summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c')
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c363
1 files changed, 190 insertions, 173 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index dbe7442fb25c..407074f958f4 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -56,10 +56,6 @@
#define GFX10_NUM_GFX_RINGS_Sienna_Cichlid 1
#define GFX10_MEC_HPD_SIZE 2048
-#define RLCG_VFGATE_DISABLED 0x4000000
-#define RLCG_WRONG_OPERATION_TYPE 0x2000000
-#define RLCG_NOT_IN_RANGE 0x1000000
-
#define F32_CE_PROGRAM_RAM_SIZE 65536
#define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L
@@ -110,6 +106,12 @@
#define mmGOLDEN_TSC_COUNT_UPPER_Vangogh_BASE_IDX 1
#define mmGOLDEN_TSC_COUNT_LOWER_Vangogh 0x0026
#define mmGOLDEN_TSC_COUNT_LOWER_Vangogh_BASE_IDX 1
+
+#define mmGOLDEN_TSC_COUNT_UPPER_GC_10_3_6 0x002d
+#define mmGOLDEN_TSC_COUNT_UPPER_GC_10_3_6_BASE_IDX 1
+#define mmGOLDEN_TSC_COUNT_LOWER_GC_10_3_6 0x002e
+#define mmGOLDEN_TSC_COUNT_LOWER_GC_10_3_6_BASE_IDX 1
+
#define mmSPI_CONFIG_CNTL_1_Vangogh 0x2441
#define mmSPI_CONFIG_CNTL_1_Vangogh_BASE_IDX 1
#define mmVGT_TF_MEMORY_BASE_HI_Vangogh 0x2261
@@ -180,14 +182,6 @@
#define mmRLC_SPARE_INT_0_Sienna_Cichlid 0x4ca5
#define mmRLC_SPARE_INT_0_Sienna_Cichlid_BASE_IDX 1
-#define GFX_RLCG_GC_WRITE_OLD (0x8 << 28)
-#define GFX_RLCG_GC_WRITE (0x0 << 28)
-#define GFX_RLCG_GC_READ (0x1 << 28)
-#define GFX_RLCG_MMHUB_WRITE (0x2 << 28)
-
-#define RLCG_ERROR_REPORT_ENABLED(adev) \
- (amdgpu_sriov_reg_indirect_mmhub(adev) || amdgpu_sriov_reg_indirect_gc(adev))
-
MODULE_FIRMWARE("amdgpu/navi10_ce.bin");
MODULE_FIRMWARE("amdgpu/navi10_pfp.bin");
MODULE_FIRMWARE("amdgpu/navi10_me.bin");
@@ -256,13 +250,6 @@ MODULE_FIRMWARE("amdgpu/yellow_carp_mec.bin");
MODULE_FIRMWARE("amdgpu/yellow_carp_mec2.bin");
MODULE_FIRMWARE("amdgpu/yellow_carp_rlc.bin");
-MODULE_FIRMWARE("amdgpu/cyan_skillfish_ce.bin");
-MODULE_FIRMWARE("amdgpu/cyan_skillfish_pfp.bin");
-MODULE_FIRMWARE("amdgpu/cyan_skillfish_me.bin");
-MODULE_FIRMWARE("amdgpu/cyan_skillfish_mec.bin");
-MODULE_FIRMWARE("amdgpu/cyan_skillfish_mec2.bin");
-MODULE_FIRMWARE("amdgpu/cyan_skillfish_rlc.bin");
-
MODULE_FIRMWARE("amdgpu/cyan_skillfish2_ce.bin");
MODULE_FIRMWARE("amdgpu/cyan_skillfish2_pfp.bin");
MODULE_FIRMWARE("amdgpu/cyan_skillfish2_me.bin");
@@ -270,6 +257,20 @@ MODULE_FIRMWARE("amdgpu/cyan_skillfish2_mec.bin");
MODULE_FIRMWARE("amdgpu/cyan_skillfish2_mec2.bin");
MODULE_FIRMWARE("amdgpu/cyan_skillfish2_rlc.bin");
+MODULE_FIRMWARE("amdgpu/gc_10_3_6_ce.bin");
+MODULE_FIRMWARE("amdgpu/gc_10_3_6_pfp.bin");
+MODULE_FIRMWARE("amdgpu/gc_10_3_6_me.bin");
+MODULE_FIRMWARE("amdgpu/gc_10_3_6_mec.bin");
+MODULE_FIRMWARE("amdgpu/gc_10_3_6_mec2.bin");
+MODULE_FIRMWARE("amdgpu/gc_10_3_6_rlc.bin");
+
+MODULE_FIRMWARE("amdgpu/gc_10_3_7_ce.bin");
+MODULE_FIRMWARE("amdgpu/gc_10_3_7_pfp.bin");
+MODULE_FIRMWARE("amdgpu/gc_10_3_7_me.bin");
+MODULE_FIRMWARE("amdgpu/gc_10_3_7_mec.bin");
+MODULE_FIRMWARE("amdgpu/gc_10_3_7_mec2.bin");
+MODULE_FIRMWARE("amdgpu/gc_10_3_7_rlc.bin");
+
static const struct soc15_reg_golden golden_settings_gc_10_1[] =
{
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_4, 0xffffffff, 0x00400014),
@@ -1463,143 +1464,6 @@ static const struct soc15_reg_golden golden_settings_gc_10_1_2[] =
SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffffffff, 0x00c00000)
};
-static bool gfx_v10_get_rlcg_flag(struct amdgpu_device *adev, u32 acc_flags, u32 hwip,
- int write, u32 *rlcg_flag)
-{
- switch (hwip) {
- case GC_HWIP:
- if (amdgpu_sriov_reg_indirect_gc(adev)) {
- *rlcg_flag = write ? GFX_RLCG_GC_WRITE : GFX_RLCG_GC_READ;
-
- return true;
- /* only in new version, AMDGPU_REGS_NO_KIQ and AMDGPU_REGS_RLC enabled simultaneously */
- } else if ((acc_flags & AMDGPU_REGS_RLC) && !(acc_flags & AMDGPU_REGS_NO_KIQ)) {
- *rlcg_flag = GFX_RLCG_GC_WRITE_OLD;
-
- return true;
- }
-
- break;
- case MMHUB_HWIP:
- if (amdgpu_sriov_reg_indirect_mmhub(adev) &&
- (acc_flags & AMDGPU_REGS_RLC) && write) {
- *rlcg_flag = GFX_RLCG_MMHUB_WRITE;
- return true;
- }
-
- break;
- default:
- DRM_DEBUG("Not program register by RLCG\n");
- }
-
- return false;
-}
-
-static u32 gfx_v10_rlcg_rw(struct amdgpu_device *adev, u32 offset, u32 v, uint32_t flag)
-{
- static void *scratch_reg0;
- static void *scratch_reg1;
- static void *scratch_reg2;
- static void *scratch_reg3;
- static void *spare_int;
- static uint32_t grbm_cntl;
- static uint32_t grbm_idx;
- uint32_t i = 0;
- uint32_t retries = 50000;
- u32 ret = 0;
- u32 tmp;
-
- scratch_reg0 = adev->rmmio +
- (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG0_BASE_IDX] + mmSCRATCH_REG0) * 4;
- scratch_reg1 = adev->rmmio +
- (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG1) * 4;
- scratch_reg2 = adev->rmmio +
- (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG0_BASE_IDX] + mmSCRATCH_REG2) * 4;
- scratch_reg3 = adev->rmmio +
- (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG3) * 4;
-
- if (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(10, 3, 0)) {
- spare_int = adev->rmmio +
- (adev->reg_offset[GC_HWIP][0][mmRLC_SPARE_INT_0_Sienna_Cichlid_BASE_IDX]
- + mmRLC_SPARE_INT_0_Sienna_Cichlid) * 4;
- } else {
- spare_int = adev->rmmio +
- (adev->reg_offset[GC_HWIP][0][mmRLC_SPARE_INT_BASE_IDX] + mmRLC_SPARE_INT) * 4;
- }
-
- grbm_cntl = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_CNTL_BASE_IDX] + mmGRBM_GFX_CNTL;
- grbm_idx = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_INDEX_BASE_IDX] + mmGRBM_GFX_INDEX;
-
- if (offset == grbm_cntl || offset == grbm_idx) {
- if (offset == grbm_cntl)
- writel(v, scratch_reg2);
- else if (offset == grbm_idx)
- writel(v, scratch_reg3);
-
- writel(v, ((void __iomem *)adev->rmmio) + (offset * 4));
- } else {
- writel(v, scratch_reg0);
- writel(offset | flag, scratch_reg1);
- writel(1, spare_int);
-
- for (i = 0; i < retries; i++) {
- tmp = readl(scratch_reg1);
- if (!(tmp & flag))
- break;
-
- udelay(10);
- }
-
- if (i >= retries) {
- if (RLCG_ERROR_REPORT_ENABLED(adev)) {
- if (tmp & RLCG_VFGATE_DISABLED)
- pr_err("The vfgate is disabled, program reg:0x%05x failed!\n", offset);
- else if (tmp & RLCG_WRONG_OPERATION_TYPE)
- pr_err("Wrong operation type, program reg:0x%05x failed!\n", offset);
- else if (tmp & RLCG_NOT_IN_RANGE)
- pr_err("The register is not in range, program reg:0x%05x failed!\n", offset);
- else
- pr_err("Unknown error type, program reg:0x%05x failed!\n", offset);
- } else
- pr_err("timeout: rlcg program reg:0x%05x failed!\n", offset);
- }
- }
-
- ret = readl(scratch_reg0);
-
- return ret;
-}
-
-static void gfx_v10_sriov_wreg(struct amdgpu_device *adev, u32 offset, u32 value, u32 acc_flags, u32 hwip)
-{
- u32 rlcg_flag;
-
- if (!amdgpu_sriov_runtime(adev) &&
- gfx_v10_get_rlcg_flag(adev, acc_flags, hwip, 1, &rlcg_flag)) {
- gfx_v10_rlcg_rw(adev, offset, value, rlcg_flag);
- return;
- }
-
- if (acc_flags & AMDGPU_REGS_NO_KIQ)
- WREG32_NO_KIQ(offset, value);
- else
- WREG32(offset, value);
-}
-
-static u32 gfx_v10_sriov_rreg(struct amdgpu_device *adev, u32 offset, u32 acc_flags, u32 hwip)
-{
- u32 rlcg_flag;
-
- if (!amdgpu_sriov_runtime(adev) &&
- gfx_v10_get_rlcg_flag(adev, acc_flags, hwip, 0, &rlcg_flag))
- return gfx_v10_rlcg_rw(adev, offset, 0, rlcg_flag);
-
- if (acc_flags & AMDGPU_REGS_NO_KIQ)
- return RREG32_NO_KIQ(offset);
- else
- return RREG32(offset);
-}
-
static const struct soc15_reg_golden golden_settings_gc_10_1_nv14[] =
{
/* Pending on emulation bring up */
@@ -3429,7 +3293,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_3_3[] =
SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG3, 0xffffffff, 0x00000280),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG4, 0xffffffff, 0x00800000),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG, 0x0c1807ff, 0x00000242),
- SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCR_GENERAL_CNTL, 0x1ff1ffff, 0x00000500),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCR_GENERAL_CNTL_Vangogh, 0x1ff1ffff, 0x00000500),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL1_PIPE_STEER, 0x000000ff, 0x000000e4),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2_PIPE_STEER_0, 0x77777777, 0x32103210),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2_PIPE_STEER_1, 0x77777777, 0x32103210),
@@ -3557,6 +3421,57 @@ static const struct soc15_reg_golden golden_settings_gc_10_0_cyan_skillfish[] =
SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0x00800000, 0x00800000)
};
+static const struct soc15_reg_golden golden_settings_gc_10_3_6[] =
+{
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CS_CLK_CTRL, 0xff7f0fff, 0x78000100),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmCH_PIPE_STEER, 0x000000ff, 0x00000044),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPF_GCR_CNTL, 0x0007ffff, 0x0000c200),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG3, 0xffffffff, 0x00000280),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG4, 0xffffffff, 0x00800000),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG, 0x0c1807ff, 0x00000042),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCR_GENERAL_CNTL_Vangogh, 0x1ff1ffff, 0x00000500),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL1_PIPE_STEER, 0x000000ff, 0x00000044),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2_PIPE_STEER_0, 0x77777777, 0x32103210),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2_PIPE_STEER_1, 0x77777777, 0x32103210),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2A_ADDR_MATCH_MASK, 0xffffffff, 0xfffffff3),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_ADDR_MATCH_MASK, 0xffffffff, 0xfffffff3),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CM_CTRL1, 0xff8fff0f, 0x580f1008),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CTRL3, 0xf7ffffff, 0x00f80988),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmLDS_CONFIG, 0x000001ff, 0x00000020),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_CL_ENHANCE, 0xf17fffff, 0x01200007),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_BINNER_TIMEOUT_COUNTER, 0xffffffff, 0x00000800),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_2, 0xffffffbf, 0x00000820),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQG_CONFIG, 0x000017ff, 0x00001000),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmSX_DEBUG_1, 0xffffff7f, 0x00010020),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfff7ffff, 0x01030000),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffffffff, 0x00100000)
+};
+
+static const struct soc15_reg_golden golden_settings_gc_10_3_7[] = {
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CS_CLK_CTRL, 0xff7f0fff, 0x78000100),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmCH_PIPE_STEER, 0x000000ff, 0x000000e4),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPF_GCR_CNTL, 0x0007ffff, 0x0000c200),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG3, 0xffffffff, 0x00000280),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG4, 0xffffffff, 0x00800000),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG, 0x0c1807ff, 0x00000041),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCR_GENERAL_CNTL_Vangogh, 0x1ff1ffff, 0x00000500),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL1_PIPE_STEER, 0x000000ff, 0x000000e4),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2_PIPE_STEER_0, 0x77777777, 0x32103210),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2_PIPE_STEER_1, 0x77777777, 0x32103210),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2A_ADDR_MATCH_MASK, 0xffffffff, 0xffffffff),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_ADDR_MATCH_MASK, 0xffffffff, 0xffffffff),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CM_CTRL1, 0xff8fff0f, 0x580f1008),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CTRL3, 0xf7ffffff, 0x00f80988),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmLDS_CONFIG, 0x000001ff, 0x00000020),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_CL_ENHANCE, 0xf000003f, 0x01200007),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_BINNER_TIMEOUT_COUNTER, 0xffffffff, 0x00000800),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_2, 0xffffffbf, 0x00000820),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQG_CONFIG, 0x000017ff, 0x00001000),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmSX_DEBUG_1, 0xffffff7f, 0x00010020),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfff7ffff, 0x01030000),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffffffff, 0x00100000)
+};
+
#define DEFAULT_SH_MEM_CONFIG \
((SH_MEM_ADDRESS_MODE_64 << SH_MEM_CONFIG__ADDRESS_MODE__SHIFT) | \
(SH_MEM_ALIGNMENT_MODE_UNALIGNED << SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT) | \
@@ -3790,10 +3705,21 @@ static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev)
(const u32)ARRAY_SIZE(golden_settings_gc_10_3_5));
break;
case IP_VERSION(10, 1, 3):
+ case IP_VERSION(10, 1, 4):
soc15_program_register_sequence(adev,
golden_settings_gc_10_0_cyan_skillfish,
(const u32)ARRAY_SIZE(golden_settings_gc_10_0_cyan_skillfish));
break;
+ case IP_VERSION(10, 3, 6):
+ soc15_program_register_sequence(adev,
+ golden_settings_gc_10_3_6,
+ (const u32)ARRAY_SIZE(golden_settings_gc_10_3_6));
+ break;
+ case IP_VERSION(10, 3, 7):
+ soc15_program_register_sequence(adev,
+ golden_settings_gc_10_3_7,
+ (const u32)ARRAY_SIZE(golden_settings_gc_10_3_7));
+ break;
default:
break;
}
@@ -3904,8 +3830,7 @@ static int gfx_v10_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
gpu_addr = adev->wb.gpu_addr + (index * 4);
adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD);
memset(&ib, 0, sizeof(ib));
- r = amdgpu_ib_get(adev, NULL, 16,
- AMDGPU_IB_POOL_DIRECT, &ib);
+ r = amdgpu_ib_get(adev, NULL, 20, AMDGPU_IB_POOL_DIRECT, &ib);
if (r)
goto err1;
@@ -3968,6 +3893,7 @@ static void gfx_v10_0_check_fw_write_wait(struct amdgpu_device *adev)
case IP_VERSION(10, 1, 2):
case IP_VERSION(10, 1, 1):
case IP_VERSION(10, 1, 3):
+ case IP_VERSION(10, 1, 4):
if ((adev->gfx.me_fw_version >= 0x00000046) &&
(adev->gfx.me_feature_version >= 27) &&
(adev->gfx.pfp_fw_version >= 0x00000068) &&
@@ -3981,7 +3907,9 @@ static void gfx_v10_0_check_fw_write_wait(struct amdgpu_device *adev)
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
adev->gfx.cp_fw_write_wait = true;
break;
default:
@@ -4102,11 +4030,15 @@ static int gfx_v10_0_init_microcode(struct amdgpu_device *adev)
case IP_VERSION(10, 3, 3):
chip_name = "yellow_carp";
break;
+ case IP_VERSION(10, 3, 6):
+ chip_name = "gc_10_3_6";
+ break;
case IP_VERSION(10, 1, 3):
- if (adev->apu_flags & AMD_APU_IS_CYAN_SKILLFISH2)
- chip_name = "cyan_skillfish2";
- else
- chip_name = "cyan_skillfish";
+ case IP_VERSION(10, 1, 4):
+ chip_name = "cyan_skillfish2";
+ break;
+ case IP_VERSION(10, 3, 7):
+ chip_name = "gc_10_3_7";
break;
default:
BUG();
@@ -4448,6 +4380,30 @@ static void gfx_v10_0_rlc_fini(struct amdgpu_device *adev)
(void **)&adev->gfx.rlc.cp_table_ptr);
}
+static void gfx_v10_0_init_rlcg_reg_access_ctrl(struct amdgpu_device *adev)
+{
+ struct amdgpu_rlcg_reg_access_ctrl *reg_access_ctrl;
+
+ reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl;
+ reg_access_ctrl->scratch_reg0 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG0);
+ reg_access_ctrl->scratch_reg1 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG1);
+ reg_access_ctrl->scratch_reg2 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG2);
+ reg_access_ctrl->scratch_reg3 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG3);
+ reg_access_ctrl->grbm_cntl = SOC15_REG_OFFSET(GC, 0, mmGRBM_GFX_CNTL);
+ reg_access_ctrl->grbm_idx = SOC15_REG_OFFSET(GC, 0, mmGRBM_GFX_INDEX);
+ switch (adev->ip_versions[GC_HWIP][0]) {
+ case IP_VERSION(10, 3, 0):
+ reg_access_ctrl->spare_int =
+ SOC15_REG_OFFSET(GC, 0, mmRLC_SPARE_INT_0_Sienna_Cichlid);
+ break;
+ default:
+ reg_access_ctrl->spare_int =
+ SOC15_REG_OFFSET(GC, 0, mmRLC_SPARE_INT);
+ break;
+ }
+ adev->gfx.rlc.rlcg_reg_access_supported = true;
+}
+
static int gfx_v10_0_rlc_init(struct amdgpu_device *adev)
{
const struct cs_section_def *cs_data;
@@ -4468,6 +4424,7 @@ static int gfx_v10_0_rlc_init(struct amdgpu_device *adev)
if (adev->gfx.rlc.funcs->update_spm_vmid)
adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf);
+
return 0;
}
@@ -4678,7 +4635,9 @@ static void gfx_v10_0_gpu_early_init(struct amdgpu_device *adev)
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
adev->gfx.config.max_hw_contexts = 8;
adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
@@ -4689,6 +4648,7 @@ static void gfx_v10_0_gpu_early_init(struct amdgpu_device *adev)
1 << REG_GET_FIELD(gb_addr_config, GB_ADDR_CONFIG, NUM_PKRS);
break;
case IP_VERSION(10, 1, 3):
+ case IP_VERSION(10, 1, 4):
adev->gfx.config.max_hw_contexts = 8;
adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
@@ -4780,7 +4740,7 @@ static int gfx_v10_0_compute_ring_init(struct amdgpu_device *adev, int ring_id,
+ ((ring->me - 1) * adev->gfx.mec.num_pipe_per_mec)
+ ring->pipe;
hw_prio = amdgpu_gfx_is_high_priority_compute_queue(adev, ring) ?
- AMDGPU_GFX_PIPE_PRIO_HIGH : AMDGPU_GFX_PIPE_PRIO_NORMAL;
+ AMDGPU_RING_PRIO_2 : AMDGPU_RING_PRIO_DEFAULT;
/* type-2 packets are deprecated on MEC, use type-3 instead */
r = amdgpu_ring_init(adev, ring, 1024, &adev->gfx.eop_irq, irq_type,
hw_prio, NULL);
@@ -4801,6 +4761,7 @@ static int gfx_v10_0_sw_init(void *handle)
case IP_VERSION(10, 1, 1):
case IP_VERSION(10, 1, 2):
case IP_VERSION(10, 1, 3):
+ case IP_VERSION(10, 1, 4):
adev->gfx.me.num_me = 1;
adev->gfx.me.num_pipe_per_me = 1;
adev->gfx.me.num_queue_per_pipe = 1;
@@ -4813,7 +4774,9 @@ static int gfx_v10_0_sw_init(void *handle)
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
adev->gfx.me.num_me = 1;
adev->gfx.me.num_pipe_per_me = 1;
adev->gfx.me.num_queue_per_pipe = 1;
@@ -4865,10 +4828,14 @@ static int gfx_v10_0_sw_init(void *handle)
if (r)
return r;
- r = gfx_v10_0_rlc_init(adev);
- if (r) {
- DRM_ERROR("Failed to init rlc BOs!\n");
- return r;
+ if (adev->gfx.rlc.funcs) {
+ if (adev->gfx.rlc.funcs->init) {
+ r = adev->gfx.rlc.funcs->init(adev);
+ if (r) {
+ dev_err(adev->dev, "Failed to init rlc BOs!\n");
+ return r;
+ }
+ }
}
r = gfx_v10_0_mec_init(adev);
@@ -5047,7 +5014,8 @@ static void gfx_v10_0_setup_rb(struct amdgpu_device *adev)
for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
bitmap = i * adev->gfx.config.max_sh_per_se + j;
if (((adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 0)) ||
- (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 3))) &&
+ (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 3)) ||
+ (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 6))) &&
((gfx_v10_3_get_disabled_sa(adev) >> bitmap) & 1))
continue;
gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff);
@@ -6321,7 +6289,9 @@ static void gfx_v10_0_cp_gfx_set_doorbell(struct amdgpu_device *adev,
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
tmp = REG_SET_FIELD(0, CP_RB_DOORBELL_RANGE_LOWER,
DOORBELL_RANGE_LOWER_Sienna_Cichlid, ring->doorbell_index);
WREG32_SOC15(GC, 0, mmCP_RB_DOORBELL_RANGE_LOWER, tmp);
@@ -6458,7 +6428,9 @@ static void gfx_v10_0_cp_compute_enable(struct amdgpu_device *adev, bool enable)
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
WREG32_SOC15(GC, 0, mmCP_MEC_CNTL_Sienna_Cichlid, 0);
break;
default:
@@ -6472,7 +6444,9 @@ static void gfx_v10_0_cp_compute_enable(struct amdgpu_device *adev, bool enable)
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
WREG32_SOC15(GC, 0, mmCP_MEC_CNTL_Sienna_Cichlid,
(CP_MEC_CNTL__MEC_ME1_HALT_MASK |
CP_MEC_CNTL__MEC_ME2_HALT_MASK));
@@ -6570,7 +6544,9 @@ static void gfx_v10_0_kiq_setting(struct amdgpu_ring *ring)
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
tmp = RREG32_SOC15(GC, 0, mmRLC_CP_SCHEDULERS_Sienna_Cichlid);
tmp &= 0xffffff00;
tmp |= (ring->me << 5) | (ring->pipe << 3) | (ring->queue);
@@ -7300,6 +7276,8 @@ static bool gfx_v10_0_check_grbm_cam_remapping(struct amdgpu_device *adev)
break;
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 6):
+ case IP_VERSION(10, 3, 7):
return true;
default:
data = RREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE);
@@ -7334,7 +7312,9 @@ static void gfx_v10_0_setup_grbm_cam_remapping(struct amdgpu_device *adev)
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
/* mmVGT_TF_RING_SIZE_UMD -> mmVGT_TF_RING_SIZE */
data = (SOC15_REG_OFFSET(GC, 0, mmVGT_TF_RING_SIZE_UMD) <<
GRBM_CAM_DATA__CAM_ADDR__SHIFT) |
@@ -7654,6 +7634,7 @@ static int gfx_v10_0_soft_reset(void *handle)
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
if (REG_GET_FIELD(tmp, GRBM_STATUS2, RLC_BUSY_Sienna_Cichlid))
grbm_soft_reset = REG_SET_FIELD(grbm_soft_reset,
@@ -7707,6 +7688,7 @@ static uint64_t gfx_v10_0_get_gpu_clock_counter(struct amdgpu_device *adev)
switch (adev->ip_versions[GC_HWIP][0]) {
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
preempt_disable();
clock_hi = RREG32_SOC15_NO_KIQ(SMUIO, 0, mmGOLDEN_TSC_COUNT_UPPER_Vangogh);
clock_lo = RREG32_SOC15_NO_KIQ(SMUIO, 0, mmGOLDEN_TSC_COUNT_LOWER_Vangogh);
@@ -7721,6 +7703,21 @@ static uint64_t gfx_v10_0_get_gpu_clock_counter(struct amdgpu_device *adev)
preempt_enable();
clock = clock_lo | (clock_hi << 32ULL);
break;
+ case IP_VERSION(10, 3, 6):
+ preempt_disable();
+ clock_hi = RREG32_SOC15_NO_KIQ(SMUIO, 0, mmGOLDEN_TSC_COUNT_UPPER_GC_10_3_6);
+ clock_lo = RREG32_SOC15_NO_KIQ(SMUIO, 0, mmGOLDEN_TSC_COUNT_LOWER_GC_10_3_6);
+ hi_check = RREG32_SOC15_NO_KIQ(SMUIO, 0, mmGOLDEN_TSC_COUNT_UPPER_GC_10_3_6);
+ /* The SMUIO TSC clock frequency is 100MHz, which sets 32-bit carry over
+ * roughly every 42 seconds.
+ */
+ if (hi_check != clock_hi) {
+ clock_lo = RREG32_SOC15_NO_KIQ(SMUIO, 0, mmGOLDEN_TSC_COUNT_LOWER_GC_10_3_6);
+ clock_hi = hi_check;
+ }
+ preempt_enable();
+ clock = clock_lo | (clock_hi << 32ULL);
+ break;
default:
preempt_disable();
clock_hi = RREG32_SOC15_NO_KIQ(SMUIO, 0, mmGOLDEN_TSC_COUNT_UPPER);
@@ -7778,6 +7775,7 @@ static int gfx_v10_0_early_init(void *handle)
case IP_VERSION(10, 1, 1):
case IP_VERSION(10, 1, 2):
case IP_VERSION(10, 1, 3):
+ case IP_VERSION(10, 1, 4):
adev->gfx.num_gfx_rings = GFX10_NUM_GFX_RINGS_NV1X;
break;
case IP_VERSION(10, 3, 0):
@@ -7785,7 +7783,9 @@ static int gfx_v10_0_early_init(void *handle)
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
adev->gfx.num_gfx_rings = GFX10_NUM_GFX_RINGS_Sienna_Cichlid;
break;
default:
@@ -7801,6 +7801,9 @@ static int gfx_v10_0_early_init(void *handle)
gfx_v10_0_set_gds_init(adev);
gfx_v10_0_set_rlc_funcs(adev);
+ /* init rlcg reg access ctrl */
+ gfx_v10_0_init_rlcg_reg_access_ctrl(adev);
+
return 0;
}
@@ -7843,7 +7846,9 @@ static void gfx_v10_0_set_safe_mode(struct amdgpu_device *adev)
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
WREG32_SOC15(GC, 0, mmRLC_SAFE_MODE_Sienna_Cichlid, data);
/* wait for RLC_SAFE_MODE */
@@ -7879,7 +7884,9 @@ static void gfx_v10_0_unset_safe_mode(struct amdgpu_device *adev)
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
WREG32_SOC15(GC, 0, mmRLC_SAFE_MODE_Sienna_Cichlid, data);
break;
default:
@@ -8333,6 +8340,8 @@ static void gfx_v10_cntl_power_gating(struct amdgpu_device *adev, bool enable)
switch (adev->ip_versions[GC_HWIP][0]) {
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 6):
+ case IP_VERSION(10, 3, 7):
data = 0x4E20 & RLC_PG_DELAY_3__CGCG_ACTIVE_BEFORE_CGPG_MASK_Vangogh;
WREG32_SOC15(GC, 0, mmRLC_PG_DELAY_3, data);
break;
@@ -8377,8 +8386,6 @@ static const struct amdgpu_rlc_funcs gfx_v10_0_rlc_funcs_sriov = {
.reset = gfx_v10_0_rlc_reset,
.start = gfx_v10_0_rlc_start,
.update_spm_vmid = gfx_v10_0_update_spm_vmid,
- .sriov_wreg = gfx_v10_sriov_wreg,
- .sriov_rreg = gfx_v10_sriov_rreg,
.is_rlcg_access_range = gfx_v10_0_is_rlcg_access_range,
};
@@ -8403,6 +8410,8 @@ static int gfx_v10_0_set_powergating_state(void *handle,
break;
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 6):
+ case IP_VERSION(10, 3, 7):
gfx_v10_cntl_pg(adev, enable);
amdgpu_gfx_off_ctrl(adev, enable);
break;
@@ -8429,7 +8438,9 @@ static int gfx_v10_0_set_clockgating_state(void *handle,
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
gfx_v10_0_update_gfx_clock_gating(adev,
state == AMD_CG_STATE_GATE);
break;
@@ -8439,7 +8450,7 @@ static int gfx_v10_0_set_clockgating_state(void *handle,
return 0;
}
-static void gfx_v10_0_get_clockgating_state(void *handle, u32 *flags)
+static void gfx_v10_0_get_clockgating_state(void *handle, u64 *flags)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
int data;
@@ -9366,6 +9377,7 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_gfx = {
.align_mask = 0xff,
.nop = PACKET3(PACKET3_NOP, 0x3FFF),
.support_64bit_ptrs = true,
+ .secure_submission_supported = true,
.vmhub = AMDGPU_GFXHUB_0,
.get_rptr = gfx_v10_0_ring_get_rptr_gfx,
.get_wptr = gfx_v10_0_ring_get_wptr_gfx,
@@ -9537,11 +9549,14 @@ static void gfx_v10_0_set_rlc_funcs(struct amdgpu_device *adev)
case IP_VERSION(10, 1, 10):
case IP_VERSION(10, 1, 1):
case IP_VERSION(10, 1, 3):
+ case IP_VERSION(10, 1, 4):
case IP_VERSION(10, 3, 2):
case IP_VERSION(10, 3, 1):
case IP_VERSION(10, 3, 4):
case IP_VERSION(10, 3, 5):
+ case IP_VERSION(10, 3, 6):
case IP_VERSION(10, 3, 3):
+ case IP_VERSION(10, 3, 7):
adev->gfx.rlc.funcs = &gfx_v10_0_rlc_funcs;
break;
case IP_VERSION(10, 1, 2):
@@ -9634,7 +9649,9 @@ static int gfx_v10_0_get_cu_info(struct amdgpu_device *adev,
for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
bitmap = i * adev->gfx.config.max_sh_per_se + j;
if (((adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 0)) ||
- (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 3))) &&
+ (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 3)) ||
+ (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 6)) ||
+ (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 7))) &&
((gfx_v10_3_get_disabled_sa(adev) >> bitmap) & 1))
continue;
mask = 1;
generated by cgit v1.2.3-70-g09d2 (git 2.51.0) at 2025-09-27 12:36:40 +0000