aboutsummaryrefslogtreecommitdiffstats
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.c154
1 files changed, 121 insertions, 33 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 957811b73672..ca5f0e7ea1ac 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -93,7 +93,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_1[] =
{
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_4, 0xffffffff, 0x00400014),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_CPF_CLK_CTRL, 0xfcff8fff, 0xf8000100),
- SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CLK_CTRL, 0xc0000000, 0xc0000100),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CLK_CTRL, 0xcd000000, 0x0d000100),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SQ_CLK_CTRL, 0x60000ff0, 0x60000100),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SQG_CLK_CTRL, 0x40000000, 0x40000100),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_VGT_CLK_CTRL, 0xffff8fff, 0xffff8100),
@@ -127,7 +127,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_1[] =
SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfff7ffff, 0x01030000),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmTCP_CNTL, 0x60000010, 0x479c0010),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CGTT_CLK_CTRL, 0xfeff0fff, 0x40000100),
- SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0x00800000, 0x00800000)
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0x00c00000, 0x00c00000)
};
static const struct soc15_reg_golden golden_settings_gc_10_0_nv10[] =
@@ -140,7 +140,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_1_1[] =
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_4, 0xffffffff, 0x003c0014),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_GS_NGG_CLK_CTRL, 0xffff8fff, 0xffff8100),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_IA_CLK_CTRL, 0xffff0fff, 0xffff0100),
- SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CLK_CTRL, 0xc0000000, 0xc0000100),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CLK_CTRL, 0xcd000000, 0x0d000100),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SQ_CLK_CTRL, 0xf8ff0fff, 0x60000100),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SQG_CLK_CTRL, 0x40000ff0, 0x40000100),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_VGT_CLK_CTRL, 0xffff8fff, 0xffff8100),
@@ -171,7 +171,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_1_1[] =
SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_LDS_CLK_CTRL, 0xffffffff, 0xffffffff),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfff7ffff, 0x01030000),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmTCP_CNTL, 0x60000010, 0x479c0010),
- SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0x00800000, 0x00800000),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0x00c00000, 0x00c00000),
};
static const struct soc15_reg_golden golden_settings_gc_10_1_2[] =
@@ -179,7 +179,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_1_2[] =
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_4, 0x003e001f, 0x003c0014),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_GS_NGG_CLK_CTRL, 0xffff8fff, 0xffff8100),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_IA_CLK_CTRL, 0xffff0fff, 0xffff0100),
- SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CLK_CTRL, 0xff7f0fff, 0xc0000100),
+ SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CLK_CTRL, 0xff7f0fff, 0x0d000100),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SQ_CLK_CTRL, 0xffffcfff, 0x60000100),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SQG_CLK_CTRL, 0xffff0fff, 0x40000100),
SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_VGT_CLK_CTRL, 0xffff8fff, 0xffff8100),
@@ -564,6 +564,32 @@ static void gfx_v10_0_free_microcode(struct amdgpu_device *adev)
kfree(adev->gfx.rlc.register_list_format);
}
+static void gfx_v10_0_check_fw_write_wait(struct amdgpu_device *adev)
+{
+ adev->gfx.cp_fw_write_wait = false;
+
+ switch (adev->asic_type) {
+ case CHIP_NAVI10:
+ case CHIP_NAVI12:
+ case CHIP_NAVI14:
+ if ((adev->gfx.me_fw_version >= 0x00000046) &&
+ (adev->gfx.me_feature_version >= 27) &&
+ (adev->gfx.pfp_fw_version >= 0x00000068) &&
+ (adev->gfx.pfp_feature_version >= 27) &&
+ (adev->gfx.mec_fw_version >= 0x0000005b) &&
+ (adev->gfx.mec_feature_version >= 27))
+ adev->gfx.cp_fw_write_wait = true;
+ break;
+ default:
+ break;
+ }
+
+ if (adev->gfx.cp_fw_write_wait == false)
+ DRM_WARN_ONCE("Warning: check cp_fw_version and update it to realize \
+ GRBM requires 1-cycle delay in cp firmware\n");
+}
+
+
static void gfx_v10_0_init_rlc_ext_microcode(struct amdgpu_device *adev)
{
const struct rlc_firmware_header_v2_1 *rlc_hdr;
@@ -832,6 +858,7 @@ static int gfx_v10_0_init_microcode(struct amdgpu_device *adev)
}
}
+ gfx_v10_0_check_fw_write_wait(adev);
out:
if (err) {
dev_err(adev->dev,
@@ -1442,7 +1469,7 @@ static int gfx_v10_0_sw_fini(void *handle)
amdgpu_ring_fini(&adev->gfx.compute_ring[i]);
amdgpu_gfx_mqd_sw_fini(adev);
- amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq.ring, &adev->gfx.kiq.irq);
+ amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq.ring);
amdgpu_gfx_kiq_fini(adev);
gfx_v10_0_pfp_fini(adev);
@@ -1758,27 +1785,52 @@ static void gfx_v10_0_enable_gui_idle_interrupt(struct amdgpu_device *adev,
WREG32_SOC15(GC, 0, mmCP_INT_CNTL_RING0, tmp);
}
-static void gfx_v10_0_init_csb(struct amdgpu_device *adev)
+static int gfx_v10_0_init_csb(struct amdgpu_device *adev)
{
+ int r;
+
+ if (adev->in_gpu_reset) {
+ r = amdgpu_bo_reserve(adev->gfx.rlc.clear_state_obj, false);
+ if (r)
+ return r;
+
+ r = amdgpu_bo_kmap(adev->gfx.rlc.clear_state_obj,
+ (void **)&adev->gfx.rlc.cs_ptr);
+ if (!r) {
+ adev->gfx.rlc.funcs->get_csb_buffer(adev,
+ adev->gfx.rlc.cs_ptr);
+ amdgpu_bo_kunmap(adev->gfx.rlc.clear_state_obj);
+ }
+
+ amdgpu_bo_unreserve(adev->gfx.rlc.clear_state_obj);
+ if (r)
+ return r;
+ }
+
/* csib */
WREG32_SOC15(GC, 0, mmRLC_CSIB_ADDR_HI,
adev->gfx.rlc.clear_state_gpu_addr >> 32);
WREG32_SOC15(GC, 0, mmRLC_CSIB_ADDR_LO,
adev->gfx.rlc.clear_state_gpu_addr & 0xfffffffc);
WREG32_SOC15(GC, 0, mmRLC_CSIB_LENGTH, adev->gfx.rlc.clear_state_size);
+
+ return 0;
}
-static void gfx_v10_0_init_pg(struct amdgpu_device *adev)
+static int gfx_v10_0_init_pg(struct amdgpu_device *adev)
{
int i;
+ int r;
- gfx_v10_0_init_csb(adev);
+ r = gfx_v10_0_init_csb(adev);
+ if (r)
+ return r;
for (i = 0; i < adev->num_vmhubs; i++)
amdgpu_gmc_flush_gpu_tlb(adev, 0, i, 0);
/* TODO: init power gating */
- return;
+ return 0;
}
void gfx_v10_0_rlc_stop(struct amdgpu_device *adev)
@@ -1880,7 +1932,10 @@ static int gfx_v10_0_rlc_resume(struct amdgpu_device *adev)
r = gfx_v10_0_wait_for_rlc_autoload_complete(adev);
if (r)
return r;
- gfx_v10_0_init_pg(adev);
+
+ r = gfx_v10_0_init_pg(adev);
+ if (r)
+ return r;
/* enable RLC SRM */
gfx_v10_0_rlc_enable_srm(adev);
@@ -1906,7 +1961,10 @@ static int gfx_v10_0_rlc_resume(struct amdgpu_device *adev)
return r;
}
- gfx_v10_0_init_pg(adev);
+ r = gfx_v10_0_init_pg(adev);
+ if (r)
+ return r;
+
adev->gfx.rlc.funcs->start(adev);
if (adev->firmware.load_type == AMDGPU_FW_LOAD_RLC_BACKDOOR_AUTO) {
@@ -2373,7 +2431,7 @@ static int gfx_v10_0_wait_for_rlc_autoload_complete(struct amdgpu_device *adev)
return 0;
}
-static void gfx_v10_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable)
+static int gfx_v10_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable)
{
int i;
u32 tmp = RREG32_SOC15(GC, 0, mmCP_ME_CNTL);
@@ -2386,7 +2444,17 @@ static void gfx_v10_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable)
adev->gfx.gfx_ring[i].sched.ready = false;
}
WREG32_SOC15(GC, 0, mmCP_ME_CNTL, tmp);
- udelay(50);
+
+ for (i = 0; i < adev->usec_timeout; i++) {
+ if (RREG32_SOC15(GC, 0, mmCP_STAT) == 0)
+ break;
+ udelay(1);
+ }
+
+ if (i >= adev->usec_timeout)
+ DRM_ERROR("failed to %s cp gfx\n", enable ? "unhalt" : "halt");
+
+ return 0;
}
static int gfx_v10_0_cp_gfx_load_pfp_microcode(struct amdgpu_device *adev)
@@ -2443,7 +2511,7 @@ static int gfx_v10_0_cp_gfx_load_pfp_microcode(struct amdgpu_device *adev)
}
if (amdgpu_emu_mode == 1)
- adev->nbio_funcs->hdp_flush(adev, NULL);
+ adev->nbio.funcs->hdp_flush(adev, NULL);
tmp = RREG32_SOC15(GC, 0, mmCP_PFP_IC_BASE_CNTL);
tmp = REG_SET_FIELD(tmp, CP_PFP_IC_BASE_CNTL, VMID, 0);
@@ -2513,7 +2581,7 @@ static int gfx_v10_0_cp_gfx_load_ce_microcode(struct amdgpu_device *adev)
}
if (amdgpu_emu_mode == 1)
- adev->nbio_funcs->hdp_flush(adev, NULL);
+ adev->nbio.funcs->hdp_flush(adev, NULL);
tmp = RREG32_SOC15(GC, 0, mmCP_CE_IC_BASE_CNTL);
tmp = REG_SET_FIELD(tmp, CP_CE_IC_BASE_CNTL, VMID, 0);
@@ -2582,7 +2650,7 @@ static int gfx_v10_0_cp_gfx_load_me_microcode(struct amdgpu_device *adev)
}
if (amdgpu_emu_mode == 1)
- adev->nbio_funcs->hdp_flush(adev, NULL);
+ adev->nbio.funcs->hdp_flush(adev, NULL);
tmp = RREG32_SOC15(GC, 0, mmCP_ME_IC_BASE_CNTL);
tmp = REG_SET_FIELD(tmp, CP_ME_IC_BASE_CNTL, VMID, 0);
@@ -2903,7 +2971,7 @@ static int gfx_v10_0_cp_compute_load_microcode(struct amdgpu_device *adev)
}
if (amdgpu_emu_mode == 1)
- adev->nbio_funcs->hdp_flush(adev, NULL);
+ adev->nbio.funcs->hdp_flush(adev, NULL);
tmp = RREG32_SOC15(GC, 0, mmCP_CPC_IC_BASE_CNTL);
tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, CACHE_POLICY, 0);
@@ -3087,6 +3155,7 @@ static int gfx_v10_0_gfx_init_queue(struct amdgpu_ring *ring)
{
struct amdgpu_device *adev = ring->adev;
struct v10_gfx_mqd *mqd = ring->mqd_ptr;
+ int mqd_idx = ring - &adev->gfx.gfx_ring[0];
if (!adev->in_gpu_reset && !adev->in_suspend) {
memset((void *)mqd, 0, sizeof(*mqd));
@@ -3098,14 +3167,15 @@ static int gfx_v10_0_gfx_init_queue(struct amdgpu_ring *ring)
#endif
nv_grbm_select(adev, 0, 0, 0, 0);
mutex_unlock(&adev->srbm_mutex);
- if (adev->gfx.me.mqd_backup[AMDGPU_MAX_GFX_RINGS])
- memcpy(adev->gfx.me.mqd_backup[AMDGPU_MAX_GFX_RINGS], mqd, sizeof(*mqd));
+ if (adev->gfx.me.mqd_backup[mqd_idx])
+ memcpy(adev->gfx.me.mqd_backup[mqd_idx], mqd, sizeof(*mqd));
} else if (adev->in_gpu_reset) {
/* reset mqd with the backup copy */
- if (adev->gfx.me.mqd_backup[AMDGPU_MAX_GFX_RINGS])
- memcpy(mqd, adev->gfx.me.mqd_backup[AMDGPU_MAX_GFX_RINGS], sizeof(*mqd));
+ if (adev->gfx.me.mqd_backup[mqd_idx])
+ memcpy(mqd, adev->gfx.me.mqd_backup[mqd_idx], sizeof(*mqd));
/* reset the ring */
ring->wptr = 0;
+ adev->wb.wb[ring->wptr_offs] = 0;
amdgpu_ring_clear_ring(ring);
#ifdef BRING_UP_DEBUG
mutex_lock(&adev->srbm_mutex);
@@ -4357,7 +4427,7 @@ static void gfx_v10_0_ring_emit_hdp_flush(struct amdgpu_ring *ring)
{
struct amdgpu_device *adev = ring->adev;
u32 ref_and_mask, reg_mem_engine;
- const struct nbio_hdp_flush_reg *nbio_hf_reg = adev->nbio_funcs->hdp_flush_reg;
+ const struct nbio_hdp_flush_reg *nbio_hf_reg = adev->nbio.hdp_flush_reg;
if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) {
switch (ring->me) {
@@ -4377,8 +4447,8 @@ static void gfx_v10_0_ring_emit_hdp_flush(struct amdgpu_ring *ring)
}
gfx_v10_0_wait_reg_mem(ring, reg_mem_engine, 0, 1,
- adev->nbio_funcs->get_hdp_flush_req_offset(adev),
- adev->nbio_funcs->get_hdp_flush_done_offset(adev),
+ adev->nbio.funcs->get_hdp_flush_req_offset(adev),
+ adev->nbio.funcs->get_hdp_flush_done_offset(adev),
ref_and_mask, ref_and_mask, 0x20);
}
@@ -4765,6 +4835,24 @@ static void gfx_v10_0_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg,
gfx_v10_0_wait_reg_mem(ring, 0, 0, 0, reg, 0, val, mask, 0x20);
}
+static void gfx_v10_0_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring,
+ uint32_t reg0, uint32_t reg1,
+ uint32_t ref, uint32_t mask)
+{
+ int usepfp = (ring->funcs->type == AMDGPU_RING_TYPE_GFX);
+ struct amdgpu_device *adev = ring->adev;
+ bool fw_version_ok = false;
+
+ fw_version_ok = adev->gfx.cp_fw_write_wait;
+
+ if (fw_version_ok)
+ gfx_v10_0_wait_reg_mem(ring, usepfp, 0, 1, reg0, reg1,
+ ref, mask, 0x20);
+ else
+ amdgpu_ring_emit_reg_write_reg_wait_helper(ring, reg0, reg1,
+ ref, mask);
+}
+
static void
gfx_v10_0_set_gfx_eop_interrupt_state(struct amdgpu_device *adev,
uint32_t me, uint32_t pipe,
@@ -5155,6 +5243,7 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_gfx = {
.emit_tmz = gfx_v10_0_ring_emit_tmz,
.emit_wreg = gfx_v10_0_ring_emit_wreg,
.emit_reg_wait = gfx_v10_0_ring_emit_reg_wait,
+ .emit_reg_write_reg_wait = gfx_v10_0_ring_emit_reg_write_reg_wait,
};
static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_compute = {
@@ -5188,6 +5277,7 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_compute = {
.pad_ib = amdgpu_ring_generic_pad_ib,
.emit_wreg = gfx_v10_0_ring_emit_wreg,
.emit_reg_wait = gfx_v10_0_ring_emit_reg_wait,
+ .emit_reg_write_reg_wait = gfx_v10_0_ring_emit_reg_write_reg_wait,
};
static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_kiq = {
@@ -5218,6 +5308,7 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_kiq = {
.emit_rreg = gfx_v10_0_ring_emit_rreg,
.emit_wreg = gfx_v10_0_ring_emit_wreg,
.emit_reg_wait = gfx_v10_0_ring_emit_reg_wait,
+ .emit_reg_write_reg_wait = gfx_v10_0_ring_emit_reg_write_reg_wait,
};
static void gfx_v10_0_set_ring_funcs(struct amdgpu_device *adev)
@@ -5283,15 +5374,12 @@ static void gfx_v10_0_set_rlc_funcs(struct amdgpu_device *adev)
static void gfx_v10_0_set_gds_init(struct amdgpu_device *adev)
{
- /* init asic gds info */
- switch (adev->asic_type) {
- case CHIP_NAVI10:
- default:
- adev->gds.gds_size = 0x10000;
- adev->gds.gds_compute_max_wave_id = 0x4ff;
- break;
- }
+ unsigned total_cu = adev->gfx.config.max_cu_per_sh *
+ adev->gfx.config.max_sh_per_se *
+ adev->gfx.config.max_shader_engines;
+ adev->gds.gds_size = 0x10000;
+ adev->gds.gds_compute_max_wave_id = total_cu * 32 - 1;
adev->gds.gws_size = 64;
adev->gds.oa_size = 16;
}
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.