diff options
Diffstat (limited to 'drivers/gpu')
84 files changed, 1159 insertions, 332 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index bb1099c549df..053fc2f465df 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1673,6 +1673,7 @@ struct amdgpu_uvd { struct amdgpu_bo *vcpu_bo; void *cpu_addr; uint64_t gpu_addr; + unsigned fw_version; atomic_t handles[AMDGPU_MAX_UVD_HANDLES]; struct drm_file *filp[AMDGPU_MAX_UVD_HANDLES]; struct delayed_work idle_work; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c index 8ac49812a716..5a8fbadbd27b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c @@ -63,10 +63,6 @@ bool amdgpu_has_atpx(void) { return amdgpu_atpx_priv.atpx_detected; } -bool amdgpu_has_atpx_dgpu_power_cntl(void) { - return amdgpu_atpx_priv.atpx.functions.power_cntl; -} - /** * amdgpu_atpx_call - call an ATPX method * @@ -146,6 +142,10 @@ static void amdgpu_atpx_parse_functions(struct amdgpu_atpx_functions *f, u32 mas */ static int amdgpu_atpx_validate(struct amdgpu_atpx *atpx) { + /* make sure required functions are enabled */ + /* dGPU power control is required */ + atpx->functions.power_cntl = true; + if (atpx->functions.px_params) { union acpi_object *info; struct atpx_px_params output; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c index 119cdc2c43e7..7ef2c13921b4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c @@ -194,12 +194,12 @@ int amdgpu_connector_get_monitor_bpc(struct drm_connector *connector) bpc = 8; DRM_DEBUG("%s: HDMI deep color 10 bpc exceeds max tmds clock. Using %d bpc.\n", connector->name, bpc); - } else if (bpc > 8) { - /* max_tmds_clock missing, but hdmi spec mandates it for deep color. */ - DRM_DEBUG("%s: Required max tmds clock for HDMI deep color missing. Using 8 bpc.\n", - connector->name); - bpc = 8; } + } else if (bpc > 8) { + /* max_tmds_clock missing, but hdmi spec mandates it for deep color. */ + DRM_DEBUG("%s: Required max tmds clock for HDMI deep color missing. Using 8 bpc.\n", + connector->name); + bpc = 8; } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 9d88023df836..c961fe093e12 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -61,12 +61,6 @@ static const char *amdgpu_asic_name[] = { "LAST", }; -#if defined(CONFIG_VGA_SWITCHEROO) -bool amdgpu_has_atpx_dgpu_power_cntl(void); -#else -static inline bool amdgpu_has_atpx_dgpu_power_cntl(void) { return false; } -#endif - bool amdgpu_device_is_px(struct drm_device *dev) { struct amdgpu_device *adev = dev->dev_private; @@ -1475,7 +1469,7 @@ int amdgpu_device_init(struct amdgpu_device *adev, if (amdgpu_runtime_pm == 1) runtime = true; - if (amdgpu_device_is_px(ddev) && amdgpu_has_atpx_dgpu_power_cntl()) + if (amdgpu_device_is_px(ddev)) runtime = true; vga_switcheroo_register_client(adev->pdev, &amdgpu_switcheroo_ops, runtime); if (runtime) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c index 7b7f4aba60c0..fe36caf1b7d7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c @@ -150,7 +150,7 @@ u32 amdgpu_dpm_get_vrefresh(struct amdgpu_device *adev) list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { amdgpu_crtc = to_amdgpu_crtc(crtc); if (crtc->enabled && amdgpu_crtc->enabled && amdgpu_crtc->hw_mode.clock) { - vrefresh = amdgpu_crtc->hw_mode.vrefresh; + vrefresh = drm_mode_vrefresh(&amdgpu_crtc->hw_mode); break; } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index e23843f4d877..4488e82f87b0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -303,7 +303,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file fw_info.feature = adev->vce.fb_version; break; case AMDGPU_INFO_FW_UVD: - fw_info.ver = 0; + fw_info.ver = adev->uvd.fw_version; fw_info.feature = 0; break; case AMDGPU_INFO_FW_GMC: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h index 064ebb347074..89df7871653d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h @@ -52,7 +52,7 @@ struct amdgpu_hpd; #define AMDGPU_MAX_HPD_PINS 6 #define AMDGPU_MAX_CRTCS 6 -#define AMDGPU_MAX_AFMT_BLOCKS 7 +#define AMDGPU_MAX_AFMT_BLOCKS 9 enum amdgpu_rmx_type { RMX_OFF, @@ -308,8 +308,8 @@ struct amdgpu_mode_info { struct atom_context *atom_context; struct card_info *atom_card_info; bool mode_config_initialized; - struct amdgpu_crtc *crtcs[6]; - struct amdgpu_afmt *afmt[7]; + struct amdgpu_crtc *crtcs[AMDGPU_MAX_CRTCS]; + struct amdgpu_afmt *afmt[AMDGPU_MAX_AFMT_BLOCKS]; /* DVI-I properties */ struct drm_property *coherent_mode_property; /* DAC enable load detect */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index b8fbbd7699e4..73628c7599e7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -540,6 +540,7 @@ int amdgpu_bo_set_metadata (struct amdgpu_bo *bo, void *metadata, if (!metadata_size) { if (bo->metadata_size) { kfree(bo->metadata); + bo->metadata = NULL; bo->metadata_size = 0; } return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index 53f987aeeacf..3b35ad83867c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -156,6 +156,9 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev) DRM_INFO("Found UVD firmware Version: %hu.%hu Family ID: %hu\n", version_major, version_minor, family_id); + adev->uvd.fw_version = ((version_major << 24) | (version_minor << 16) | + (family_id << 8)); + bo_size = AMDGPU_GPU_PAGE_ALIGN(le32_to_cpu(hdr->ucode_size_bytes) + 8) + AMDGPU_UVD_STACK_SIZE + AMDGPU_UVD_HEAP_SIZE; r = amdgpu_bo_create(adev, bo_size, PAGE_SIZE, true, @@ -273,6 +276,8 @@ int amdgpu_uvd_resume(struct amdgpu_device *adev) memcpy(adev->uvd.cpu_addr, (adev->uvd.fw->data) + offset, (adev->uvd.fw->size) - offset); + cancel_delayed_work_sync(&adev->uvd.idle_work); + size = amdgpu_bo_size(adev->uvd.vcpu_bo); size -= le32_to_cpu(hdr->ucode_size_bytes); ptr = adev->uvd.cpu_addr; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c index a745eeeb5d82..bb0da76051a1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c @@ -220,6 +220,7 @@ int amdgpu_vce_suspend(struct amdgpu_device *adev) if (i == AMDGPU_MAX_VCE_HANDLES) return 0; + cancel_delayed_work_sync(&adev->vce.idle_work); /* TODO: suspending running encoding sessions isn't supported */ return -EINVAL; } diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c b/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c index 1e0bba29e167..1cd6de575305 100644 --- a/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c +++ b/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c @@ -298,6 +298,10 @@ bool amdgpu_atombios_encoder_mode_fixup(struct drm_encoder *encoder, && (mode->crtc_vsync_start < (mode->crtc_vdisplay + 2))) adjusted_mode->crtc_vsync_start = adjusted_mode->crtc_vdisplay + 2; + /* vertical FP must be at least 1 */ + if (mode->crtc_vsync_start == mode->crtc_vdisplay) + adjusted_mode->crtc_vsync_start++; + /* get the native mode for scaling */ if (amdgpu_encoder->active_device & (ATOM_DEVICE_LCD_SUPPORT)) amdgpu_panel_mode_fixup(encoder, adjusted_mode); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index aa491540ba85..b57fffc2d4af 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -3628,7 +3628,7 @@ static void gfx_v7_0_ring_emit_vm_flush(struct amdgpu_ring *ring, unsigned vm_id, uint64_t pd_addr) { int usepfp = (ring->type == AMDGPU_RING_TYPE_GFX); - uint32_t seq = ring->fence_drv.sync_seq; + uint32_t seq = ring->fence_drv.sync_seq[ring->idx]; uint64_t addr = ring->fence_drv.gpu_addr; amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5)); @@ -5463,7 +5463,7 @@ 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/amdkfd/kfd_dbgdev.c b/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c index c34c393e9aea..d5e19b5fbbfb 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c @@ -513,7 +513,7 @@ static int dbgdev_wave_control_set_registers( union SQ_CMD_BITS *in_reg_sq_cmd, union GRBM_GFX_INDEX_BITS *in_reg_gfx_index) { - int status; + int status = 0; union SQ_CMD_BITS reg_sq_cmd; union GRBM_GFX_INDEX_BITS reg_gfx_index; struct HsaDbgWaveMsgAMDGen2 *pMsg; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index 9be007081b72..eb1da83c9902 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/atmel-hlcdc/atmel_hlcdc_plane.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c index d0299aed517e..59d1269626b1 100644 --- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c +++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c @@ -335,6 +335,8 @@ atmel_hlcdc_plane_update_pos_and_size(struct atmel_hlcdc_plane *plane, atmel_hlcdc_layer_update_cfg(&plane->layer, 13, 0xffffffff, factor_reg); + } else { + atmel_hlcdc_layer_update_cfg(&plane->layer, 13, 0xffffffff, 0); } } diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index aeee083c7f95..6253775b8d9c 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -150,7 +150,7 @@ void drm_atomic_state_default_clear(struct drm_atomic_state *state) for (i = 0; i < state->num_connector; i++) { struct drm_connector *connector = state->connectors[i]; - if (!connector) + if (!connector || !connector->funcs) continue; /* @@ -367,6 +367,8 @@ int drm_atomic_set_mode_prop_for_crtc(struct drm_crtc_state *state, drm_property_unreference_blob(state->mode_blob); state->mode_blob = NULL; + memset(&state->mode, 0, sizeof(state->mode)); + if (blob) { if (blob->length != sizeof(struct drm_mode_modeinfo) || drm_mode_convert_umode(&state->mode, @@ -379,7 +381,6 @@ int drm_atomic_set_mode_prop_for_crtc(struct drm_crtc_state *state, DRM_DEBUG_ATOMIC("Set [MODE:%s] for CRTC state %p\n", state->mode.name, state); } else { - memset(&state->mode, 0, sizeof(state->mode)); state->enable = false; DRM_DEBUG_ATOMIC("Set [NOMODE] for CRTC state %p\n", state); diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index 24c5434abd1c..dc84003f694e 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -2682,8 +2682,6 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data, goto out; } - drm_mode_set_crtcinfo(mode, CRTC_INTERLACE_HALVE_V); - /* * Check whether the primary plane supports the fb pixel format. * Drivers not implementing the universal planes API use a @@ -3316,6 +3314,24 @@ int drm_mode_addfb2(struct drm_device *dev, return 0; } +struct drm_mode_rmfb_work { + struct work_struct work; + struct list_head fbs; +}; + +static void drm_mode_rmfb_work_fn(struct work_struct *w) +{ + struct drm_mode_rmfb_work *arg = container_of(w, typeof(*arg), work); + + while (!list_empty(&arg->fbs)) { + struct drm_framebuffer *fb = + list_first_entry(&arg->fbs, typeof(*fb), filp_head); + + list_del_init(&fb->filp_head); + drm_framebuffer_remove(fb); + } +} + /** * drm_mode_rmfb - remove an FB from the configuration * @dev: drm device for the ioctl @@ -3356,7 +3372,25 @@ int drm_mode_rmfb(struct drm_device *dev, mutex_unlock(&dev->mode_config.fb_lock); mutex_unlock(&file_priv->fbs_lock); - drm_framebuffer_unreference(fb); + /* + * we now own the reference that was stored in the fbs list + * + * drm_framebuffer_remove may fail with -EINTR on pending signals, + * so run this in a separate stack as there's no way to correctly + * handle this after the fb is already removed from the lookup table. + */ + if (atomic_read(&fb->refcount.refcount) > 1) { + struct drm_mode_rmfb_work arg; + + INIT_WORK_ONSTACK(&arg.work, drm_mode_rmfb_work_fn); + INIT_LIST_HEAD(&arg.fbs); + list_add_tail(&fb->filp_head, &arg.fbs); + + schedule_work(&arg.work); + flush_work(&arg.work); + destroy_work_on_stack(&arg.work); + } else + drm_framebuffer_unreference(fb); return 0; @@ -3509,7 +3543,6 @@ out_err1: return ret; } - /** * drm_fb_release - remove and free the FBs on this file * @priv: drm file for the ioctl @@ -3524,6 +3557,9 @@ out_err1: void drm_fb_release(struct drm_file *priv) { struct drm_framebuffer *fb, *tfb; + struct drm_mode_rmfb_work arg; + + INIT_LIST_HEAD(&arg.fbs); /* * When the file gets released that means no one else can access the fb @@ -3536,10 +3572,22 @@ void drm_fb_release(struct drm_file *priv) * at it any more. */ list_for_each_entry_safe(fb, tfb, &priv->fbs, filp_head) { - list_del_init(&fb->filp_head); + if (atomic_read(&fb->refcount.refcount) > 1) { + list_move_tail(&fb->filp_head, &arg.fbs); + } else { + list_del_init(&fb->filp_head); - /* This drops the fpriv->fbs reference. */ - drm_framebuffer_unreference(fb); + /* This drops the fpriv->fbs reference. */ + drm_framebuffer_unreference(fb); + } + } + + if (!list_empty(&arg.fbs)) { + INIT_WORK_ONSTACK(&arg.work, drm_mode_rmfb_work_fn); + + schedule_work(&arg.work); + flush_work(&arg.work); + destroy_work_on_stack(&arg.work); } } diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index 39d7e2e15c11..2485fb652716 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -1665,13 +1665,19 @@ static int drm_dp_payload_send_msg(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_branch *mstb; int len, ret, port_num; + port = drm_dp_get_validated_port_ref(mgr, port); + if (!port) + return -EINVAL; + port_num = port->port_num; mstb = drm_dp_get_validated_mstb_ref(mgr, port->parent); if (!mstb) { mstb = drm_dp_get_last_connected_port_and_mstb(mgr, port->parent, &port_num); - if (!mstb) + if (!mstb) { + drm_dp_put_port(port); return -EINVAL; + } } txmsg = kzalloc(sizeof(*txmsg), GFP_KERNEL); @@ -1697,6 +1703,7 @@ static int drm_dp_payload_send_msg(struct drm_dp_mst_topology_mgr *mgr, kfree(txmsg); fail_put: drm_dp_put_mst_branch_device(mstb); + drm_dp_put_port(port); return ret; } @@ -1779,6 +1786,11 @@ int drm_dp_update_payload_part1(struct drm_dp_mst_topology_mgr *mgr) req_payload.start_slot = cur_slots; if (mgr->proposed_vcpis[i]) { port = container_of(mgr->proposed_vcpis[i], struct drm_dp_mst_port, vcpi); + port = drm_dp_get_validated_port_ref(mgr, port); + if (!port) { + mutex_unlock(&mgr->payload_lock); + return -EINVAL; + } req_payload.num_slots = mgr->proposed_vcpis[i]->num_slots; } else { port = NULL; @@ -1804,6 +1816,9 @@ int drm_dp_update_payload_part1(struct drm_dp_mst_topology_mgr *mgr) mgr->payloads[i].payload_state = req_payload.payload_state; } cur_slots += req_payload.num_slots; + + if (port) + drm_dp_put_port(port); } for (i = 0; i < mgr->max_payloads; i++) { @@ -2109,6 +2124,8 @@ int drm_dp_mst_topology_mgr_resume(struct drm_dp_mst_topology_mgr *mgr) if (mgr->mst_primary) { int sret; + u8 guid[16]; + sret = drm_dp_dpcd_read(mgr->aux, DP_DPCD_REV, mgr->dpcd, DP_RECEIVER_CAP_SIZE); if (sret != DP_RECEIVER_CAP_SIZE) { DRM_DEBUG_KMS("dpcd read failed - undocked during suspend?\n"); @@ -2123,6 +2140,16 @@ int drm_dp_mst_topology_mgr_resume(struct drm_dp_mst_topology_mgr *mgr) ret = -1; goto out_unlock; } + + /* Some hubs forget their guids after they resume */ + sret = drm_dp_dpcd_read(mgr->aux, DP_GUID, guid, 16); + if (sret != 16) { + DRM_DEBUG_KMS("dpcd read failed - undocked during suspend?\n"); + ret = -1; + goto out_unlock; + } + drm_dp_check_mstb_guid(mgr->mst_primary, guid); + ret = 0; } else ret = -1; @@ -2847,11 +2874,9 @@ static void drm_dp_destroy_connector_work(struct work_struct *work) drm_dp_port_teardown_pdt(port, port->pdt); if (!port->input && port->vcpi.vcpi > 0) { - if (mgr->mst_state) { - drm_dp_mst_reset_vcpi_slots(mgr, port); - drm_dp_update_payload_part1(mgr); - drm_dp_mst_put_payload_id(mgr, port->vcpi.vcpi); - } + drm_dp_mst_reset_vcpi_slots(mgr, port); + drm_dp_update_payload_part1(mgr); + drm_dp_mst_put_payload_id(mgr, port->vcpi.vcpi); } kref_put(&port->kref, drm_dp_free_mst_port); diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index 69cbab5e5c81..5ad036741b99 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -1899,7 +1899,6 @@ static int drm_pick_crtcs(struct drm_fb_helper *fb_helper, int n, int width, int height) { int c, o; - struct drm_device *dev = fb_helper->dev; struct drm_connector *connector; const struct drm_connector_helper_funcs *connector_funcs; struct drm_encoder *encoder; @@ -1918,7 +1917,7 @@ static int drm_pick_crtcs(struct drm_fb_helper *fb_helper, if (modes[n] == NULL) return best_score; - crtcs = kzalloc(dev->mode_config.num_connector * + crtcs = kzalloc(fb_helper->connector_count * sizeof(struct drm_fb_helper_crtc *), GFP_KERNEL); if (!crtcs) return best_score; @@ -1964,7 +1963,7 @@ static int drm_pick_crtcs(struct drm_fb_helper *fb_helper, if (score > best_score) { best_score = score; memcpy(best_crtcs, crtcs, - dev->mode_config.num_connector * + fb_helper->connector_count * sizeof(struct drm_fb_helper_crtc *)); } } diff --git a/drivers/gpu/drm/drm_modes.c b/drivers/gpu/drm/drm_modes.c index cd74a0953f42..39e30abddf08 100644 --- a/drivers/gpu/drm/drm_modes.c +++ b/drivers/gpu/drm/drm_modes.c @@ -1487,6 +1487,8 @@ int drm_mode_convert_umode(struct drm_display_mode *out, if (out->status != MODE_OK) goto out; + drm_mode_set_crtcinfo(out, CRTC_INTERLACE_HALVE_V); + ret = 0; out: diff --git a/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c b/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c index 6b43ae3ffd73..1616af209bfc 100644 --- a/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c +++ b/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c @@ -72,7 +72,7 @@ static const char *const dsi_errors[] = { "RX Prot Violation", "HS Generic Write FIFO Full", "LP Generic Write FIFO Full", - "Generic Read Data Avail" + "Generic Read Data Avail", "Special Packet Sent", "Tearing Effect", }; diff --git a/drivers/gpu/drm/i915/i915_gem_shrinker.c b/drivers/gpu/drm/i915/i915_gem_shrinker.c index f7df54a8ee2b..c0a96f1ee18e 100644 --- a/drivers/gpu/drm/i915/i915_gem_shrinker.c +++ b/drivers/gpu/drm/i915/i915_gem_shrinker.c @@ -39,7 +39,7 @@ static bool mutex_is_locked_by(struct mutex *mutex, struct task_struct *task) if (!mutex_is_locked(mutex)) return false; -#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_MUTEXES) +#if defined(CONFIG_DEBUG_MUTEXES) || defined(CONFIG_MUTEX_SPIN_ON_OWNER) return mutex->owner == task; #else /* Since UP may be pre-empted, we cannot assume that we own the lock */ diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index bc7b8faba84d..9ed9f6dde86f 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -2838,7 +2838,14 @@ enum skl_disp_power_wells { #define GEN6_RP_STATE_CAP (MCHBAR_MIRROR_BASE_SNB + 0x5998) #define BXT_RP_STATE_CAP 0x138170 -#define INTERVAL_1_28_US(us) (((us) * 100) >> 7) +/* + * Make these a multiple of magic 25 to avoid SNB (eg. Dell XPS + * 8300) freezing up around GPU hangs. Looks as if even + * scheduling/timer interrupts start misbehaving if the RPS + * EI/thresholds are "bad", leading to a very sluggish or even + * frozen machine. + */ +#define INTERVAL_1_28_US(us) roundup(((us) * 100) >> 7, 25) #define INTERVAL_1_33_US(us) (((us) * 3) >> 2) #define INTERVAL_0_833_US(us) (((us) * 6) / 5) #define GT_INTERVAL_FROM_US(dev_priv, us) (IS_GEN9(dev_priv) ? \ @@ -7350,6 +7357,8 @@ enum skl_disp_power_wells { #define TRANS_CLK_SEL_DISABLED (0x0<<29) #define TRANS_CLK_SEL_PORT(x) (((x)+1)<<29) +#define CDCLK_FREQ 0x46200 + #define TRANSA_MSA_MISC 0x60410 #define TRANSB_MSA_MISC 0x61410 #define TRANSC_MSA_MISC 0x62410 diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c index 6a2c76e367a5..97d1ed20418b 100644 --- a/drivers/gpu/drm/i915/intel_crt.c +++ b/drivers/gpu/drm/i915/intel_crt.c @@ -248,8 +248,14 @@ static bool intel_crt_compute_config(struct intel_encoder *encoder, pipe_config->has_pch_encoder = true; /* LPT FDI RX only supports 8bpc. */ - if (HAS_PCH_LPT(dev)) + if (HAS_PCH_LPT(dev)) { + if (pipe_config->bw_constrained && pipe_config->pipe_bpp < 24) { + DRM_DEBUG_KMS("LPT only supports 24bpp\n"); + return false; + } + pipe_config->pipe_bpp = 24; + } /* FDI must always be 2.7 GHz */ if (HAS_DDI(dev)) { diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c index 9e530a739354..fc28c512ece3 100644 --- a/drivers/gpu/drm/i915/intel_csr.c +++ b/drivers/gpu/drm/i915/intel_csr.c @@ -180,7 +180,8 @@ struct stepping_info { static const struct stepping_info skl_stepping_info[] = { {'A', '0'}, {'B', '0'}, {'C', '0'}, {'D', '0'}, {'E', '0'}, {'F', '0'}, - {'G', '0'}, {'H', '0'}, {'I', '0'} + {'G', '0'}, {'H', '0'}, {'I', '0'}, + {'J', '0'}, {'K', '0'} }; static struct stepping_info bxt_stepping_info[] = { diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c index 7e6158b889da..3c6b07683bd9 100644 --- a/drivers/gpu/drm/i915/intel_ddi.c +++ b/drivers/gpu/drm/i915/intel_ddi.c @@ -464,9 +464,17 @@ static void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port, } else if (IS_BROADWELL(dev)) { ddi_translations_fdi = bdw_ddi_translations_fdi; ddi_translations_dp = bdw_ddi_translations_dp; - ddi_translations_edp = bdw_ddi_translations_edp; + + if (dev_priv->edp_low_vswing) { + ddi_translations_edp = bdw_ddi_translations_edp; + n_edp_entries = ARRAY_SIZE(bdw_ddi_translations_edp); + } else { + ddi_translations_edp = bdw_ddi_translations_dp; + n_edp_entries = ARRAY_SIZE(bdw_ddi_translations_dp); + } + ddi_translations_hdmi = bdw_ddi_translations_hdmi; - n_edp_entries = ARRAY_SIZE(bdw_ddi_translations_edp); + n_dp_entries = ARRAY_SIZE(bdw_ddi_translations_dp); n_hdmi_entries = ARRAY_SIZE(bdw_ddi_translations_hdmi); hdmi_default_entry = 7; @@ -3188,12 +3196,6 @@ void intel_ddi_get_config(struct intel_encoder *encoder, intel_ddi_clock_get(encoder, pipe_config); } -static void intel_ddi_destroy(struct drm_encoder *encoder) -{ - /* HDMI has nothing special to destroy, so we can go with this. */ - intel_dp_encoder_destroy(encoder); -} - static bool intel_ddi_compute_config(struct intel_encoder *encoder, struct intel_crtc_state *pipe_config) { @@ -3212,7 +3214,8 @@ static bool intel_ddi_compute_config(struct intel_encoder *encoder, } static const struct drm_encoder_funcs intel_ddi_funcs = { - .destroy = intel_ddi_destroy, + .reset = intel_dp_encoder_reset, + .destroy = intel_dp_encoder_destroy, }; static struct intel_connector * @@ -3284,6 +3287,7 @@ void intel_ddi_init(struct drm_device *dev, enum port port) intel_encoder->post_disable = intel_ddi_post_disable; intel_encoder->get_hw_state = intel_ddi_get_hw_state; intel_encoder->get_config = intel_ddi_get_config; + intel_encoder->suspend = intel_dp_encoder_suspend; intel_dig_port->port = port; intel_dig_port->saved_port_bits = I915_READ(DDI_BUF_CTL(port)) & diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index f859a5b87ed4..c41bc42b6fa7 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -4447,7 +4447,7 @@ int skl_update_scaler_crtc(struct intel_crtc_state *state) intel_crtc->base.base.id, intel_crtc->pipe, SKL_CRTC_INDEX); return skl_update_scaler(state, !state->base.active, SKL_CRTC_INDEX, - &state->scaler_state.scaler_id, DRM_ROTATE_0, + &state->scaler_state.scaler_id, BIT(DRM_ROTATE_0), state->pipe_src_w, state->pipe_src_h, adjusted_mode->crtc_hdisplay, adjusted_mode->crtc_vdisplay); } @@ -8228,12 +8228,14 @@ static void ironlake_init_pch_refclk(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; struct intel_encoder *encoder; + int i; u32 val, final; bool has_lvds = false; bool has_cpu_edp = false; bool has_panel = false; bool has_ck505 = false; bool can_ssc = false; + bool using_ssc_source = false; /* We need to take the global config into account */ for_each_intel_encoder(dev, encoder) { @@ -8260,8 +8262,22 @@ static void ironlake_init_pch_refclk(struct drm_device *dev) can_ssc = true; } - DRM_DEBUG_KMS("has_panel %d has_lvds %d has_ck505 %d\n", - has_panel, has_lvds, has_ck505); + /* Check if any DPLLs are using the SSC source */ + for (i = 0; i < dev_priv->num_shared_dpll; i++) { + u32 temp = I915_READ(PCH_DPLL(i)); + + if (!(temp & DPLL_VCO_ENABLE)) + continue; + + if ((temp & PLL_REF_INPUT_MASK) == + PLLB_REF_INPUT_SPREADSPECTRUMIN) { + using_ssc_source = true; + break; + } + } + + DRM_DEBUG_KMS("has_panel %d has_lvds %d has_ck505 %d using_ssc_source %d\n", + has_panel, has_lvds, has_ck505, using_ssc_source); /* Ironlake: try to setup display ref clock before DPLL * enabling. This is only under driver's control after @@ -8298,9 +8314,9 @@ static void ironlake_init_pch_refclk(struct drm_device *dev) final |= DREF_CPU_SOURCE_OUTPUT_NONSPREAD; } else final |= DREF_CPU_SOURCE_OUTPUT_DISABLE; - } else { - final |= DREF_SSC_SOURCE_DISABLE; - final |= DREF_CPU_SOURCE_OUTPUT_DISABLE; + } else if (using_ssc_source) { + final |= DREF_SSC_SOURCE_ENABLE; + final |= DREF_SSC1_ENABLE; } if (final == val) @@ -8346,7 +8362,7 @@ static void ironlake_init_pch_refclk(struct drm_device *dev) POSTING_READ(PCH_DREF_CONTROL); udelay(200); } else { - DRM_DEBUG_KMS("Disabling SSC entirely\n"); + DRM_DEBUG_KMS("Disabling CPU source output\n"); val &= ~DREF_CPU_SOURCE_OUTPUT_MASK; @@ -8357,16 +8373,20 @@ static void ironlake_init_pch_refclk(struct drm_device *dev) POSTING_READ(PCH_DREF_CONTROL); udelay(200); - /* Turn off the SSC source */ - val &= ~DREF_SSC_SOURCE_MASK; - val |= DREF_SSC_SOURCE_DISABLE; + if (!using_ssc_source) { + DRM_DEBUG_KMS("Disabling SSC source\n"); - /* Turn off SSC1 */ - val &= ~DREF_SSC1_ENABLE; + /* Turn off the SSC source */ + val &= ~DREF_SSC_SOURCE_MASK; + val |= DREF_SSC_SOURCE_DISABLE; - I915_WRITE(PCH_DREF_CONTROL, val); - POSTING_READ(PCH_DREF_CONTROL); - udelay(200); + /* Turn off SSC1 */ + val &= ~DREF_SSC1_ENABLE; + + I915_WRITE(PCH_DREF_CONTROL, val); + POSTING_READ(PCH_DREF_CONTROL); + udelay(200); + } } BUG_ON(val != final); @@ -9669,6 +9689,8 @@ static void broadwell_set_cdclk(struct drm_device *dev, int cdclk) sandybridge_pcode_write(dev_priv, HSW_PCODE_DE_WRITE_FREQ_REQ, data); mutex_unlock(&dev_priv->rps.hw_lock); + I915_WRITE(CDCLK_FREQ, DIV_ROUND_CLOSEST(cdclk, 1000) - 1); + intel_update_cdclk(dev); WARN(cdclk != dev_priv->cdclk_freq, diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 78b8ec84d576..8e1d6d74c203 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -3628,8 +3628,7 @@ static bool intel_dp_reset_link_train(struct intel_dp *intel_dp, uint32_t *DP, uint8_t dp_train_pat) { - if (!intel_dp->train_set_valid) - memset(intel_dp->train_set, 0, sizeof(intel_dp->train_set)); + memset(intel_dp->train_set, 0, sizeof(intel_dp->train_set)); intel_dp_set_signal_levels(intel_dp, DP); return intel_dp_set_link_train(intel_dp, DP, dp_train_pat); } @@ -3746,22 +3745,6 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp) break; } - /* - * if we used previously trained voltage and pre-emphasis values - * and we don't get clock recovery, reset link training values - */ - if (intel_dp->train_set_valid) { - DRM_DEBUG_KMS("clock recovery not ok, reset"); - /* clear the flag as we are not reusing train set */ - intel_dp->train_set_valid = false; - if (!intel_dp_reset_link_train(intel_dp, &DP, - DP_TRAINING_PATTERN_1 | - DP_LINK_SCRAMBLING_DISABLE)) { - DRM_ERROR("failed to enable link training\n"); - return; - } - continue; - } /* Check to see if we've tried the max voltage */ for (i = 0; i < intel_dp->lane_count; i++) @@ -3854,7 +3837,6 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp) /* Make sure clock is still ok */ if (!drm_dp_clock_recovery_ok(link_status, intel_dp->lane_count)) { - intel_dp->train_set_valid = false; intel_dp_link_training_clock_recovery(intel_dp); intel_dp_set_link_train(intel_dp, &DP, training_pattern | @@ -3871,7 +3853,6 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp) /* Try 5 times, then try clock recovery if that fails */ if (tries > 5) { - intel_dp->train_set_valid = false; intel_dp_link_training_clock_recovery(intel_dp); intel_dp_set_link_train(intel_dp, &DP, training_pattern | @@ -3893,10 +3874,8 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp) intel_dp->DP = DP; - if (channel_eq) { - intel_dp->train_set_valid = true; + if (channel_eq) DRM_DEBUG_KMS("Channel EQ done. DP Training successful\n"); - } } void intel_dp_stop_link_train(struct intel_dp *intel_dp) @@ -5035,7 +5014,7 @@ void intel_dp_encoder_destroy(struct drm_encoder *encoder) kfree(intel_dig_port); } -static void intel_dp_encoder_suspend(struct intel_encoder *intel_encoder) +void intel_dp_encoder_suspend(struct intel_encoder *intel_encoder) { struct intel_dp *intel_dp = enc_to_intel_dp(&intel_encoder->base); @@ -5077,15 +5056,17 @@ static void intel_edp_panel_vdd_sanitize(struct intel_dp *intel_dp) edp_panel_vdd_schedule_off(intel_dp); } -static void intel_dp_encoder_reset(struct drm_encoder *encoder) +void intel_dp_encoder_reset(struct drm_encoder *encoder) { - struct intel_dp *intel_dp; + struct drm_i915_private *dev_priv = to_i915(encoder->dev); + struct intel_dp *intel_dp = enc_to_intel_dp(encoder); + + if (!HAS_DDI(dev_priv)) + intel_dp->DP = I915_READ(intel_dp->output_reg); if (to_intel_encoder(encoder)->type != INTEL_OUTPUT_EDP) return; - intel_dp = enc_to_intel_dp(encoder); - pps_lock(intel_dp); /* @@ -5157,9 +5138,6 @@ intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, bool long_hpd) intel_display_power_get(dev_priv, power_domain); if (long_hpd) { - /* indicate that we need to restart link training */ - intel_dp->train_set_valid = false; - if (!intel_digital_port_connected(dev_priv, intel_dig_port)) goto mst_fail; diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c index 0639275fc471..06bd9257acdc 100644 --- a/drivers/gpu/drm/i915/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/intel_dp_mst.c @@ -477,6 +477,8 @@ static void intel_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr, struct intel_connector *intel_connector = to_intel_connector(connector); struct drm_device *dev = connector->dev; + intel_connector->unregister(intel_connector); + /* need to nuke the connector */ drm_modeset_lock_all(dev); if (connector->state->crtc) { @@ -490,11 +492,7 @@ static void intel_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr, WARN(ret, "Disabling mst crtc failed with %i\n", ret); } - drm_modeset_unlock_all(dev); - intel_connector->unregister(intel_connector); - - drm_modeset_lock_all(dev); intel_connector_remove_from_fbdev(intel_connector); drm_connector_cleanup(connector); drm_modeset_unlock_all(dev); diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 0d00f07b7163..c5f11e0c5d5b 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -783,7 +783,6 @@ struct intel_dp { bool has_aux_irq, int send_bytes, uint32_t aux_clock_divider); - bool train_set_valid; /* Displayport compliance testing */ unsigned long compliance_test_type; @@ -1204,6 +1203,8 @@ void intel_dp_set_link_params(struct intel_dp *intel_dp, void intel_dp_start_link_train(struct intel_dp *intel_dp); void intel_dp_stop_link_train(struct intel_dp *intel_dp); void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode); +void intel_dp_encoder_reset(struct drm_encoder *encoder); +void intel_dp_encoder_suspend(struct intel_encoder *intel_encoder); void intel_dp_encoder_destroy(struct drm_encoder *encoder); int intel_dp_sink_crc(struct intel_dp *intel_dp, u8 *crc); bool intel_dp_compute_config(struct intel_encoder *encoder, diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c index 4fd5fdfef6bd..c0c094d5b822 100644 --- a/drivers/gpu/drm/i915/intel_fbdev.c +++ b/drivers/gpu/drm/i915/intel_fbdev.c @@ -362,12 +362,12 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper, uint64_t conn_configured = 0, mask; int pass = 0; - save_enabled = kcalloc(dev->mode_config.num_connector, sizeof(bool), + save_enabled = kcalloc(fb_helper->connector_count, sizeof(bool), GFP_KERNEL); if (!save_enabled) return false; - memcpy(save_enabled, enabled, dev->mode_config.num_connector); + memcpy(save_enabled, enabled, fb_helper->connector_count); mask = (1 << fb_helper->connector_count) - 1; retry: for (i = 0; i < fb_helper->connector_count; i++) { @@ -501,7 +501,7 @@ retry: if (fallback) { bail: DRM_DEBUG_KMS("Not using firmware configuration\n"); - memcpy(enabled, save_enabled, dev->mode_config.num_connector); + memcpy(enabled, save_enabled, fb_helper->connector_count); kfree(save_enabled); return false; } diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c index e6c035b0fc1c..4b8ed9f2dabc 100644 --- a/drivers/gpu/drm/i915/intel_hdmi.c +++ b/drivers/gpu/drm/i915/intel_hdmi.c @@ -1388,8 +1388,16 @@ intel_hdmi_detect(struct drm_connector *connector, bool force) hdmi_to_dig_port(intel_hdmi)); } - if (!live_status) - DRM_DEBUG_KMS("Live status not up!"); + if (!live_status) { + DRM_DEBUG_KMS("HDMI live status down\n"); + /* + * Live status register is not reliable on all intel platforms. + * So consider live_status only for certain platforms, for + * others, read EDID to determine presence of sink. + */ + if (INTEL_INFO(dev_priv)->gen < 7 || IS_IVYBRIDGE(dev_priv)) + live_status = true; + } intel_hdmi_unset_edid(connector); diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index d69547a65dbb..7058f75c7b42 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -776,11 +776,11 @@ static int logical_ring_prepare(struct drm_i915_gem_request *req, int bytes) if (unlikely(total_bytes > remain_usable)) { /* * The base request will fit but the reserved space - * falls off the end. So only need to to wait for the - * reserved size after flushing out the remainder. + * falls off the end. So don't need an immediate wrap + * and only need to effectively wait for the reserved + * size space from the start of ringbuffer. */ wait_bytes = remain_actual + ringbuf->reserved_size; - need_wrap = true; } else if (total_bytes > ringbuf->space) { /* No wrapping required, just waiting. */ wait_bytes = total_bytes; diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index f091ad12d694..62284e45d531 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -3880,6 +3880,8 @@ static void ilk_pipe_wm_get_hw_state(struct drm_crtc *crtc) if (IS_HASWELL(dev) || IS_BROADWELL(dev)) hw->wm_linetime[pipe] = I915_READ(PIPE_WM_LINETIME(pipe)); + memset(active, 0, sizeof(*active)); + active->pipe_enabled = intel_crtc->active; if (active->pipe_enabled) { @@ -6620,6 +6622,12 @@ static void broadwell_init_clock_gating(struct drm_device *dev) misccpctl = I915_READ(GEN7_MISCCPCTL); I915_WRITE(GEN7_MISCCPCTL, misccpctl & ~GEN7_DOP_CLOCK_GATE_ENABLE); I915_WRITE(GEN8_L3SQCREG1, BDW_WA_L3SQCREG1_DEFAULT); + /* + * Wait at least 100 clocks before re-enabling clock gating. See + * the definition of L3SQCREG1 in BSpec. + */ + POSTING_READ(GEN8_L3SQCREG1); + udelay(1); I915_WRITE(GEN7_MISCCPCTL, misccpctl); /* diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index f6b2a814e629..9d48443bca2e 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1922,6 +1922,17 @@ i915_dispatch_execbuffer(struct drm_i915_gem_request *req, return 0; } +static void cleanup_phys_status_page(struct intel_engine_cs *ring) +{ + struct drm_i915_private *dev_priv = to_i915(ring->dev); + + if (!dev_priv->status_page_dmah) + return; + + drm_pci_free(ring->dev, dev_priv->status_page_dmah); + ring->status_page.page_addr = NULL; +} + static void cleanup_status_page(struct intel_engine_cs *ring) { struct drm_i915_gem_object *obj; @@ -1938,9 +1949,9 @@ static void cleanup_status_page(struct intel_engine_cs *ring) static int init_status_page(struct intel_engine_cs *ring) { - struct drm_i915_gem_object *obj; + struct drm_i915_gem_object *obj = ring->status_page.obj; - if ((obj = ring->status_page.obj) == NULL) { + if (obj == NULL) { unsigned flags; int ret; @@ -2134,7 +2145,7 @@ static int intel_init_ring_buffer(struct drm_device *dev, if (ret) goto error; } else { - BUG_ON(ring->id != RCS); + WARN_ON(ring->id != RCS); ret = init_phys_status_page(ring); if (ret) goto error; @@ -2179,7 +2190,12 @@ void intel_cleanup_ring_buffer(struct intel_engine_cs *ring) if (ring->cleanup) ring->cleanup(ring); - cleanup_status_page(ring); + if (I915_NEED_GFX_HWS(ring->dev)) { + cleanup_status_page(ring); + } else { + WARN_ON(ring->id != RCS); + cleanup_phys_status_page(ring); + } i915_cmd_parser_fini_ring(ring); i915_gem_batch_pool_fini(&ring->batch_pool); @@ -2341,11 +2357,11 @@ static int __intel_ring_prepare(struct intel_engine_cs *ring, int bytes) if (unlikely(total_bytes > remain_usable)) { /* * The base request will fit but the reserved space - * falls off the end. So only need to to wait for the - * reserved size after flushing out the remainder. + * falls off the end. So don't need an immediate wrap + * and only need to effectively wait for the reserved + * size space from the start of ringbuffer. */ wait_bytes = remain_actual + ringbuf->reserved_size; - need_wrap = true; } else if (total_bytes > ringbuf->space) { /* No wrapping required, just waiting. */ wait_bytes = total_bytes; diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index 43cba129a0c0..cc91ae832ffb 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -1132,7 +1132,11 @@ static void intel_uncore_fw_domains_init(struct drm_device *dev) } else if (IS_HASWELL(dev) || IS_BROADWELL(dev)) { dev_priv->uncore.funcs.force_wake_get = fw_domains_get_with_thread_status; - dev_priv->uncore.funcs.force_wake_put = fw_domains_put; + if (IS_HASWELL(dev)) + dev_priv->uncore.funcs.force_wake_put = + fw_domains_put_with_fifo; + else + dev_priv->uncore.funcs.force_wake_put = fw_domains_put; fw_domain_init(dev_priv, FW_DOMAIN_ID_RENDER, FORCEWAKE_MT, FORCEWAKE_ACK_HSW); } else if (IS_IVYBRIDGE(dev)) { diff --git a/drivers/gpu/drm/imx/imx-drm-core.c b/drivers/gpu/drm/imx/imx-drm-core.c index 7b990b4e96d2..5378bdc3bbf9 100644 --- a/drivers/gpu/drm/imx/imx-drm-core.c +++ b/drivers/gpu/drm/imx/imx-drm-core.c @@ -26,6 +26,7 @@ #include <drm/drm_fb_cma_helper.h> #include <drm/drm_plane_helper.h> #include <drm/drm_of.h> +#include <video/imx-ipu-v3.h> #include "imx-drm.h" @@ -504,6 +505,13 @@ static int compare_of(struct device *dev, void *data) { struct device_node *np = data; + /* Special case for DI, dev->of_node may not be set yet */ + if (strcmp(dev->driver->name, "imx-ipuv3-crtc") == 0) { + struct ipu_client_platformdata *pdata = dev->platform_data; + + return pdata->of_node == np; + } + /* Special case for LDB, one device for two channels */ if (of_node_cmp(np->name, "lvds-channel") == 0) { np = of_get_parent(np); diff --git a/drivers/gpu/drm/imx/ipuv3-crtc.c b/drivers/gpu/drm/imx/ipuv3-crtc.c index 4ab841eebee1..9b0abd44b751 100644 --- a/drivers/gpu/drm/imx/ipuv3-crtc.c +++ b/drivers/gpu/drm/imx/ipuv3-crtc.c @@ -369,7 +369,7 @@ static int ipu_crtc_init(struct ipu_crtc *ipu_crtc, ret = imx_drm_add_crtc(drm, &ipu_crtc->base, &ipu_crtc->imx_crtc, &ipu_crtc->plane[0]->base, &ipu_crtc_helper_funcs, - ipu_crtc->dev->of_node); + pdata->of_node); if (ret) { dev_err(ipu_crtc->dev, "adding crtc failed with %d.\n", ret); goto err_put_resources; diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c index c99d3fe12881..e5bb40e58020 100644 --- a/drivers/gpu/drm/mgag200/mgag200_mode.c +++ b/drivers/gpu/drm/mgag200/mgag200_mode.c @@ -194,7 +194,7 @@ static int mga_g200se_set_plls(struct mga_device *mdev, long clock) } } - fvv = pllreffreq * testn / testm; + fvv = pllreffreq * (n + 1) / (m + 1); fvv = (fvv - 800000) / 50000; if (fvv > 15) @@ -214,6 +214,14 @@ static int mga_g200se_set_plls(struct mga_device *mdev, long clock) WREG_DAC(MGA1064_PIX_PLLC_M, m); WREG_DAC(MGA1064_PIX_PLLC_N, n); WREG_DAC(MGA1064_PIX_PLLC_P, p); + + if (mdev->unique_rev_id >= 0x04) { + WREG_DAC(0x1a, 0x09); + msleep(20); + WREG_DAC(0x1a, 0x01); + + } + return 0; } diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c index 59f27e774acb..e40a1b07a014 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c +++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c @@ -557,6 +557,8 @@ nouveau_fbcon_init(struct drm_device *dev) if (ret) goto fini; + if (fbcon->helper.fbdev) + fbcon->helper.fbdev->pixmap.buf_align = 4; return 0; fini: diff --git a/drivers/gpu/drm/nouveau/nv04_fbcon.c b/drivers/gpu/drm/nouveau/nv04_fbcon.c index 789dc2993b0d..8f715feadf56 100644 --- a/drivers/gpu/drm/nouveau/nv04_fbcon.c +++ b/drivers/gpu/drm/nouveau/nv04_fbcon.c @@ -82,7 +82,6 @@ nv04_fbcon_imageblit(struct fb_info *info, const struct fb_image *image) uint32_t fg; uint32_t bg; uint32_t dsize; - uint32_t width; uint32_t *data = (uint32_t *)image->data; int ret; @@ -93,9 +92,6 @@ nv04_fbcon_imageblit(struct fb_info *info, const struct fb_image *image) if (ret) return ret; - width = ALIGN(image->width, 8); - dsize = ALIGN(width * image->height, 32) >> 5; - if (info->fix.visual == FB_VISUAL_TRUECOLOR || info->fix.visual == FB_VISUAL_DIRECTCOLOR) { fg = ((uint32_t *) info->pseudo_palette)[image->fg_color]; @@ -111,10 +107,11 @@ nv04_fbcon_imageblit(struct fb_info *info, const struct fb_image *image) ((image->dx + image->width) & 0xffff)); OUT_RING(chan, bg); OUT_RING(chan, fg); - OUT_RING(chan, (image->height << 16) | width); + OUT_RING(chan, (image->height << 16) | image->width); OUT_RING(chan, (image->height << 16) | image->width); OUT_RING(chan, (image->dy << 16) | (image->dx & 0xffff)); + dsize = ALIGN(image->width * image->height, 32) >> 5; while (dsize) { int iter_len = dsize > 128 ? 128 : dsize; diff --git a/drivers/gpu/drm/nouveau/nv50_fbcon.c b/drivers/gpu/drm/nouveau/nv50_fbcon.c index e05499d6ed83..a4e259a00430 100644 --- a/drivers/gpu/drm/nouveau/nv50_fbcon.c +++ b/drivers/gpu/drm/nouveau/nv50_fbcon.c @@ -95,7 +95,7 @@ nv50_fbcon_imageblit(struct fb_info *info, const struct fb_image *image) struct nouveau_fbdev *nfbdev = info->par; struct nouveau_drm *drm = nouveau_drm(nfbdev->dev); struct nouveau_channel *chan = drm->channel; - uint32_t width, dwords, *data = (uint32_t *)image->data; + uint32_t dwords, *data = (uint32_t *)image->data; uint32_t mask = ~(~0 >> (32 - info->var.bits_per_pixel)); uint32_t *palette = info->pseudo_palette; int ret; @@ -107,9 +107,6 @@ nv50_fbcon_imageblit(struct fb_info *info, const struct fb_image *image) if (ret) return ret; - width = ALIGN(image->width, 32); - dwords = (width * image->height) >> 5; - BEGIN_NV04(chan, NvSub2D, 0x0814, 2); if (info->fix.visual == FB_VISUAL_TRUECOLOR || info->fix.visual == FB_VISUAL_DIRECTCOLOR) { @@ -128,6 +125,7 @@ nv50_fbcon_imageblit(struct fb_info *info, const struct fb_image *image) OUT_RING(chan, 0); OUT_RING(chan, image->dy); + dwords = ALIGN(image->width * image->height, 32) >> 5; while (dwords) { int push = dwords > 2047 ? 2047 : dwords; diff --git a/drivers/gpu/drm/nouveau/nvc0_fbcon.c b/drivers/gpu/drm/nouveau/nvc0_fbcon.c index c97395b4a312..f28315e865a5 100644 --- a/drivers/gpu/drm/nouveau/nvc0_fbcon.c +++ b/drivers/gpu/drm/nouveau/nvc0_fbcon.c @@ -95,7 +95,7 @@ nvc0_fbcon_imageblit(struct fb_info *info, const struct fb_image *image) struct nouveau_fbdev *nfbdev = info->par; struct nouveau_drm *drm = nouveau_drm(nfbdev->dev); struct nouveau_channel *chan = drm->channel; - uint32_t width, dwords, *data = (uint32_t *)image->data; + uint32_t dwords, *data = (uint32_t *)image->data; uint32_t mask = ~(~0 >> (32 - info->var.bits_per_pixel)); uint32_t *palette = info->pseudo_palette; int ret; @@ -107,9 +107,6 @@ nvc0_fbcon_imageblit(struct fb_info *info, const struct fb_image *image) if (ret) return ret; - width = ALIGN(image->width, 32); - dwords = (width * image->height) >> 5; - BEGIN_NVC0(chan, NvSub2D, 0x0814, 2); if (info->fix.visual == FB_VISUAL_TRUECOLOR || info->fix.visual == FB_VISUAL_DIRECTCOLOR) { @@ -128,6 +125,7 @@ nvc0_fbcon_imageblit(struct fb_info *info, const struct fb_image *image) OUT_RING (chan, 0); OUT_RING (chan, image->dy); + dwords = ALIGN(image->width * image->height, 32) >> 5; while (dwords) { int push = dwords > 2047 ? 2047 : dwords; diff --git a/drivers/gpu/drm/nouveau/nvkm/core/ramht.c b/drivers/gpu/drm/nouveau/nvkm/core/ramht.c index 3216e157a8a0..89da47234016 100644 --- a/drivers/gpu/drm/nouveau/nvkm/core/ramht.c +++ b/drivers/gpu/drm/nouveau/nvkm/core/ramht.c @@ -131,7 +131,7 @@ nvkm_ramht_del(struct nvkm_ramht **pramht) struct nvkm_ramht *ramht = *pramht; if (ramht) { nvkm_gpuobj_del(&ramht->gpuobj); - kfree(*pramht); + vfree(*pramht); *pramht = NULL; } } @@ -143,8 +143,8 @@ nvkm_ramht_new(struct nvkm_device *device, u32 size, u32 align, struct nvkm_ramht *ramht; int ret, i; - if (!(ramht = *pramht = kzalloc(sizeof(*ramht) + (size >> 3) * - sizeof(*ramht->data), GFP_KERNEL))) + if (!(ramht = *pramht = vzalloc(sizeof(*ramht) + + (size >> 3) * sizeof(*ramht->data)))) return -ENOMEM; ramht->device = device; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgf119.c index b4b41b135643..2aaf0dd19a55 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgf119.c @@ -40,8 +40,8 @@ static int gf119_sor_dp_pattern(struct nvkm_output_dp *outp, int pattern) { struct nvkm_device *device = outp->base.disp->engine.subdev.device; - const u32 loff = gf119_sor_loff(outp); - nvkm_mask(device, 0x61c110 + loff, 0x0f0f0f0f, 0x01010101 * pattern); + const u32 soff = gf119_sor_soff(outp); + nvkm_mask(device, 0x61c110 + soff, 0x0f0f0f0f, 0x01010101 * pattern); return 0; } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c index 9f5dfc85147a..eeeea1c2ca23 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c @@ -874,22 +874,41 @@ gf100_gr_trap_gpc_rop(struct gf100_gr *gr, int gpc) } static const struct nvkm_enum gf100_mp_warp_error[] = { - { 0x00, "NO_ERROR" }, - { 0x01, "STACK_MISMATCH" }, + { 0x01, "STACK_ERROR" }, + { 0x02, "API_STACK_ERROR" }, + { 0x03, "RET_EMPTY_STACK_ERROR" }, + { 0x04, "PC_WRAP" }, { 0x05, "MISALIGNED_PC" }, - { 0x08, "MISALIGNED_GPR" }, - { 0x09, "INVALID_OPCODE" }, - { 0x0d, "GPR_OUT_OF_BOUNDS" }, - { 0x0e, "MEM_OUT_OF_BOUNDS" }, - { 0x0f, "UNALIGNED_MEM_ACCESS" }, + { 0x06, "PC_OVERFLOW" }, + { 0x07, "MISALIGNED_IMMC_ADDR" }, + { 0x08, "MISALIGNED_REG" }, + { 0x09, "ILLEGAL_INSTR_ENCODING" }, + { 0x0a, "ILLEGAL_SPH_INSTR_COMBO" }, + { 0x0b, "ILLEGAL_INSTR_PARAM" }, + { 0x0c, "INVALID_CONST_ADDR" }, + { 0x0d, "OOR_REG" }, + { 0x0e, "OOR_ADDR" }, + { 0x0f, "MISALIGNED_ADDR" }, { 0x10, "INVALID_ADDR_SPACE" }, - { 0x11, "INVALID_PARAM" }, + { 0x11, "ILLEGAL_INSTR_PARAM2" }, + { 0x12, "INVALID_CONST_ADDR_LDC" }, + { 0x13, "GEOMETRY_SM_ERROR" }, + { 0x14, "DIVERGENT" }, + { 0x15, "WARP_EXIT" }, {} }; static const struct nvkm_bitfield gf100_mp_global_error[] = { + { 0x00000001, "SM_TO_SM_FAULT" }, + { 0x00000002, "L1_ERROR" }, { 0x00000004, "MULTIPLE_WARP_ERRORS" }, - { 0x00000008, "OUT_OF_STACK_SPACE" }, + { 0x00000008, "PHYSICAL_STACK_OVERFLOW" }, + { 0x00000010, "BPT_INT" }, + { 0x00000020, "BPT_PAUSE" }, + { 0x00000040, "SINGLE_STEP_COMPLETE" }, + { 0x20000000, "ECC_SEC_ERROR" }, + { 0x40000000, "ECC_DED_ERROR" }, + { 0x80000000, "TIMEOUT" }, {} }; @@ -1717,6 +1736,8 @@ gf100_gr_init(struct gf100_gr *gr) gf100_gr_mmio(gr, gr->func->mmio); + nvkm_mask(device, TPC_UNIT(0, 0, 0x05c), 0x00000001, 0x00000001); + memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr)); for (i = 0, gpc = -1; i < gr->tpc_total; i++) { do { diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c index 183aea1abebc..5edebf495c07 100644 --- a/drivers/gpu/drm/qxl/qxl_display.c +++ b/drivers/gpu/drm/qxl/qxl_display.c @@ -375,10 +375,15 @@ static int qxl_crtc_cursor_set2(struct drm_crtc *crtc, qxl_bo_kunmap(user_bo); + qcrtc->cur_x += qcrtc->hot_spot_x - hot_x; + qcrtc->cur_y += qcrtc->hot_spot_y - hot_y; + qcrtc->hot_spot_x = hot_x; + qcrtc->hot_spot_y = hot_y; + cmd = (struct qxl_cursor_cmd *)qxl_release_map(qdev, release); cmd->type = QXL_CURSOR_SET; - cmd->u.set.position.x = qcrtc->cur_x; - cmd->u.set.position.y = qcrtc->cur_y; + cmd->u.set.position.x = qcrtc->cur_x + qcrtc->hot_spot_x; + cmd->u.set.position.y = qcrtc->cur_y + qcrtc->hot_spot_y; cmd->u.set.shape = qxl_bo_physical_address(qdev, cursor_bo, 0); @@ -441,8 +446,8 @@ static int qxl_crtc_cursor_move(struct drm_crtc *crtc, cmd = (struct qxl_cursor_cmd *)qxl_release_map(qdev, release); cmd->type = QXL_CURSOR_MOVE; - cmd->u.position.x = qcrtc->cur_x; - cmd->u.position.y = qcrtc->cur_y; + cmd->u.position.x = qcrtc->cur_x + qcrtc->hot_spot_x; + cmd->u.position.y = qcrtc->cur_y + qcrtc->hot_spot_y; qxl_release_unmap(qdev, release, &cmd->release_info); qxl_push_cursor_ring_release(qdev, release, QXL_CMD_CURSOR, false); diff --git a/drivers/gpu/drm/qxl/qxl_drv.h b/drivers/gpu/drm/qxl/qxl_drv.h index 01a86948eb8c..3ab90179e9ab 100644 --- a/drivers/gpu/drm/qxl/qxl_drv.h +++ b/drivers/gpu/drm/qxl/qxl_drv.h @@ -135,6 +135,8 @@ struct qxl_crtc { int index; int cur_x; int cur_y; + int hot_spot_x; + int hot_spot_y; }; struct qxl_output { diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index dac78ad24b31..79bab6fd76bb 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -1739,6 +1739,7 @@ static u32 radeon_get_pll_use_mask(struct drm_crtc *crtc) static int radeon_get_shared_dp_ppll(struct drm_crtc *crtc) { struct drm_device *dev = crtc->dev; + struct radeon_device *rdev = dev->dev_private; struct drm_crtc *test_crtc; struct radeon_crtc *test_radeon_crtc; @@ -1748,6 +1749,10 @@ static int radeon_get_shared_dp_ppll(struct drm_crtc *crtc) test_radeon_crtc = to_radeon_crtc(test_crtc); if (test_radeon_crtc->encoder && ENCODER_MODE_IS_DP(atombios_get_encoder_mode(test_radeon_crtc->encoder))) { + /* PPLL2 is exclusive to UNIPHYA on DCE61 */ + if (ASIC_IS_DCE61(rdev) && !ASIC_IS_DCE8(rdev) && + test_radeon_crtc->pll_id == ATOM_PPLL2) + continue; /* for DP use the same PLL for all */ if (test_radeon_crtc->pll_id != ATOM_PPLL_INVALID) return test_radeon_crtc->pll_id; @@ -1769,6 +1774,7 @@ static int radeon_get_shared_nondp_ppll(struct drm_crtc *crtc) { struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); struct drm_device *dev = crtc->dev; + struct radeon_device *rdev = dev->dev_private; struct drm_crtc *test_crtc; struct radeon_crtc *test_radeon_crtc; u32 adjusted_clock, test_adjusted_clock; @@ -1784,6 +1790,10 @@ static int radeon_get_shared_nondp_ppll(struct drm_crtc *crtc) test_radeon_crtc = to_radeon_crtc(test_crtc); if (test_radeon_crtc->encoder && !ENCODER_MODE_IS_DP(atombios_get_encoder_mode(test_radeon_crtc->encoder))) { + /* PPLL2 is exclusive to UNIPHYA on DCE61 */ + if (ASIC_IS_DCE61(rdev) && !ASIC_IS_DCE8(rdev) && + test_radeon_crtc->pll_id == ATOM_PPLL2) + continue; /* check if we are already driving this connector with another crtc */ if (test_radeon_crtc->connector == radeon_crtc->connector) { /* if we are, return that pll */ diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c index adf74f4366bb..0b04b9282f56 100644 --- a/drivers/gpu/drm/radeon/atombios_encoders.c +++ b/drivers/gpu/drm/radeon/atombios_encoders.c @@ -310,6 +310,10 @@ static bool radeon_atom_mode_fixup(struct drm_encoder *encoder, && (mode->crtc_vsync_start < (mode->crtc_vdisplay + 2))) adjusted_mode->crtc_vsync_start = adjusted_mode->crtc_vdisplay + 2; + /* vertical FP must be at least 1 */ + if (mode->crtc_vsync_start == mode->crtc_vdisplay) + adjusted_mode->crtc_vsync_start++; + /* get the native mode for scaling */ if (radeon_encoder->active_device & (ATOM_DEVICE_LCD_SUPPORT)) { radeon_panel_mode_fixup(encoder, adjusted_mode); diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index 2ad462896896..32491355a1d4 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -2608,10 +2608,152 @@ static void evergreen_agp_enable(struct radeon_device *rdev) WREG32(VM_CONTEXT1_CNTL, 0); } +static const unsigned ni_dig_offsets[] = +{ + NI_DIG0_REGISTER_OFFSET, + NI_DIG1_REGISTER_OFFSET, + NI_DIG2_REGISTER_OFFSET, + NI_DIG3_REGISTER_OFFSET, + NI_DIG4_REGISTER_OFFSET, + NI_DIG5_REGISTER_OFFSET +}; + +static const unsigned ni_tx_offsets[] = +{ + NI_DCIO_UNIPHY0_UNIPHY_TX_CONTROL1, + NI_DCIO_UNIPHY1_UNIPHY_TX_CONTROL1, + NI_DCIO_UNIPHY2_UNIPHY_TX_CONTROL1, + NI_DCIO_UNIPHY3_UNIPHY_TX_CONTROL1, + NI_DCIO_UNIPHY4_UNIPHY_TX_CONTROL1, + NI_DCIO_UNIPHY5_UNIPHY_TX_CONTROL1 +}; + +static const unsigned evergreen_dp_offsets[] = +{ + EVERGREEN_DP0_REGISTER_OFFSET, + EVERGREEN_DP1_REGISTER_OFFSET, + EVERGREEN_DP2_REGISTER_OFFSET, + EVERGREEN_DP3_REGISTER_OFFSET, + EVERGREEN_DP4_REGISTER_OFFSET, + EVERGREEN_DP5_REGISTER_OFFSET +}; + + +/* + * Assumption is that EVERGREEN_CRTC_MASTER_EN enable for requested crtc + * We go from crtc to connector and it is not relible since it + * should be an opposite direction .If crtc is enable then + * find the dig_fe which selects this crtc and insure that it enable. + * if such dig_fe is found then find dig_be which selects found dig_be and + * insure that it enable and in DP_SST mode. + * if UNIPHY_PLL_CONTROL1.enable then we should disconnect timing + * from dp symbols clocks . + */ +static bool evergreen_is_dp_sst_stream_enabled(struct radeon_device *rdev, + unsigned crtc_id, unsigned *ret_dig_fe) +{ + unsigned i; + unsigned dig_fe; + unsigned dig_be; + unsigned dig_en_be; + unsigned uniphy_pll; + unsigned digs_fe_selected; + unsigned dig_be_mode; + unsigned dig_fe_mask; + bool is_enabled = false; + bool found_crtc = false; + + /* loop through all running dig_fe to find selected crtc */ + for (i = 0; i < ARRAY_SIZE(ni_dig_offsets); i++) { + dig_fe = RREG32(NI_DIG_FE_CNTL + ni_dig_offsets[i]); + if (dig_fe & NI_DIG_FE_CNTL_SYMCLK_FE_ON && + crtc_id == NI_DIG_FE_CNTL_SOURCE_SELECT(dig_fe)) { + /* found running pipe */ + found_crtc = true; + dig_fe_mask = 1 << i; + dig_fe = i; + break; + } + } + + if (found_crtc) { + /* loop through all running dig_be to find selected dig_fe */ + for (i = 0; i < ARRAY_SIZE(ni_dig_offsets); i++) { + dig_be = RREG32(NI_DIG_BE_CNTL + ni_dig_offsets[i]); + /* if dig_fe_selected by dig_be? */ + digs_fe_selected = NI_DIG_BE_CNTL_FE_SOURCE_SELECT(dig_be); + dig_be_mode = NI_DIG_FE_CNTL_MODE(dig_be); + if (dig_fe_mask & digs_fe_selected && + /* if dig_be in sst mode? */ + dig_be_mode == NI_DIG_BE_DPSST) { + dig_en_be = RREG32(NI_DIG_BE_EN_CNTL + + ni_dig_offsets[i]); + uniphy_pll = RREG32(NI_DCIO_UNIPHY0_PLL_CONTROL1 + + ni_tx_offsets[i]); + /* dig_be enable and tx is running */ + if (dig_en_be & NI_DIG_BE_EN_CNTL_ENABLE && + dig_en_be & NI_DIG_BE_EN_CNTL_SYMBCLK_ON && + uniphy_pll & NI_DCIO_UNIPHY0_PLL_CONTROL1_ENABLE) { + is_enabled = true; + *ret_dig_fe = dig_fe; + break; + } + } + } + } + + return is_enabled; +} + +/* + * Blank dig when in dp sst mode + * Dig ignores crtc timing + */ +static void evergreen_blank_dp_output(struct radeon_device *rdev, + unsigned dig_fe) +{ + unsigned stream_ctrl; + unsigned fifo_ctrl; + unsigned counter = 0; + + if (dig_fe >= ARRAY_SIZE(evergreen_dp_offsets)) { + DRM_ERROR("invalid dig_fe %d\n", dig_fe); + return; + } + + stream_ctrl = RREG32(EVERGREEN_DP_VID_STREAM_CNTL + + evergreen_dp_offsets[dig_fe]); + if (!(stream_ctrl & EVERGREEN_DP_VID_STREAM_CNTL_ENABLE)) { + DRM_ERROR("dig %d , should be enable\n", dig_fe); + return; + } + + stream_ctrl &=~EVERGREEN_DP_VID_STREAM_CNTL_ENABLE; + WREG32(EVERGREEN_DP_VID_STREAM_CNTL + + evergreen_dp_offsets[dig_fe], stream_ctrl); + + stream_ctrl = RREG32(EVERGREEN_DP_VID_STREAM_CNTL + + evergreen_dp_offsets[dig_fe]); + while (counter < 32 && stream_ctrl & EVERGREEN_DP_VID_STREAM_STATUS) { + msleep(1); + counter++; + stream_ctrl = RREG32(EVERGREEN_DP_VID_STREAM_CNTL + + evergreen_dp_offsets[dig_fe]); + } + if (counter >= 32 ) + DRM_ERROR("counter exceeds %d\n", counter); + + fifo_ctrl = RREG32(EVERGREEN_DP_STEER_FIFO + evergreen_dp_offsets[dig_fe]); + fifo_ctrl |= EVERGREEN_DP_STEER_FIFO_RESET; + WREG32(EVERGREEN_DP_STEER_FIFO + evergreen_dp_offsets[dig_fe], fifo_ctrl); + +} + void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_save *save) { u32 crtc_enabled, tmp, frame_count, blackout; int i, j; + unsigned dig_fe; if (!ASIC_IS_NODCE(rdev)) { save->vga_render_control = RREG32(VGA_RENDER_CONTROL); @@ -2651,7 +2793,17 @@ void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_save *sav break; udelay(1); } - + /*we should disable dig if it drives dp sst*/ + /*but we are in radeon_device_init and the topology is unknown*/ + /*and it is available after radeon_modeset_init*/ + /*the following method radeon_atom_encoder_dpms_dig*/ + /*does the job if we initialize it properly*/ + /*for now we do it this manually*/ + /**/ + if (ASIC_IS_DCE5(rdev) && + evergreen_is_dp_sst_stream_enabled(rdev, i ,&dig_fe)) + evergreen_blank_dp_output(rdev, dig_fe); + /*we could remove 6 lines below*/ /* XXX this is a hack to avoid strange behavior with EFI on certain systems */ WREG32(EVERGREEN_CRTC_UPDATE_LOCK + crtc_offsets[i], 1); tmp = RREG32(EVERGREEN_CRTC_CONTROL + crtc_offsets[i]); diff --git a/drivers/gpu/drm/radeon/evergreen_reg.h b/drivers/gpu/drm/radeon/evergreen_reg.h index aa939dfed3a3..b436badf9efa 100644 --- a/drivers/gpu/drm/radeon/evergreen_reg.h +++ b/drivers/gpu/drm/radeon/evergreen_reg.h @@ -250,8 +250,43 @@ /* HDMI blocks at 0x7030, 0x7c30, 0x10830, 0x11430, 0x12030, 0x12c30 */ #define EVERGREEN_HDMI_BASE 0x7030 +/*DIG block*/ +#define NI_DIG0_REGISTER_OFFSET (0x7000 - 0x7000) +#define NI_DIG1_REGISTER_OFFSET (0x7C00 - 0x7000) +#define NI_DIG2_REGISTER_OFFSET (0x10800 - 0x7000) +#define NI_DIG3_REGISTER_OFFSET (0x11400 - 0x7000) +#define NI_DIG4_REGISTER_OFFSET (0x12000 - 0x7000) +#define NI_DIG5_REGISTER_OFFSET (0x12C00 - 0x7000) + + +#define NI_DIG_FE_CNTL 0x7000 +# define NI_DIG_FE_CNTL_SOURCE_SELECT(x) ((x) & 0x3) +# define NI_DIG_FE_CNTL_SYMCLK_FE_ON (1<<24) + + +#define NI_DIG_BE_CNTL 0x7140 +# define NI_DIG_BE_CNTL_FE_SOURCE_SELECT(x) (((x) >> 8 ) & 0x3F) +# define NI_DIG_FE_CNTL_MODE(x) (((x) >> 16) & 0x7 ) + +#define NI_DIG_BE_EN_CNTL 0x7144 +# define NI_DIG_BE_EN_CNTL_ENABLE (1 << 0) +# define NI_DIG_BE_EN_CNTL_SYMBCLK_ON (1 << 8) +# define NI_DIG_BE_DPSST 0 /* Display Port block */ +#define EVERGREEN_DP0_REGISTER_OFFSET (0x730C - 0x730C) +#define EVERGREEN_DP1_REGISTER_OFFSET (0x7F0C - 0x730C) +#define EVERGREEN_DP2_REGISTER_OFFSET (0x10B0C - 0x730C) +#define EVERGREEN_DP3_REGISTER_OFFSET (0x1170C - 0x730C) +#define EVERGREEN_DP4_REGISTER_OFFSET (0x1230C - 0x730C) +#define EVERGREEN_DP5_REGISTER_OFFSET (0x12F0C - 0x730C) + + +#define EVERGREEN_DP_VID_STREAM_CNTL 0x730C +# define EVERGREEN_DP_VID_STREAM_CNTL_ENABLE (1 << 0) +# define EVERGREEN_DP_VID_STREAM_STATUS (1 <<16) +#define EVERGREEN_DP_STEER_FIFO 0x7310 +# define EVERGREEN_DP_STEER_FIFO_RESET (1 << 0) #define EVERGREEN_DP_SEC_CNTL 0x7280 # define EVERGREEN_DP_SEC_STREAM_ENABLE (1 << 0) # define EVERGREEN_DP_SEC_ASP_ENABLE (1 << 4) @@ -266,4 +301,15 @@ # define EVERGREEN_DP_SEC_N_BASE_MULTIPLE(x) (((x) & 0xf) << 24) # define EVERGREEN_DP_SEC_SS_EN (1 << 28) +/*DCIO_UNIPHY block*/ +#define NI_DCIO_UNIPHY0_UNIPHY_TX_CONTROL1 (0x6600 -0x6600) +#define NI_DCIO_UNIPHY1_UNIPHY_TX_CONTROL1 (0x6640 -0x6600) +#define NI_DCIO_UNIPHY2_UNIPHY_TX_CONTROL1 (0x6680 - 0x6600) +#define NI_DCIO_UNIPHY3_UNIPHY_TX_CONTROL1 (0x66C0 - 0x6600) +#define NI_DCIO_UNIPHY4_UNIPHY_TX_CONTROL1 (0x6700 - 0x6600) +#define NI_DCIO_UNIPHY5_UNIPHY_TX_CONTROL1 (0x6740 - 0x6600) + +#define NI_DCIO_UNIPHY0_PLL_CONTROL1 0x6618 +# define NI_DCIO_UNIPHY0_PLL_CONTROL1_ENABLE (1 << 0) + #endif diff --git a/drivers/gpu/drm/radeon/radeon_atpx_handler.c b/drivers/gpu/drm/radeon/radeon_atpx_handler.c index 9bc408c9f9f6..c4b4f298a283 100644 --- a/drivers/gpu/drm/radeon/radeon_atpx_handler.c +++ b/drivers/gpu/drm/radeon/radeon_atpx_handler.c @@ -62,10 +62,6 @@ bool radeon_has_atpx(void) { return radeon_atpx_priv.atpx_detected; } -bool radeon_has_atpx_dgpu_power_cntl(void) { - return radeon_atpx_priv.atpx.functions.power_cntl; -} - /** * radeon_atpx_call - call an ATPX method * @@ -145,6 +141,10 @@ static void radeon_atpx_parse_functions(struct radeon_atpx_functions *f, u32 mas */ static int radeon_atpx_validate(struct radeon_atpx *atpx) { + /* make sure required functions are enabled */ + /* dGPU power control is required */ + atpx->functions.power_cntl = true; + if (atpx->functions.px_params) { union acpi_object *info; struct atpx_px_params output; diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index 340f3f549f29..9cfc1c3e1965 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -1996,10 +1996,12 @@ radeon_add_atom_connector(struct drm_device *dev, rdev->mode_info.dither_property, RADEON_FMT_DITHER_DISABLE); - if (radeon_audio != 0) + if (radeon_audio != 0) { drm_object_attach_property(&radeon_connector->base.base, rdev->mode_info.audio_property, RADEON_AUDIO_AUTO); + radeon_connector->audio = RADEON_AUDIO_AUTO; + } if (ASIC_IS_DCE5(rdev)) drm_object_attach_property(&radeon_connector->base.base, rdev->mode_info.output_csc_property, @@ -2124,6 +2126,7 @@ radeon_add_atom_connector(struct drm_device *dev, drm_object_attach_property(&radeon_connector->base.base, rdev->mode_info.audio_property, RADEON_AUDIO_AUTO); + radeon_connector->audio = RADEON_AUDIO_AUTO; } if (connector_type == DRM_MODE_CONNECTOR_DVII) { radeon_connector->dac_load_detect = true; @@ -2179,6 +2182,7 @@ radeon_add_atom_connector(struct drm_device *dev, drm_object_attach_property(&radeon_connector->base.base, rdev->mode_info.audio_property, RADEON_AUDIO_AUTO); + radeon_connector->audio = RADEON_AUDIO_AUTO; } if (ASIC_IS_DCE5(rdev)) drm_object_attach_property(&radeon_connector->base.base, @@ -2231,6 +2235,7 @@ radeon_add_atom_connector(struct drm_device *dev, drm_object_attach_property(&radeon_connector->base.base, rdev->mode_info.audio_property, RADEON_AUDIO_AUTO); + radeon_connector->audio = RADEON_AUDIO_AUTO; } if (ASIC_IS_DCE5(rdev)) drm_object_attach_property(&radeon_connector->base.base, diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index f78f111e68de..e2dd5d19c32c 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -103,12 +103,6 @@ static const char radeon_family_name[][16] = { "LAST", }; -#if defined(CONFIG_VGA_SWITCHEROO) -bool radeon_has_atpx_dgpu_power_cntl(void); -#else -static inline bool radeon_has_atpx_dgpu_power_cntl(void) { return false; } -#endif - #define RADEON_PX_QUIRK_DISABLE_PX (1 << 0) #define RADEON_PX_QUIRK_LONG_WAKEUP (1 << 1) @@ -636,6 +630,23 @@ void radeon_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc) /* * GPU helpers function. */ + +/** + * radeon_device_is_virtual - check if we are running is a virtual environment + * + * Check if the asic has been passed through to a VM (all asics). + * Used at driver startup. + * Returns true if virtual or false if not. + */ +static bool radeon_device_is_virtual(void) +{ +#ifdef CONFIG_X86 + return boot_cpu_has(X86_FEATURE_HYPERVISOR); +#else + return false; +#endif +} + /** * radeon_card_posted - check if the hw has already been initialized * @@ -649,6 +660,10 @@ bool radeon_card_posted(struct radeon_device *rdev) { uint32_t reg; + /* for pass through, always force asic_init */ + if (radeon_device_is_virtual()) + return false; + /* required for EFI mode on macbook2,1 which uses an r5xx asic */ if (efi_enabled(EFI_BOOT) && (rdev->pdev->subsystem_vendor == PCI_VENDOR_ID_APPLE) && @@ -1439,7 +1454,7 @@ int radeon_device_init(struct radeon_device *rdev, * ignore it */ vga_client_register(rdev->pdev, rdev, NULL, radeon_vga_set_decode); - if ((rdev->flags & RADEON_IS_PX) && radeon_has_atpx_dgpu_power_cntl()) + if (rdev->flags & RADEON_IS_PX) runtime = true; vga_switcheroo_register_client(rdev->pdev, &radeon_switcheroo_ops, runtime); if (runtime) diff --git a/drivers/gpu/drm/radeon/radeon_dp_auxch.c b/drivers/gpu/drm/radeon/radeon_dp_auxch.c index 3b0c229d7dcd..db64e0062689 100644 --- a/drivers/gpu/drm/radeon/radeon_dp_auxch.c +++ b/drivers/gpu/drm/radeon/radeon_dp_auxch.c @@ -105,7 +105,7 @@ radeon_dp_aux_transfer_native(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg tmp &= AUX_HPD_SEL(0x7); tmp |= AUX_HPD_SEL(chan->rec.hpd); - tmp |= AUX_EN | AUX_LS_READ_EN; + tmp |= AUX_EN | AUX_LS_READ_EN | AUX_HPD_DISCON(0x1); WREG32(AUX_CONTROL + aux_offset[instance], tmp); diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index e06ac546a90f..f342aad79cc6 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -235,6 +235,8 @@ static int radeon_verify_access(struct ttm_buffer_object *bo, struct file *filp) { struct radeon_bo *rbo = container_of(bo, struct radeon_bo, tbo); + if (radeon_ttm_tt_has_userptr(bo->ttm)) + return -EPERM; return drm_vma_node_verify_access(&rbo->gem_base.vma_node, filp); } diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index 7285adb27099..caa73de584a5 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -2931,6 +2931,7 @@ static struct si_dpm_quirk si_dpm_quirk_list[] = { { PCI_VENDOR_ID_ATI, 0x6811, 0x1462, 0x2015, 0, 120000 }, { PCI_VENDOR_ID_ATI, 0x6811, 0x1043, 0x2015, 0, 120000 }, { PCI_VENDOR_ID_ATI, 0x6811, 0x148c, 0x2015, 0, 120000 }, + { PCI_VENDOR_ID_ATI, 0x6810, 0x1682, 0x9275, 0, 120000 }, { 0, 0, 0, 0 }, }; diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index 745e996d2dbc..4ae8b56b1847 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -1004,9 +1004,9 @@ out_unlock: return ret; } -static bool ttm_bo_mem_compat(struct ttm_placement *placement, - struct ttm_mem_reg *mem, - uint32_t *new_flags) +bool ttm_bo_mem_compat(struct ttm_placement *placement, + struct ttm_mem_reg *mem, + uint32_t *new_flags) { int i; @@ -1038,6 +1038,7 @@ static bool ttm_bo_mem_compat(struct ttm_placement *placement, return false; } +EXPORT_SYMBOL(ttm_bo_mem_compat); int ttm_bo_validate(struct ttm_buffer_object *bo, struct ttm_placement *placement, diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_dmabuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_dmabuf.c index 299925a1f6c6..eadc981ee79a 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_dmabuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_dmabuf.c @@ -49,6 +49,7 @@ int vmw_dmabuf_pin_in_placement(struct vmw_private *dev_priv, { struct ttm_buffer_object *bo = &buf->base; int ret; + uint32_t new_flags; ret = ttm_write_lock(&dev_priv->reservation_sem, interruptible); if (unlikely(ret != 0)) @@ -60,7 +61,12 @@ int vmw_dmabuf_pin_in_placement(struct vmw_private *dev_priv, if (unlikely(ret != 0)) goto err; - ret = ttm_bo_validate(bo, placement, interruptible, false); + if (buf->pin_count > 0) + ret = ttm_bo_mem_compat(placement, &bo->mem, + &new_flags) == true ? 0 : -EINVAL; + else + ret = ttm_bo_validate(bo, placement, interruptible, false); + if (!ret) vmw_bo_pin_reserved(buf, true); @@ -91,6 +97,7 @@ int vmw_dmabuf_pin_in_vram_or_gmr(struct vmw_private *dev_priv, { struct ttm_buffer_object *bo = &buf->base; int ret; + uint32_t new_flags; ret = ttm_write_lock(&dev_priv->reservation_sem, interruptible); if (unlikely(ret != 0)) @@ -102,6 +109,12 @@ int vmw_dmabuf_pin_in_vram_or_gmr(struct vmw_private *dev_priv, if (unlikely(ret != 0)) goto err; + if (buf->pin_count > 0) { + ret = ttm_bo_mem_compat(&vmw_vram_gmr_placement, &bo->mem, + &new_flags) == true ? 0 : -EINVAL; + goto out_unreserve; + } + ret = ttm_bo_validate(bo, &vmw_vram_gmr_placement, interruptible, false); if (likely(ret == 0) || ret == -ERESTARTSYS) @@ -161,6 +174,7 @@ int vmw_dmabuf_pin_in_start_of_vram(struct vmw_private *dev_priv, struct ttm_placement placement; struct ttm_place place; int ret = 0; + uint32_t new_flags; place = vmw_vram_placement.placement[0]; place.lpfn = bo->num_pages; @@ -185,10 +199,15 @@ int vmw_dmabuf_pin_in_start_of_vram(struct vmw_private *dev_priv, */ if (bo->mem.mem_type == TTM_PL_VRAM && bo->mem.start < bo->num_pages && - bo->mem.start > 0) + bo->mem.start > 0 && + buf->pin_count == 0) (void) ttm_bo_validate(bo, &vmw_sys_placement, false, false); - ret = ttm_bo_validate(bo, &placement, interruptible, false); + if (buf->pin_count > 0) + ret = ttm_bo_mem_compat(&placement, &bo->mem, + &new_flags) == true ? 0 : -EINVAL; + else + ret = ttm_bo_validate(bo, &placement, interruptible, false); /* For some reason we didn't end up at the start of vram */ WARN_ON(ret == 0 && bo->offset != 0); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index 24fb348a44e1..f3f31f995878 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -227,6 +227,7 @@ static int vmw_force_iommu; static int vmw_restrict_iommu; static int vmw_force_coherent; static int vmw_restrict_dma_mask; +static int vmw_assume_16bpp; static int vmw_probe(struct pci_dev *, const struct pci_device_id *); static void vmw_master_init(struct vmw_master *); @@ -243,6 +244,8 @@ MODULE_PARM_DESC(force_coherent, "Force coherent TTM pages"); module_param_named(force_coherent, vmw_force_coherent, int, 0600); MODULE_PARM_DESC(restrict_dma_mask, "Restrict DMA mask to 44 bits with IOMMU"); module_param_named(restrict_dma_mask, vmw_restrict_dma_mask, int, 0600); +MODULE_PARM_DESC(assume_16bpp, "Assume 16-bpp when filtering modes"); +module_param_named(assume_16bpp, vmw_assume_16bpp, int, 0600); static void vmw_print_capabilities(uint32_t capabilities) @@ -652,6 +655,8 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) dev_priv->vram_start = pci_resource_start(dev->pdev, 1); dev_priv->mmio_start = pci_resource_start(dev->pdev, 2); + dev_priv->assume_16bpp = !!vmw_assume_16bpp; + dev_priv->enable_fb = enable_fbdev; vmw_write(dev_priv, SVGA_REG_ID, SVGA_ID_2); @@ -698,6 +703,13 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) vmw_read(dev_priv, SVGA_REG_SUGGESTED_GBOBJECT_MEM_SIZE_KB); + /* + * Workaround for low memory 2D VMs to compensate for the + * allocation taken by fbdev + */ + if (!(dev_priv->capabilities & SVGA_CAP_3D)) + mem_size *= 2; + dev_priv->max_mob_pages = mem_size * 1024 / PAGE_SIZE; dev_priv->prim_bb_mem = vmw_read(dev_priv, diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index 469cdd520615..2e94fe27b3f6 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -387,6 +387,7 @@ struct vmw_private { spinlock_t hw_lock; spinlock_t cap_lock; bool has_dx; + bool assume_16bpp; /* * VGA registers. diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 5da5de0cb522..4948c1529836 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -3273,19 +3273,19 @@ static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = { &vmw_cmd_dx_cid_check, true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_DX_DEFINE_QUERY, &vmw_cmd_dx_define_query, true, false, true), - VMW_CMD_DEF(SVGA_3D_CMD_DX_DESTROY_QUERY, &vmw_cmd_ok, + VMW_CMD_DEF(SVGA_3D_CMD_DX_DESTROY_QUERY, &vmw_cmd_dx_cid_check, true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_DX_BIND_QUERY, &vmw_cmd_dx_bind_query, true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_QUERY_OFFSET, - &vmw_cmd_ok, true, false, true), - VMW_CMD_DEF(SVGA_3D_CMD_DX_BEGIN_QUERY, &vmw_cmd_ok, + &vmw_cmd_dx_cid_check, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_BEGIN_QUERY, &vmw_cmd_dx_cid_check, true, false, true), - VMW_CMD_DEF(SVGA_3D_CMD_DX_END_QUERY, &vmw_cmd_ok, + VMW_CMD_DEF(SVGA_3D_CMD_DX_END_QUERY, &vmw_cmd_dx_cid_check, true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_DX_READBACK_QUERY, &vmw_cmd_invalid, true, false, true), - VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_PREDICATION, &vmw_cmd_invalid, + VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_PREDICATION, &vmw_cmd_dx_cid_check, true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_DX_SET_VIEWPORTS, &vmw_cmd_dx_cid_check, true, false, true), diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c index 3b1faf7862a5..d2d93959b119 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c @@ -517,28 +517,6 @@ static int vmw_fb_kms_framebuffer(struct fb_info *info) par->set_fb = &vfb->base; - if (!par->bo_ptr) { - /* - * Pin before mapping. Since we don't know in what placement - * to pin, call into KMS to do it for us. - */ - ret = vfb->pin(vfb); - if (ret) { - DRM_ERROR("Could not pin the fbdev framebuffer.\n"); - return ret; - } - - ret = ttm_bo_kmap(&par->vmw_bo->base, 0, - par->vmw_bo->base.num_pages, &par->map); - if (ret) { - vfb->unpin(vfb); - DRM_ERROR("Could not map the fbdev framebuffer.\n"); - return ret; - } - - par->bo_ptr = ttm_kmap_obj_virtual(&par->map, &par->bo_iowrite); - } - return 0; } @@ -573,9 +551,9 @@ static int vmw_fb_set_par(struct fb_info *info) mode = old_mode; old_mode = NULL; } else if (!vmw_kms_validate_mode_vram(vmw_priv, - mode->hdisplay * - (var->bits_per_pixel + 7) / 8, - mode->vdisplay)) { + mode->hdisplay * + DIV_ROUND_UP(var->bits_per_pixel, 8), + mode->vdisplay)) { drm_mode_destroy(vmw_priv->dev, mode); return -EINVAL; } @@ -601,6 +579,31 @@ static int vmw_fb_set_par(struct fb_info *info) if (ret) goto out_unlock; + if (!par->bo_ptr) { + struct vmw_framebuffer *vfb = vmw_framebuffer_to_vfb(set.fb); + + /* + * Pin before mapping. Since we don't know in what placement + * to pin, call into KMS to do it for us. + */ + ret = vfb->pin(vfb); + if (ret) { + DRM_ERROR("Could not pin the fbdev framebuffer.\n"); + goto out_unlock; + } + + ret = ttm_bo_kmap(&par->vmw_bo->base, 0, + par->vmw_bo->base.num_pages, &par->map); + if (ret) { + vfb->unpin(vfb); + DRM_ERROR("Could not map the fbdev framebuffer.\n"); + goto out_unlock; + } + + par->bo_ptr = ttm_kmap_obj_virtual(&par->map, &par->bo_iowrite); + } + + vmw_fb_dirty_mark(par, par->fb_x, par->fb_y, par->set_fb->width, par->set_fb->height); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index 7c2e118a77b0..060e5c6f4446 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -1538,14 +1538,10 @@ int vmw_du_connector_fill_modes(struct drm_connector *connector, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC) }; int i; - u32 assumed_bpp = 2; + u32 assumed_bpp = 4; - /* - * If using screen objects, then assume 32-bpp because that's what the - * SVGA device is assuming - */ - if (dev_priv->active_display_unit == vmw_du_screen_object) - assumed_bpp = 4; + if (dev_priv->assume_16bpp) + assumed_bpp = 2; if (dev_priv->active_display_unit == vmw_du_screen_target) { max_width = min(max_width, dev_priv->stdu_max_width); diff --git a/drivers/gpu/ipu-v3/ipu-common.c b/drivers/gpu/ipu-v3/ipu-common.c index a0e28f3a278d..5030cba4a581 100644 --- a/drivers/gpu/ipu-v3/ipu-common.c +++ b/drivers/gpu/ipu-v3/ipu-common.c @@ -997,7 +997,7 @@ struct ipu_platform_reg { }; /* These must be in the order of the corresponding device tree port nodes */ -static const struct ipu_platform_reg client_reg[] = { +static struct ipu_platform_reg client_reg[] = { { .pdata = { .csi = 0, @@ -1048,7 +1048,7 @@ static int ipu_add_client_devices(struct ipu_soc *ipu, unsigned long ipu_base) mutex_unlock(&ipu_client_id_mutex); for (i = 0; i < ARRAY_SIZE(client_reg); i++) { - const struct ipu_platform_reg *reg = &client_reg[i]; + struct ipu_platform_reg *reg = &client_reg[i]; struct platform_device *pdev; struct device_node *of_node; @@ -1068,9 +1068,9 @@ static int ipu_add_client_devices(struct ipu_soc *ipu, unsigned long ipu_base) goto err_register; } - pdev->dev.of_node = of_node; pdev->dev.parent = dev; + reg->pdata.of_node = of_node; ret = platform_device_add_data(pdev, ®->pdata, sizeof(reg->pdata)); if (!ret) @@ -1079,6 +1079,12 @@ static int ipu_add_client_devices(struct ipu_soc *ipu, unsigned long ipu_base) platform_device_put(pdev); goto err_register; } + + /* + * Set of_node only after calling platform_device_add. Otherwise + * the platform:imx-ipuv3-crtc modalias won't be used. + */ + pdev->dev.of_node = of_node; } return 0; diff --git a/drivers/gpu/msm/a5xx_reg.h b/drivers/gpu/msm/a5xx_reg.h index f3b4e6622043..ef2861ca96dd 100644 --- a/drivers/gpu/msm/a5xx_reg.h +++ b/drivers/gpu/msm/a5xx_reg.h @@ -608,6 +608,7 @@ #define A5XX_PC_PERFCTR_PC_SEL_7 0xD17 /* HLSQ registers */ +#define A5XX_HLSQ_DBG_ECO_CNTL 0xE04 #define A5XX_HLSQ_ADDR_MODE_CNTL 0xE05 #define A5XX_HLSQ_PERFCTR_HLSQ_SEL_0 0xE10 #define A5XX_HLSQ_PERFCTR_HLSQ_SEL_1 0xE11 @@ -632,6 +633,7 @@ #define A5XX_VFD_PERFCTR_VFD_SEL_7 0xE57 /* VPC registers */ +#define A5XX_VPC_DBG_ECO_CNTL 0xE60 #define A5XX_VPC_ADDR_MODE_CNTL 0xE61 #define A5XX_VPC_PERFCTR_VPC_SEL_0 0xE64 #define A5XX_VPC_PERFCTR_VPC_SEL_1 0xE65 @@ -891,6 +893,11 @@ #define A5XX_GDPM_INT_MASK 0xB811 #define A5XX_GPMU_BEC_ENABLE 0xB9A0 +/* ISENSE registers */ +#define A5XX_GPU_CS_DECIMAL_ALIGN 0xC16A +#define A5XX_GPU_CS_SENSOR_PARAM_CORE_1 0xC126 +#define A5XX_GPU_CS_SENSOR_PARAM_CORE_2 0xC127 +#define A5XX_GPU_CS_SW_OV_FUSE_EN 0xC168 #define A5XX_GPU_CS_SENSOR_GENERAL_STATUS 0xC41A #define A5XX_GPU_CS_AMP_CALIBRATION_STATUS1_0 0xC41D #define A5XX_GPU_CS_AMP_CALIBRATION_STATUS1_2 0xC41F @@ -898,5 +905,6 @@ #define A5XX_GPU_CS_ENABLE_REG 0xC520 #define A5XX_GPU_CS_AMP_CALIBRATION_CONTROL1 0xC557 #define A5XX_GPU_CS_AMP_CALIBRATION_DONE 0xC565 +#define A5XX_GPU_CS_ENDPOINT_CALIBRATION_DONE 0xC556 #endif /* _A5XX_REG_H */ diff --git a/drivers/gpu/msm/adreno-gpulist.h b/drivers/gpu/msm/adreno-gpulist.h index 3615be45b6d9..a02ed40ba9d5 100644 --- a/drivers/gpu/msm/adreno-gpulist.h +++ b/drivers/gpu/msm/adreno-gpulist.h @@ -269,7 +269,7 @@ static const struct adreno_gpu_core adreno_gpulist[] = { .patchid = ANY_ID, .features = ADRENO_PREEMPTION | ADRENO_64BIT | ADRENO_CONTENT_PROTECTION | - ADRENO_GPMU | ADRENO_SPTP_PC, + ADRENO_GPMU | ADRENO_SPTP_PC | ADRENO_LM, .pm4fw_name = "a530_pm4.fw", .pfpfw_name = "a530_pfp.fw", .zap_name = "a540_zap", diff --git a/drivers/gpu/msm/adreno.c b/drivers/gpu/msm/adreno.c index 9940f7a7c2b7..11226472d801 100644 --- a/drivers/gpu/msm/adreno.c +++ b/drivers/gpu/msm/adreno.c @@ -589,11 +589,21 @@ static irqreturn_t adreno_irq_handler(struct kgsl_device *device) struct adreno_gpudev *gpudev = ADRENO_GPU_DEVICE(adreno_dev); struct adreno_irq *irq_params = gpudev->irq; irqreturn_t ret = IRQ_NONE; - unsigned int status = 0, tmp; + unsigned int status = 0, tmp, int_bit; int i; adreno_readreg(adreno_dev, ADRENO_REG_RBBM_INT_0_STATUS, &status); + /* + * Clear all the interrupt bits but ADRENO_INT_RBBM_AHB_ERROR. Because + * even if we clear it here, it will stay high until it is cleared + * in its respective handler. Otherwise, the interrupt handler will + * fire again. + */ + int_bit = ADRENO_INT_BIT(adreno_dev, ADRENO_INT_RBBM_AHB_ERROR); + adreno_writereg(adreno_dev, ADRENO_REG_RBBM_INT_CLEAR_CMD, + status & ~int_bit); + /* Loop through all set interrupts and call respective handlers */ for (tmp = status; tmp != 0;) { i = fls(tmp) - 1; @@ -612,9 +622,14 @@ static irqreturn_t adreno_irq_handler(struct kgsl_device *device) gpudev->irq_trace(adreno_dev, status); - if (status) + /* + * Clear ADRENO_INT_RBBM_AHB_ERROR bit after this interrupt has been + * cleared in its respective handler + */ + if (status & int_bit) adreno_writereg(adreno_dev, ADRENO_REG_RBBM_INT_CLEAR_CMD, - status); + int_bit); + return ret; } @@ -828,6 +843,8 @@ static struct { { ADRENO_QUIRK_FAULT_DETECT_MASK, "qcom,gpu-quirk-fault-detect-mask" }, { ADRENO_QUIRK_DISABLE_RB_DP2CLOCKGATING, "qcom,gpu-quirk-dp2clockgating-disable" }, + { ADRENO_QUIRK_DISABLE_LMLOADKILL, + "qcom,gpu-quirk-lmloadkill-disable" }, }; static int adreno_of_get_power(struct adreno_device *adreno_dev, @@ -2094,8 +2111,6 @@ static int adreno_soft_reset(struct kgsl_device *device) adreno_support_64bit(adreno_dev)) gpudev->enable_64bit(adreno_dev); - /* Restore physical performance counter values after soft reset */ - adreno_perfcounter_restore(adreno_dev); /* Reinitialize the GPU */ gpudev->start(adreno_dev); @@ -2122,6 +2137,9 @@ static int adreno_soft_reset(struct kgsl_device *device) set_bit(ADRENO_DEVICE_STARTED, &adreno_dev->priv); } + /* Restore physical performance counter values after soft reset */ + adreno_perfcounter_restore(adreno_dev); + return ret; } @@ -2276,9 +2294,9 @@ static void adreno_read(struct kgsl_device *device, void __iomem *base, unsigned int mem_len) { - unsigned int __iomem *reg; + void __iomem *reg; BUG_ON(offsetwords*sizeof(uint32_t) >= mem_len); - reg = (unsigned int __iomem *)(base + (offsetwords << 2)); + reg = (base + (offsetwords << 2)); if (!in_interrupt()) kgsl_pre_hwaccess(device); @@ -2318,7 +2336,7 @@ static void adreno_regwrite(struct kgsl_device *device, unsigned int offsetwords, unsigned int value) { - unsigned int __iomem *reg; + void __iomem *reg; BUG_ON(offsetwords*sizeof(uint32_t) >= device->reg_len); @@ -2328,7 +2346,7 @@ static void adreno_regwrite(struct kgsl_device *device, trace_kgsl_regwrite(device, offsetwords, value); kgsl_cffdump_regwrite(device, offsetwords << 2, value); - reg = (unsigned int __iomem *)(device->reg_virt + (offsetwords << 2)); + reg = (device->reg_virt + (offsetwords << 2)); /*ensure previous writes post before this one, * i.e. act like normal writel() */ diff --git a/drivers/gpu/msm/adreno.h b/drivers/gpu/msm/adreno.h index 0f3403cb0095..d81142db5b58 100644 --- a/drivers/gpu/msm/adreno.h +++ b/drivers/gpu/msm/adreno.h @@ -123,6 +123,8 @@ #define ADRENO_QUIRK_FAULT_DETECT_MASK BIT(3) /* Disable RB sampler datapath clock gating optimization */ #define ADRENO_QUIRK_DISABLE_RB_DP2CLOCKGATING BIT(4) +/* Disable local memory(LM) feature to avoid corner case error */ +#define ADRENO_QUIRK_DISABLE_LMLOADKILL BIT(5) /* Flags to control command packet settings */ #define KGSL_CMD_FLAGS_NONE 0 @@ -198,6 +200,10 @@ struct adreno_gpudev; /* Time to allow preemption to complete (in ms) */ #define ADRENO_PREEMPT_TIMEOUT 10000 +#define ADRENO_INT_BIT(a, _bit) (((a)->gpucore->gpudev->int_bits) ? \ + (adreno_get_int(a, _bit) < 0 ? 0 : \ + BIT(adreno_get_int(a, _bit))) : 0) + /** * enum adreno_preempt_states * ADRENO_PREEMPT_NONE: No preemption is scheduled @@ -574,6 +580,11 @@ enum adreno_regs { ADRENO_REG_REGISTER_MAX, }; +enum adreno_int_bits { + ADRENO_INT_RBBM_AHB_ERROR, + ADRENO_INT_BITS_MAX, +}; + /** * adreno_reg_offsets: Holds array of register offsets * @offsets: Offset array of size defined by enum adreno_regs @@ -589,6 +600,7 @@ struct adreno_reg_offsets { #define ADRENO_REG_UNUSED 0xFFFFFFFF #define ADRENO_REG_SKIP 0xFFFFFFFE #define ADRENO_REG_DEFINE(_offset, _reg) [_offset] = _reg +#define ADRENO_INT_DEFINE(_offset, _val) ADRENO_REG_DEFINE(_offset, _val) /* * struct adreno_vbif_data - Describes vbif register value pair @@ -726,6 +738,7 @@ struct adreno_gpudev { * so define them in the structure and use them as variables. */ const struct adreno_reg_offsets *reg_offsets; + unsigned int *const int_bits; const struct adreno_ft_perf_counters *ft_perf_counters; unsigned int ft_perf_counters_count; @@ -1101,6 +1114,23 @@ static inline unsigned int adreno_getreg(struct adreno_device *adreno_dev, return gpudev->reg_offsets->offsets[offset_name]; } +/* + * adreno_get_int() - Returns the offset value of an interrupt bit from + * the interrupt bit array in the gpudev node + * @adreno_dev: Pointer to the the adreno device + * @bit_name: The interrupt bit enum whose bit is returned + */ +static inline unsigned int adreno_get_int(struct adreno_device *adreno_dev, + enum adreno_int_bits bit_name) +{ + struct adreno_gpudev *gpudev = ADRENO_GPU_DEVICE(adreno_dev); + + if (bit_name >= ADRENO_INT_BITS_MAX) + return -ERANGE; + + return gpudev->int_bits[bit_name]; +} + /** * adreno_gpu_fault() - Return the current state of the GPU * @adreno_dev: A pointer to the adreno_device to query diff --git a/drivers/gpu/msm/adreno_a3xx.c b/drivers/gpu/msm/adreno_a3xx.c index 97e71464c2df..423071811b43 100644 --- a/drivers/gpu/msm/adreno_a3xx.c +++ b/drivers/gpu/msm/adreno_a3xx.c @@ -151,6 +151,43 @@ static const unsigned int _a3xx_pwron_fixup_fs_instructions[] = { 0x00000000, 0x03000000, 0x00000000, 0x00000000, }; +static void a3xx_efuse_speed_bin(struct adreno_device *adreno_dev) +{ + unsigned int val; + unsigned int speed_bin[3]; + struct kgsl_device *device = &adreno_dev->dev; + + if (of_property_read_u32_array(device->pdev->dev.of_node, + "qcom,gpu-speed-bin", speed_bin, 3)) + return; + + adreno_efuse_read_u32(adreno_dev, speed_bin[0], &val); + + adreno_dev->speed_bin = (val & speed_bin[1]) >> speed_bin[2]; +} + +static const struct { + int (*check)(struct adreno_device *adreno_dev); + void (*func)(struct adreno_device *adreno_dev); +} a3xx_efuse_funcs[] = { + { adreno_is_a306a, a3xx_efuse_speed_bin }, +}; + +static void a3xx_check_features(struct adreno_device *adreno_dev) +{ + unsigned int i; + + if (adreno_efuse_map(adreno_dev)) + return; + + for (i = 0; i < ARRAY_SIZE(a3xx_efuse_funcs); i++) { + if (a3xx_efuse_funcs[i].check(adreno_dev)) + a3xx_efuse_funcs[i].func(adreno_dev); + } + + adreno_efuse_unmap(adreno_dev); +} + /** * _a3xx_pwron_fixup() - Initialize a special command buffer to run a * post-power collapse shader workaround @@ -604,6 +641,9 @@ static void a3xx_platform_setup(struct adreno_device *adreno_dev) gpudev->vbif_xin_halt_ctrl0_mask = A30X_VBIF_XIN_HALT_CTRL0_MASK; } + + /* Check efuse bits for various capabilties */ + a3xx_check_features(adreno_dev); } static int a3xx_send_me_init(struct adreno_device *adreno_dev, @@ -1425,6 +1465,10 @@ static struct adreno_coresight a3xx_coresight = { .groups = a3xx_coresight_groups, }; +static unsigned int a3xx_int_bits[ADRENO_INT_BITS_MAX] = { + ADRENO_INT_DEFINE(ADRENO_INT_RBBM_AHB_ERROR, A3XX_INT_RBBM_AHB_ERROR), +}; + /* Register offset defines for A3XX */ static unsigned int a3xx_register_offsets[ADRENO_REG_REGISTER_MAX] = { ADRENO_REG_DEFINE(ADRENO_REG_CP_ME_RAM_WADDR, A3XX_CP_ME_RAM_WADDR), @@ -1853,6 +1897,7 @@ int a3xx_microcode_load(struct adreno_device *adreno_dev, struct adreno_gpudev adreno_a3xx_gpudev = { .reg_offsets = &a3xx_reg_offsets, + .int_bits = a3xx_int_bits, .ft_perf_counters = a3xx_ft_perf_counters, .ft_perf_counters_count = ARRAY_SIZE(a3xx_ft_perf_counters), .perfcounters = &a3xx_perfcounters, diff --git a/drivers/gpu/msm/adreno_a4xx.c b/drivers/gpu/msm/adreno_a4xx.c index bfbdb0e7ac1f..5ca04e522270 100644 --- a/drivers/gpu/msm/adreno_a4xx.c +++ b/drivers/gpu/msm/adreno_a4xx.c @@ -739,6 +739,10 @@ static void a4xx_err_callback(struct adreno_device *adreno_dev, int bit) } } +static unsigned int a4xx_int_bits[ADRENO_INT_BITS_MAX] = { + ADRENO_INT_DEFINE(ADRENO_INT_RBBM_AHB_ERROR, A4XX_INT_RBBM_AHB_ERROR), +}; + /* Register offset defines for A4XX, in order of enum adreno_regs */ static unsigned int a4xx_register_offsets[ADRENO_REG_REGISTER_MAX] = { ADRENO_REG_DEFINE(ADRENO_REG_CP_ME_RAM_WADDR, A4XX_CP_ME_RAM_WADDR), @@ -1765,6 +1769,7 @@ static struct adreno_snapshot_data a4xx_snapshot_data = { struct adreno_gpudev adreno_a4xx_gpudev = { .reg_offsets = &a4xx_reg_offsets, + .int_bits = a4xx_int_bits, .ft_perf_counters = a4xx_ft_perf_counters, .ft_perf_counters_count = ARRAY_SIZE(a4xx_ft_perf_counters), .perfcounters = &a4xx_perfcounters, diff --git a/drivers/gpu/msm/adreno_a4xx_preempt.c b/drivers/gpu/msm/adreno_a4xx_preempt.c index 4087ac60c89e..ef837dc4b7ea 100644 --- a/drivers/gpu/msm/adreno_a4xx_preempt.c +++ b/drivers/gpu/msm/adreno_a4xx_preempt.c @@ -146,6 +146,8 @@ static int a4xx_submit_preempt_token(struct adreno_ringbuffer *rb, &ptname, PT_INFO_OFFSET(current_rb_ptname)); pt = kgsl_mmu_get_pt_from_ptname(&(device->mmu), ptname); + if (IS_ERR_OR_NULL(pt)) + return (pt == NULL) ? -ENOENT : PTR_ERR(pt); /* set the ringbuffer for incoming RB */ pt_switch_sizedwords = adreno_iommu_set_pt_generate_cmds(incoming_rb, diff --git a/drivers/gpu/msm/adreno_a5xx.c b/drivers/gpu/msm/adreno_a5xx.c index 38ed2d1e8a99..dc35c0080d11 100644 --- a/drivers/gpu/msm/adreno_a5xx.c +++ b/drivers/gpu/msm/adreno_a5xx.c @@ -196,6 +196,8 @@ static void a5xx_platform_setup(struct adreno_device *adreno_dev) /* A510 has 3 XIN ports in VBIF */ gpudev->vbif_xin_halt_ctrl0_mask = A510_VBIF_XIN_HALT_CTRL0_MASK; + } else if (adreno_is_a540(adreno_dev)) { + gpudev->snapshot_data->sect_sizes->cp_merciu = 1024; } /* Calculate SP local and private mem addresses */ @@ -429,6 +431,43 @@ static int _poll_gdsc_status(struct adreno_device *adreno_dev, return 0; } +static void a5xx_restore_isense_regs(struct adreno_device *adreno_dev) +{ + struct kgsl_device *device = KGSL_DEVICE(adreno_dev); + unsigned int reg, i, ramp = GPMU_ISENSE_SAVE; + static unsigned int isense_regs[6] = {0xFFFF}, isense_reg_addr[] = { + A5XX_GPU_CS_DECIMAL_ALIGN, + A5XX_GPU_CS_SENSOR_PARAM_CORE_1, + A5XX_GPU_CS_SENSOR_PARAM_CORE_2, + A5XX_GPU_CS_SW_OV_FUSE_EN, + A5XX_GPU_CS_ENDPOINT_CALIBRATION_DONE, + A5XX_GPMU_TEMP_SENSOR_CONFIG}; + + if (!adreno_is_a540(adreno_dev)) + return; + + /* read signature */ + kgsl_regread(device, ramp++, ®); + + if (reg == 0xBABEFACE) { + /* store memory locations in buffer */ + for (i = 0; i < ARRAY_SIZE(isense_regs); i++) + kgsl_regread(device, ramp + i, isense_regs + i); + + /* clear signature */ + kgsl_regwrite(device, GPMU_ISENSE_SAVE, 0x0); + } + + /* if we never stored memory locations - do nothing */ + if (isense_regs[0] == 0xFFFF) + return; + + /* restore registers from memory */ + for (i = 0; i < ARRAY_SIZE(isense_reg_addr); i++) + kgsl_regwrite(device, isense_reg_addr[i], isense_regs[i]); + +} + /* * a5xx_regulator_enable() - Enable any necessary HW regulators * @adreno_dev: The adreno device pointer @@ -478,6 +517,7 @@ static int a5xx_regulator_enable(struct adreno_device *adreno_dev) kgsl_regrmw(device, A5XX_GPMU_GPMU_SP_CLOCK_CONTROL, CNTL_IP_CLK_ENABLE, 1); + a5xx_restore_isense_regs(adreno_dev); return 0; } @@ -1534,12 +1574,12 @@ static void a5xx_clk_set_options(struct adreno_device *adreno_dev, const char *name, struct clk *clk) { if (adreno_is_a540(adreno_dev)) { - if (!strcmp(name, "mem_iface_clk")) - clk_set_flags(clk, CLKFLAG_NORETAIN_PERIPH); - clk_set_flags(clk, CLKFLAG_NORETAIN_MEM); - if (!strcmp(name, "core_clk")) { + if (!strcmp(name, "mem_iface_clk")) { clk_set_flags(clk, CLKFLAG_NORETAIN_PERIPH); clk_set_flags(clk, CLKFLAG_NORETAIN_MEM); + } else if (!strcmp(name, "core_clk")) { + clk_set_flags(clk, CLKFLAG_RETAIN_PERIPH); + clk_set_flags(clk, CLKFLAG_RETAIN_MEM); } } } @@ -1781,11 +1821,11 @@ static void a5xx_start(struct adreno_device *adreno_dev) set_bit(ADRENO_DEVICE_HANG_INTR, &adreno_dev->priv); gpudev->irq->mask |= (1 << A5XX_INT_MISC_HANG_DETECT); /* - * Set hang detection threshold to 1 million cycles - * (0xFFFF*16) + * Set hang detection threshold to 4 million cycles + * (0x3FFFF*16) */ kgsl_regwrite(device, A5XX_RBBM_INTERFACE_HANG_INT_CNTL, - (1 << 30) | 0xFFFF); + (1 << 30) | 0x3FFFF); } @@ -1944,6 +1984,16 @@ static void a5xx_start(struct adreno_device *adreno_dev) } + /* + * VPC corner case with local memory load kill leads to corrupt + * internal state. Normal Disable does not work for all a5x chips. + * So do the following setting to disable it. + */ + if (ADRENO_QUIRK(adreno_dev, ADRENO_QUIRK_DISABLE_LMLOADKILL)) { + kgsl_regrmw(device, A5XX_VPC_DBG_ECO_CNTL, 0, 0x1 << 23); + kgsl_regrmw(device, A5XX_HLSQ_DBG_ECO_CNTL, 0x1 << 18, 0); + } + a5xx_preemption_start(adreno_dev); a5xx_protect_init(adreno_dev); } @@ -2872,6 +2922,10 @@ static struct adreno_ft_perf_counters a5xx_ft_perf_counters[] = { {KGSL_PERFCOUNTER_GROUP_TSE, A5XX_TSE_INPUT_PRIM_NUM}, }; +static unsigned int a5xx_int_bits[ADRENO_INT_BITS_MAX] = { + ADRENO_INT_DEFINE(ADRENO_INT_RBBM_AHB_ERROR, A5XX_INT_RBBM_AHB_ERROR), +}; + /* Register offset defines for A5XX, in order of enum adreno_regs */ static unsigned int a5xx_register_offsets[ADRENO_REG_REGISTER_MAX] = { ADRENO_REG_DEFINE(ADRENO_REG_CP_WFI_PEND_CTR, A5XX_CP_WFI_PEND_CTR), @@ -3504,6 +3558,7 @@ static struct adreno_coresight a5xx_coresight = { struct adreno_gpudev adreno_a5xx_gpudev = { .reg_offsets = &a5xx_reg_offsets, + .int_bits = a5xx_int_bits, .ft_perf_counters = a5xx_ft_perf_counters, .ft_perf_counters_count = ARRAY_SIZE(a5xx_ft_perf_counters), .coresight = &a5xx_coresight, diff --git a/drivers/gpu/msm/adreno_a5xx.h b/drivers/gpu/msm/adreno_a5xx.h index e424e7a9f228..08fd16a83f30 100644 --- a/drivers/gpu/msm/adreno_a5xx.h +++ b/drivers/gpu/msm/adreno_a5xx.h @@ -228,6 +228,7 @@ void a5xx_hwcg_set(struct adreno_device *adreno_dev, bool on); #define LM_SEQUENCE_ID 1 #define MAX_SEQUENCE_ID 3 +#define GPMU_ISENSE_SAVE (A5XX_GPMU_DATA_RAM_BASE + 200/4) /* LM defaults */ #define LM_DEFAULT_LIMIT 6000 #define A530_DEFAULT_LEAKAGE 0x004E001A diff --git a/drivers/gpu/msm/adreno_perfcounter.c b/drivers/gpu/msm/adreno_perfcounter.c index 42f8119ad8b4..f5f99c3ebb4a 100644 --- a/drivers/gpu/msm/adreno_perfcounter.c +++ b/drivers/gpu/msm/adreno_perfcounter.c @@ -522,12 +522,18 @@ int adreno_perfcounter_get(struct adreno_device *adreno_dev, if (empty == -1) return -EBUSY; + /* initialize the new counter */ + group->regs[empty].countable = countable; + /* enable the new counter */ ret = adreno_perfcounter_enable(adreno_dev, groupid, empty, countable); - if (ret) + if (ret) { + /* Put back the perfcounter */ + if (!(group->flags & ADRENO_PERFCOUNTER_GROUP_FIXED)) + group->regs[empty].countable = + KGSL_PERFCOUNTER_NOT_USED; return ret; - /* initialize the new counter */ - group->regs[empty].countable = countable; + } /* set initial kernel and user count */ if (flags & PERFCOUNTER_FLAG_KERNEL) { @@ -720,10 +726,22 @@ static int _perfcounter_enable_default(struct adreno_device *adreno_dev, /* wait for the above commands submitted to complete */ ret = adreno_ringbuffer_waittimestamp(rb, rb->timestamp, ADRENO_IDLE_TIMEOUT); - if (ret) - KGSL_DRV_ERR(device, - "Perfcounter %u/%u/%u start via commands failed %d\n", - group, counter, countable, ret); + if (ret) { + /* + * If we were woken up because of cancelling rb events + * either due to soft reset or adreno_stop, ignore the + * error and return 0 here. The perfcounter is already + * set up in software and it will be programmed in + * hardware when we wake up or come up after soft reset, + * by adreno_perfcounter_restore. + */ + if (ret == -EAGAIN) + ret = 0; + else + KGSL_DRV_ERR(device, + "Perfcounter %u/%u/%u start via commands failed %d\n", + group, counter, countable, ret); + } } else { /* Select the desired perfcounter */ kgsl_regwrite(device, reg->select, countable); diff --git a/drivers/gpu/msm/kgsl.c b/drivers/gpu/msm/kgsl.c index d1e9b74ac49c..699d99651f2c 100644 --- a/drivers/gpu/msm/kgsl.c +++ b/drivers/gpu/msm/kgsl.c @@ -2465,6 +2465,8 @@ static int kgsl_setup_dma_buf(struct kgsl_device *device, meta->dmabuf = dmabuf; meta->attach = attach; + attach->priv = entry; + entry->priv_data = meta; entry->memdesc.pagetable = pagetable; entry->memdesc.size = 0; @@ -2515,6 +2517,45 @@ out: } #endif +#ifdef CONFIG_DMA_SHARED_BUFFER +void kgsl_get_egl_counts(struct kgsl_mem_entry *entry, + int *egl_surface_count, int *egl_image_count) +{ + struct kgsl_dma_buf_meta *meta = entry->priv_data; + struct dma_buf *dmabuf = meta->dmabuf; + struct dma_buf_attachment *mem_entry_buf_attachment = meta->attach; + struct device *buf_attachment_dev = mem_entry_buf_attachment->dev; + struct dma_buf_attachment *attachment = NULL; + + mutex_lock(&dmabuf->lock); + list_for_each_entry(attachment, &dmabuf->attachments, node) { + struct kgsl_mem_entry *scan_mem_entry = NULL; + + if (attachment->dev != buf_attachment_dev) + continue; + + scan_mem_entry = attachment->priv; + if (!scan_mem_entry) + continue; + + switch (kgsl_memdesc_get_memtype(&scan_mem_entry->memdesc)) { + case KGSL_MEMTYPE_EGL_SURFACE: + (*egl_surface_count)++; + break; + case KGSL_MEMTYPE_EGL_IMAGE: + (*egl_image_count)++; + break; + } + } + mutex_unlock(&dmabuf->lock); +} +#else +void kgsl_get_egl_counts(struct kgsl_mem_entry *entry, + int *egl_surface_count, int *egl_image_count) +{ +} +#endif + long kgsl_ioctl_map_user_mem(struct kgsl_device_private *dev_priv, unsigned int cmd, void *data) { @@ -3851,7 +3892,7 @@ kgsl_mmap_memstore(struct kgsl_device *device, struct vm_area_struct *vma) return -EINVAL; } - vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); + vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); result = remap_pfn_range(vma, vma->vm_start, device->memstore.physaddr >> PAGE_SHIFT, @@ -4487,6 +4528,9 @@ int kgsl_device_platform_probe(struct kgsl_device *device) if (status) goto error_close_mmu; + /* Initialize the memory pools */ + kgsl_init_page_pools(device->pdev); + status = kgsl_allocate_global(device, &device->memstore, KGSL_MEMSTORE_SIZE, 0, KGSL_MEMDESC_CONTIG, "memstore"); @@ -4541,9 +4585,6 @@ int kgsl_device_platform_probe(struct kgsl_device *device) /* Initialize common sysfs entries */ kgsl_pwrctrl_init_sysfs(device); - /* Initialize the memory pools */ - kgsl_init_page_pools(); - return 0; error_free_memstore: diff --git a/drivers/gpu/msm/kgsl.h b/drivers/gpu/msm/kgsl.h index 826c4edb3582..fbf9197b6d1b 100644 --- a/drivers/gpu/msm/kgsl.h +++ b/drivers/gpu/msm/kgsl.h @@ -435,6 +435,9 @@ long kgsl_ioctl_sparse_unbind(struct kgsl_device_private *dev_priv, void kgsl_mem_entry_destroy(struct kref *kref); +void kgsl_get_egl_counts(struct kgsl_mem_entry *entry, + int *egl_surface_count, int *egl_image_count); + struct kgsl_mem_entry * __must_check kgsl_sharedmem_find(struct kgsl_process_private *private, uint64_t gpuaddr); diff --git a/drivers/gpu/msm/kgsl_debugfs.c b/drivers/gpu/msm/kgsl_debugfs.c index 2f293e4da398..7758fc956055 100644 --- a/drivers/gpu/msm/kgsl_debugfs.c +++ b/drivers/gpu/msm/kgsl_debugfs.c @@ -125,13 +125,15 @@ static char get_cacheflag(const struct kgsl_memdesc *m) } -static int print_mem_entry(int id, void *ptr, void *data) +static int print_mem_entry(void *data, void *ptr) { struct seq_file *s = data; struct kgsl_mem_entry *entry = ptr; char flags[10]; char usage[16]; struct kgsl_memdesc *m = &entry->memdesc; + unsigned int usermem_type = kgsl_memdesc_usermem_type(m); + int egl_surface_count = 0, egl_image_count = 0; if (m->flags & KGSL_MEMFLAGS_SPARSE_VIRT) return 0; @@ -149,12 +151,17 @@ static int print_mem_entry(int id, void *ptr, void *data) kgsl_get_memory_usage(usage, sizeof(usage), m->flags); - seq_printf(s, "%pK %pK %16llu %5d %9s %10s %16s %5d %16llu", + if (usermem_type == KGSL_MEM_ENTRY_ION) + kgsl_get_egl_counts(entry, &egl_surface_count, + &egl_image_count); + + seq_printf(s, "%pK %pK %16llu %5d %9s %10s %16s %5d %16llu %6d %6d", (uint64_t *)(uintptr_t) m->gpuaddr, (unsigned long *) m->useraddr, m->size, entry->id, flags, - memtype_str(kgsl_memdesc_usermem_type(m)), - usage, (m->sgt ? m->sgt->nents : 0), m->mapsize); + memtype_str(usermem_type), + usage, (m->sgt ? m->sgt->nents : 0), m->mapsize, + egl_surface_count, egl_image_count); if (entry->metadata[0] != 0) seq_printf(s, " %s", entry->metadata); @@ -164,25 +171,83 @@ static int print_mem_entry(int id, void *ptr, void *data) return 0; } -static int process_mem_print(struct seq_file *s, void *unused) +static struct kgsl_mem_entry *process_mem_seq_find(struct seq_file *s, + void *ptr, loff_t pos) { + struct kgsl_mem_entry *entry = ptr; struct kgsl_process_private *private = s->private; + int id = 0; + loff_t temp_pos = 1; - seq_printf(s, "%16s %16s %16s %5s %9s %10s %16s %5s %16s\n", - "gpuaddr", "useraddr", "size", "id", "flags", "type", - "usage", "sglen", "mapsize"); + if (entry != SEQ_START_TOKEN) + id = entry->id + 1; spin_lock(&private->mem_lock); - idr_for_each(&private->mem_idr, print_mem_entry, s); + for (entry = idr_get_next(&private->mem_idr, &id); entry; + id++, entry = idr_get_next(&private->mem_idr, &id), + temp_pos++) { + if (temp_pos == pos && kgsl_mem_entry_get(entry)) { + spin_unlock(&private->mem_lock); + goto found; + } + } spin_unlock(&private->mem_lock); - return 0; + entry = NULL; +found: + if (ptr != SEQ_START_TOKEN) + kgsl_mem_entry_put(ptr); + + return entry; +} + +static void *process_mem_seq_start(struct seq_file *s, loff_t *pos) +{ + loff_t seq_file_offset = *pos; + + if (seq_file_offset == 0) + return SEQ_START_TOKEN; + else + return process_mem_seq_find(s, SEQ_START_TOKEN, + seq_file_offset); +} + +static void process_mem_seq_stop(struct seq_file *s, void *ptr) +{ + if (ptr && ptr != SEQ_START_TOKEN) + kgsl_mem_entry_put(ptr); } +static void *process_mem_seq_next(struct seq_file *s, void *ptr, + loff_t *pos) +{ + ++*pos; + return process_mem_seq_find(s, ptr, 1); +} + +static int process_mem_seq_show(struct seq_file *s, void *ptr) +{ + if (ptr == SEQ_START_TOKEN) { + seq_printf(s, "%16s %16s %16s %5s %9s %10s %16s %5s %16s %6s %6s\n", + "gpuaddr", "useraddr", "size", "id", "flags", "type", + "usage", "sglen", "mapsize", "eglsrf", "eglimg"); + return 0; + } else + return print_mem_entry(s, ptr); +} + +static const struct seq_operations process_mem_seq_fops = { + .start = process_mem_seq_start, + .stop = process_mem_seq_stop, + .next = process_mem_seq_next, + .show = process_mem_seq_show, +}; + static int process_mem_open(struct inode *inode, struct file *file) { int ret; pid_t pid = (pid_t) (unsigned long) inode->i_private; + struct seq_file *s = NULL; struct kgsl_process_private *private = NULL; private = kgsl_process_private_find(pid); @@ -190,9 +255,13 @@ static int process_mem_open(struct inode *inode, struct file *file) if (!private) return -ENODEV; - ret = single_open(file, process_mem_print, private); + ret = seq_open(file, &process_mem_seq_fops); if (ret) kgsl_process_private_put(private); + else { + s = file->private_data; + s->private = private; + } return ret; } @@ -205,7 +274,7 @@ static int process_mem_release(struct inode *inode, struct file *file) if (private) kgsl_process_private_put(private); - return single_release(inode, file); + return seq_release(inode, file); } static const struct file_operations process_mem_fops = { diff --git a/drivers/gpu/msm/kgsl_mmu.c b/drivers/gpu/msm/kgsl_mmu.c index 0df1ad02f16e..46bb6f4656fb 100644 --- a/drivers/gpu/msm/kgsl_mmu.c +++ b/drivers/gpu/msm/kgsl_mmu.c @@ -390,6 +390,13 @@ kgsl_mmu_map(struct kgsl_pagetable *pagetable, if (!memdesc->gpuaddr) return -EINVAL; + if (!(memdesc->flags & (KGSL_MEMFLAGS_SPARSE_VIRT | + KGSL_MEMFLAGS_SPARSE_PHYS))) { + /* Only global mappings should be mapped multiple times */ + if (!kgsl_memdesc_is_global(memdesc) && + (KGSL_MEMDESC_MAPPED & memdesc->priv)) + return -EINVAL; + } size = kgsl_memdesc_footprint(memdesc); @@ -403,6 +410,9 @@ kgsl_mmu_map(struct kgsl_pagetable *pagetable, atomic_inc(&pagetable->stats.entries); KGSL_STATS_ADD(size, &pagetable->stats.mapped, &pagetable->stats.max_mapped); + + /* This is needed for non-sparse mappings */ + memdesc->priv |= KGSL_MEMDESC_MAPPED; } return 0; @@ -467,6 +477,13 @@ kgsl_mmu_unmap(struct kgsl_pagetable *pagetable, if (memdesc->size == 0) return -EINVAL; + if (!(memdesc->flags & (KGSL_MEMFLAGS_SPARSE_VIRT | + KGSL_MEMFLAGS_SPARSE_PHYS))) { + /* Only global mappings should be mapped multiple times */ + if (!(KGSL_MEMDESC_MAPPED & memdesc->priv)) + return -EINVAL; + } + if (PT_OP_VALID(pagetable, mmu_unmap)) { uint64_t size; @@ -476,6 +493,9 @@ kgsl_mmu_unmap(struct kgsl_pagetable *pagetable, atomic_dec(&pagetable->stats.entries); atomic_long_sub(size, &pagetable->stats.mapped); + + if (!kgsl_memdesc_is_global(memdesc)) + memdesc->priv &= ~KGSL_MEMDESC_MAPPED; } return ret; diff --git a/drivers/gpu/msm/kgsl_pool.c b/drivers/gpu/msm/kgsl_pool.c index f5402fdc7e57..6ecbab466c7c 100644 --- a/drivers/gpu/msm/kgsl_pool.c +++ b/drivers/gpu/msm/kgsl_pool.c @@ -21,6 +21,10 @@ #include "kgsl_device.h" #include "kgsl_pool.h" +#define KGSL_MAX_POOLS 4 +#define KGSL_MAX_POOL_ORDER 8 +#define KGSL_MAX_RESERVED_PAGES 4096 + /** * struct kgsl_page_pool - Structure to hold information for the pool * @pool_order: Page order describing the size of the page @@ -40,41 +44,10 @@ struct kgsl_page_pool { struct list_head page_list; }; -static struct kgsl_page_pool kgsl_pools[] = { - { - .pool_order = 0, - .reserved_pages = 2048, - .allocation_allowed = true, - .list_lock = __SPIN_LOCK_UNLOCKED(kgsl_pools[0].list_lock), - .page_list = LIST_HEAD_INIT(kgsl_pools[0].page_list), - }, -#ifndef CONFIG_ALLOC_BUFFERS_IN_4K_CHUNKS - { - .pool_order = 1, - .reserved_pages = 1024, - .allocation_allowed = true, - .list_lock = __SPIN_LOCK_UNLOCKED(kgsl_pools[1].list_lock), - .page_list = LIST_HEAD_INIT(kgsl_pools[1].page_list), - }, - { - .pool_order = 4, - .reserved_pages = 256, - .allocation_allowed = false, - .list_lock = __SPIN_LOCK_UNLOCKED(kgsl_pools[2].list_lock), - .page_list = LIST_HEAD_INIT(kgsl_pools[2].page_list), - }, - { - .pool_order = 8, - .reserved_pages = 32, - .allocation_allowed = false, - .list_lock = __SPIN_LOCK_UNLOCKED(kgsl_pools[3].list_lock), - .page_list = LIST_HEAD_INIT(kgsl_pools[3].page_list), - }, +static struct kgsl_page_pool kgsl_pools[KGSL_MAX_POOLS]; +static int kgsl_num_pools; +static int kgsl_pool_max_pages; -#endif -}; - -#define KGSL_NUM_POOLS ARRAY_SIZE(kgsl_pools) /* Returns KGSL pool corresponding to input page order*/ static struct kgsl_page_pool * @@ -82,7 +55,7 @@ _kgsl_get_pool_from_order(unsigned int order) { int i; - for (i = 0; i < KGSL_NUM_POOLS; i++) { + for (i = 0; i < kgsl_num_pools; i++) { if (kgsl_pools[i].pool_order == order) return &kgsl_pools[i]; } @@ -154,7 +127,7 @@ static int kgsl_pool_size_total(void) int i; int total = 0; - for (i = 0; i < KGSL_NUM_POOLS; i++) + for (i = 0; i < kgsl_num_pools; i++) total += kgsl_pool_size(&kgsl_pools[i]); return total; } @@ -207,7 +180,7 @@ kgsl_pool_reduce(unsigned int target_pages, bool exit) total_pages = kgsl_pool_size_total(); - for (i = (KGSL_NUM_POOLS - 1); i >= 0; i--) { + for (i = (kgsl_num_pools - 1); i >= 0; i--) { pool = &kgsl_pools[i]; /* @@ -300,7 +273,7 @@ static int kgsl_pool_idx_lookup(unsigned int order) { int i; - for (i = 0; i < KGSL_NUM_POOLS; i++) + for (i = 0; i < kgsl_num_pools; i++) if (order == kgsl_pools[i].pool_order) return i; @@ -384,10 +357,13 @@ void kgsl_pool_free_page(struct page *page) page_order = compound_order(page); - pool = _kgsl_get_pool_from_order(page_order); - if (pool != NULL) { - _kgsl_pool_add_page(pool, page); - return; + if (!kgsl_pool_max_pages || + (kgsl_pool_size_total() < kgsl_pool_max_pages)) { + pool = _kgsl_get_pool_from_order(page_order); + if (pool != NULL) { + _kgsl_pool_add_page(pool, page); + return; + } } /* Give back to system as not added to pool */ @@ -398,7 +374,7 @@ static void kgsl_pool_reserve_pages(void) { int i, j; - for (i = 0; i < KGSL_NUM_POOLS; i++) { + for (i = 0; i < kgsl_num_pools; i++) { struct page *page; for (j = 0; j < kgsl_pools[i].reserved_pages; j++) { @@ -445,8 +421,76 @@ static struct shrinker kgsl_pool_shrinker = { .batch = 0, }; -void kgsl_init_page_pools(void) +static void kgsl_pool_config(unsigned int order, unsigned int reserved_pages, + bool allocation_allowed) { +#ifdef CONFIG_ALLOC_BUFFERS_IN_4K_CHUNKS + if (order > 0) { + pr_info("%s: Pool order:%d not supprted.!!\n", __func__, order); + return; + } +#endif + if ((order > KGSL_MAX_POOL_ORDER) || + (reserved_pages > KGSL_MAX_RESERVED_PAGES)) + return; + + kgsl_pools[kgsl_num_pools].pool_order = order; + kgsl_pools[kgsl_num_pools].reserved_pages = reserved_pages; + kgsl_pools[kgsl_num_pools].allocation_allowed = allocation_allowed; + spin_lock_init(&kgsl_pools[kgsl_num_pools].list_lock); + INIT_LIST_HEAD(&kgsl_pools[kgsl_num_pools].page_list); + kgsl_num_pools++; +} + +static void kgsl_of_parse_mempools(struct device_node *node) +{ + struct device_node *child; + unsigned int page_size, reserved_pages = 0; + bool allocation_allowed; + + for_each_child_of_node(node, child) { + unsigned int index; + + if (of_property_read_u32(child, "reg", &index)) + return; + + if (index >= KGSL_MAX_POOLS) + continue; + + if (of_property_read_u32(child, "qcom,mempool-page-size", + &page_size)) + return; + + of_property_read_u32(child, "qcom,mempool-reserved", + &reserved_pages); + + allocation_allowed = of_property_read_bool(child, + "qcom,mempool-allocate"); + + kgsl_pool_config(ilog2(page_size >> PAGE_SHIFT), reserved_pages, + allocation_allowed); + } +} + +static void kgsl_of_get_mempools(struct device_node *parent) +{ + struct device_node *node; + + node = of_find_compatible_node(parent, NULL, "qcom,gpu-mempools"); + if (node != NULL) { + /* Get Max pages limit for mempool */ + of_property_read_u32(node, "qcom,mempool-max-pages", + &kgsl_pool_max_pages); + kgsl_of_parse_mempools(node); + } +} + +void kgsl_init_page_pools(struct platform_device *pdev) +{ + + /* Get GPU mempools data and configure pools */ + kgsl_of_get_mempools(pdev->dev.of_node); + /* Reserve the appropriate number of pages for each pool */ kgsl_pool_reserve_pages(); diff --git a/drivers/gpu/msm/kgsl_pool.h b/drivers/gpu/msm/kgsl_pool.h index efbfa96f1498..d55e1ada123b 100644 --- a/drivers/gpu/msm/kgsl_pool.h +++ b/drivers/gpu/msm/kgsl_pool.h @@ -35,7 +35,7 @@ kgsl_gfp_mask(unsigned int page_order) void kgsl_pool_free_sgt(struct sg_table *sgt); void kgsl_pool_free_pages(struct page **pages, unsigned int page_count); -void kgsl_init_page_pools(void); +void kgsl_init_page_pools(struct platform_device *pdev); void kgsl_exit_page_pools(void); int kgsl_pool_alloc_page(int *page_size, struct page **pages, unsigned int pages_len, unsigned int *align); diff --git a/drivers/gpu/msm/kgsl_sharedmem.c b/drivers/gpu/msm/kgsl_sharedmem.c index b18934fb66ae..283b72c22db4 100644 --- a/drivers/gpu/msm/kgsl_sharedmem.c +++ b/drivers/gpu/msm/kgsl_sharedmem.c @@ -571,12 +571,11 @@ int kgsl_cache_range_op(struct kgsl_memdesc *memdesc, uint64_t offset, void *addr = (memdesc->hostptr) ? memdesc->hostptr : (void *) memdesc->useraddr; - /* Make sure that size is non-zero */ - if (!size) + if (size == 0 || size > UINT_MAX) return -EINVAL; - /* Make sure that the offset + size isn't bigger than we can handle */ - if ((offset + size) > ULONG_MAX) + /* Make sure that the offset + size does not overflow */ + if ((offset + size < offset) || (offset + size < size)) return -ERANGE; /* Make sure the offset + size do not overflow the address */ diff --git a/drivers/gpu/msm/kgsl_snapshot.c b/drivers/gpu/msm/kgsl_snapshot.c index a2e4a909062f..13dc3017072d 100644 --- a/drivers/gpu/msm/kgsl_snapshot.c +++ b/drivers/gpu/msm/kgsl_snapshot.c @@ -1067,6 +1067,9 @@ void kgsl_snapshot_save_frozen_objs(struct work_struct *work) size_t size = 0; void *ptr; + if (IS_ERR_OR_NULL(device)) + return; + kgsl_snapshot_process_ib_obj_list(snapshot); list_for_each_entry(obj, &snapshot->obj_list, node) { |