summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/amd
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/amd')
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu.h11
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_device.c17
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c9
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c6
-rw-r--r--drivers/gpu/drm/amd/amdgpu/cik.c7
-rw-r--r--drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c2
-rw-r--r--drivers/gpu/drm/amd/amdgpu/vi.c15
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_process.c70
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_topology.c2
-rw-r--r--drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr_ppt.h1
-rw-r--r--drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_hwmgr.c28
-rw-r--r--drivers/gpu/drm/amd/powerplay/hwmgr/tonga_pptable.h16
-rw-r--r--drivers/gpu/drm/amd/powerplay/hwmgr/tonga_processpptables.c91
13 files changed, 194 insertions, 81 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 01c36b8d6222..e055d5be1c3c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -799,7 +799,6 @@ struct amdgpu_ring {
unsigned cond_exe_offs;
u64 cond_exe_gpu_addr;
volatile u32 *cond_exe_cpu_addr;
- int vmid;
};
/*
@@ -937,8 +936,7 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring,
unsigned vm_id, uint64_t pd_addr,
uint32_t gds_base, uint32_t gds_size,
uint32_t gws_base, uint32_t gws_size,
- uint32_t oa_base, uint32_t oa_size,
- bool vmid_switch);
+ uint32_t oa_base, uint32_t oa_size);
void amdgpu_vm_reset_id(struct amdgpu_device *adev, unsigned vm_id);
uint64_t amdgpu_vm_map_gart(const dma_addr_t *pages_addr, uint64_t addr);
int amdgpu_vm_update_page_directory(struct amdgpu_device *adev,
@@ -1822,6 +1820,8 @@ struct amdgpu_asic_funcs {
/* MM block clocks */
int (*set_uvd_clocks)(struct amdgpu_device *adev, u32 vclk, u32 dclk);
int (*set_vce_clocks)(struct amdgpu_device *adev, u32 evclk, u32 ecclk);
+ /* query virtual capabilities */
+ u32 (*get_virtual_caps)(struct amdgpu_device *adev);
};
/*
@@ -1916,8 +1916,12 @@ void amdgpu_cgs_destroy_device(struct cgs_device *cgs_device);
/* GPU virtualization */
+#define AMDGPU_VIRT_CAPS_SRIOV_EN (1 << 0)
+#define AMDGPU_VIRT_CAPS_IS_VF (1 << 1)
struct amdgpu_virtualization {
bool supports_sr_iov;
+ bool is_virtual;
+ u32 caps;
};
/*
@@ -2206,6 +2210,7 @@ amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
#define amdgpu_asic_get_xclk(adev) (adev)->asic_funcs->get_xclk((adev))
#define amdgpu_asic_set_uvd_clocks(adev, v, d) (adev)->asic_funcs->set_uvd_clocks((adev), (v), (d))
#define amdgpu_asic_set_vce_clocks(adev, ev, ec) (adev)->asic_funcs->set_vce_clocks((adev), (ev), (ec))
+#define amdgpu_asic_get_virtual_caps(adev) ((adev)->asic_funcs->get_virtual_caps((adev)))
#define amdgpu_asic_get_gpu_clock_counter(adev) (adev)->asic_funcs->get_gpu_clock_counter((adev))
#define amdgpu_asic_read_disabled_bios(adev) (adev)->asic_funcs->read_disabled_bios((adev))
#define amdgpu_asic_read_bios_from_rom(adev, b, l) (adev)->asic_funcs->read_bios_from_rom((adev), (b), (l))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 964f31404f17..66482b429458 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1385,6 +1385,15 @@ static int amdgpu_resume(struct amdgpu_device *adev)
return 0;
}
+static bool amdgpu_device_is_virtual(void)
+{
+#ifdef CONFIG_X86
+ return boot_cpu_has(X86_FEATURE_HYPERVISOR);
+#else
+ return false;
+#endif
+}
+
/**
* amdgpu_device_init - initialize the driver
*
@@ -1519,8 +1528,14 @@ int amdgpu_device_init(struct amdgpu_device *adev,
adev->virtualization.supports_sr_iov =
amdgpu_atombios_has_gpu_virtualization_table(adev);
+ /* Check if we are executing in a virtualized environment */
+ adev->virtualization.is_virtual = amdgpu_device_is_virtual();
+ adev->virtualization.caps = amdgpu_asic_get_virtual_caps(adev);
+
/* Post card if necessary */
- if (!amdgpu_card_posted(adev)) {
+ if (!amdgpu_card_posted(adev) ||
+ (adev->virtualization.is_virtual &&
+ !adev->virtualization.caps & AMDGPU_VIRT_CAPS_SRIOV_EN)) {
if (!adev->bios) {
dev_err(adev->dev, "Card not posted and no BIOS - ignoring\n");
return -EINVAL;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index 7a0b1e50f293..34e35423b78e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -122,7 +122,6 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
bool skip_preamble, need_ctx_switch;
unsigned patch_offset = ~0;
struct amdgpu_vm *vm;
- int vmid = 0, old_vmid = ring->vmid;
struct fence *hwf;
uint64_t ctx;
@@ -136,11 +135,9 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
if (job) {
vm = job->vm;
ctx = job->ctx;
- vmid = job->vm_id;
} else {
vm = NULL;
ctx = 0;
- vmid = 0;
}
if (!ring->ready) {
@@ -166,8 +163,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
r = amdgpu_vm_flush(ring, job->vm_id, job->vm_pd_addr,
job->gds_base, job->gds_size,
job->gws_base, job->gws_size,
- job->oa_base, job->oa_size,
- (ring->current_ctx == ctx) && (old_vmid != vmid));
+ job->oa_base, job->oa_size);
if (r) {
amdgpu_ring_undo(ring);
return r;
@@ -184,6 +180,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
need_ctx_switch = ring->current_ctx != ctx;
for (i = 0; i < num_ibs; ++i) {
ib = &ibs[i];
+
/* drop preamble IBs if we don't have a context switch */
if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && skip_preamble)
continue;
@@ -191,7 +188,6 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
amdgpu_ring_emit_ib(ring, ib, job ? job->vm_id : 0,
need_ctx_switch);
need_ctx_switch = false;
- ring->vmid = vmid;
}
if (ring->funcs->emit_hdp_invalidate)
@@ -202,7 +198,6 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
dev_err(adev->dev, "failed to emit fence (%d)\n", r);
if (job && job->vm_id)
amdgpu_vm_reset_id(adev, job->vm_id);
- ring->vmid = old_vmid;
amdgpu_ring_undo(ring);
return r;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 62a4c127620f..9f36ed30ba11 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -298,8 +298,7 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring,
unsigned vm_id, uint64_t pd_addr,
uint32_t gds_base, uint32_t gds_size,
uint32_t gws_base, uint32_t gws_size,
- uint32_t oa_base, uint32_t oa_size,
- bool vmid_switch)
+ uint32_t oa_base, uint32_t oa_size)
{
struct amdgpu_device *adev = ring->adev;
struct amdgpu_vm_id *id = &adev->vm_manager.ids[vm_id];
@@ -313,7 +312,8 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring,
int r;
if (ring->funcs->emit_pipeline_sync && (
- pd_addr != AMDGPU_VM_NO_FLUSH || gds_switch_needed || vmid_switch))
+ pd_addr != AMDGPU_VM_NO_FLUSH || gds_switch_needed ||
+ ring->type == AMDGPU_RING_TYPE_COMPUTE))
amdgpu_ring_emit_pipeline_sync(ring);
if (ring->funcs->emit_vm_flush &&
diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c
index 07bc795a4ca9..910431808542 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik.c
@@ -962,6 +962,12 @@ static bool cik_read_bios_from_rom(struct amdgpu_device *adev,
return true;
}
+static u32 cik_get_virtual_caps(struct amdgpu_device *adev)
+{
+ /* CIK does not support SR-IOV */
+ return 0;
+}
+
static const struct amdgpu_allowed_register_entry cik_allowed_read_registers[] = {
{mmGRBM_STATUS, false},
{mmGB_ADDR_CONFIG, false},
@@ -2007,6 +2013,7 @@ static const struct amdgpu_asic_funcs cik_asic_funcs =
.get_xclk = &cik_get_xclk,
.set_uvd_clocks = &cik_set_uvd_clocks,
.set_vce_clocks = &cik_set_vce_clocks,
+ .get_virtual_caps = &cik_get_virtual_caps,
/* these should be moved to their own ip modules */
.get_gpu_clock_counter = &gfx_v7_0_get_gpu_clock_counter,
.wait_for_mc_idle = &gmc_v7_0_mc_wait_for_idle,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 8c6ad1e72f02..fc8ff4d3ccf8 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -4833,7 +4833,7 @@ static int gfx_v7_0_eop_irq(struct amdgpu_device *adev,
case 2:
for (i = 0; i < adev->gfx.num_compute_rings; i++) {
ring = &adev->gfx.compute_ring[i];
- if ((ring->me == me_id) & (ring->pipe == pipe_id))
+ if ((ring->me == me_id) && (ring->pipe == pipe_id))
amdgpu_fence_process(ring);
}
break;
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 2c88d0b66cf3..a65c96029476 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -421,6 +421,20 @@ static bool vi_read_bios_from_rom(struct amdgpu_device *adev,
return true;
}
+static u32 vi_get_virtual_caps(struct amdgpu_device *adev)
+{
+ u32 caps = 0;
+ u32 reg = RREG32(mmBIF_IOV_FUNC_IDENTIFIER);
+
+ if (REG_GET_FIELD(reg, BIF_IOV_FUNC_IDENTIFIER, IOV_ENABLE))
+ caps |= AMDGPU_VIRT_CAPS_SRIOV_EN;
+
+ if (REG_GET_FIELD(reg, BIF_IOV_FUNC_IDENTIFIER, FUNC_IDENTIFIER))
+ caps |= AMDGPU_VIRT_CAPS_IS_VF;
+
+ return caps;
+}
+
static const struct amdgpu_allowed_register_entry tonga_allowed_read_registers[] = {
{mmGB_MACROTILE_MODE7, true},
};
@@ -1118,6 +1132,7 @@ static const struct amdgpu_asic_funcs vi_asic_funcs =
.get_xclk = &vi_get_xclk,
.set_uvd_clocks = &vi_set_uvd_clocks,
.set_vce_clocks = &vi_set_vce_clocks,
+ .get_virtual_caps = &vi_get_virtual_caps,
/* these should be moved to their own ip modules */
.get_gpu_clock_counter = &gfx_v8_0_get_gpu_clock_counter,
.wait_for_mc_idle = &gmc_v8_0_mc_wait_for_idle,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index ac005796b71c..7708d90b9da9 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -242,13 +242,19 @@ static void kfd_process_notifier_release(struct mmu_notifier *mn,
pqm_uninit(&p->pqm);
/* Iterate over all process device data structure and check
- * if we should reset all wavefronts */
- list_for_each_entry(pdd, &p->per_device_data, per_device_list)
+ * if we should delete debug managers and reset all wavefronts
+ */
+ list_for_each_entry(pdd, &p->per_device_data, per_device_list) {
+ if ((pdd->dev->dbgmgr) &&
+ (pdd->dev->dbgmgr->pasid == p->pasid))
+ kfd_dbgmgr_destroy(pdd->dev->dbgmgr);
+
if (pdd->reset_wavefronts) {
pr_warn("amdkfd: Resetting all wave fronts\n");
dbgdev_wave_reset_wavefronts(pdd->dev, p);
pdd->reset_wavefronts = false;
}
+ }
mutex_unlock(&p->mutex);
@@ -404,42 +410,52 @@ void kfd_unbind_process_from_device(struct kfd_dev *dev, unsigned int pasid)
idx = srcu_read_lock(&kfd_processes_srcu);
+ /*
+ * Look for the process that matches the pasid. If there is no such
+ * process, we either released it in amdkfd's own notifier, or there
+ * is a bug. Unfortunately, there is no way to tell...
+ */
hash_for_each_rcu(kfd_processes_table, i, p, kfd_processes)
- if (p->pasid == pasid)
- break;
+ if (p->pasid == pasid) {
- srcu_read_unlock(&kfd_processes_srcu, idx);
+ srcu_read_unlock(&kfd_processes_srcu, idx);
- BUG_ON(p->pasid != pasid);
+ pr_debug("Unbinding process %d from IOMMU\n", pasid);
- mutex_lock(&p->mutex);
+ mutex_lock(&p->mutex);
- if ((dev->dbgmgr) && (dev->dbgmgr->pasid == p->pasid))
- kfd_dbgmgr_destroy(dev->dbgmgr);
+ if ((dev->dbgmgr) && (dev->dbgmgr->pasid == p->pasid))
+ kfd_dbgmgr_destroy(dev->dbgmgr);
- pqm_uninit(&p->pqm);
+ pqm_uninit(&p->pqm);
- pdd = kfd_get_process_device_data(dev, p);
+ pdd = kfd_get_process_device_data(dev, p);
- if (!pdd) {
- mutex_unlock(&p->mutex);
- return;
- }
+ if (!pdd) {
+ mutex_unlock(&p->mutex);
+ return;
+ }
- if (pdd->reset_wavefronts) {
- dbgdev_wave_reset_wavefronts(pdd->dev, p);
- pdd->reset_wavefronts = false;
- }
+ if (pdd->reset_wavefronts) {
+ dbgdev_wave_reset_wavefronts(pdd->dev, p);
+ pdd->reset_wavefronts = false;
+ }
- /*
- * Just mark pdd as unbound, because we still need it to call
- * amd_iommu_unbind_pasid() in when the process exits.
- * We don't call amd_iommu_unbind_pasid() here
- * because the IOMMU called us.
- */
- pdd->bound = false;
+ /*
+ * Just mark pdd as unbound, because we still need it
+ * to call amd_iommu_unbind_pasid() in when the
+ * process exits.
+ * We don't call amd_iommu_unbind_pasid() here
+ * because the IOMMU called us.
+ */
+ pdd->bound = false;
- mutex_unlock(&p->mutex);
+ mutex_unlock(&p->mutex);
+
+ return;
+ }
+
+ srcu_read_unlock(&kfd_processes_srcu, idx);
}
struct kfd_process_device *kfd_get_first_process_device_data(struct kfd_process *p)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 74909e72a009..884c96f50c3d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -666,7 +666,7 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
dev->node_props.simd_count);
if (dev->mem_bank_count < dev->node_props.mem_banks_count) {
- pr_warn("kfd: mem_banks_count truncated from %d to %d\n",
+ pr_info_once("kfd: mem_banks_count truncated from %d to %d\n",
dev->node_props.mem_banks_count,
dev->mem_bank_count);
sysfs_show_32bit_prop(buffer, "mem_banks_count",
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr_ppt.h b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr_ppt.h
index 347fef127ce9..2930a3355948 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr_ppt.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr_ppt.h
@@ -39,6 +39,7 @@ struct phm_ppt_v1_clock_voltage_dependency_record {
uint8_t phases;
uint8_t cks_enable;
uint8_t cks_voffset;
+ uint32_t sclk_offset;
};
typedef struct phm_ppt_v1_clock_voltage_dependency_record phm_ppt_v1_clock_voltage_dependency_record;
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_hwmgr.c
index aa6be033f21b..1400bc420881 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_hwmgr.c
@@ -999,7 +999,7 @@ static int polaris10_get_dependency_volt_by_clk(struct pp_hwmgr *hwmgr,
vddci = phm_find_closest_vddci(&(data->vddci_voltage_table),
(dep_table->entries[i].vddc -
(uint16_t)data->vddc_vddci_delta));
- *voltage |= (vddci * VOLTAGE_SCALE) << VDDCI_SHIFT;
+ *voltage |= (vddci * VOLTAGE_SCALE) << VDDCI_SHIFT;
}
if (POLARIS10_VOLTAGE_CONTROL_NONE == data->mvdd_control)
@@ -3520,10 +3520,11 @@ static int polaris10_get_pp_table_entry_callback_func(struct pp_hwmgr *hwmgr,
ATOM_Tonga_State *state_entry = (ATOM_Tonga_State *)state;
ATOM_Tonga_POWERPLAYTABLE *powerplay_table =
(ATOM_Tonga_POWERPLAYTABLE *)pp_table;
- ATOM_Tonga_SCLK_Dependency_Table *sclk_dep_table =
- (ATOM_Tonga_SCLK_Dependency_Table *)
+ PPTable_Generic_SubTable_Header *sclk_dep_table =
+ (PPTable_Generic_SubTable_Header *)
(((unsigned long)powerplay_table) +
le16_to_cpu(powerplay_table->usSclkDependencyTableOffset));
+
ATOM_Tonga_MCLK_Dependency_Table *mclk_dep_table =
(ATOM_Tonga_MCLK_Dependency_Table *)
(((unsigned long)powerplay_table) +
@@ -3575,7 +3576,11 @@ static int polaris10_get_pp_table_entry_callback_func(struct pp_hwmgr *hwmgr,
/* Performance levels are arranged from low to high. */
performance_level->memory_clock = mclk_dep_table->entries
[state_entry->ucMemoryClockIndexLow].ulMclk;
- performance_level->engine_clock = sclk_dep_table->entries
+ if (sclk_dep_table->ucRevId == 0)
+ performance_level->engine_clock = ((ATOM_Tonga_SCLK_Dependency_Table *)sclk_dep_table)->entries
+ [state_entry->ucEngineClockIndexLow].ulSclk;
+ else if (sclk_dep_table->ucRevId == 1)
+ performance_level->engine_clock = ((ATOM_Polaris_SCLK_Dependency_Table *)sclk_dep_table)->entries
[state_entry->ucEngineClockIndexLow].ulSclk;
performance_level->pcie_gen = get_pcie_gen_support(data->pcie_gen_cap,
state_entry->ucPCIEGenLow);
@@ -3586,8 +3591,14 @@ static int polaris10_get_pp_table_entry_callback_func(struct pp_hwmgr *hwmgr,
[polaris10_power_state->performance_level_count++]);
performance_level->memory_clock = mclk_dep_table->entries
[state_entry->ucMemoryClockIndexHigh].ulMclk;
- performance_level->engine_clock = sclk_dep_table->entries
+
+ if (sclk_dep_table->ucRevId == 0)
+ performance_level->engine_clock = ((ATOM_Tonga_SCLK_Dependency_Table *)sclk_dep_table)->entries
[state_entry->ucEngineClockIndexHigh].ulSclk;
+ else if (sclk_dep_table->ucRevId == 1)
+ performance_level->engine_clock = ((ATOM_Polaris_SCLK_Dependency_Table *)sclk_dep_table)->entries
+ [state_entry->ucEngineClockIndexHigh].ulSclk;
+
performance_level->pcie_gen = get_pcie_gen_support(data->pcie_gen_cap,
state_entry->ucPCIEGenHigh);
performance_level->pcie_lane = get_pcie_lane_support(data->pcie_lane_cap,
@@ -3645,7 +3656,6 @@ static int polaris10_get_pp_table_entry(struct pp_hwmgr *hwmgr,
switch (state->classification.ui_label) {
case PP_StateUILabel_Performance:
data->use_pcie_performance_levels = true;
-
for (i = 0; i < ps->performance_level_count; i++) {
if (data->pcie_gen_performance.max <
ps->performance_levels[i].pcie_gen)
@@ -3661,7 +3671,6 @@ static int polaris10_get_pp_table_entry(struct pp_hwmgr *hwmgr,
ps->performance_levels[i].pcie_lane)
data->pcie_lane_performance.max =
ps->performance_levels[i].pcie_lane;
-
if (data->pcie_lane_performance.min >
ps->performance_levels[i].pcie_lane)
data->pcie_lane_performance.min =
@@ -4187,12 +4196,9 @@ int polaris10_update_samu_dpm(struct pp_hwmgr *hwmgr, bool bgate)
{
struct polaris10_hwmgr *data = (struct polaris10_hwmgr *)(hwmgr->backend);
uint32_t mm_boot_level_offset, mm_boot_level_value;
- struct phm_ppt_v1_information *table_info =
- (struct phm_ppt_v1_information *)(hwmgr->pptable);
if (!bgate) {
- data->smc_state_table.SamuBootLevel =
- (uint8_t) (table_info->mm_dep_table->count - 1);
+ data->smc_state_table.SamuBootLevel = 0;
mm_boot_level_offset = data->dpm_table_start +
offsetof(SMU74_Discrete_DpmTable, SamuBootLevel);
mm_boot_level_offset /= 4;
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_pptable.h b/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_pptable.h
index 1b44f4e9b8f5..f127198aafc4 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_pptable.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_pptable.h
@@ -197,6 +197,22 @@ typedef struct _ATOM_Tonga_SCLK_Dependency_Table {
ATOM_Tonga_SCLK_Dependency_Record entries[1]; /* Dynamically allocate entries. */
} ATOM_Tonga_SCLK_Dependency_Table;
+typedef struct _ATOM_Polaris_SCLK_Dependency_Record {
+ UCHAR ucVddInd; /* Base voltage */
+ USHORT usVddcOffset; /* Offset relative to base voltage */
+ ULONG ulSclk;
+ USHORT usEdcCurrent;
+ UCHAR ucReliabilityTemperature;
+ UCHAR ucCKSVOffsetandDisable; /* Bits 0~6: Voltage offset for CKS, Bit 7: Disable/enable for the SCLK level. */
+ ULONG ulSclkOffset;
+} ATOM_Polaris_SCLK_Dependency_Record;
+
+typedef struct _ATOM_Polaris_SCLK_Dependency_Table {
+ UCHAR ucRevId;
+ UCHAR ucNumEntries; /* Number of entries. */
+ ATOM_Polaris_SCLK_Dependency_Record entries[1]; /* Dynamically allocate entries. */
+} ATOM_Polaris_SCLK_Dependency_Table;
+
typedef struct _ATOM_Tonga_PCIE_Record {
UCHAR ucPCIEGenSpeed;
UCHAR usPCIELaneWidth;
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_processpptables.c b/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_processpptables.c
index 296ec7ef6d45..671fdb4d615a 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_processpptables.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_processpptables.c
@@ -408,41 +408,78 @@ static int get_mclk_voltage_dependency_table(
static int get_sclk_voltage_dependency_table(
struct pp_hwmgr *hwmgr,
phm_ppt_v1_clock_voltage_dependency_table **pp_tonga_sclk_dep_table,
- const ATOM_Tonga_SCLK_Dependency_Table * sclk_dep_table
+ const PPTable_Generic_SubTable_Header *sclk_dep_table
)
{
uint32_t table_size, i;
phm_ppt_v1_clock_voltage_dependency_table *sclk_table;
- PP_ASSERT_WITH_CODE((0 != sclk_dep_table->ucNumEntries),
- "Invalid PowerPlay Table!", return -1);
+ if (sclk_dep_table->ucRevId < 1) {
+ const ATOM_Tonga_SCLK_Dependency_Table *tonga_table =
+ (ATOM_Tonga_SCLK_Dependency_Table *)sclk_dep_table;
- table_size = sizeof(uint32_t) + sizeof(phm_ppt_v1_clock_voltage_dependency_record)
- * sclk_dep_table->ucNumEntries;
+ PP_ASSERT_WITH_CODE((0 != tonga_table->ucNumEntries),
+ "Invalid PowerPlay Table!", return -1);
- sclk_table = (phm_ppt_v1_clock_voltage_dependency_table *)
- kzalloc(table_size, GFP_KERNEL);
+ table_size = sizeof(uint32_t) + sizeof(phm_ppt_v1_clock_voltage_dependency_record)
+ * tonga_table->ucNumEntries;
- if (NULL == sclk_table)
- return -ENOMEM;
+ sclk_table = (phm_ppt_v1_clock_voltage_dependency_table *)
+ kzalloc(table_size, GFP_KERNEL);
- memset(sclk_table, 0x00, table_size);
-
- sclk_table->count = (uint32_t)sclk_dep_table->ucNumEntries;
-
- for (i = 0; i < sclk_dep_table->ucNumEntries; i++) {
- sclk_table->entries[i].vddInd =
- sclk_dep_table->entries[i].ucVddInd;
- sclk_table->entries[i].vdd_offset =
- sclk_dep_table->entries[i].usVddcOffset;
- sclk_table->entries[i].clk =
- sclk_dep_table->entries[i].ulSclk;
- sclk_table->entries[i].cks_enable =
- (((sclk_dep_table->entries[i].ucCKSVOffsetandDisable & 0x80) >> 7) == 0) ? 1 : 0;
- sclk_table->entries[i].cks_voffset =
- (sclk_dep_table->entries[i].ucCKSVOffsetandDisable & 0x7F);
- }
+ if (NULL == sclk_table)
+ return -ENOMEM;
+
+ memset(sclk_table, 0x00, table_size);
+
+ sclk_table->count = (uint32_t)tonga_table->ucNumEntries;
+
+ for (i = 0; i < tonga_table->ucNumEntries; i++) {
+ sclk_table->entries[i].vddInd =
+ tonga_table->entries[i].ucVddInd;
+ sclk_table->entries[i].vdd_offset =
+ tonga_table->entries[i].usVddcOffset;
+ sclk_table->entries[i].clk =
+ tonga_table->entries[i].ulSclk;
+ sclk_table->entries[i].cks_enable =
+ (((tonga_table->entries[i].ucCKSVOffsetandDisable & 0x80) >> 7) == 0) ? 1 : 0;
+ sclk_table->entries[i].cks_voffset =
+ (tonga_table->entries[i].ucCKSVOffsetandDisable & 0x7F);
+ }
+ } else {
+ const ATOM_Polaris_SCLK_Dependency_Table *polaris_table =
+ (ATOM_Polaris_SCLK_Dependency_Table *)sclk_dep_table;
+ PP_ASSERT_WITH_CODE((0 != polaris_table->ucNumEntries),
+ "Invalid PowerPlay Table!", return -1);
+
+ table_size = sizeof(uint32_t) + sizeof(phm_ppt_v1_clock_voltage_dependency_record)
+ * polaris_table->ucNumEntries;
+
+ sclk_table = (phm_ppt_v1_clock_voltage_dependency_table *)
+ kzalloc(table_size, GFP_KERNEL);
+
+ if (NULL == sclk_table)
+ return -ENOMEM;
+
+ memset(sclk_table, 0x00, table_size);
+
+ sclk_table->count = (uint32_t)polaris_table->ucNumEntries;
+
+ for (i = 0; i < polaris_table->ucNumEntries; i++) {
+ sclk_table->entries[i].vddInd =
+ polaris_table->entries[i].ucVddInd;
+ sclk_table->entries[i].vdd_offset =
+ polaris_table->entries[i].usVddcOffset;
+ sclk_table->entries[i].clk =
+ polaris_table->entries[i].ulSclk;
+ sclk_table->entries[i].cks_enable =
+ (((polaris_table->entries[i].ucCKSVOffsetandDisable & 0x80) >> 7) == 0) ? 1 : 0;
+ sclk_table->entries[i].cks_voffset =
+ (polaris_table->entries[i].ucCKSVOffsetandDisable & 0x7F);
+ sclk_table->entries[i].sclk_offset = polaris_table->entries[i].ulSclkOffset;
+ }
+ }
*pp_tonga_sclk_dep_table = sclk_table;
return 0;
@@ -708,8 +745,8 @@ static int init_clock_voltage_dependency(
const ATOM_Tonga_MCLK_Dependency_Table *mclk_dep_table =
(const ATOM_Tonga_MCLK_Dependency_Table *)(((unsigned long) powerplay_table) +
le16_to_cpu(powerplay_table->usMclkDependencyTableOffset));
- const ATOM_Tonga_SCLK_Dependency_Table *sclk_dep_table =
- (const ATOM_Tonga_SCLK_Dependency_Table *)(((unsigned long) powerplay_table) +
+ const PPTable_Generic_SubTable_Header *sclk_dep_table =
+ (const PPTable_Generic_SubTable_Header *)(((unsigned long) powerplay_table) +
le16_to_cpu(powerplay_table->usSclkDependencyTableOffset));
const ATOM_Tonga_Hard_Limit_Table *pHardLimits =
(const ATOM_Tonga_Hard_Limit_Table *)(((unsigned long) powerplay_table) +