nbio_v7_4.c 8.8 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
/*
 * Copyright 2018 Advanced Micro Devices, Inc.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
 * to deal in the Software without restriction, including without limitation
 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
 * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
 * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
 * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
 * OTHER DEALINGS IN THE SOFTWARE.
 *
 */
#include "amdgpu.h"
#include "amdgpu_atombios.h"
#include "nbio_v7_4.h"

#include "nbio/nbio_7_4_offset.h"
#include "nbio/nbio_7_4_sh_mask.h"

#define smnNBIF_MGCG_CTRL_LCLK	0x1013a21c

#define smnCPM_CONTROL                                                                                  0x11180460
#define smnPCIE_CNTL2                                                                                   0x11180070
34
#define smnPCIE_CI_CNTL                                                                                 0x11180080
35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79

static u32 nbio_v7_4_get_rev_id(struct amdgpu_device *adev)
{
    u32 tmp = RREG32_SOC15(NBIO, 0, mmRCC_DEV0_EPF0_STRAP0);

	tmp &= RCC_DEV0_EPF0_STRAP0__STRAP_ATI_REV_ID_DEV0_F0_MASK;
	tmp >>= RCC_DEV0_EPF0_STRAP0__STRAP_ATI_REV_ID_DEV0_F0__SHIFT;

	return tmp;
}

static void nbio_v7_4_mc_access_enable(struct amdgpu_device *adev, bool enable)
{
	if (enable)
		WREG32_SOC15(NBIO, 0, mmBIF_FB_EN,
			BIF_FB_EN__FB_READ_EN_MASK | BIF_FB_EN__FB_WRITE_EN_MASK);
	else
		WREG32_SOC15(NBIO, 0, mmBIF_FB_EN, 0);
}

static void nbio_v7_4_hdp_flush(struct amdgpu_device *adev,
				struct amdgpu_ring *ring)
{
	if (!ring || !ring->funcs->emit_wreg)
		WREG32_SOC15_NO_KIQ(NBIO, 0, mmHDP_MEM_COHERENCY_FLUSH_CNTL, 0);
	else
		amdgpu_ring_emit_wreg(ring, SOC15_REG_OFFSET(
			NBIO, 0, mmHDP_MEM_COHERENCY_FLUSH_CNTL), 0);
}

static u32 nbio_v7_4_get_memsize(struct amdgpu_device *adev)
{
	return RREG32_SOC15(NBIO, 0, mmRCC_CONFIG_MEMSIZE);
}

static void nbio_v7_4_sdma_doorbell_range(struct amdgpu_device *adev, int instance,
					  bool use_doorbell, int doorbell_index)
{
	u32 reg = instance == 0 ? SOC15_REG_OFFSET(NBIO, 0, mmBIF_SDMA0_DOORBELL_RANGE) :
			SOC15_REG_OFFSET(NBIO, 0, mmBIF_SDMA1_DOORBELL_RANGE);

	u32 doorbell_range = RREG32(reg);

	if (use_doorbell) {
		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, OFFSET, doorbell_index);
80
		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, 2);
81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208
	} else
		doorbell_range = REG_SET_FIELD(doorbell_range, BIF_SDMA0_DOORBELL_RANGE, SIZE, 0);

	WREG32(reg, doorbell_range);
}

static void nbio_v7_4_enable_doorbell_aperture(struct amdgpu_device *adev,
					       bool enable)
{
	WREG32_FIELD15(NBIO, 0, RCC_DOORBELL_APER_EN, BIF_DOORBELL_APER_EN, enable ? 1 : 0);
}

static void nbio_v7_4_enable_doorbell_selfring_aperture(struct amdgpu_device *adev,
							bool enable)
{

}

static void nbio_v7_4_ih_doorbell_range(struct amdgpu_device *adev,
					bool use_doorbell, int doorbell_index)
{
	u32 ih_doorbell_range = RREG32_SOC15(NBIO, 0 , mmBIF_IH_DOORBELL_RANGE);

	if (use_doorbell) {
		ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range, BIF_IH_DOORBELL_RANGE, OFFSET, doorbell_index);
		ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range, BIF_IH_DOORBELL_RANGE, SIZE, 2);
	} else
		ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range, BIF_IH_DOORBELL_RANGE, SIZE, 0);

	WREG32_SOC15(NBIO, 0, mmBIF_IH_DOORBELL_RANGE, ih_doorbell_range);
}


static void nbio_v7_4_update_medium_grain_clock_gating(struct amdgpu_device *adev,
						       bool enable)
{
	//TODO: Add support for v7.4
}

static void nbio_v7_4_update_medium_grain_light_sleep(struct amdgpu_device *adev,
						      bool enable)
{
	uint32_t def, data;

	def = data = RREG32_PCIE(smnPCIE_CNTL2);
	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_BIF_LS)) {
		data |= (PCIE_CNTL2__SLV_MEM_LS_EN_MASK |
			 PCIE_CNTL2__MST_MEM_LS_EN_MASK |
			 PCIE_CNTL2__REPLAY_MEM_LS_EN_MASK);
	} else {
		data &= ~(PCIE_CNTL2__SLV_MEM_LS_EN_MASK |
			  PCIE_CNTL2__MST_MEM_LS_EN_MASK |
			  PCIE_CNTL2__REPLAY_MEM_LS_EN_MASK);
	}

	if (def != data)
		WREG32_PCIE(smnPCIE_CNTL2, data);
}

static void nbio_v7_4_get_clockgating_state(struct amdgpu_device *adev,
					    u32 *flags)
{
	int data;

	/* AMD_CG_SUPPORT_BIF_MGCG */
	data = RREG32_PCIE(smnCPM_CONTROL);
	if (data & CPM_CONTROL__LCLK_DYN_GATE_ENABLE_MASK)
		*flags |= AMD_CG_SUPPORT_BIF_MGCG;

	/* AMD_CG_SUPPORT_BIF_LS */
	data = RREG32_PCIE(smnPCIE_CNTL2);
	if (data & PCIE_CNTL2__SLV_MEM_LS_EN_MASK)
		*flags |= AMD_CG_SUPPORT_BIF_LS;
}

static void nbio_v7_4_ih_control(struct amdgpu_device *adev)
{
	u32 interrupt_cntl;

	/* setup interrupt control */
	WREG32_SOC15(NBIO, 0, mmINTERRUPT_CNTL2, adev->dummy_page_addr >> 8);
	interrupt_cntl = RREG32_SOC15(NBIO, 0, mmINTERRUPT_CNTL);
	/* INTERRUPT_CNTL__IH_DUMMY_RD_OVERRIDE_MASK=0 - dummy read disabled with msi, enabled without msi
	 * INTERRUPT_CNTL__IH_DUMMY_RD_OVERRIDE_MASK=1 - dummy read controlled by IH_DUMMY_RD_EN
	 */
	interrupt_cntl = REG_SET_FIELD(interrupt_cntl, INTERRUPT_CNTL, IH_DUMMY_RD_OVERRIDE, 0);
	/* INTERRUPT_CNTL__IH_REQ_NONSNOOP_EN_MASK=1 if ring is in non-cacheable memory, e.g., vram */
	interrupt_cntl = REG_SET_FIELD(interrupt_cntl, INTERRUPT_CNTL, IH_REQ_NONSNOOP_EN, 0);
	WREG32_SOC15(NBIO, 0, mmINTERRUPT_CNTL, interrupt_cntl);
}

static u32 nbio_v7_4_get_hdp_flush_req_offset(struct amdgpu_device *adev)
{
	return SOC15_REG_OFFSET(NBIO, 0, mmGPU_HDP_FLUSH_REQ);
}

static u32 nbio_v7_4_get_hdp_flush_done_offset(struct amdgpu_device *adev)
{
	return SOC15_REG_OFFSET(NBIO, 0, mmGPU_HDP_FLUSH_DONE);
}

static u32 nbio_v7_4_get_pcie_index_offset(struct amdgpu_device *adev)
{
	return SOC15_REG_OFFSET(NBIO, 0, mmPCIE_INDEX2);
}

static u32 nbio_v7_4_get_pcie_data_offset(struct amdgpu_device *adev)
{
	return SOC15_REG_OFFSET(NBIO, 0, mmPCIE_DATA2);
}

static const struct nbio_hdp_flush_reg nbio_v7_4_hdp_flush_reg = {
	.ref_and_mask_cp0 = GPU_HDP_FLUSH_DONE__CP0_MASK,
	.ref_and_mask_cp1 = GPU_HDP_FLUSH_DONE__CP1_MASK,
	.ref_and_mask_cp2 = GPU_HDP_FLUSH_DONE__CP2_MASK,
	.ref_and_mask_cp3 = GPU_HDP_FLUSH_DONE__CP3_MASK,
	.ref_and_mask_cp4 = GPU_HDP_FLUSH_DONE__CP4_MASK,
	.ref_and_mask_cp5 = GPU_HDP_FLUSH_DONE__CP5_MASK,
	.ref_and_mask_cp6 = GPU_HDP_FLUSH_DONE__CP6_MASK,
	.ref_and_mask_cp7 = GPU_HDP_FLUSH_DONE__CP7_MASK,
	.ref_and_mask_cp8 = GPU_HDP_FLUSH_DONE__CP8_MASK,
	.ref_and_mask_cp9 = GPU_HDP_FLUSH_DONE__CP9_MASK,
	.ref_and_mask_sdma0 = GPU_HDP_FLUSH_DONE__SDMA0_MASK,
	.ref_and_mask_sdma1 = GPU_HDP_FLUSH_DONE__SDMA1_MASK,
};

static void nbio_v7_4_detect_hw_virt(struct amdgpu_device *adev)
{
209 210 211 212 213 214 215 216 217 218 219 220 221
	uint32_t reg;

	reg = RREG32_SOC15(NBIO, 0, mmRCC_IOV_FUNC_IDENTIFIER);
	if (reg & 1)
		adev->virt.caps |= AMDGPU_SRIOV_CAPS_IS_VF;

	if (reg & 0x80000000)
		adev->virt.caps |= AMDGPU_SRIOV_CAPS_ENABLE_IOV;

	if (!reg) {
		if (is_virtual_machine())	/* passthrough mode exclus sriov mod */
			adev->virt.caps |= AMDGPU_PASSTHROUGH_MODE;
	}
222 223 224 225
}

static void nbio_v7_4_init_registers(struct amdgpu_device *adev)
{
226 227 228 229
	uint32_t def, data;

	def = data = RREG32_PCIE(smnPCIE_CI_CNTL);
	data = REG_SET_FIELD(data, PCIE_CI_CNTL, CI_SLV_ORDERING_DIS, 1);
230

231 232
	if (def != data)
		WREG32_PCIE(smnPCIE_CI_CNTL, data);
233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255
}

const struct amdgpu_nbio_funcs nbio_v7_4_funcs = {
	.hdp_flush_reg = &nbio_v7_4_hdp_flush_reg,
	.get_hdp_flush_req_offset = nbio_v7_4_get_hdp_flush_req_offset,
	.get_hdp_flush_done_offset = nbio_v7_4_get_hdp_flush_done_offset,
	.get_pcie_index_offset = nbio_v7_4_get_pcie_index_offset,
	.get_pcie_data_offset = nbio_v7_4_get_pcie_data_offset,
	.get_rev_id = nbio_v7_4_get_rev_id,
	.mc_access_enable = nbio_v7_4_mc_access_enable,
	.hdp_flush = nbio_v7_4_hdp_flush,
	.get_memsize = nbio_v7_4_get_memsize,
	.sdma_doorbell_range = nbio_v7_4_sdma_doorbell_range,
	.enable_doorbell_aperture = nbio_v7_4_enable_doorbell_aperture,
	.enable_doorbell_selfring_aperture = nbio_v7_4_enable_doorbell_selfring_aperture,
	.ih_doorbell_range = nbio_v7_4_ih_doorbell_range,
	.update_medium_grain_clock_gating = nbio_v7_4_update_medium_grain_clock_gating,
	.update_medium_grain_light_sleep = nbio_v7_4_update_medium_grain_light_sleep,
	.get_clockgating_state = nbio_v7_4_get_clockgating_state,
	.ih_control = nbio_v7_4_ih_control,
	.init_registers = nbio_v7_4_init_registers,
	.detect_hw_virt = nbio_v7_4_detect_hw_virt,
};