value |= SUPPORTED_FREQUENCIES_STEREO(sad->freq); - break; + stereo_freqs |= sad->freq; + else + break; } } + + value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs); + WREG32_ENDPOINT(offset, eld_reg_to_type[i][0], value); } @@ -296,7 +308,9 @@ int dce6_audio_init(struct radeon_device *rdev) rdev->audio.enabled = true; if (ASIC_IS_DCE8(rdev)) - rdev->audio.num_pins = 7; + rdev->audio.num_pins = 6; + else if (ASIC_IS_DCE61(rdev)) + rdev->audio.num_pins = 4; else rdev->audio.num_pins = 6; diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index 7a84d0cdeda7..9702e55e924e 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -1174,23 +1174,16 @@ int evergreen_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk) void evergreen_fix_pci_max_read_req_size(struct radeon_device *rdev) { - u16 ctl, v; - int err; - - err = pcie_capability_read_word(rdev->pdev, PCI_EXP_DEVCTL, &ctl); - if (err) - return; - - v = (ctl & PCI_EXP_DEVCTL_READRQ) >> 12; + int readrq; + u16 v; + readrq = pcie_get_readrq(rdev->pdev); + v = ffs(readrq) - 8; /* if bios or OS sets MAX_READ_REQUEST_SIZE to an invalid value, fix it * to avoid hangs or perfomance issues */ - if ((v == 0) || (v == 6) || (v == 7)) { - ctl &= ~PCI_EXP_DEVCTL_READRQ; - ctl |= (2 << 12); - pcie_capability_write_word(rdev->pdev, PCI_EXP_DEVCTL, ctl); - } + if ((v == 0) || (v == 6) || (v == 7)) + pcie_set_readrq(rdev->pdev, 512); } void dce4_program_fmt(struct drm_encoder *encoder) @@ -3187,7 +3180,7 @@ static void evergreen_gpu_init(struct radeon_device *rdev) rdev->config.evergreen.sx_max_export_size = 256; rdev->config.evergreen.sx_max_export_pos_size = 64; rdev->config.evergreen.sx_max_export_smx_size = 192; - rdev->config.evergreen.max_hw_contexts = 8; + rdev->config.evergreen.max_hw_contexts = 4; rdev->config.evergreen.sq_num_cf_insts = 2; rdev->config.evergreen.sc_prim_fifo_size = 0x40; diff --git a/drivers/gpu/drm/radeon/evergreen_dma.c b/drivers/gpu/drm/radeon/evergreen_dma.c index 6a0656d00ed0..a37b54436382 100644 --- a/drivers/gpu/drm/radeon/evergreen_dma.c +++ b/drivers/gpu/drm/radeon/evergreen_dma.c @@ -131,13 +131,8 @@ int evergreen_copy_dma(struct radeon_device *rdev, return r; } - if (radeon_fence_need_sync(*fence, ring->idx)) { - radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring, - ring->idx); - radeon_fence_note_sync(*fence, ring->idx); - } else { - radeon_semaphore_free(rdev, &sem, NULL); - } + radeon_semaphore_sync_to(sem, *fence); + radeon_semaphore_sync_rings(rdev, sem, ring->idx); for (i = 0; i < num_loops; i++) { cur_size_in_dw = size_in_dw; diff --git a/drivers/gpu/drm/radeon/evergreen_hdmi.c b/drivers/gpu/drm/radeon/evergreen_hdmi.c index 678736542ed8..0c6d5cef4cf1 100644 --- a/drivers/gpu/drm/radeon/evergreen_hdmi.c +++ b/drivers/gpu/drm/radeon/evergreen_hdmi.c @@ -118,7 +118,7 @@ static void dce4_afmt_write_speaker_allocation(struct drm_encoder *encoder) } sad_count = drm_edid_to_speaker_allocation(radeon_connector->edid, &sadb); - if (sad_count < 0) { + if (sad_count <= 0) { DRM_ERROR("Couldn't read Speaker Allocation Data Block: %d\n", sad_count); return; } @@ -173,7 +173,7 @@ static void evergreen_hdmi_write_sad_regs(struct drm_encoder *encoder) } sad_count = drm_edid_to_sad(radeon_connector->edid, &sads); - if (sad_count < 0) { + if (sad_count <= 0) { DRM_ERROR("Couldn't read SADs: %d\n", sad_count); return; } @@ -181,20 +181,30 @@ static void evergreen_hdmi_write_sad_regs(struct drm_encoder *encoder) for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) { u32 value = 0; + u8 stereo_freqs = 0; + int max_channels = -1; int j; for (j = 0; j < sad_count; j++) { struct cea_sad *sad = &sads[j]; if (sad->format == eld_reg_to_type[i][1]) { - value = MAX_CHANNELS(sad->channels) | - DESCRIPTOR_BYTE_2(sad->byte2) | - SUPPORTED_FREQUENCIES(sad->freq); + if (sad->channels > max_channels) { + value = MAX_CHANNELS(sad->channels) | + DESCRIPTOR_BYTE_2(sad->byte2) | + SUPPORTED_FREQUENCIES(sad->freq); + max_channels = sad->channels; + } + if (sad->format == HDMI_AUDIO_CODING_TYPE_PCM) - value |= SUPPORTED_FREQUENCIES_STEREO(sad->freq); - break; + stereo_freqs |= sad->freq; + else + break; } } + + value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs); + WREG32(eld_reg_to_type[i][0], value); } @@ -330,8 +340,8 @@ void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode /* fglrx clears sth in AFMT_AUDIO_PACKET_CONTROL2 here */ WREG32(HDMI_ACR_PACKET_CONTROL + offset, - HDMI_ACR_AUTO_SEND | /* allow hw to sent ACR packets when required */ - HDMI_ACR_SOURCE); /* select SW CTS value */ + HDMI_ACR_SOURCE | /* select SW CTS value */ + HDMI_ACR_AUTO_SEND); /* allow hw to sent ACR packets when required */ evergreen_hdmi_update_ACR(encoder, mode->clock); diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h index 11e002a47b55..17f990798992 100644 --- a/drivers/gpu/drm/radeon/evergreend.h +++ b/drivers/gpu/drm/radeon/evergreend.h @@ -1571,7 +1571,7 @@ * 6. COMMAND [29:22] | BYTE_COUNT [20:0] */ # define PACKET3_CP_DMA_DST_SEL(x) ((x) << 20) - /* 0 - SRC_ADDR + /* 0 - DST_ADDR * 1 - GDS */ # define PACKET3_CP_DMA_ENGINE(x) ((x) << 27) @@ -1586,7 +1586,7 @@ # define PACKET3_CP_DMA_CP_SYNC (1 << 31) /* COMMAND */ # define PACKET3_CP_DMA_DIS_WC (1 << 21) -# define PACKET3_CP_DMA_CMD_SRC_SWAP(x) ((x) << 23) +# define PACKET3_CP_DMA_CMD_SRC_SWAP(x) ((x) << 22) /* 0 - none * 1 - 8 in 16 * 2 - 8 in 32 diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c index 71399065db04..b41905573cd2 100644 --- a/drivers/gpu/drm/radeon/kv_dpm.c +++ b/drivers/gpu/drm/radeon/kv_dpm.c @@ -2635,7 +2635,7 @@ int kv_dpm_init(struct radeon_device *rdev) pi->caps_sclk_ds = true; pi->enable_auto_thermal_throttling = true; pi->disable_nb_ps3_in_battery = false; - pi->bapm_enable = true; + pi->bapm_enable = false; pi->voltage_drop_t = 0; pi->caps_sclk_throttle_low_notification = false; pi->caps_fps = false; /* true? */ diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c index e299a38e683a..f59a9e9fccf8 100644 --- a/drivers/gpu/drm/radeon/ni.c +++ b/drivers/gpu/drm/radeon/ni.c @@ -799,6 +799,7 @@ int ni_init_microcode(struct radeon_device *rdev) fw_name); release_firmware(rdev->smc_fw); rdev->smc_fw = NULL; + err = 0; } else if (rdev->smc_fw->size != smc_req_size) { printk(KERN_ERR "ni_mc: Bogus length %zu in firmware \"%s\"\n", @@ -894,6 +895,10 @@ static void cayman_gpu_init(struct radeon_device *rdev) (rdev->pdev->device == 0x999C)) { rdev->config.cayman.max_simds_per_se = 6; rdev->config.cayman.max_backends_per_se = 2; + rdev->config.cayman.max_hw_contexts = 8; + rdev->config.cayman.sx_max_export_size = 256; + rdev->config.cayman.sx_max_export_pos_size = 64; + rdev->config.cayman.sx_max_export_smx_size = 192; } else if ((rdev->pdev->device == 0x9903) || (rdev->pdev->device == 0x9904) || (rdev->pdev->device == 0x990A) || @@ -904,6 +909,10 @@ static void cayman_gpu_init(struct radeon_device *rdev) (rdev->pdev->device == 0x999D)) { rdev->config.cayman.max_simds_per_se = 4; rdev->config.cayman.max_backends_per_se = 2; + rdev->config.cayman.max_hw_contexts = 8; + rdev->config.cayman.sx_max_export_size = 256; + rdev->config.cayman.sx_max_export_pos_size = 64; + rdev->config.cayman.sx_max_export_smx_size = 192; } else if ((rdev->pdev->device == 0x9919) || (rdev->pdev->device == 0x9990) || (rdev->pdev->device == 0x9991) || @@ -914,9 +923,17 @@ static void cayman_gpu_init(struct radeon_device *rdev) (rdev->pdev->device == 0x99A0)) { rdev->config.cayman.max_simds_per_se = 3; rdev->config.cayman.max_backends_per_se = 1; + rdev->config.cayman.max_hw_contexts = 4; + rdev->config.cayman.sx_max_export_size = 128; + rdev->config.cayman.sx_max_export_pos_size = 32; + rdev->config.cayman.sx_max_export_smx_size = 96; } else { rdev->config.cayman.max_simds_per_se = 2; rdev->config.cayman.max_backends_per_se = 1; + rdev->config.cayman.max_hw_contexts = 4; + rdev->config.cayman.sx_max_export_size = 128; + rdev->config.cayman.sx_max_export_pos_size = 32; + rdev->config.cayman.sx_max_export_smx_size = 96; } rdev->config.cayman.max_texture_channel_caches = 2; rdev->config.cayman.max_gprs = 256; @@ -924,10 +941,6 @@ static void cayman_gpu_init(struct radeon_device *rdev) rdev->config.cayman.max_gs_threads = 32; rdev->config.cayman.max_stack_entries = 512; rdev->config.cayman.sx_num_of_sets = 8; - rdev->config.cayman.sx_max_export_size = 256; - rdev->config.cayman.sx_max_export_pos_size = 64; - rdev->config.cayman.sx_max_export_smx_size = 192; - rdev->config.cayman.max_hw_contexts = 8; rdev->config.cayman.sq_num_cf_insts = 2; rdev->config.cayman.sc_prim_fifo_size = 0x40; diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c index f26339028154..49c4d48f54d6 100644 --- a/drivers/gpu/drm/radeon/ni_dpm.c +++ b/drivers/gpu/drm/radeon/ni_dpm.c @@ -785,8 +785,8 @@ static void ni_apply_state_adjust_rules(struct radeon_device *rdev, struct ni_ps *ps = ni_get_ps(rps); struct radeon_clock_and_voltage_limits *max_limits; bool disable_mclk_switching; - u32 mclk, sclk; - u16 vddc, vddci; + u32 mclk; + u16 vddci; u32 max_sclk_vddc, max_mclk_vddci, max_mclk_vddc; int i; @@ -839,24 +839,14 @@ static void ni_apply_state_adjust_rules(struct radeon_device *rdev, /* XXX validate the min clocks required for display */ + /* adjust low state */ if (disable_mclk_switching) { - mclk = ps->performance_levels[ps->performance_level_count - 1].mclk; - sclk = ps->performance_levels[0].sclk; - vddc = ps->performance_levels[0].vddc; - vddci = ps->performance_levels[ps->performance_level_count - 1].vddci; - } else { - sclk = ps->performance_levels[0].sclk; - mclk = ps->performance_levels[0].mclk; - vddc = ps->performance_levels[0].vddc; - vddci = ps->performance_levels[0].vddci; + ps->performance_levels[0].mclk = + ps->performance_levels[ps->performance_level_count - 1].mclk; + ps->performance_levels[0].vddci = + ps->performance_levels[ps->performance_level_count - 1].vddci; } - /* adjusted low state */ - ps->performance_levels[0].sclk = sclk; - ps->performance_levels[0].mclk = mclk; - ps->performance_levels[0].vddc = vddc; - ps->performance_levels[0].vddci = vddci; - btc_skip_blacklist_clocks(rdev, max_limits->sclk, max_limits->mclk, &ps->performance_levels[0].sclk, &ps->performance_levels[0].mclk); @@ -868,11 +858,15 @@ static void ni_apply_state_adjust_rules(struct radeon_device *rdev, ps->performance_levels[i].vddc = ps->performance_levels[i - 1].vddc; } + /* adjust remaining states */ if (disable_mclk_switching) { mclk = ps->performance_levels[0].mclk; + vddci = ps->performance_levels[0].vddci; for (i = 1; i < ps->performance_level_count; i++) { if (mclk < ps->performance_levels[i].mclk) mclk = ps->performance_levels[i].mclk; + if (vddci < ps->performance_levels[i].vddci) + vddci = ps->performance_levels[i].vddci; } for (i = 0; i < ps->performance_level_count; i++) { ps->performance_levels[i].mclk = mclk; @@ -3445,9 +3439,9 @@ static int ni_enable_smc_cac(struct radeon_device *rdev, static int ni_pcie_performance_request(struct radeon_device *rdev, u8 perf_req, bool advertise) { +#if defined(CONFIG_ACPI) struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev); -#if defined(CONFIG_ACPI) if ((perf_req == PCIE_PERF_REQ_PECI_GEN1) || (perf_req == PCIE_PERF_REQ_PECI_GEN2)) { if (eg_pi->pcie_performance_request_registered == false) diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index 784983d78158..10abc4d5a6cc 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -869,13 +869,14 @@ void r100_fence_ring_emit(struct radeon_device *rdev, radeon_ring_write(ring, RADEON_SW_INT_FIRE); } -void r100_semaphore_ring_emit(struct radeon_device *rdev, +bool r100_semaphore_ring_emit(struct radeon_device *rdev, struct radeon_ring *ring, struct radeon_semaphore *semaphore, bool emit_wait) { /* Unused on older asics, since we don't have semaphores or multiple rings */ BUG(); + return false; } int r100_copy_blit(struct radeon_device *rdev, diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index af922e297a12..9ad06732a78b 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -2355,6 +2355,7 @@ int r600_init_microcode(struct radeon_device *rdev) fw_name); release_firmware(rdev->smc_fw); rdev->smc_fw = NULL; + err = 0; } else if (rdev->smc_fw->size != smc_req_size) { printk(KERN_ERR "smc: Bogus length %zu in firmware \"%s\"\n", @@ -2649,7 +2650,7 @@ void r600_fence_ring_emit(struct radeon_device *rdev, } } -void r600_semaphore_ring_emit(struct radeon_device *rdev, +bool r600_semaphore_ring_emit(struct radeon_device *rdev, struct radeon_ring *ring, struct radeon_semaphore *semaphore, bool emit_wait) @@ -2663,6 +2664,8 @@ void r600_semaphore_ring_emit(struct radeon_device *rdev, radeon_ring_write(ring, PACKET3(PACKET3_MEM_SEMAPHORE, 1)); radeon_ring_write(ring, addr & 0xffffffff); radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | sel); + + return true; } /** @@ -2705,13 +2708,8 @@ int r600_copy_cpdma(struct radeon_device *rdev, return r; } - if (radeon_fence_need_sync(*fence, ring->idx)) { - radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring, - ring->idx); - radeon_fence_note_sync(*fence, ring->idx); - } else { - radeon_semaphore_free(rdev, &sem, NULL); - } + radeon_semaphore_sync_to(sem, *fence); + radeon_semaphore_sync_rings(rdev, sem, ring->idx); radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1)); radeon_ring_write(ring, (WAIT_UNTIL - PACKET3_SET_CONFIG_REG_OFFSET) >> 2); diff --git a/drivers/gpu/drm/radeon/r600_dma.c b/drivers/gpu/drm/radeon/r600_dma.c index 3b317456512a..7844d15c139f 100644 --- a/drivers/gpu/drm/radeon/r600_dma.c +++ b/drivers/gpu/drm/radeon/r600_dma.c @@ -311,7 +311,7 @@ void r600_dma_fence_ring_emit(struct radeon_device *rdev, * Add a DMA semaphore packet to the ring wait on or signal * other rings (r6xx-SI). */ -void r600_dma_semaphore_ring_emit(struct radeon_device *rdev, +bool r600_dma_semaphore_ring_emit(struct radeon_device *rdev, struct radeon_ring *ring, struct radeon_semaphore *semaphore, bool emit_wait) @@ -322,6 +322,8 @@ void r600_dma_semaphore_ring_emit(struct radeon_device *rdev, radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_SEMAPHORE, 0, s, 0)); radeon_ring_write(ring, addr & 0xfffffffc); radeon_ring_write(ring, upper_32_bits(addr) & 0xff); + + return true; } /** @@ -462,13 +464,8 @@ int r600_copy_dma(struct radeon_device *rdev, return r; } - if (radeon_fence_need_sync(*fence, ring->idx)) { - radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring, - ring->idx); - radeon_fence_note_sync(*fence, ring->idx); - } else { - radeon_semaphore_free(rdev, &sem, NULL); - } + radeon_semaphore_sync_to(sem, *fence); + radeon_semaphore_sync_rings(rdev, sem, ring->idx); for (i = 0; i < num_loops; i++) { cur_size_in_dw = size_in_dw; diff --git a/drivers/gpu/drm/radeon/r600_hdmi.c b/drivers/gpu/drm/radeon/r600_hdmi.c index 21f2b74e20e6..b7d3ecba43e3 100644 --- a/drivers/gpu/drm/radeon/r600_hdmi.c +++ b/drivers/gpu/drm/radeon/r600_hdmi.c @@ -24,6 +24,7 @@ * Authors: Christian König */ #include <linux/hdmi.h> +#include <linux/gcd.h> #include <drm/drmP.h> #include <drm/radeon_drm.h> #include "radeon.h" @@ -57,28 +58,57 @@ enum r600_hdmi_iec_status_bits { static const struct radeon_hdmi_acr r600_hdmi_predefined_acr[] = { /* 32kHz 44.1kHz 48kHz */ /* Clock N CTS N CTS N CTS */ - { 25174, 4576, 28125, 7007, 31250, 6864, 28125 }, /* 25,20/1.001 MHz */ + { 25175, 4096, 25175, 28224, 125875, 6144, 25175 }, /* 25,20/1.001 MHz */ { 25200, 4096, 25200, 6272, 28000, 6144, 25200 }, /* 25.20 MHz */ { 27000, 4096, 27000, 6272, 30000, 6144, 27000 }, /* 27.00 MHz */ { 27027, 4096, 27027, 6272, 30030, 6144, 27027 }, /* 27.00*1.001 MHz */ { 54000, 4096, 54000, 6272, 60000, 6144, 54000 }, /* 54.00 MHz */ { 54054, 4096, 54054, 6272, 60060, 6144, 54054 }, /* 54.00*1.001 MHz */ - { 74175, 11648, 210937, 17836, 234375, 11648, 140625 }, /* 74.25/1.001 MHz */ + { 74176, 4096, 74176, 5733, 75335, 6144, 74176 }, /* 74.25/1.001 MHz */ { 74250, 4096, 74250, 6272, 82500, 6144, 74250 }, /* 74.25 MHz */ - { 148351, 11648, 421875, 8918, 234375, 5824, 140625 }, /* 148.50/1.001 MHz */ + { 148352, 4096, 148352, 5733, 150670, 6144, 148352 }, /* 148.50/1.001 MHz */ { 148500, 4096, 148500, 6272, 165000, 6144, 148500 }, /* 148.50 MHz */ - { 0, 4096, 0, 6272, 0, 6144, 0 } /* Other */ }; + /* - * calculate CTS value if it's not found in the table + * calculate CTS and N values if they are not found in the table */ -static void r600_hdmi_calc_cts(uint32_t clock, int *CTS, int N, int freq) +static void r600_hdmi_calc_cts(uint32_t clock, int *CTS, int *N, int freq) { - if (*CTS == 0) - *CTS = clock * N / (128 * freq) * 1000; - DRM_DEBUG("Using ACR timing N=%d CTS=%d for frequency %d\n", - N, *CTS, freq); + int n, cts; + unsigned long div, mul; + + /* Safe, but overly large values */ + n = 128 * freq; + cts = clock * 1000; + + /* Smallest valid fraction */ + div = gcd(n, cts); + + n /= div; + cts /= div; + + /* + * The optimal N is 128*freq/1000. Calculate the closest larger + * value that doesn't truncate any bits. + */ + mul = ((128*freq/1000) + (n-1))/n; + + n *= mul; + cts *= mul; + + /* Check that we are in spec (not always possible) */ + if (n < (128*freq/1500)) + printk(KERN_WARNING "Calculated ACR N value is too small. You may experience audio problems.\n"); + if (n > (128*freq/300)) + printk(KERN_WARNING "Calculated ACR N value is too large. You may experience audio problems.\n"); + + *N = n; + *CTS = cts; + + DRM_DEBUG("Calculated ACR timing N=%d CTS=%d for frequency %d\n", + *N, *CTS, freq); } struct radeon_hdmi_acr r600_hdmi_acr(uint32_t clock) @@ -86,15 +116,16 @@ struct radeon_hdmi_acr r600_hdmi_acr(uint32_t clock) struct radeon_hdmi_acr res; u8 i; - for (i = 0; r600_hdmi_predefined_acr[i].clock != clock && - r600_hdmi_predefined_acr[i].clock != 0; i++) - ; - res = r600_hdmi_predefined_acr[i]; + /* Precalculated values for common clocks */ + for (i = 0; i < ARRAY_SIZE(r600_hdmi_predefined_acr); i++) { + if (r600_hdmi_predefined_acr[i].clock == clock) + return r600_hdmi_predefined_acr[i]; + } - /* In case some CTS are missing */ - r600_hdmi_calc_cts(clock, &res.cts_32khz, res.n_32khz, 32000); - r600_hdmi_calc_cts(clock, &res.cts_44_1khz, res.n_44_1khz, 44100); - r600_hdmi_calc_cts(clock, &res.cts_48khz, res.n_48khz, 48000); + /* And odd clocks get manually calculated */ + r600_hdmi_calc_cts(clock, &res.cts_32khz, &res.n_32khz, 32000); + r600_hdmi_calc_cts(clock, &res.cts_44_1khz, &res.n_44_1khz, 44100); + r600_hdmi_calc_cts(clock, &res.cts_48khz, &res.n_48khz, 48000); return res; } @@ -273,9 +304,9 @@ void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock) WREG32(DCCG_AUDIO_DTO1_MODULE, dto_modulo); WREG32(DCCG_AUDIO_DTO_SELECT, 1); /* select DTO1 */ } - } else if (ASIC_IS_DCE3(rdev)) { + } else { /* according to the reg specs, this should DCE3.2 only, but in - * practice it seems to cover DCE3.0/3.1 as well. + * practice it seems to cover DCE2.0/3.0/3.1 as well. */ if (dig->dig_encoder == 0) { WREG32(DCCG_AUDIO_DTO0_PHASE, base_rate * 100); @@ -286,10 +317,6 @@ void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock) WREG32(DCCG_AUDIO_DTO1_MODULE, clock * 100); WREG32(DCCG_AUDIO_DTO_SELECT, 1); /* select DTO1 */ } - } else { - /* according to the reg specs, this should be DCE2.0 and DCE3.0/3.1 */ - WREG32(AUDIO_DTO, AUDIO_DTO_PHASE(base_rate / 10) | - AUDIO_DTO_MODULE(clock / 10)); } } @@ -302,6 +329,9 @@ static void dce3_2_afmt_write_speaker_allocation(struct drm_encoder *encoder) u8 *sadb; int sad_count; + /* XXX: setting this register causes hangs on some asics */ + return; + list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) { if (connector->encoder == encoder) { radeon_connector = to_radeon_connector(connector); @@ -378,20 +408,30 @@ static void dce3_2_afmt_write_sad_regs(struct drm_encoder *encoder) for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) { u32 value = 0; + u8 stereo_freqs = 0; + int max_channels = -1; int j; for (j = 0; j < sad_count; j++) { struct cea_sad *sad = &sads[j]; if (sad->format == eld_reg_to_type[i][1]) { - value = MAX_CHANNELS(sad->channels) | - DESCRIPTOR_BYTE_2(sad->byte2) | - SUPPORTED_FREQUENCIES(sad->freq); + if (sad->channels > max_channels) { + value = MAX_CHANNELS(sad->channels) | + DESCRIPTOR_BYTE_2(sad->byte2) | + SUPPORTED_FREQUENCIES(sad->freq); + max_channels = sad->channels; + } + if (sad->format == HDMI_AUDIO_CODING_TYPE_PCM) - value |= SUPPORTED_FREQUENCIES_STEREO(sad->freq); - break; + stereo_freqs |= sad->freq; + else + break; } } + + value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs); + WREG32(eld_reg_to_type[i][0], value); } @@ -448,8 +488,8 @@ void r600_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mod } WREG32(HDMI0_ACR_PACKET_CONTROL + offset, - HDMI0_ACR_AUTO_SEND | /* allow hw to sent ACR packets when required */ - HDMI0_ACR_SOURCE); /* select SW CTS value */ + HDMI0_ACR_SOURCE | /* select SW CTS value - XXX verify that hw CTS works on all families */ + HDMI0_ACR_AUTO_SEND); /* allow hw to sent ACR packets when required */ WREG32(HDMI0_VBI_PACKET_CONTROL + offset, HDMI0_NULL_SEND | /* send null packets when required */ diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h index 8e01b126aaeb..ebe38724a976 100644 --- a/drivers/gpu/drm/radeon/r600d.h +++ b/drivers/gpu/drm/radeon/r600d.h @@ -1551,7 +1551,7 @@ */ # define PACKET3_CP_DMA_CP_SYNC (1 << 31) /* COMMAND */ -# define PACKET3_CP_DMA_CMD_SRC_SWAP(x) ((x) << 23) +# define PACKET3_CP_DMA_CMD_SRC_SWAP(x) ((x) << 22) /* 0 - none * 1 - 8 in 16 * 2 - 8 in 32 diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index b264af6e8b9c..45e1f447bc79 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -348,6 +348,7 @@ int radeon_fence_emit(struct radeon_device *rdev, struct radeon_fence **fence, i void radeon_fence_process(struct radeon_device *rdev, int ring); bool radeon_fence_signaled(struct radeon_fence *fence); int radeon_fence_wait(struct radeon_fence *fence, bool interruptible); +int radeon_fence_wait_locked(struct radeon_fence *fence); int radeon_fence_wait_next_locked(struct radeon_device *rdev, int ring); int radeon_fence_wait_empty_locked(struct radeon_device *rdev, int ring); int radeon_fence_wait_any(struct radeon_device *rdev, @@ -548,17 +549,20 @@ struct radeon_semaphore { struct radeon_sa_bo *sa_bo; signed waiters; uint64_t gpu_addr; + struct radeon_fence *sync_to[RADEON_NUM_RINGS]; }; int radeon_semaphore_create(struct radeon_device *rdev, struct radeon_semaphore **semaphore); -void radeon_semaphore_emit_signal(struct radeon_device *rdev, int ring, +bool radeon_semaphore_emit_signal(struct radeon_device *rdev, int ring, struct radeon_semaphore *semaphore); -void radeon_semaphore_emit_wait(struct radeon_device *rdev, int ring, +bool radeon_semaphore_emit_wait(struct radeon_device *rdev, int ring, struct radeon_semaphore *semaphore); +void radeon_semaphore_sync_to(struct radeon_semaphore *semaphore, + struct radeon_fence *fence); int radeon_semaphore_sync_rings(struct radeon_device *rdev, struct radeon_semaphore *semaphore, - int signaler, int waiter); + int waiting_ring); void radeon_semaphore_free(struct radeon_device *rdev, struct radeon_semaphore **semaphore, struct radeon_fence *fence); @@ -645,13 +649,15 @@ void radeon_scratch_free(struct radeon_device *rdev, uint32_t reg); /* * GPU doorbell structures, functions & helpers */ +#define RADEON_MAX_DOORBELLS 1024 /* Reserve at most 1024 doorbell slots for radeon-owned rings. */ + struct radeon_doorbell { - u32 num_pages; - bool free[1024]; /* doorbell mmio */ - resource_size_t base; - resource_size_t size; - void __iomem *ptr; + resource_size_t base; + resource_size_t size; + u32 __iomem *ptr; + u32 num_doorbells; /* Number of doorbells actually reserved for radeon. */ + unsigned long used[DIV_ROUND_UP(RADEON_MAX_DOORBELLS, BITS_PER_LONG)]; }; int radeon_doorbell_get(struct radeon_device *rdev, u32 *page); @@ -765,7 +771,6 @@ struct radeon_ib { struct radeon_fence *fence; struct radeon_vm *vm; bool is_const_ib; - struct radeon_fence *sync_to[RADEON_NUM_RINGS]; struct radeon_semaphore *semaphore; }; @@ -799,8 +804,7 @@ struct radeon_ring { u32 pipe; u32 queue; struct radeon_bo *mqd_obj; - u32 doorbell_page_num; - u32 doorbell_offset; + u32 doorbell_index; unsigned wptr_offs; }; @@ -921,7 +925,6 @@ int radeon_ib_get(struct radeon_device *rdev, int ring, struct radeon_ib *ib, struct radeon_vm *vm, unsigned size); void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib); -void radeon_ib_sync_to(struct radeon_ib *ib, struct radeon_fence *fence); int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib, struct radeon_ib *const_ib); int radeon_ib_pool_init(struct radeon_device *rdev); @@ -1281,8 +1284,8 @@ struct radeon_blacklist_clocks struct radeon_clock_and_voltage_limits { u32 sclk; u32 mclk; - u32 vddc; - u32 vddci; + u16 vddc; + u16 vddci; }; struct radeon_clock_array { @@ -1638,7 +1641,7 @@ struct radeon_asic_ring { /* command emmit functions */ void (*ib_execute)(struct radeon_device *rdev, struct radeon_ib *ib); void (*emit_fence)(struct radeon_device *rdev, struct radeon_fence *fence); - void (*emit_semaphore)(struct radeon_device *rdev, struct radeon_ring *cp, + bool (*emit_semaphore)(struct radeon_device *rdev, struct radeon_ring *cp, struct radeon_semaphore *semaphore, bool emit_wait); void (*vm_flush)(struct radeon_device *rdev, int ridx, struct radeon_vm *vm); @@ -1937,7 +1940,7 @@ struct si_asic { unsigned sc_earlyz_tile_fifo_size; unsigned num_tile_pipes; - unsigned num_backends_per_se; + unsigned backend_enable_mask; unsigned backend_disable_mask_per_asic; unsigned backend_map; unsigned num_texture_channel_caches; @@ -1967,7 +1970,7 @@ struct cik_asic { unsigned sc_earlyz_tile_fifo_size; unsigned num_tile_pipes; - unsigned num_backends_per_se; + unsigned backend_enable_mask; unsigned backend_disable_mask_per_asic; unsigned backend_map; unsigned num_texture_channel_caches; @@ -1979,6 +1982,7 @@ struct cik_asic { unsigned tile_config; uint32_t tile_mode_array[32]; + uint32_t macrotile_mode_array[16]; }; union radeon_asic_config { @@ -2239,8 +2243,8 @@ void r100_mm_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v, u32 r100_io_rreg(struct radeon_device *rdev, u32 reg); void r100_io_wreg(struct radeon_device *rdev, u32 reg, u32 v); -u32 cik_mm_rdoorbell(struct radeon_device *rdev, u32 offset); -void cik_mm_wdoorbell(struct radeon_device *rdev, u32 offset, u32 v); +u32 cik_mm_rdoorbell(struct radeon_device *rdev, u32 index); +void cik_mm_wdoorbell(struct radeon_device *rdev, u32 index, u32 v); /* * Cast helper @@ -2303,8 +2307,8 @@ void cik_mm_wdoorbell(struct radeon_device *rdev, u32 offset, u32 v); #define RREG32_IO(reg) r100_io_rreg(rdev, (reg)) #define WREG32_IO(reg, v) r100_io_wreg(rdev, (reg), (v)) -#define RDOORBELL32(offset) cik_mm_rdoorbell(rdev, (offset)) -#define WDOORBELL32(offset, v) cik_mm_wdoorbell(rdev, (offset), (v)) +#define RDOORBELL32(index) cik_mm_rdoorbell(rdev, (index)) +#define WDOORBELL32(index, v) cik_mm_wdoorbell(rdev, (index), (v)) /* * Indirect registers accessor @@ -2706,10 +2710,10 @@ void radeon_vm_fence(struct radeon_device *rdev, struct radeon_vm *vm, struct radeon_fence *fence); uint64_t radeon_vm_map_gart(struct radeon_device *rdev, uint64_t addr); -int radeon_vm_bo_update_pte(struct radeon_device *rdev, - struct radeon_vm *vm, - struct radeon_bo *bo, - struct ttm_mem_reg *mem); +int radeon_vm_bo_update(struct radeon_device *rdev, + struct radeon_vm *vm, + struct radeon_bo *bo, + struct ttm_mem_reg *mem); void radeon_vm_bo_invalidate(struct radeon_device *rdev, struct radeon_bo *bo); struct radeon_bo_va *radeon_vm_bo_find(struct radeon_vm *vm, diff --git a/drivers/gpu/drm/radeon/radeon_acpi.c b/drivers/gpu/drm/radeon/radeon_acpi.c index 10f98c7742d8..98a9074b306b 100644 --- a/drivers/gpu/drm/radeon/radeon_acpi.c +++ b/drivers/gpu/drm/radeon/radeon_acpi.c @@ -369,7 +369,7 @@ int radeon_atif_handler(struct radeon_device *rdev, return NOTIFY_DONE; /* Check pending SBIOS requests */ - handle = DEVICE_ACPI_HANDLE(&rdev->pdev->dev); + handle = ACPI_HANDLE(&rdev->pdev->dev); count = radeon_atif_get_sbios_requests(handle, &req); if (count <= 0) @@ -556,7 +556,7 @@ int radeon_acpi_pcie_notify_device_ready(struct radeon_device *rdev) struct radeon_atcs *atcs = &rdev->atcs; /* Get the device handle */ - handle = DEVICE_ACPI_HANDLE(&rdev->pdev->dev); + handle = ACPI_HANDLE(&rdev->pdev->dev); if (!handle) return -EINVAL; @@ -596,7 +596,7 @@ int radeon_acpi_pcie_performance_request(struct radeon_device *rdev, u32 retry = 3; /* Get the device handle */ - handle = DEVICE_ACPI_HANDLE(&rdev->pdev->dev); + handle = ACPI_HANDLE(&rdev->pdev->dev); if (!handle) return -EINVAL; @@ -699,7 +699,7 @@ int radeon_acpi_init(struct radeon_device *rdev) int ret; /* Get the device handle */ - handle = DEVICE_ACPI_HANDLE(&rdev->pdev->dev); + handle = ACPI_HANDLE(&rdev->pdev->dev); /* No need to proceed if we're sure that ATIF is not supported */ if (!ASIC_IS_AVIVO(rdev) || !rdev->bios || !handle) diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c index d4b91675671d..c0425bb6223a 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.c +++ b/drivers/gpu/drm/radeon/radeon_asic.c @@ -2015,11 +2015,13 @@ static struct radeon_asic ci_asic = { .bandwidth_update = &dce8_bandwidth_update, .get_vblank_counter = &evergreen_get_vblank_counter, .wait_for_vblank = &dce4_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, .hdmi_enable = &evergreen_hdmi_enable, .hdmi_setmode = &evergreen_hdmi_setmode, }, .copy = { - .blit = NULL, + .blit = &cik_copy_cpdma, .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, .dma = &cik_copy_dma, .dma_ring_index = R600_RING_TYPE_DMA_INDEX, @@ -2114,11 +2116,13 @@ static struct radeon_asic kv_asic = { .bandwidth_update = &dce8_bandwidth_update, .get_vblank_counter = &evergreen_get_vblank_counter, .wait_for_vblank = &dce4_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, .hdmi_enable = &evergreen_hdmi_enable, .hdmi_setmode = &evergreen_hdmi_setmode, }, .copy = { - .blit = NULL, + .blit = &cik_copy_cpdma, .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, .dma = &cik_copy_dma, .dma_ring_index = R600_RING_TYPE_DMA_INDEX, @@ -2437,27 +2441,48 @@ int radeon_asic_init(struct radeon_device *rdev) } break; case CHIP_BONAIRE: + case CHIP_HAWAII: rdev->asic = &ci_asic; rdev->num_crtc = 6; rdev->has_uvd = true; - rdev->cg_flags = - RADEON_CG_SUPPORT_GFX_MGCG | - RADEON_CG_SUPPORT_GFX_MGLS | - /*RADEON_CG_SUPPORT_GFX_CGCG |*/ - RADEON_CG_SUPPORT_GFX_CGLS | - RADEON_CG_SUPPORT_GFX_CGTS | - RADEON_CG_SUPPORT_GFX_CGTS_LS | - RADEON_CG_SUPPORT_GFX_CP_LS | - RADEON_CG_SUPPORT_MC_LS | - RADEON_CG_SUPPORT_MC_MGCG | - RADEON_CG_SUPPORT_SDMA_MGCG | - RADEON_CG_SUPPORT_SDMA_LS | - RADEON_CG_SUPPORT_BIF_LS | - RADEON_CG_SUPPORT_VCE_MGCG | - RADEON_CG_SUPPORT_UVD_MGCG | - RADEON_CG_SUPPORT_HDP_LS | - RADEON_CG_SUPPORT_HDP_MGCG; - rdev->pg_flags = 0; + if (rdev->family == CHIP_BONAIRE) { + rdev->cg_flags = + RADEON_CG_SUPPORT_GFX_MGCG | + RADEON_CG_SUPPORT_GFX_MGLS | + /*RADEON_CG_SUPPORT_GFX_CGCG |*/ + RADEON_CG_SUPPORT_GFX_CGLS | + RADEON_CG_SUPPORT_GFX_CGTS | + RADEON_CG_SUPPORT_GFX_CGTS_LS | + RADEON_CG_SUPPORT_GFX_CP_LS | + RADEON_CG_SUPPORT_MC_LS | + RADEON_CG_SUPPORT_MC_MGCG | + RADEON_CG_SUPPORT_SDMA_MGCG | + RADEON_CG_SUPPORT_SDMA_LS | + RADEON_CG_SUPPORT_BIF_LS | + RADEON_CG_SUPPORT_VCE_MGCG | + RADEON_CG_SUPPORT_UVD_MGCG | + RADEON_CG_SUPPORT_HDP_LS | + RADEON_CG_SUPPORT_HDP_MGCG; + rdev->pg_flags = 0; + } else { + rdev->cg_flags = + RADEON_CG_SUPPORT_GFX_MGCG | + RADEON_CG_SUPPORT_GFX_MGLS | + /*RADEON_CG_SUPPORT_GFX_CGCG |*/ + RADEON_CG_SUPPORT_GFX_CGLS | + RADEON_CG_SUPPORT_GFX_CGTS | + RADEON_CG_SUPPORT_GFX_CP_LS | + RADEON_CG_SUPPORT_MC_LS | + RADEON_CG_SUPPORT_MC_MGCG | + RADEON_CG_SUPPORT_SDMA_MGCG | + RADEON_CG_SUPPORT_SDMA_LS | + RADEON_CG_SUPPORT_BIF_LS | + RADEON_CG_SUPPORT_VCE_MGCG | + RADEON_CG_SUPPORT_UVD_MGCG | + RADEON_CG_SUPPORT_HDP_LS | + RADEON_CG_SUPPORT_HDP_MGCG; + rdev->pg_flags = 0; + } break; case CHIP_KAVERI: case CHIP_KABINI: diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h index f2833ee3a613..c9fd97b58076 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.h +++ b/drivers/gpu/drm/radeon/radeon_asic.h @@ -80,7 +80,7 @@ int r100_irq_set(struct radeon_device *rdev); int r100_irq_process(struct radeon_device *rdev); void r100_fence_ring_emit(struct radeon_device *rdev, struct radeon_fence *fence); -void r100_semaphore_ring_emit(struct radeon_device *rdev, +bool r100_semaphore_ring_emit(struct radeon_device *rdev, struct radeon_ring *cp, struct radeon_semaphore *semaphore, bool emit_wait); @@ -313,13 +313,13 @@ int r600_cs_parse(struct radeon_cs_parser *p); int r600_dma_cs_parse(struct radeon_cs_parser *p); void r600_fence_ring_emit(struct radeon_device *rdev, struct radeon_fence *fence); -void r600_semaphore_ring_emit(struct radeon_device *rdev, +bool r600_semaphore_ring_emit(struct radeon_device *rdev, struct radeon_ring *cp, struct radeon_semaphore *semaphore, bool emit_wait); void r600_dma_fence_ring_emit(struct radeon_device *rdev, struct radeon_fence *fence); -void r600_dma_semaphore_ring_emit(struct radeon_device *rdev, +bool r600_dma_semaphore_ring_emit(struct radeon_device *rdev, struct radeon_ring *ring, struct radeon_semaphore *semaphore, bool emit_wait); @@ -566,10 +566,6 @@ int sumo_dpm_force_performance_level(struct radeon_device *rdev, */ void cayman_fence_ring_emit(struct radeon_device *rdev, struct radeon_fence *fence); -void cayman_uvd_semaphore_emit(struct radeon_device *rdev, - struct radeon_ring *ring, - struct radeon_semaphore *semaphore, - bool emit_wait); void cayman_pcie_gart_tlb_flush(struct radeon_device *rdev); int cayman_init(struct radeon_device *rdev); void cayman_fini(struct radeon_device *rdev); @@ -697,7 +693,7 @@ void cik_pciep_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v); int cik_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk); void cik_sdma_fence_ring_emit(struct radeon_device *rdev, struct radeon_fence *fence); -void cik_sdma_semaphore_ring_emit(struct radeon_device *rdev, +bool cik_sdma_semaphore_ring_emit(struct radeon_device *rdev, struct radeon_ring *ring, struct radeon_semaphore *semaphore, bool emit_wait); @@ -717,7 +713,7 @@ void cik_fence_gfx_ring_emit(struct radeon_device *rdev, struct radeon_fence *fence); void cik_fence_compute_ring_emit(struct radeon_device *rdev, struct radeon_fence *fence); -void cik_semaphore_ring_emit(struct radeon_device *rdev, +bool cik_semaphore_ring_emit(struct radeon_device *rdev, struct radeon_ring *cp, struct radeon_semaphore *semaphore, bool emit_wait); @@ -807,7 +803,7 @@ void uvd_v1_0_stop(struct radeon_device *rdev); int uvd_v1_0_ring_test(struct radeon_device *rdev, struct radeon_ring *ring); int uvd_v1_0_ib_test(struct radeon_device *rdev, struct radeon_ring *ring); -void uvd_v1_0_semaphore_emit(struct radeon_device *rdev, +bool uvd_v1_0_semaphore_emit(struct radeon_device *rdev, struct radeon_ring *ring, struct radeon_semaphore *semaphore, bool emit_wait); @@ -819,7 +815,7 @@ void uvd_v2_2_fence_emit(struct radeon_device *rdev, struct radeon_fence *fence); /* uvd v3.1 */ -void uvd_v3_1_semaphore_emit(struct radeon_device *rdev, +bool uvd_v3_1_semaphore_emit(struct radeon_device *rdev, struct radeon_ring *ring, struct radeon_semaphore *semaphore, bool emit_wait); diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c index f79ee184ffd5..5c39bf7c3d88 100644 --- a/drivers/gpu/drm/radeon/radeon_atombios.c +++ b/drivers/gpu/drm/radeon/radeon_atombios.c @@ -2918,7 +2918,7 @@ int radeon_atom_get_memory_pll_dividers(struct radeon_device *rdev, mpll_param->dll_speed = args.ucDllSpeed; mpll_param->bwcntl = args.ucBWCntl; mpll_param->vco_mode = - (args.ucPllCntlFlag & MPLL_CNTL_FLAG_VCO_MODE_MASK) ? 1 : 0; + (args.ucPllCntlFlag & MPLL_CNTL_FLAG_VCO_MODE_MASK); mpll_param->yclk_sel = (args.ucPllCntlFlag & MPLL_CNTL_FLAG_BYPASS_DQ_PLL) ? 1 : 0; mpll_param->qdr = diff --git a/drivers/gpu/drm/radeon/radeon_atpx_handler.c b/drivers/gpu/drm/radeon/radeon_atpx_handler.c index 6153ec18943a..485848f889f5 100644 --- a/drivers/gpu/drm/radeon/radeon_atpx_handler.c +++ b/drivers/gpu/drm/radeon/radeon_atpx_handler.c @@ -8,8 +8,7 @@ */ #include <linux/vga_switcheroo.h> #include <linux/slab.h> -#include <acpi/acpi.h> -#include <acpi/acpi_bus.h> +#include <linux/acpi.h> #include <linux/pci.h> #include "radeon_acpi.h" @@ -34,6 +33,7 @@ static struct radeon_atpx_priv { bool atpx_detected; /* handle for device - and atpx */ acpi_handle dhandle; + acpi_handle other_handle; struct radeon_atpx atpx; } radeon_atpx_priv; @@ -447,14 +447,15 @@ static bool radeon_atpx_pci_probe_handle(struct pci_dev *pdev) acpi_handle dhandle, atpx_handle; acpi_status status; - dhandle = DEVICE_ACPI_HANDLE(&pdev->dev); + dhandle = ACPI_HANDLE(&pdev->dev); if (!dhandle) return false; status = acpi_get_handle(dhandle, "ATPX", &atpx_handle); - if (ACPI_FAILURE(status)) + if (ACPI_FAILURE(status)) { + radeon_atpx_priv.other_handle = dhandle; return false; - + } radeon_atpx_priv.dhandle = dhandle; radeon_atpx_priv.atpx.handle = atpx_handle; return true; @@ -493,7 +494,7 @@ static int radeon_atpx_init(void) */ static int radeon_atpx_get_client_id(struct pci_dev *pdev) { - if (radeon_atpx_priv.dhandle == DEVICE_ACPI_HANDLE(&pdev->dev)) + if (radeon_atpx_priv.dhandle == ACPI_HANDLE(&pdev->dev)) return VGA_SWITCHEROO_IGD; else return VGA_SWITCHEROO_DIS; @@ -531,6 +532,16 @@ static bool radeon_atpx_detect(void) printk(KERN_INFO "VGA switcheroo: detected switching method %s handle\n", acpi_method_name); radeon_atpx_priv.atpx_detected = true; + /* + * On some systems hotplug events are generated for the device + * being switched off when ATPX is executed. They cause ACPI + * hotplug to trigger and attempt to remove the device from + * the system, which causes it to break down. Prevent that from + * happening by setting the no_hotplug flag for the involved + * ACPI device objects. + */ + acpi_bus_no_hotplug(radeon_atpx_priv.dhandle); + acpi_bus_no_hotplug(radeon_atpx_priv.other_handle); return true; } return false; diff --git a/drivers/gpu/drm/radeon/radeon_bios.c b/drivers/gpu/drm/radeon/radeon_bios.c index c155d6f3fa68..b3633d9a5317 100644 --- a/drivers/gpu/drm/radeon/radeon_bios.c +++ b/drivers/gpu/drm/radeon/radeon_bios.c @@ -185,7 +185,7 @@ static bool radeon_atrm_get_bios(struct radeon_device *rdev) return false; while ((pdev = pci_get_class(PCI_CLASS_DISPLAY_VGA << 8, pdev)) != NULL) { - dhandle = DEVICE_ACPI_HANDLE(&pdev->dev); + dhandle = ACPI_HANDLE(&pdev->dev); if (!dhandle) continue; diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index e972143e5a36..20a768ac89a8 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -1720,12 +1720,16 @@ radeon_add_atom_connector(struct drm_device *dev, drm_object_attach_property(&radeon_connector->base.base, rdev->mode_info.underscan_vborder_property, 0); - drm_object_attach_property(&radeon_connector->base.base, - rdev->mode_info.audio_property, - RADEON_AUDIO_AUTO); + drm_object_attach_property(&radeon_connector->base.base, rdev->mode_info.dither_property, RADEON_FMT_DITHER_DISABLE); + + if (radeon_audio != 0) + drm_object_attach_property(&radeon_connector->base.base, + rdev->mode_info.audio_property, + RADEON_AUDIO_AUTO); + subpixel_order = SubPixelHorizontalRGB; connector->interlace_allowed = true; if (connector_type == DRM_MODE_CONNECTOR_HDMIB) @@ -1819,10 +1823,10 @@ radeon_add_atom_connector(struct drm_device *dev, rdev->mode_info.underscan_vborder_property, 0); } - if (ASIC_IS_DCE2(rdev)) { + if (ASIC_IS_DCE2(rdev) && (radeon_audio != 0)) { drm_object_attach_property(&radeon_connector->base.base, - rdev->mode_info.audio_property, - RADEON_AUDIO_AUTO); + rdev->mode_info.audio_property, + RADEON_AUDIO_AUTO); } if (ASIC_IS_AVIVO(rdev)) { drm_object_attach_property(&radeon_connector->base.base, @@ -1869,10 +1873,10 @@ radeon_add_atom_connector(struct drm_device *dev, rdev->mode_info.underscan_vborder_property, 0); } - if (ASIC_IS_DCE2(rdev)) { + if (ASIC_IS_DCE2(rdev) && (radeon_audio != 0)) { drm_object_attach_property(&radeon_connector->base.base, - rdev->mode_info.audio_property, - RADEON_AUDIO_AUTO); + rdev->mode_info.audio_property, + RADEON_AUDIO_AUTO); } if (ASIC_IS_AVIVO(rdev)) { drm_object_attach_property(&radeon_connector->base.base, @@ -1918,15 +1922,16 @@ radeon_add_atom_connector(struct drm_device *dev, rdev->mode_info.underscan_vborder_property, 0); } - if (ASIC_IS_DCE2(rdev)) { + if (ASIC_IS_DCE2(rdev) && (radeon_audio != 0)) { drm_object_attach_property(&radeon_connector->base.base, - rdev->mode_info.audio_property, - RADEON_AUDIO_AUTO); + rdev->mode_info.audio_property, + RADEON_AUDIO_AUTO); } if (ASIC_IS_AVIVO(rdev)) { drm_object_attach_property(&radeon_connector->base.base, rdev->mode_info.dither_property, RADEON_FMT_DITHER_DISABLE); + } connector->interlace_allowed = true; /* in theory with a DP to VGA converter... */ diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c index c8ab019a980e..0b366169d64d 100644 --- a/drivers/gpu/drm/radeon/radeon_cs.c +++ b/drivers/gpu/drm/radeon/radeon_cs.c @@ -85,9 +85,8 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) VRAM, also but everything into VRAM on AGP cards to avoid image corruptions */ if (p->ring == R600_RING_TYPE_UVD_INDEX && - p->rdev->family < CHIP_PALM && (i == 0 || drm_pci_device_is_agp(p->rdev->ddev))) { - + /* TODO: is this still needed for NI+ ? */ p->relocs[i].lobj.domain = RADEON_GEM_DOMAIN_VRAM; @@ -160,7 +159,8 @@ static void radeon_cs_sync_rings(struct radeon_cs_parser *p) if (!p->relocs[i].robj) continue; - radeon_ib_sync_to(&p->ib, p->relocs[i].robj->tbo.sync_obj); + radeon_semaphore_sync_to(p->ib.semaphore, + p->relocs[i].robj->tbo.sync_obj); } } @@ -360,13 +360,13 @@ static int radeon_bo_vm_update_pte(struct radeon_cs_parser *parser, struct radeon_bo *bo; int r; - r = radeon_vm_bo_update_pte(rdev, vm, rdev->ring_tmp_bo.bo, &rdev->ring_tmp_bo.bo->tbo.mem); + r = radeon_vm_bo_update(rdev, vm, rdev->ring_tmp_bo.bo, &rdev->ring_tmp_bo.bo->tbo.mem); if (r) { return r; } list_for_each_entry(lobj, &parser->validated, tv.head) { bo = lobj->bo; - r = radeon_vm_bo_update_pte(parser->rdev, vm, bo, &bo->tbo.mem); + r = radeon_vm_bo_update(parser->rdev, vm, bo, &bo->tbo.mem); if (r) { return r; } @@ -412,9 +412,9 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev, goto out; } radeon_cs_sync_rings(parser); - radeon_ib_sync_to(&parser->ib, vm->fence); - radeon_ib_sync_to(&parser->ib, radeon_vm_grab_id( - rdev, vm, parser->ring)); + radeon_semaphore_sync_to(parser->ib.semaphore, vm->fence); + radeon_semaphore_sync_to(parser->ib.semaphore, + radeon_vm_grab_id(rdev, vm, parser->ring)); if ((rdev->family >= CHIP_TAHITI) && (parser->chunk_const_ib_idx != -1)) { diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index 61dbdd938133..39b033b441d2 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -98,6 +98,7 @@ static const char radeon_family_name[][16] = { "BONAIRE", "KAVERI", "KABINI", + "HAWAII", "LAST", }; @@ -250,28 +251,23 @@ void radeon_scratch_free(struct radeon_device *rdev, uint32_t reg) */ int radeon_doorbell_init(struct radeon_device *rdev) { - int i; - /* doorbell bar mapping */ rdev->doorbell.base = pci_resource_start(rdev->pdev, 2); rdev->doorbell.size = pci_resource_len(rdev->pdev, 2); - /* limit to 4 MB for now */ - if (rdev->doorbell.size > (4 * 1024 * 1024)) - rdev->doorbell.size = 4 * 1024 * 1024; + rdev->doorbell.num_doorbells = min_t(u32, rdev->doorbell.size / sizeof(u32), RADEON_MAX_DOORBELLS); + if (rdev->doorbell.num_doorbells == 0) + return -EINVAL; - rdev->doorbell.ptr = ioremap(rdev->doorbell.base, rdev->doorbell.size); + rdev->doorbell.ptr = ioremap(rdev->doorbell.base, rdev->doorbell.num_doorbells * sizeof(u32)); if (rdev->doorbell.ptr == NULL) { return -ENOMEM; } DRM_INFO("doorbell mmio base: 0x%08X\n", (uint32_t)rdev->doorbell.base); DRM_INFO("doorbell mmio size: %u\n", (unsigned)rdev->doorbell.size); - rdev->doorbell.num_pages = rdev->doorbell.size / PAGE_SIZE; + memset(&rdev->doorbell.used, 0, sizeof(rdev->doorbell.used)); - for (i = 0; i < rdev->doorbell.num_pages; i++) { - rdev->doorbell.free[i] = true; - } return 0; } @@ -289,40 +285,38 @@ void radeon_doorbell_fini(struct radeon_device *rdev) } /** - * radeon_doorbell_get - Allocate a doorbell page + * radeon_doorbell_get - Allocate a doorbell entry * * @rdev: radeon_device pointer - * @doorbell: doorbell page number + * @doorbell: doorbell index * - * Allocate a doorbell page for use by the driver (all asics). + * Allocate a doorbell for use by the driver (all asics). * Returns 0 on success or -EINVAL on failure. */ int radeon_doorbell_get(struct radeon_device *rdev, u32 *doorbell) { - int i; - - for (i = 0; i < rdev->doorbell.num_pages; i++) { - if (rdev->doorbell.free[i]) { - rdev->doorbell.free[i] = false; - *doorbell = i; - return 0; - } + unsigned long offset = find_first_zero_bit(rdev->doorbell.used, rdev->doorbell.num_doorbells); + if (offset < rdev->doorbell.num_doorbells) { + __set_bit(offset, rdev->doorbell.used); + *doorbell = offset; + return 0; + } else { + return -EINVAL; } - return -EINVAL; } /** - * radeon_doorbell_free - Free a doorbell page + * radeon_doorbell_free - Free a doorbell entry * * @rdev: radeon_device pointer - * @doorbell: doorbell page number + * @doorbell: doorbell index * - * Free a doorbell page allocated for use by the driver (all asics) + * Free a doorbell allocated for use by the driver (all asics) */ void radeon_doorbell_free(struct radeon_device *rdev, u32 doorbell) { - if (doorbell < rdev->doorbell.num_pages) - rdev->doorbell.free[doorbell] = true; + if (doorbell < rdev->doorbell.num_doorbells) + __clear_bit(doorbell, rdev->doorbell.used); } /* diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index aab24173d2e9..db39ea36bf22 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -76,9 +76,11 @@ * 2.32.0 - new info request for rings working * 2.33.0 - Add SI tiling mode array query * 2.34.0 - Add CIK tiling mode array query + * 2.35.0 - Add CIK macrotile mode array query + * 2.36.0 - Fix CIK DCE tiling setup */ #define KMS_DRIVER_MAJOR 2 -#define KMS_DRIVER_MINOR 34 +#define KMS_DRIVER_MINOR 36 #define KMS_DRIVER_PATCHLEVEL 0 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags); int radeon_driver_unload_kms(struct drm_device *dev); @@ -156,7 +158,7 @@ int radeon_benchmarking = 0; int radeon_testing = 0; int radeon_connector_table = 0; int radeon_tv = 1; -int radeon_audio = 1; +int radeon_audio = -1; int radeon_disp_priority = 0; int radeon_hw_i2c = 0; int radeon_pcie_gen2 = -1; @@ -200,7 +202,7 @@ module_param_named(connector_table, radeon_connector_table, int, 0444); MODULE_PARM_DESC(tv, "TV enable (0 = disable)"); module_param_named(tv, radeon_tv, int, 0444); -MODULE_PARM_DESC(audio, "Audio enable (1 = enable)"); +MODULE_PARM_DESC(audio, "Audio enable (-1 = auto, 0 = disable, 1 = enable)"); module_param_named(audio, radeon_audio, int, 0444); MODULE_PARM_DESC(disp_priority, "Display Priority (0 = auto, 1 = normal, 2 = high)"); @@ -507,15 +509,6 @@ static const struct file_operations radeon_driver_kms_fops = { #endif }; - -static void -radeon_pci_shutdown(struct pci_dev *pdev) -{ - struct drm_device *dev = pci_get_drvdata(pdev); - - radeon_driver_unload_kms(dev); -} - static struct drm_driver kms_driver = { .driver_features = DRIVER_USE_AGP | @@ -585,7 +578,6 @@ static struct pci_driver radeon_kms_pci_driver = { .probe = radeon_pci_probe, .remove = radeon_pci_remove, .driver.pm = &radeon_pm_ops, - .shutdown = radeon_pci_shutdown, }; static int __init radeon_init(void) diff --git a/drivers/gpu/drm/radeon/radeon_drv.h b/drivers/gpu/drm/radeon/radeon_drv.h index 543dcfae7e6f..00e0d449021c 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.h +++ b/drivers/gpu/drm/radeon/radeon_drv.h @@ -108,9 +108,10 @@ * 1.31- Add support for num Z pipes from GET_PARAM * 1.32- fixes for rv740 setup * 1.33- Add r6xx/r7xx const buffer support + * 1.34- fix evergreen/cayman GS register */ #define DRIVER_MAJOR 1 -#define DRIVER_MINOR 33 +#define DRIVER_MINOR 34 #define DRIVER_PATCHLEVEL 0 long radeon_drm_ioctl(struct file *filp, diff --git a/drivers/gpu/drm/radeon/radeon_family.h b/drivers/gpu/drm/radeon/radeon_family.h index 3c8289083f9d..614ad549297f 100644 --- a/drivers/gpu/drm/radeon/radeon_family.h +++ b/drivers/gpu/drm/radeon/radeon_family.h @@ -96,6 +96,7 @@ enum radeon_family { CHIP_BONAIRE, CHIP_KAVERI, CHIP_KABINI, + CHIP_HAWAII, CHIP_LAST, }; diff --git a/drivers/gpu/drm/radeon/radeon_fence.c b/drivers/gpu/drm/radeon/radeon_fence.c index 281d14c22a47..d3a86e43c012 100644 --- a/drivers/gpu/drm/radeon/radeon_fence.c +++ b/drivers/gpu/drm/radeon/radeon_fence.c @@ -472,6 +472,36 @@ int radeon_fence_wait_any(struct radeon_device *rdev, } /** + * radeon_fence_wait_locked - wait for a fence to signal + * + * @fence: radeon fence object + * + * Wait for the requested fence to signal (all asics). + * Returns 0 if the fence has passed, error for all other cases. + */ +int radeon_fence_wait_locked(struct radeon_fence *fence) +{ + uint64_t seq[RADEON_NUM_RINGS] = {}; + int r; + + if (fence == NULL) { + WARN(1, "Querying an invalid fence : %p !\n", fence); + return -EINVAL; + } + + seq[fence->ring] = fence->seq; + if (seq[fence->ring] == RADEON_FENCE_SIGNALED_SEQ) + return 0; + + r = radeon_fence_wait_seq(fence->rdev, seq, false, false); + if (r) + return r; + + fence->seq = RADEON_FENCE_SIGNALED_SEQ; + return 0; +} + +/** * radeon_fence_wait_next_locked - wait for the next fence to signal * * @rdev: radeon device pointer diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c index 8a83b89d4709..96e440061bdb 100644 --- a/drivers/gpu/drm/radeon/radeon_gart.c +++ b/drivers/gpu/drm/radeon/radeon_gart.c @@ -29,6 +29,7 @@ #include <drm/radeon_drm.h> #include "radeon.h" #include "radeon_reg.h" +#include "radeon_trace.h" /* * GART @@ -651,7 +652,7 @@ retry: radeon_asic_vm_set_page(rdev, &ib, vm->pd_gpu_addr, 0, pd_entries, 0, 0); - radeon_ib_sync_to(&ib, vm->fence); + radeon_semaphore_sync_to(ib.semaphore, vm->fence); r = radeon_ib_schedule(rdev, &ib, NULL); if (r) { radeon_ib_free(rdev, &ib); @@ -737,6 +738,7 @@ struct radeon_fence *radeon_vm_grab_id(struct radeon_device *rdev, for (i = 0; i < 2; ++i) { if (choices[i]) { vm->id = choices[i]; + trace_radeon_vm_grab_id(vm->id, ring); return rdev->vm_manager.active[choices[i]]; } } @@ -1116,7 +1118,7 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev, } /** - * radeon_vm_bo_update_pte - map a bo into the vm page table + * radeon_vm_bo_update - map a bo into the vm page table * * @rdev: radeon_device pointer * @vm: requested vm @@ -1128,10 +1130,10 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev, * * Object have to be reserved & global and local mutex must be locked! */ -int radeon_vm_bo_update_pte(struct radeon_device *rdev, - struct radeon_vm *vm, - struct radeon_bo *bo, - struct ttm_mem_reg *mem) +int radeon_vm_bo_update(struct radeon_device *rdev, + struct radeon_vm *vm, + struct radeon_bo *bo, + struct ttm_mem_reg *mem) { struct radeon_ib ib; struct radeon_bo_va *bo_va; @@ -1176,6 +1178,8 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev, bo_va->valid = false; } + trace_radeon_vm_bo_update(bo_va); + nptes = radeon_bo_ngpu_pages(bo); /* assume two extra pdes in case the mapping overlaps the borders */ @@ -1209,6 +1213,8 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev, return -ENOMEM; r = radeon_ib_get(rdev, R600_RING_TYPE_DMA_INDEX, &ib, NULL, ndw * 4); + if (r) + return r; ib.length_dw = 0; r = radeon_vm_update_pdes(rdev, vm, &ib, bo_va->soffset, bo_va->eoffset); @@ -1220,7 +1226,7 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev, radeon_vm_update_ptes(rdev, vm, &ib, bo_va->soffset, bo_va->eoffset, addr, radeon_vm_page_flags(bo_va->flags)); - radeon_ib_sync_to(&ib, vm->fence); + radeon_semaphore_sync_to(ib.semaphore, vm->fence); r = radeon_ib_schedule(rdev, &ib, NULL); if (r) { radeon_ib_free(rdev, &ib); @@ -1255,7 +1261,7 @@ int radeon_vm_bo_rmv(struct radeon_device *rdev, mutex_lock(&rdev->vm_manager.lock); mutex_lock(&bo_va->vm->mutex); if (bo_va->soffset) { - r = radeon_vm_bo_update_pte(rdev, bo_va->vm, bo_va->bo, NULL); + r = radeon_vm_bo_update(rdev, bo_va->vm, bo_va->bo, NULL); } mutex_unlock(&rdev->vm_manager.lock); list_del(&bo_va->vm_list); diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index bb8710531a1b..21d593c0ecaf 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -340,7 +340,7 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) break; case RADEON_INFO_BACKEND_MAP: if (rdev->family >= CHIP_BONAIRE) - return -EINVAL; + *value = rdev->config.cik.backend_map; else if (rdev->family >= CHIP_TAHITI) *value = rdev->config.si.backend_map; else if (rdev->family >= CHIP_CAYMAN) @@ -449,9 +449,27 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) return -EINVAL; } break; + case RADEON_INFO_CIK_MACROTILE_MODE_ARRAY: + if (rdev->family >= CHIP_BONAIRE) { + value = rdev->config.cik.macrotile_mode_array; + value_size = sizeof(uint32_t)*16; + } else { + DRM_DEBUG_KMS("macrotile mode array is cik+ only!\n"); + return -EINVAL; + } + break; case RADEON_INFO_SI_CP_DMA_COMPUTE: *value = 1; break; + case RADEON_INFO_SI_BACKEND_ENABLED_MASK: + if (rdev->family >= CHIP_BONAIRE) { + *value = rdev->config.cik.backend_enable_mask; + } else if (rdev->family >= CHIP_TAHITI) { + *value = rdev->config.si.backend_enable_mask; + } else { + DRM_DEBUG_KMS("BACKEND_ENABLED_MASK is si+ only!\n"); + } + break; default: DRM_DEBUG_KMS("Invalid request %d\n", info->request); return -EINVAL; diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c index 7cb178a34a0f..0b158f98d287 100644 --- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c +++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c @@ -422,6 +422,7 @@ int radeon_crtc_do_set_base(struct drm_crtc *crtc, /* Pin framebuffer & get tilling informations */ obj = radeon_fb->obj; rbo = gem_to_radeon_bo(obj); +retry: r = radeon_bo_reserve(rbo, false); if (unlikely(r != 0)) return r; @@ -430,6 +431,33 @@ int radeon_crtc_do_set_base(struct drm_crtc *crtc, &base); if (unlikely(r != 0)) { radeon_bo_unreserve(rbo); + + /* On old GPU like RN50 with little vram pining can fails because + * current fb is taking all space needed. So instead of unpining + * the old buffer after pining the new one, first unpin old one + * and then retry pining new one. + * + * As only master can set mode only master can pin and it is + * unlikely the master client will race with itself especialy + * on those old gpu with single crtc. + * + * We don't shutdown the display controller because new buffer + * will end up in same spot. + */ + if (!atomic && fb && fb != crtc->fb) { + struct radeon_bo *old_rbo; + unsigned long nsize, osize; + + old_rbo = gem_to_radeon_bo(to_radeon_framebuffer(fb)->obj); + osize = radeon_bo_size(old_rbo); + nsize = radeon_bo_size(rbo); + if (nsize <= osize && !radeon_bo_reserve(old_rbo, false)) { + radeon_bo_unpin(old_rbo); + radeon_bo_unreserve(old_rbo); + fb = NULL; + goto retry; + } + } return -EINVAL; } radeon_bo_get_tiling_flags(rbo, &tiling_flags, NULL); @@ -1056,6 +1084,26 @@ static void radeon_crtc_commit(struct drm_crtc *crtc) } } +static void radeon_crtc_disable(struct drm_crtc *crtc) +{ + radeon_crtc_dpms(crtc, DRM_MODE_DPMS_OFF); + if (crtc->fb) { + int r; + struct radeon_framebuffer *radeon_fb; + struct radeon_bo *rbo; + + radeon_fb = to_radeon_framebuffer(crtc->fb); + rbo = gem_to_radeon_bo(radeon_fb->obj); + r = radeon_bo_reserve(rbo, false); + if (unlikely(r)) + DRM_ERROR("failed to reserve rbo before unpin\n"); + else { + radeon_bo_unpin(rbo); + radeon_bo_unreserve(rbo); + } + } +} + static const struct drm_crtc_helper_funcs legacy_helper_funcs = { .dpms = radeon_crtc_dpms, .mode_fixup = radeon_crtc_mode_fixup, @@ -1065,6 +1113,7 @@ static const struct drm_crtc_helper_funcs legacy_helper_funcs = { .prepare = radeon_crtc_prepare, .commit = radeon_crtc_commit, .load_lut = radeon_crtc_load_lut, + .disable = radeon_crtc_disable }; diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c index 981fd06e1f9a..984097b907ef 100644 --- a/drivers/gpu/drm/radeon/radeon_pm.c +++ b/drivers/gpu/drm/radeon/radeon_pm.c @@ -537,8 +537,7 @@ static ssize_t radeon_hwmon_show_temp(struct device *dev, struct device_attribute *attr, char *buf) { - struct drm_device *ddev = dev_get_drvdata(dev); - struct radeon_device *rdev = ddev->dev_private; + struct radeon_device *rdev = dev_get_drvdata(dev); int temp; if (rdev->asic->pm.get_temperature) @@ -553,8 +552,7 @@ static ssize_t radeon_hwmon_show_temp_thresh(struct device *dev, struct device_attribute *attr, char *buf) { - struct drm_device *ddev = dev_get_drvdata(dev); - struct radeon_device *rdev = ddev->dev_private; + struct radeon_device *rdev = dev_get_drvdata(dev); int hyst = to_sensor_dev_attr(attr)->index; int temp; @@ -566,23 +564,14 @@ static ssize_t radeon_hwmon_show_temp_thresh(struct device *dev, return snprintf(buf, PAGE_SIZE, "%d\n", temp); } -static ssize_t radeon_hwmon_show_name(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - return sprintf(buf, "radeon\n"); -} - static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, radeon_hwmon_show_temp, NULL, 0); static SENSOR_DEVICE_ATTR(temp1_crit, S_IRUGO, radeon_hwmon_show_temp_thresh, NULL, 0); static SENSOR_DEVICE_ATTR(temp1_crit_hyst, S_IRUGO, radeon_hwmon_show_temp_thresh, NULL, 1); -static SENSOR_DEVICE_ATTR(name, S_IRUGO, radeon_hwmon_show_name, NULL, 0); static struct attribute *hwmon_attributes[] = { &sensor_dev_attr_temp1_input.dev_attr.attr, &sensor_dev_attr_temp1_crit.dev_attr.attr, &sensor_dev_attr_temp1_crit_hyst.dev_attr.attr, - &sensor_dev_attr_name.dev_attr.attr, NULL }; @@ -590,8 +579,7 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, struct attribute *attr, int index) { struct device *dev = container_of(kobj, struct device, kobj); - struct drm_device *ddev = dev_get_drvdata(dev); - struct radeon_device *rdev = ddev->dev_private; + struct radeon_device *rdev = dev_get_drvdata(dev); /* Skip limit attributes if DPM is not enabled */ if (rdev->pm.pm_method != PM_METHOD_DPM && @@ -607,11 +595,15 @@ static const struct attribute_group hwmon_attrgroup = { .is_visible = hwmon_attributes_visible, }; +static const struct attribute_group *hwmon_groups[] = { + &hwmon_attrgroup, + NULL +}; + static int radeon_hwmon_init(struct radeon_device *rdev) { int err = 0; - - rdev->pm.int_hwmon_dev = NULL; + struct device *hwmon_dev; switch (rdev->pm.int_thermal_type) { case THERMAL_TYPE_RV6XX: @@ -624,20 +616,13 @@ static int radeon_hwmon_init(struct radeon_device *rdev) case THERMAL_TYPE_KV: if (rdev->asic->pm.get_temperature == NULL) return err; - rdev->pm.int_hwmon_dev = hwmon_device_register(rdev->dev); - if (IS_ERR(rdev->pm.int_hwmon_dev)) { - err = PTR_ERR(rdev->pm.int_hwmon_dev); + hwmon_dev = hwmon_device_register_with_groups(rdev->dev, + "radeon", rdev, + hwmon_groups); + if (IS_ERR(hwmon_dev)) { + err = PTR_ERR(hwmon_dev); dev_err(rdev->dev, "Unable to register hwmon device: %d\n", err); - break; - } - dev_set_drvdata(rdev->pm.int_hwmon_dev, rdev->ddev); - err = sysfs_create_group(&rdev->pm.int_hwmon_dev->kobj, - &hwmon_attrgroup); - if (err) { - dev_err(rdev->dev, - "Unable to create hwmon sysfs file: %d\n", err); - hwmon_device_unregister(rdev->dev); } break; default: @@ -647,14 +632,6 @@ static int radeon_hwmon_init(struct radeon_device *rdev) return err; } -static void radeon_hwmon_fini(struct radeon_device *rdev) -{ - if (rdev->pm.int_hwmon_dev) { - sysfs_remove_group(&rdev->pm.int_hwmon_dev->kobj, &hwmon_attrgroup); - hwmon_device_unregister(rdev->pm.int_hwmon_dev); - } -} - static void radeon_dpm_thermal_work_handler(struct work_struct *work) { struct radeon_device *rdev = @@ -954,6 +931,8 @@ void radeon_dpm_enable_uvd(struct radeon_device *rdev, bool enable) if (enable) { mutex_lock(&rdev->pm.mutex); rdev->pm.dpm.uvd_active = true; + /* disable this for now */ +#if 0 if ((rdev->pm.dpm.sd == 1) && (rdev->pm.dpm.hd == 0)) dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_SD; else if ((rdev->pm.dpm.sd == 2) && (rdev->pm.dpm.hd == 0)) @@ -963,6 +942,7 @@ void radeon_dpm_enable_uvd(struct radeon_device *rdev, bool enable) else if ((rdev->pm.dpm.sd == 0) && (rdev->pm.dpm.hd == 2)) dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD2; else +#endif dpm_state = POWER_STATE_TYPE_INTERNAL_UVD; rdev->pm.dpm.state = dpm_state; mutex_unlock(&rdev->pm.mutex); @@ -1249,10 +1229,10 @@ int radeon_pm_init(struct radeon_device *rdev) case CHIP_RS780: case CHIP_RS880: case CHIP_CAYMAN: - case CHIP_ARUBA: case CHIP_BONAIRE: case CHIP_KABINI: case CHIP_KAVERI: + case CHIP_HAWAII: /* DPM requires the RLC, RV770+ dGPU requires SMC */ if (!rdev->rlc_fw) rdev->pm.pm_method = PM_METHOD_PROFILE; @@ -1280,6 +1260,7 @@ int radeon_pm_init(struct radeon_device *rdev) case CHIP_BARTS: case CHIP_TURKS: case CHIP_CAICOS: + case CHIP_ARUBA: case CHIP_TAHITI: case CHIP_PITCAIRN: case CHIP_VERDE: @@ -1333,8 +1314,6 @@ static void radeon_pm_fini_old(struct radeon_device *rdev) if (rdev->pm.power_state) kfree(rdev->pm.power_state); - - radeon_hwmon_fini(rdev); } static void radeon_pm_fini_dpm(struct radeon_device *rdev) @@ -1354,8 +1333,6 @@ static void radeon_pm_fini_dpm(struct radeon_device *rdev) if (rdev->pm.power_state) kfree(rdev->pm.power_state); - - radeon_hwmon_fini(rdev); } void radeon_pm_fini(struct radeon_device *rdev) diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c index 18254e1c3e71..9214403ae173 100644 --- a/drivers/gpu/drm/radeon/radeon_ring.c +++ b/drivers/gpu/drm/radeon/radeon_ring.c @@ -61,7 +61,7 @@ int radeon_ib_get(struct radeon_device *rdev, int ring, struct radeon_ib *ib, struct radeon_vm *vm, unsigned size) { - int i, r; + int r; r = radeon_sa_bo_new(rdev, &rdev->ring_tmp_bo, &ib->sa_bo, size, 256, true); if (r) { @@ -87,8 +87,6 @@ int radeon_ib_get(struct radeon_device *rdev, int ring, ib->gpu_addr = radeon_sa_bo_gpu_addr(ib->sa_bo); } ib->is_const_ib = false; - for (i = 0; i < RADEON_NUM_RINGS; ++i) - ib->sync_to[i] = NULL; return 0; } @@ -109,25 +107,6 @@ void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib) } /** - * radeon_ib_sync_to - sync to fence before executing the IB - * - * @ib: IB object to add fence to - * @fence: fence to sync to - * - * Sync to the fence before executing the IB - */ -void radeon_ib_sync_to(struct radeon_ib *ib, struct radeon_fence *fence) -{ - struct radeon_fence *other; - - if (!fence) - return; - - other = ib->sync_to[fence->ring]; - ib->sync_to[fence->ring] = radeon_fence_later(fence, other); -} - -/** * radeon_ib_schedule - schedule an IB (Indirect Buffer) on the ring * * @rdev: radeon_device pointer @@ -151,8 +130,7 @@ int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib, struct radeon_ib *const_ib) { struct radeon_ring *ring = &rdev->ring[ib->ring]; - bool need_sync = false; - int i, r = 0; + int r = 0; if (!ib->length_dw || !ring->ready) { /* TODO: Nothings in the ib we should report. */ @@ -166,19 +144,15 @@ int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib, dev_err(rdev->dev, "scheduling IB failed (%d).\n", r); return r; } - for (i = 0; i < RADEON_NUM_RINGS; ++i) { - struct radeon_fence *fence = ib->sync_to[i]; - if (radeon_fence_need_sync(fence, ib->ring)) { - need_sync = true; - radeon_semaphore_sync_rings(rdev, ib->semaphore, - fence->ring, ib->ring); - radeon_fence_note_sync(fence, ib->ring); - } - } - /* immediately free semaphore when we don't need to sync */ - if (!need_sync) { - radeon_semaphore_free(rdev, &ib->semaphore, NULL); + + /* sync with other rings */ + r = radeon_semaphore_sync_rings(rdev, ib->semaphore, ib->ring); + if (r) { + dev_err(rdev->dev, "failed to sync rings (%d)\n", r); + radeon_ring_unlock_undo(rdev, ring); + return r; } + /* if we can't remember our last VM flush then flush now! */ /* XXX figure out why we have to flush for every IB */ if (ib->vm /*&& !ib->vm->last_flush*/) { diff --git a/drivers/gpu/drm/radeon/radeon_semaphore.c b/drivers/gpu/drm/radeon/radeon_semaphore.c index 8dcc20f53d73..2b42aa1914f2 100644 --- a/drivers/gpu/drm/radeon/radeon_semaphore.c +++ b/drivers/gpu/drm/radeon/radeon_semaphore.c @@ -29,12 +29,12 @@ */ #include <drm/drmP.h> #include "radeon.h" - +#include "radeon_trace.h" int radeon_semaphore_create(struct radeon_device *rdev, struct radeon_semaphore **semaphore) { - int r; + int i, r; *semaphore = kmalloc(sizeof(struct radeon_semaphore), GFP_KERNEL); if (*semaphore == NULL) { @@ -50,54 +50,121 @@ int radeon_semaphore_create(struct radeon_device *rdev, (*semaphore)->waiters = 0; (*semaphore)->gpu_addr = radeon_sa_bo_gpu_addr((*semaphore)->sa_bo); *((uint64_t*)radeon_sa_bo_cpu_addr((*semaphore)->sa_bo)) = 0; + + for (i = 0; i < RADEON_NUM_RINGS; ++i) + (*semaphore)->sync_to[i] = NULL; + return 0; } -void radeon_semaphore_emit_signal(struct radeon_device *rdev, int ring, +bool radeon_semaphore_emit_signal(struct radeon_device *rdev, int ridx, struct radeon_semaphore *semaphore) { - --semaphore->waiters; - radeon_semaphore_ring_emit(rdev, ring, &rdev->ring[ring], semaphore, false); + struct radeon_ring *ring = &rdev->ring[ridx]; + + trace_radeon_semaphore_signale(ridx, semaphore); + + if (radeon_semaphore_ring_emit(rdev, ridx, ring, semaphore, false)) { + --semaphore->waiters; + + /* for debugging lockup only, used by sysfs debug files */ + ring->last_semaphore_signal_addr = semaphore->gpu_addr; + return true; + } + return false; } -void radeon_semaphore_emit_wait(struct radeon_device *rdev, int ring, +bool radeon_semaphore_emit_wait(struct radeon_device *rdev, int ridx, struct radeon_semaphore *semaphore) { - ++semaphore->waiters; - radeon_semaphore_ring_emit(rdev, ring, &rdev->ring[ring], semaphore, true); + struct radeon_ring *ring = &rdev->ring[ridx]; + + trace_radeon_semaphore_wait(ridx, semaphore); + + if (radeon_semaphore_ring_emit(rdev, ridx, ring, semaphore, true)) { + ++semaphore->waiters; + + /* for debugging lockup only, used by sysfs debug files */ + ring->last_semaphore_wait_addr = semaphore->gpu_addr; + return true; + } + return false; +} + +/** + * radeon_semaphore_sync_to - use the semaphore to sync to a fence + * + * @semaphore: semaphore object to add fence to + * @fence: fence to sync to + * + * Sync to the fence using this semaphore object + */ +void radeon_semaphore_sync_to(struct radeon_semaphore *semaphore, + struct radeon_fence *fence) +{ + struct radeon_fence *other; + + if (!fence) + return; + + other = semaphore->sync_to[fence->ring]; + semaphore->sync_to[fence->ring] = radeon_fence_later(fence, other); } -/* caller must hold ring lock */ +/** + * radeon_semaphore_sync_rings - sync ring to all registered fences + * + * @rdev: radeon_device pointer + * @semaphore: semaphore object to use for sync + * @ring: ring that needs sync + * + * Ensure that all registered fences are signaled before letting + * the ring continue. The caller must hold the ring lock. + */ int radeon_semaphore_sync_rings(struct radeon_device *rdev, struct radeon_semaphore *semaphore, - int signaler, int waiter) + int ring) { - int r; + int i, r; - /* no need to signal and wait on the same ring */ - if (signaler == waiter) { - return 0; - } + for (i = 0; i < RADEON_NUM_RINGS; ++i) { + struct radeon_fence *fence = semaphore->sync_to[i]; - /* prevent GPU deadlocks */ - if (!rdev->ring[signaler].ready) { - dev_err(rdev->dev, "Trying to sync to a disabled ring!"); - return -EINVAL; - } + /* check if we really need to sync */ + if (!radeon_fence_need_sync(fence, ring)) + continue; - r = radeon_ring_alloc(rdev, &rdev->ring[signaler], 8); - if (r) { - return r; - } - radeon_semaphore_emit_signal(rdev, signaler, semaphore); - radeon_ring_commit(rdev, &rdev->ring[signaler]); + /* prevent GPU deadlocks */ + if (!rdev->ring[i].ready) { + dev_err(rdev->dev, "Syncing to a disabled ring!"); + return -EINVAL; + } - /* we assume caller has already allocated space on waiters ring */ - radeon_semaphore_emit_wait(rdev, waiter, semaphore); + /* allocate enough space for sync command */ + r = radeon_ring_alloc(rdev, &rdev->ring[i], 16); + if (r) { + return r; + } - /* for debugging lockup only, used by sysfs debug files */ - rdev->ring[signaler].last_semaphore_signal_addr = semaphore->gpu_addr; - rdev->ring[waiter].last_semaphore_wait_addr = semaphore->gpu_addr; + /* emit the signal semaphore */ + if (!radeon_semaphore_emit_signal(rdev, i, semaphore)) { + /* signaling wasn't successful wait manually */ + radeon_ring_undo(&rdev->ring[i]); + radeon_fence_wait_locked(fence); + continue; + } + + /* we assume caller has already allocated space on waiters ring */ + if (!radeon_semaphore_emit_wait(rdev, ring, semaphore)) { + /* waiting wasn't successful wait manually */ + radeon_ring_undo(&rdev->ring[i]); + radeon_fence_wait_locked(fence); + continue; + } + + radeon_ring_commit(rdev, &rdev->ring[i]); + radeon_fence_note_sync(fence, ring); + } return 0; } diff --git a/drivers/gpu/drm/radeon/radeon_test.c b/drivers/gpu/drm/radeon/radeon_test.c index f4d6bcee9006..12e8099a0823 100644 --- a/drivers/gpu/drm/radeon/radeon_test.c +++ b/drivers/gpu/drm/radeon/radeon_test.c @@ -36,8 +36,8 @@ static void radeon_do_test_moves(struct radeon_device *rdev, int flag) struct radeon_bo *vram_obj = NULL; struct radeon_bo **gtt_obj = NULL; uint64_t gtt_addr, vram_addr; - unsigned i, n, size; - int r, ring; + unsigned n, size; + int i, r, ring; switch (flag) { case RADEON_TEST_COPY_DMA: diff --git a/drivers/gpu/drm/radeon/radeon_trace.h b/drivers/gpu/drm/radeon/radeon_trace.h index 811bca691b36..0473257d4078 100644 --- a/drivers/gpu/drm/radeon/radeon_trace.h +++ b/drivers/gpu/drm/radeon/radeon_trace.h @@ -47,6 +47,39 @@ TRACE_EVENT(radeon_cs, __entry->fences) ); +TRACE_EVENT(radeon_vm_grab_id, + TP_PROTO(unsigned vmid, int ring), + TP_ARGS(vmid, ring), + TP_STRUCT__entry( + __field(u32, vmid) + __field(u32, ring) + ), + + TP_fast_assign( + __entry->vmid = vmid; + __entry->ring = ring; + ), + TP_printk("vmid=%u, ring=%u", __entry->vmid, __entry->ring) +); + +TRACE_EVENT(radeon_vm_bo_update, + TP_PROTO(struct radeon_bo_va *bo_va), + TP_ARGS(bo_va), + TP_STRUCT__entry( + __field(u64, soffset) + __field(u64, eoffset) + __field(u32, flags) + ), + + TP_fast_assign( + __entry->soffset = bo_va->soffset; + __entry->eoffset = bo_va->eoffset; + __entry->flags = bo_va->flags; + ), + TP_printk("soffs=%010llx, eoffs=%010llx, flags=%08x", + __entry->soffset, __entry->eoffset, __entry->flags) +); + TRACE_EVENT(radeon_vm_set_page, TP_PROTO(uint64_t pe, uint64_t addr, unsigned count, uint32_t incr, uint32_t flags), @@ -111,6 +144,42 @@ DEFINE_EVENT(radeon_fence_request, radeon_fence_wait_end, TP_ARGS(dev, seqno) ); +DECLARE_EVENT_CLASS(radeon_semaphore_request, + + TP_PROTO(int ring, struct radeon_semaphore *sem), + + TP_ARGS(ring, sem), + + TP_STRUCT__entry( + __field(int, ring) + __field(signed, waiters) + __field(uint64_t, gpu_addr) + ), + + TP_fast_assign( + __entry->ring = ring; + __entry->waiters = sem->waiters; + __entry->gpu_addr = sem->gpu_addr; + ), + + TP_printk("ring=%u, waiters=%d, addr=%010Lx", __entry->ring, + __entry->waiters, __entry->gpu_addr) +); + +DEFINE_EVENT(radeon_semaphore_request, radeon_semaphore_signale, + + TP_PROTO(int ring, struct radeon_semaphore *sem), + + TP_ARGS(ring, sem) +); + +DEFINE_EVENT(radeon_semaphore_request, radeon_semaphore_wait, + + TP_PROTO(int ring, struct radeon_semaphore *sem), + + TP_ARGS(ring, sem) +); + #endif /* This part must be outside protection */ diff --git a/drivers/gpu/drm/radeon/radeon_ucode.h b/drivers/gpu/drm/radeon/radeon_ucode.h index 33858364fe89..a77cd274dfc3 100644 --- a/drivers/gpu/drm/radeon/radeon_ucode.h +++ b/drivers/gpu/drm/radeon/radeon_ucode.h @@ -59,6 +59,7 @@ #define SI_MC_UCODE_SIZE 7769 #define OLAND_MC_UCODE_SIZE 7863 #define CIK_MC_UCODE_SIZE 7866 +#define HAWAII_MC_UCODE_SIZE 7933 /* SDMA */ #define CIK_SDMA_UCODE_SIZE 1050 @@ -143,4 +144,7 @@ #define BONAIRE_SMC_UCODE_START 0x20000 #define BONAIRE_SMC_UCODE_SIZE 0x1FDEC +#define HAWAII_SMC_UCODE_START 0x20000 +#define HAWAII_SMC_UCODE_SIZE 0x1FDEC + #endif diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c index 234b2a3b8add..b9c0529b4a2e 100644 --- a/drivers/gpu/drm/radeon/radeon_uvd.c +++ b/drivers/gpu/drm/radeon/radeon_uvd.c @@ -97,6 +97,7 @@ int radeon_uvd_init(struct radeon_device *rdev) case CHIP_BONAIRE: case CHIP_KABINI: case CHIP_KAVERI: + case CHIP_HAWAII: fw_name = FIRMWARE_BONAIRE; break; @@ -472,13 +473,14 @@ static int radeon_uvd_cs_reloc(struct radeon_cs_parser *p, return -EINVAL; } - if ((start >> 28) != (end >> 28)) { + if ((start >> 28) != ((end - 1) >> 28)) { DRM_ERROR("reloc %LX-%LX crossing 256MB boundary!\n", start, end); return -EINVAL; } - if (p->rdev->family < CHIP_PALM && (cmd == 0 || cmd == 0x3) && + /* TODO: is this still necessary on NI+ ? */ + if ((cmd == 0 || cmd == 0x3) && (start >> 28) != (p->rdev->uvd.gpu_addr >> 28)) { DRM_ERROR("msg/fb buffer %LX-%LX out of 256MB segment!\n", start, end); @@ -800,7 +802,8 @@ void radeon_uvd_note_usage(struct radeon_device *rdev) (rdev->pm.dpm.hd != hd)) { rdev->pm.dpm.sd = sd; rdev->pm.dpm.hd = hd; - streams_changed = true; + /* disable this for now */ + /*streams_changed = true;*/ } } diff --git a/drivers/gpu/drm/radeon/reg_srcs/cayman b/drivers/gpu/drm/radeon/reg_srcs/cayman index a072fa8c46b0..d46b58d078aa 100644 --- a/drivers/gpu/drm/radeon/reg_srcs/cayman +++ b/drivers/gpu/drm/radeon/reg_srcs/cayman @@ -21,7 +21,7 @@ cayman 0x9400 0x000089AC VGT_COMPUTE_THREAD_GOURP_SIZE 0x000089B0 VGT_HS_OFFCHIP_PARAM 0x00008A14 PA_CL_ENHANCE -0x00008A60 PA_SC_LINE_STIPPLE_VALUE +0x00008A60 PA_SU_LINE_STIPPLE_VALUE 0x00008B10 PA_SC_LINE_STIPPLE_STATE 0x00008BF0 PA_SC_ENHANCE 0x00008D8C SQ_DYN_GPR_CNTL_PS_FLUSH_REQ @@ -532,7 +532,7 @@ cayman 0x9400 0x00028B84 PA_SU_POLY_OFFSET_FRONT_OFFSET 0x00028B88 PA_SU_POLY_OFFSET_BACK_SCALE 0x00028B8C PA_SU_POLY_OFFSET_BACK_OFFSET -0x00028B74 VGT_GS_INSTANCE_CNT +0x00028B90 VGT_GS_INSTANCE_CNT 0x00028BD4 PA_SC_CENTROID_PRIORITY_0 0x00028BD8 PA_SC_CENTROID_PRIORITY_1 0x00028BDC PA_SC_LINE_CNTL diff --git a/drivers/gpu/drm/radeon/reg_srcs/evergreen b/drivers/gpu/drm/radeon/reg_srcs/evergreen index b912a37689bf..57745c8761c8 100644 --- a/drivers/gpu/drm/radeon/reg_srcs/evergreen +++ b/drivers/gpu/drm/radeon/reg_srcs/evergreen @@ -22,7 +22,7 @@ evergreen 0x9400 0x000089A4 VGT_COMPUTE_START_Z 0x000089AC VGT_COMPUTE_THREAD_GOURP_SIZE 0x00008A14 PA_CL_ENHANCE -0x00008A60 PA_SC_LINE_STIPPLE_VALUE +0x00008A60 PA_SU_LINE_STIPPLE_VALUE 0x00008B10 PA_SC_LINE_STIPPLE_STATE 0x00008BF0 PA_SC_ENHANCE 0x00008D8C SQ_DYN_GPR_CNTL_PS_FLUSH_REQ @@ -545,7 +545,7 @@ evergreen 0x9400 0x00028B84 PA_SU_POLY_OFFSET_FRONT_OFFSET 0x00028B88 PA_SU_POLY_OFFSET_BACK_SCALE 0x00028B8C PA_SU_POLY_OFFSET_BACK_OFFSET -0x00028B74 VGT_GS_INSTANCE_CNT +0x00028B90 VGT_GS_INSTANCE_CNT 0x00028C00 PA_SC_LINE_CNTL 0x00028C08 PA_SU_VTX_CNTL 0x00028C0C PA_CL_GB_VERT_CLIP_ADJ diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c index 1447d794c22a..e7dab069cccf 100644 --- a/drivers/gpu/drm/radeon/rs690.c +++ b/drivers/gpu/drm/radeon/rs690.c @@ -162,6 +162,16 @@ static void rs690_mc_init(struct radeon_device *rdev) base = RREG32_MC(R_000100_MCCFG_FB_LOCATION); base = G_000100_MC_FB_START(base) << 16; rdev->mc.igp_sideport_enabled = radeon_atombios_sideport_present(rdev); + /* Some boards seem to be configured for 128MB of sideport memory, + * but really only have 64MB. Just skip the sideport and use + * UMA memory. + */ + if (rdev->mc.igp_sideport_enabled && + (rdev->mc.real_vram_size == (384 * 1024 * 1024))) { + base += 128 * 1024 * 1024; + rdev->mc.real_vram_size -= 128 * 1024 * 1024; + rdev->mc.mc_vram_size = rdev->mc.real_vram_size; + } /* Use K8 direct mapping for fast fb access. */ rdev->fastfb_working = false; @@ -345,9 +355,11 @@ static void rs690_crtc_bandwidth_compute(struct radeon_device *rdev, if (max_bandwidth.full > rdev->pm.sideport_bandwidth.full && rdev->pm.sideport_bandwidth.full) max_bandwidth = rdev->pm.sideport_bandwidth; - read_delay_latency.full = dfixed_const(370 * 800 * 1000); - read_delay_latency.full = dfixed_div(read_delay_latency, - rdev->pm.igp_sideport_mclk); + read_delay_latency.full = dfixed_const(370 * 800); + a.full = dfixed_const(1000); + b.full = dfixed_div(rdev->pm.igp_sideport_mclk, a); + read_delay_latency.full = dfixed_div(read_delay_latency, b); + read_delay_latency.full = dfixed_mul(read_delay_latency, a); } else { if (max_bandwidth.full > rdev->pm.k8_bandwidth.full && rdev->pm.k8_bandwidth.full) @@ -488,14 +500,10 @@ static void rs690_compute_mode_priority(struct radeon_device *rdev, } if (wm0->priority_mark.full > priority_mark02.full) priority_mark02.full = wm0->priority_mark.full; - if (dfixed_trunc(priority_mark02) < 0) - priority_mark02.full = 0; if (wm0->priority_mark_max.full > priority_mark02.full) priority_mark02.full = wm0->priority_mark_max.full; if (wm1->priority_mark.full > priority_mark12.full) priority_mark12.full = wm1->priority_mark.full; - if (dfixed_trunc(priority_mark12) < 0) - priority_mark12.full = 0; if (wm1->priority_mark_max.full > priority_mark12.full) priority_mark12.full = wm1->priority_mark_max.full; *d1mode_priority_a_cnt = dfixed_trunc(priority_mark02); @@ -526,8 +534,6 @@ static void rs690_compute_mode_priority(struct radeon_device *rdev, } if (wm0->priority_mark.full > priority_mark02.full) priority_mark02.full = wm0->priority_mark.full; - if (dfixed_trunc(priority_mark02) < 0) - priority_mark02.full = 0; if (wm0->priority_mark_max.full > priority_mark02.full) priority_mark02.full = wm0->priority_mark_max.full; *d1mode_priority_a_cnt = dfixed_trunc(priority_mark02); @@ -555,8 +561,6 @@ static void rs690_compute_mode_priority(struct radeon_device *rdev, } if (wm1->priority_mark.full > priority_mark12.full) priority_mark12.full = wm1->priority_mark.full; - if (dfixed_trunc(priority_mark12) < 0) - priority_mark12.full = 0; if (wm1->priority_mark_max.full > priority_mark12.full) priority_mark12.full = wm1->priority_mark_max.full; *d2mode_priority_a_cnt = dfixed_trunc(priority_mark12); diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c index 873eb4b193b4..5d1c316115ef 100644 --- a/drivers/gpu/drm/radeon/rv515.c +++ b/drivers/gpu/drm/radeon/rv515.c @@ -1155,14 +1155,10 @@ static void rv515_compute_mode_priority(struct radeon_device *rdev, } if (wm0->priority_mark.full > priority_mark02.full) priority_mark02.full = wm0->priority_mark.full; - if (dfixed_trunc(priority_mark02) < 0) - priority_mark02.full = 0; if (wm0->priority_mark_max.full > priority_mark02.full) priority_mark02.full = wm0->priority_mark_max.full; if (wm1->priority_mark.full > priority_mark12.full) priority_mark12.full = wm1->priority_mark.full; - if (dfixed_trunc(priority_mark12) < 0) - priority_mark12.full = 0; if (wm1->priority_mark_max.full > priority_mark12.full) priority_mark12.full = wm1->priority_mark_max.full; *d1mode_priority_a_cnt = dfixed_trunc(priority_mark02); @@ -1193,8 +1189,6 @@ static void rv515_compute_mode_priority(struct radeon_device *rdev, } if (wm0->priority_mark.full > priority_mark02.full) priority_mark02.full = wm0->priority_mark.full; - if (dfixed_trunc(priority_mark02) < 0) - priority_mark02.full = 0; if (wm0->priority_mark_max.full > priority_mark02.full) priority_mark02.full = wm0->priority_mark_max.full; *d1mode_priority_a_cnt = dfixed_trunc(priority_mark02); @@ -1222,8 +1216,6 @@ static void rv515_compute_mode_priority(struct radeon_device *rdev, } if (wm1->priority_mark.full > priority_mark12.full) priority_mark12.full = wm1->priority_mark.full; - if (dfixed_trunc(priority_mark12) < 0) - priority_mark12.full = 0; if (wm1->priority_mark_max.full > priority_mark12.full) priority_mark12.full = wm1->priority_mark_max.full; *d2mode_priority_a_cnt = dfixed_trunc(priority_mark12); diff --git a/drivers/gpu/drm/radeon/rv770_dma.c b/drivers/gpu/drm/radeon/rv770_dma.c index f9b02e3d6830..aca8cbe8a335 100644 --- a/drivers/gpu/drm/radeon/rv770_dma.c +++ b/drivers/gpu/drm/radeon/rv770_dma.c @@ -66,13 +66,8 @@ int rv770_copy_dma(struct radeon_device *rdev, return r; } - if (radeon_fence_need_sync(*fence, ring->idx)) { - radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring, - ring->idx); - radeon_fence_note_sync(*fence, ring->idx); - } else { - radeon_semaphore_free(rdev, &sem, NULL); - } + radeon_semaphore_sync_to(sem, *fence); + radeon_semaphore_sync_rings(rdev, sem, ring->idx); for (i = 0; i < num_loops; i++) { cur_size_in_dw = size_in_dw; diff --git a/drivers/gpu/drm/radeon/rv770_dpm.c b/drivers/gpu/drm/radeon/rv770_dpm.c index 913b025ae9b3..374499db20c7 100644 --- a/drivers/gpu/drm/radeon/rv770_dpm.c +++ b/drivers/gpu/drm/radeon/rv770_dpm.c @@ -2328,6 +2328,12 @@ void rv770_get_engine_memory_ss(struct radeon_device *rdev) pi->mclk_ss = radeon_atombios_get_asic_ss_info(rdev, &ss, ASIC_INTERNAL_MEMORY_SS, 0); + /* disable ss, causes hangs on some cayman boards */ + if (rdev->family == CHIP_CAYMAN) { + pi->sclk_ss = false; + pi->mclk_ss = false; + } + if (pi->sclk_ss || pi->mclk_ss) pi->dynamic_ss = true; else diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index 2acfe561796f..85e1edfaa3be 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -80,6 +80,9 @@ extern void evergreen_print_gpu_status_regs(struct radeon_device *rdev); extern bool evergreen_is_display_hung(struct radeon_device *rdev); static void si_enable_gui_idle_interrupt(struct radeon_device *rdev, bool enable); +static void si_fini_pg(struct radeon_device *rdev); +static void si_fini_cg(struct radeon_device *rdev); +static void si_rlc_stop(struct radeon_device *rdev); static const u32 verde_rlc_save_restore_register_list[] = { @@ -1673,6 +1676,7 @@ static int si_init_microcode(struct radeon_device *rdev) fw_name); release_firmware(rdev->smc_fw); rdev->smc_fw = NULL; + err = 0; } else if (rdev->smc_fw->size != smc_req_size) { printk(KERN_ERR "si_smc: Bogus length %zu in firmware \"%s\"\n", @@ -2807,7 +2811,7 @@ static void si_setup_spi(struct radeon_device *rdev, } static u32 si_get_rb_disabled(struct radeon_device *rdev, - u32 max_rb_num, u32 se_num, + u32 max_rb_num_per_se, u32 sh_per_se) { u32 data, mask; @@ -2821,14 +2825,14 @@ static u32 si_get_rb_disabled(struct radeon_device *rdev, data >>= BACKEND_DISABLE_SHIFT; - mask = si_create_bitmask(max_rb_num / se_num / sh_per_se); + mask = si_create_bitmask(max_rb_num_per_se / sh_per_se); return data & mask; } static void si_setup_rb(struct radeon_device *rdev, u32 se_num, u32 sh_per_se, - u32 max_rb_num) + u32 max_rb_num_per_se) { int i, j; u32 data, mask; @@ -2838,19 +2842,21 @@ static void si_setup_rb(struct radeon_device *rdev, for (i = 0; i < se_num; i++) { for (j = 0; j < sh_per_se; j++) { si_select_se_sh(rdev, i, j); - data = si_get_rb_disabled(rdev, max_rb_num, se_num, sh_per_se); + data = si_get_rb_disabled(rdev, max_rb_num_per_se, sh_per_se); disabled_rbs |= data << ((i * sh_per_se + j) * TAHITI_RB_BITMAP_WIDTH_PER_SH); } } si_select_se_sh(rdev, 0xffffffff, 0xffffffff); mask = 1; - for (i = 0; i < max_rb_num; i++) { + for (i = 0; i < max_rb_num_per_se * se_num; i++) { if (!(disabled_rbs & mask)) enabled_rbs |= mask; mask <<= 1; } + rdev->config.si.backend_enable_mask = enabled_rbs; + for (i = 0; i < se_num; i++) { si_select_se_sh(rdev, i, 0xffffffff); data = 0; @@ -3603,6 +3609,13 @@ static void si_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask) dev_info(rdev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n", RREG32(VM_CONTEXT1_PROTECTION_FAULT_STATUS)); + /* disable PG/CG */ + si_fini_pg(rdev); + si_fini_cg(rdev); + + /* stop the rlc */ + si_rlc_stop(rdev); + /* Disable CP parsing/prefetching */ WREG32(CP_ME_CNTL, CP_ME_HALT | CP_PFP_HALT | CP_CE_HALT); @@ -3871,8 +3884,15 @@ static int si_mc_init(struct radeon_device *rdev) rdev->mc.aper_base = pci_resource_start(rdev->pdev, 0); rdev->mc.aper_size = pci_resource_len(rdev->pdev, 0); /* size in MB on si */ - rdev->mc.mc_vram_size = RREG32(CONFIG_MEMSIZE) * 1024ULL * 1024ULL; - rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE) * 1024ULL * 1024ULL; + tmp = RREG32(CONFIG_MEMSIZE); + /* some boards may have garbage in the upper 16 bits */ + if (tmp & 0xffff0000) { + DRM_INFO("Probable bad vram size: 0x%08x\n", tmp); + if (tmp & 0xffff) + tmp &= 0xffff; + } + rdev->mc.mc_vram_size = tmp * 1024ULL * 1024ULL; + rdev->mc.real_vram_size = rdev->mc.mc_vram_size; rdev->mc.visible_vram_size = rdev->mc.aper_size; si_vram_gtt_location(rdev, &rdev->mc); radeon_update_bandwidth_info(rdev); diff --git a/drivers/gpu/drm/radeon/si_dma.c b/drivers/gpu/drm/radeon/si_dma.c index 8e8f46133532..59be2cfcbb47 100644 --- a/drivers/gpu/drm/radeon/si_dma.c +++ b/drivers/gpu/drm/radeon/si_dma.c @@ -195,13 +195,8 @@ int si_copy_dma(struct radeon_device *rdev, return r; } - if (radeon_fence_need_sync(*fence, ring->idx)) { - radeon_semaphore_sync_rings(rdev, sem, (*fence)->ring, - ring->idx); - radeon_fence_note_sync(*fence, ring->idx); - } else { - radeon_semaphore_free(rdev, &sem, NULL); - } + radeon_semaphore_sync_to(sem, *fence); + radeon_semaphore_sync_rings(rdev, sem, ring->idx); for (i = 0; i < num_loops; i++) { cur_size_in_bytes = size_in_bytes; diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index a24ada406046..0b00c790fb77 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -5213,7 +5213,7 @@ static int si_set_mc_special_registers(struct radeon_device *rdev, table->mc_reg_table_entry[k].mc_data[j] |= 0x100; } j++; - if (j > SMC_SISLANDS_MC_REGISTER_ARRAY_SIZE) + if (j >= SMC_SISLANDS_MC_REGISTER_ARRAY_SIZE) return -EINVAL; if (!pi->mem_gddr5) { @@ -5223,7 +5223,7 @@ static int si_set_mc_special_registers(struct radeon_device *rdev, table->mc_reg_table_entry[k].mc_data[j] = (table->mc_reg_table_entry[k].mc_data[i] & 0xffff0000) >> 16; j++; - if (j > SMC_SISLANDS_MC_REGISTER_ARRAY_SIZE) + if (j >= SMC_SISLANDS_MC_REGISTER_ARRAY_SIZE) return -EINVAL; } break; @@ -5236,7 +5236,7 @@ static int si_set_mc_special_registers(struct radeon_device *rdev, (temp_reg & 0xffff0000) | (table->mc_reg_table_entry[k].mc_data[i] & 0x0000ffff); j++; - if (j > SMC_SISLANDS_MC_REGISTER_ARRAY_SIZE) + if (j >= SMC_SISLANDS_MC_REGISTER_ARRAY_SIZE) return -EINVAL; break; default: diff --git a/drivers/gpu/drm/radeon/sid.h b/drivers/gpu/drm/radeon/sid.h index 5691a7c30686..b322acc48097 100644 --- a/drivers/gpu/drm/radeon/sid.h +++ b/drivers/gpu/drm/radeon/sid.h @@ -1598,7 +1598,7 @@ * 6. COMMAND [30:21] | BYTE_COUNT [20:0] */ # define PACKET3_CP_DMA_DST_SEL(x) ((x) << 20) - /* 0 - SRC_ADDR + /* 0 - DST_ADDR * 1 - GDS */ # define PACKET3_CP_DMA_ENGINE(x) ((x) << 27) @@ -1613,7 +1613,7 @@ # define PACKET3_CP_DMA_CP_SYNC (1 << 31) /* COMMAND */ # define PACKET3_CP_DMA_DIS_WC (1 << 21) -# define PACKET3_CP_DMA_CMD_SRC_SWAP(x) ((x) << 23) +# define PACKET3_CP_DMA_CMD_SRC_SWAP(x) ((x) << 22) /* 0 - none * 1 - 8 in 16 * 2 - 8 in 32 diff --git a/drivers/gpu/drm/radeon/trinity_dpm.c b/drivers/gpu/drm/radeon/trinity_dpm.c index 7f998bf1cc9d..d700698a1f22 100644 --- a/drivers/gpu/drm/radeon/trinity_dpm.c +++ b/drivers/gpu/drm/radeon/trinity_dpm.c @@ -1868,14 +1868,14 @@ int trinity_dpm_init(struct radeon_device *rdev) for (i = 0; i < SUMO_MAX_HARDWARE_POWERLEVELS; i++) pi->at[i] = TRINITY_AT_DFLT; - pi->enable_bapm = true; + pi->enable_bapm = false; pi->enable_nbps_policy = true; pi->enable_sclk_ds = true; pi->enable_gfx_power_gating = true; pi->enable_gfx_clock_gating = true; - pi->enable_mg_clock_gating = true; - pi->enable_gfx_dynamic_mgpg = true; /* ??? */ - pi->override_dynamic_mgpg = true; + pi->enable_mg_clock_gating = false; + pi->enable_gfx_dynamic_mgpg = false; + pi->override_dynamic_mgpg = false; pi->enable_auto_thermal_throttling = true; pi->voltage_drop_in_dce = false; /* need to restructure dpm/modeset interaction */ pi->uvd_dpm = true; /* ??? */ diff --git a/drivers/gpu/drm/radeon/uvd_v1_0.c b/drivers/gpu/drm/radeon/uvd_v1_0.c index 3100fa9cb52f..d4a68af1a279 100644 --- a/drivers/gpu/drm/radeon/uvd_v1_0.c +++ b/drivers/gpu/drm/radeon/uvd_v1_0.c @@ -212,8 +212,8 @@ int uvd_v1_0_start(struct radeon_device *rdev) /* enable VCPU clock */ WREG32(UVD_VCPU_CNTL, 1 << 9); - /* enable UMC and NC0 */ - WREG32_P(UVD_LMI_CTRL2, 1 << 13, ~((1 << 8) | (1 << 13))); + /* enable UMC */ + WREG32_P(UVD_LMI_CTRL2, 0, ~(1 << 8)); /* boot up the VCPU */ WREG32(UVD_SOFT_RESET, 0); @@ -357,7 +357,7 @@ int uvd_v1_0_ring_test(struct radeon_device *rdev, struct radeon_ring *ring) * * Emit a semaphore command (either wait or signal) to the UVD ring. */ -void uvd_v1_0_semaphore_emit(struct radeon_device *rdev, +bool uvd_v1_0_semaphore_emit(struct radeon_device *rdev, struct radeon_ring *ring, struct radeon_semaphore *semaphore, bool emit_wait) @@ -372,6 +372,8 @@ void uvd_v1_0_semaphore_emit(struct radeon_device *rdev, radeon_ring_write(ring, PACKET0(UVD_SEMA_CMD, 0)); radeon_ring_write(ring, emit_wait ? 1 : 0); + + return true; } /** diff --git a/drivers/gpu/drm/radeon/uvd_v3_1.c b/drivers/gpu/drm/radeon/uvd_v3_1.c index 5b6fa1f62d4e..d722db2cf340 100644 --- a/drivers/gpu/drm/radeon/uvd_v3_1.c +++ b/drivers/gpu/drm/radeon/uvd_v3_1.c @@ -37,7 +37,7 @@ * * Emit a semaphore command (either wait or signal) to the UVD ring. */ -void uvd_v3_1_semaphore_emit(struct radeon_device *rdev, +bool uvd_v3_1_semaphore_emit(struct radeon_device *rdev, struct radeon_ring *ring, struct radeon_semaphore *semaphore, bool emit_wait) @@ -52,4 +52,6 @@ void uvd_v3_1_semaphore_emit(struct radeon_device *rdev, radeon_ring_write(ring, PACKET0(UVD_SEMA_CMD, 0)); radeon_ring_write(ring, 0x80 | (emit_wait ? 1 : 0)); + + return true; } |