diff options
Diffstat (limited to 'drivers/gpu/drm/i915')
36 files changed, 681 insertions, 279 deletions
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c index c7c2b349858d..2a27fb5d7dc6 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.c +++ b/drivers/gpu/drm/i915/display/intel_ddi.c @@ -3986,6 +3986,7 @@ static void intel_enable_ddi(struct intel_encoder *encoder, if (conn_state->content_protection == DRM_MODE_CONTENT_PROTECTION_DESIRED) intel_hdcp_enable(to_intel_connector(conn_state->connector), + crtc_state->cpu_transcoder, (u8)conn_state->hdcp_content_type); } @@ -4089,7 +4090,9 @@ static void intel_ddi_update_pipe(struct intel_encoder *encoder, if (conn_state->content_protection == DRM_MODE_CONTENT_PROTECTION_DESIRED || content_protection_type_changed) - intel_hdcp_enable(connector, (u8)conn_state->hdcp_content_type); + intel_hdcp_enable(connector, + crtc_state->cpu_transcoder, + (u8)conn_state->hdcp_content_type); } static void diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 6f5e3bd13ad1..effc4250b230 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -15112,7 +15112,7 @@ intel_prepare_plane_fb(struct drm_plane *plane, return ret; fb_obj_bump_render_priority(obj); - intel_frontbuffer_flush(obj->frontbuffer, ORIGIN_DIRTYFB); + i915_gem_object_flush_frontbuffer(obj, ORIGIN_DIRTYFB); if (!new_plane_state->base.fence) { /* implicit fencing */ struct dma_fence *fence; diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c index ce1b64f4dd44..12ba74788cce 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power.c +++ b/drivers/gpu/drm/i915/display/intel_display_power.c @@ -3688,6 +3688,151 @@ static const struct i915_power_well_desc icl_power_wells[] = { }, }; +static const struct i915_power_well_desc ehl_power_wells[] = { + { + .name = "always-on", + .always_on = true, + .domains = POWER_DOMAIN_MASK, + .ops = &i9xx_always_on_power_well_ops, + .id = DISP_PW_ID_NONE, + }, + { + .name = "power well 1", + /* Handled by the DMC firmware */ + .always_on = true, + .domains = 0, + .ops = &hsw_power_well_ops, + .id = SKL_DISP_PW_1, + { + .hsw.regs = &hsw_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_PW_1, + .hsw.has_fuses = true, + }, + }, + { + .name = "DC off", + .domains = ICL_DISPLAY_DC_OFF_POWER_DOMAINS, + .ops = &gen9_dc_off_power_well_ops, + .id = SKL_DISP_DC_OFF, + }, + { + .name = "power well 2", + .domains = ICL_PW_2_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = SKL_DISP_PW_2, + { + .hsw.regs = &hsw_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_PW_2, + .hsw.has_fuses = true, + }, + }, + { + .name = "power well 3", + .domains = ICL_PW_3_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &hsw_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_PW_3, + .hsw.irq_pipe_mask = BIT(PIPE_B), + .hsw.has_vga = true, + .hsw.has_fuses = true, + }, + }, + { + .name = "DDI A IO", + .domains = ICL_DDI_IO_A_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_ddi_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_DDI_A, + }, + }, + { + .name = "DDI B IO", + .domains = ICL_DDI_IO_B_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_ddi_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_DDI_B, + }, + }, + { + .name = "DDI C IO", + .domains = ICL_DDI_IO_C_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_ddi_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_DDI_C, + }, + }, + { + .name = "DDI D IO", + .domains = ICL_DDI_IO_D_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_ddi_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_DDI_D, + }, + }, + { + .name = "AUX A", + .domains = ICL_AUX_A_IO_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_aux_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_AUX_A, + }, + }, + { + .name = "AUX B", + .domains = ICL_AUX_B_IO_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_aux_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_AUX_B, + }, + }, + { + .name = "AUX C", + .domains = ICL_AUX_C_TC1_IO_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_aux_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_AUX_C, + }, + }, + { + .name = "AUX D", + .domains = ICL_AUX_D_TC2_IO_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_aux_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_AUX_D, + }, + }, + { + .name = "power well 4", + .domains = ICL_PW_4_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &hsw_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_PW_4, + .hsw.has_fuses = true, + .hsw.irq_pipe_mask = BIT(PIPE_C), + }, + }, +}; + static const struct i915_power_well_desc tgl_power_wells[] = { { .name = "always-on", @@ -3832,7 +3977,7 @@ static const struct i915_power_well_desc tgl_power_wells[] = { { .name = "AUX A", .domains = TGL_AUX_A_IO_POWER_DOMAINS, - .ops = &icl_combo_phy_aux_power_well_ops, + .ops = &hsw_power_well_ops, .id = DISP_PW_ID_NONE, { .hsw.regs = &icl_aux_power_well_regs, @@ -3842,7 +3987,7 @@ static const struct i915_power_well_desc tgl_power_wells[] = { { .name = "AUX B", .domains = TGL_AUX_B_IO_POWER_DOMAINS, - .ops = &icl_combo_phy_aux_power_well_ops, + .ops = &hsw_power_well_ops, .id = DISP_PW_ID_NONE, { .hsw.regs = &icl_aux_power_well_regs, @@ -3852,7 +3997,7 @@ static const struct i915_power_well_desc tgl_power_wells[] = { { .name = "AUX C", .domains = TGL_AUX_C_IO_POWER_DOMAINS, - .ops = &icl_combo_phy_aux_power_well_ops, + .ops = &hsw_power_well_ops, .id = DISP_PW_ID_NONE, { .hsw.regs = &icl_aux_power_well_regs, @@ -4162,6 +4307,8 @@ int intel_power_domains_init(struct drm_i915_private *dev_priv) */ if (IS_GEN(dev_priv, 12)) { err = set_power_wells(power_domains, tgl_power_wells); + } else if (IS_ELKHARTLAKE(dev_priv)) { + err = set_power_wells(power_domains, ehl_power_wells); } else if (IS_GEN(dev_priv, 11)) { err = set_power_wells(power_domains, icl_power_wells); } else if (IS_CANNONLAKE(dev_priv)) { diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index 050655a1a3d8..b05b2191b919 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -2414,9 +2414,6 @@ intel_dp_compute_config(struct intel_encoder *encoder, intel_psr_compute_config(intel_dp, pipe_config); - intel_hdcp_transcoder_config(intel_connector, - pipe_config->cpu_transcoder); - return 0; } diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c index 3111ecaeabd0..20616639b8ab 100644 --- a/drivers/gpu/drm/i915/display/intel_fbc.c +++ b/drivers/gpu/drm/i915/display/intel_fbc.c @@ -1284,7 +1284,7 @@ static int intel_sanitize_fbc_option(struct drm_i915_private *dev_priv) return 0; /* https://bugs.freedesktop.org/show_bug.cgi?id=108085 */ - if (IS_GEMINILAKE(dev_priv)) + if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) return 0; if (IS_BROADWELL(dev_priv) || INTEL_GEN(dev_priv) >= 9) diff --git a/drivers/gpu/drm/i915/display/intel_frontbuffer.c b/drivers/gpu/drm/i915/display/intel_frontbuffer.c index 84b164f31895..6cb02c912acc 100644 --- a/drivers/gpu/drm/i915/display/intel_frontbuffer.c +++ b/drivers/gpu/drm/i915/display/intel_frontbuffer.c @@ -229,11 +229,11 @@ static void frontbuffer_release(struct kref *ref) vma->display_alignment = I915_GTT_MIN_ALIGNMENT; spin_unlock(&obj->vma.lock); - obj->frontbuffer = NULL; + RCU_INIT_POINTER(obj->frontbuffer, NULL); spin_unlock(&to_i915(obj->base.dev)->fb_tracking.lock); i915_gem_object_put(obj); - kfree(front); + kfree_rcu(front, rcu); } struct intel_frontbuffer * @@ -242,11 +242,7 @@ intel_frontbuffer_get(struct drm_i915_gem_object *obj) struct drm_i915_private *i915 = to_i915(obj->base.dev); struct intel_frontbuffer *front; - spin_lock(&i915->fb_tracking.lock); - front = obj->frontbuffer; - if (front) - kref_get(&front->ref); - spin_unlock(&i915->fb_tracking.lock); + front = __intel_frontbuffer_get(obj); if (front) return front; @@ -262,13 +258,13 @@ intel_frontbuffer_get(struct drm_i915_gem_object *obj) i915_active_may_sleep(frontbuffer_retire)); spin_lock(&i915->fb_tracking.lock); - if (obj->frontbuffer) { + if (rcu_access_pointer(obj->frontbuffer)) { kfree(front); - front = obj->frontbuffer; + front = rcu_dereference_protected(obj->frontbuffer, true); kref_get(&front->ref); } else { i915_gem_object_get(obj); - obj->frontbuffer = front; + rcu_assign_pointer(obj->frontbuffer, front); } spin_unlock(&i915->fb_tracking.lock); diff --git a/drivers/gpu/drm/i915/display/intel_frontbuffer.h b/drivers/gpu/drm/i915/display/intel_frontbuffer.h index adc64d61a4a5..6d41f5394425 100644 --- a/drivers/gpu/drm/i915/display/intel_frontbuffer.h +++ b/drivers/gpu/drm/i915/display/intel_frontbuffer.h @@ -27,10 +27,10 @@ #include <linux/atomic.h> #include <linux/kref.h> +#include "gem/i915_gem_object_types.h" #include "i915_active.h" struct drm_i915_private; -struct drm_i915_gem_object; enum fb_op_origin { ORIGIN_GTT, @@ -45,6 +45,7 @@ struct intel_frontbuffer { atomic_t bits; struct i915_active write; struct drm_i915_gem_object *obj; + struct rcu_head rcu; }; void intel_frontbuffer_flip_prepare(struct drm_i915_private *i915, @@ -54,6 +55,35 @@ void intel_frontbuffer_flip_complete(struct drm_i915_private *i915, void intel_frontbuffer_flip(struct drm_i915_private *i915, unsigned frontbuffer_bits); +void intel_frontbuffer_put(struct intel_frontbuffer *front); + +static inline struct intel_frontbuffer * +__intel_frontbuffer_get(const struct drm_i915_gem_object *obj) +{ + struct intel_frontbuffer *front; + + if (likely(!rcu_access_pointer(obj->frontbuffer))) + return NULL; + + rcu_read_lock(); + do { + front = rcu_dereference(obj->frontbuffer); + if (!front) + break; + + if (unlikely(!kref_get_unless_zero(&front->ref))) + continue; + + if (likely(front == rcu_access_pointer(obj->frontbuffer))) + break; + + intel_frontbuffer_put(front); + } while (1); + rcu_read_unlock(); + + return front; +} + struct intel_frontbuffer * intel_frontbuffer_get(struct drm_i915_gem_object *obj); @@ -119,6 +149,4 @@ void intel_frontbuffer_track(struct intel_frontbuffer *old, struct intel_frontbuffer *new, unsigned int frontbuffer_bits); -void intel_frontbuffer_put(struct intel_frontbuffer *front); - #endif /* __INTEL_FRONTBUFFER_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c index f1f41ca8402b..a448815d8fc2 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp.c @@ -1821,23 +1821,6 @@ enum mei_fw_tc intel_get_mei_fw_tc(enum transcoder cpu_transcoder) } } -void intel_hdcp_transcoder_config(struct intel_connector *connector, - enum transcoder cpu_transcoder) -{ - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct intel_hdcp *hdcp = &connector->hdcp; - - if (!hdcp->shim) - return; - - if (INTEL_GEN(dev_priv) >= 12) { - mutex_lock(&hdcp->mutex); - hdcp->cpu_transcoder = cpu_transcoder; - hdcp->port_data.fw_tc = intel_get_mei_fw_tc(cpu_transcoder); - mutex_unlock(&hdcp->mutex); - } -} - static inline int initialize_hdcp_port_data(struct intel_connector *connector, const struct intel_hdcp_shim *shim) { @@ -1959,8 +1942,10 @@ int intel_hdcp_init(struct intel_connector *connector, return 0; } -int intel_hdcp_enable(struct intel_connector *connector, u8 content_type) +int intel_hdcp_enable(struct intel_connector *connector, + enum transcoder cpu_transcoder, u8 content_type) { + struct drm_i915_private *dev_priv = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; unsigned long check_link_interval = DRM_HDCP_CHECK_PERIOD_MS; int ret = -EINVAL; @@ -1972,6 +1957,11 @@ int intel_hdcp_enable(struct intel_connector *connector, u8 content_type) WARN_ON(hdcp->value == DRM_MODE_CONTENT_PROTECTION_ENABLED); hdcp->content_type = content_type; + if (INTEL_GEN(dev_priv) >= 12) { + hdcp->cpu_transcoder = cpu_transcoder; + hdcp->port_data.fw_tc = intel_get_mei_fw_tc(cpu_transcoder); + } + /* * Considering that HDCP2.2 is more secure than HDCP1.4, If the setup * is capable of HDCP2.2, it is preferred to use HDCP2.2. diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.h b/drivers/gpu/drm/i915/display/intel_hdcp.h index 41c1053d9e38..f3c3272e712a 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.h +++ b/drivers/gpu/drm/i915/display/intel_hdcp.h @@ -21,11 +21,10 @@ enum transcoder; void intel_hdcp_atomic_check(struct drm_connector *connector, struct drm_connector_state *old_state, struct drm_connector_state *new_state); -void intel_hdcp_transcoder_config(struct intel_connector *connector, - enum transcoder cpu_transcoder); int intel_hdcp_init(struct intel_connector *connector, const struct intel_hdcp_shim *hdcp_shim); -int intel_hdcp_enable(struct intel_connector *connector, u8 content_type); +int intel_hdcp_enable(struct intel_connector *connector, + enum transcoder cpu_transcoder, u8 content_type); int intel_hdcp_disable(struct intel_connector *connector); bool is_hdcp_supported(struct drm_i915_private *dev_priv, enum port port); bool intel_hdcp_capable(struct intel_connector *connector); diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c index f6f5312205c4..f56fffc474fa 100644 --- a/drivers/gpu/drm/i915/display/intel_hdmi.c +++ b/drivers/gpu/drm/i915/display/intel_hdmi.c @@ -2489,9 +2489,6 @@ int intel_hdmi_compute_config(struct intel_encoder *encoder, return -EINVAL; } - intel_hdcp_transcoder_config(intel_hdmi->attached_connector, - pipe_config->cpu_transcoder); - return 0; } diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c index 848ce07a8ec2..8a98a1aa7adc 100644 --- a/drivers/gpu/drm/i915/display/intel_overlay.c +++ b/drivers/gpu/drm/i915/display/intel_overlay.c @@ -279,12 +279,21 @@ static void intel_overlay_flip_prepare(struct intel_overlay *overlay, struct i915_vma *vma) { enum pipe pipe = overlay->crtc->pipe; + struct intel_frontbuffer *from = NULL, *to = NULL; WARN_ON(overlay->old_vma); - intel_frontbuffer_track(overlay->vma ? overlay->vma->obj->frontbuffer : NULL, - vma ? vma->obj->frontbuffer : NULL, - INTEL_FRONTBUFFER_OVERLAY(pipe)); + if (overlay->vma) + from = intel_frontbuffer_get(overlay->vma->obj); + if (vma) + to = intel_frontbuffer_get(vma->obj); + + intel_frontbuffer_track(from, to, INTEL_FRONTBUFFER_OVERLAY(pipe)); + + if (to) + intel_frontbuffer_put(to); + if (from) + intel_frontbuffer_put(from); intel_frontbuffer_flip_prepare(overlay->i915, INTEL_FRONTBUFFER_OVERLAY(pipe)); @@ -766,7 +775,7 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay, ret = PTR_ERR(vma); goto out_pin_section; } - intel_frontbuffer_flush(new_bo->frontbuffer, ORIGIN_DIRTYFB); + i915_gem_object_flush_frontbuffer(new_bo, ORIGIN_DIRTYFB); if (!overlay->active) { u32 oconfig; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_clflush.c b/drivers/gpu/drm/i915/gem/i915_gem_clflush.c index b9f504ba3b32..18ee708585a9 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_clflush.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_clflush.c @@ -20,7 +20,8 @@ static void __do_clflush(struct drm_i915_gem_object *obj) { GEM_BUG_ON(!i915_gem_object_has_pages(obj)); drm_clflush_sg(obj->mm.pages); - intel_frontbuffer_flush(obj->frontbuffer, ORIGIN_CPU); + + i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU); } static int clflush_work(struct dma_fence_work *base) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.c b/drivers/gpu/drm/i915/gem/i915_gem_context.c index 337ba17b1e0e..42385277c684 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_context.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_context.c @@ -2167,8 +2167,7 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data, ext_data.fpriv = file->driver_priv; if (client_is_banned(ext_data.fpriv)) { DRM_DEBUG("client %s[%d] banned from creating ctx\n", - current->comm, - pid_nr(get_task_pid(current, PIDTYPE_PID))); + current->comm, task_pid_nr(current)); return -EIO; } diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c index 9937b4c341f1..f86400a191b0 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c @@ -664,7 +664,7 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data, i915_gem_object_unlock(obj); if (write_domain) - intel_frontbuffer_invalidate(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_invalidate_frontbuffer(obj, ORIGIN_CPU); out_unpin: i915_gem_object_unpin_pages(obj); @@ -784,7 +784,7 @@ int i915_gem_object_prepare_write(struct drm_i915_gem_object *obj, } out: - intel_frontbuffer_invalidate(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_invalidate_frontbuffer(obj, ORIGIN_CPU); obj->mm.dirty = true; /* return with the pages pinned */ return 0; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index f0998f1225af..bc3a67226163 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -2694,6 +2694,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, err = eb_submit(&eb); err_request: add_to_client(eb.request, file); + i915_request_get(eb.request); i915_request_add(eb.request); if (fences) @@ -2709,6 +2710,7 @@ err_request: fput(out_fence->file); } } + i915_request_put(eb.request); err_batch_unpin: if (eb.batch_flags & I915_DISPATCH_SECURE) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c index a50296cce0d8..a596548c07bf 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c @@ -280,7 +280,7 @@ i915_gem_object_flush_write_domain(struct drm_i915_gem_object *obj, for_each_ggtt_vma(vma, obj) intel_gt_flush_ggtt_writes(vma->vm->gt); - intel_frontbuffer_flush(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU); for_each_ggtt_vma(vma, obj) { if (vma->iomap) @@ -308,6 +308,30 @@ i915_gem_object_flush_write_domain(struct drm_i915_gem_object *obj, obj->write_domain = 0; } +void __i915_gem_object_flush_frontbuffer(struct drm_i915_gem_object *obj, + enum fb_op_origin origin) +{ + struct intel_frontbuffer *front; + + front = __intel_frontbuffer_get(obj); + if (front) { + intel_frontbuffer_flush(front, origin); + intel_frontbuffer_put(front); + } +} + +void __i915_gem_object_invalidate_frontbuffer(struct drm_i915_gem_object *obj, + enum fb_op_origin origin) +{ + struct intel_frontbuffer *front; + + front = __intel_frontbuffer_get(obj); + if (front) { + intel_frontbuffer_invalidate(front, origin); + intel_frontbuffer_put(front); + } +} + void i915_gem_init__objects(struct drm_i915_private *i915) { INIT_WORK(&i915->mm.free_work, __i915_gem_free_work); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h index 458cd51331f1..4b93591fd5c7 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h @@ -13,8 +13,8 @@ #include <drm/i915_drm.h> +#include "display/intel_frontbuffer.h" #include "i915_gem_object_types.h" - #include "i915_gem_gtt.h" void i915_gem_init__objects(struct drm_i915_private *i915); @@ -463,4 +463,25 @@ int i915_gem_object_wait_priority(struct drm_i915_gem_object *obj, unsigned int flags, const struct i915_sched_attr *attr); +void __i915_gem_object_flush_frontbuffer(struct drm_i915_gem_object *obj, + enum fb_op_origin origin); +void __i915_gem_object_invalidate_frontbuffer(struct drm_i915_gem_object *obj, + enum fb_op_origin origin); + +static inline void +i915_gem_object_flush_frontbuffer(struct drm_i915_gem_object *obj, + enum fb_op_origin origin) +{ + if (unlikely(rcu_access_pointer(obj->frontbuffer))) + __i915_gem_object_flush_frontbuffer(obj, origin); +} + +static inline void +i915_gem_object_invalidate_frontbuffer(struct drm_i915_gem_object *obj, + enum fb_op_origin origin) +{ + if (unlikely(rcu_access_pointer(obj->frontbuffer))) + __i915_gem_object_invalidate_frontbuffer(obj, origin); +} + #endif diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h index 96008374a412..e3f3944fbd90 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h @@ -150,7 +150,7 @@ struct drm_i915_gem_object { */ u16 write_domain; - struct intel_frontbuffer *frontbuffer; + struct intel_frontbuffer __rcu *frontbuffer; /** Current tiling stride for the object, if it's tiled. */ unsigned int tiling_and_stride; diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm.c b/drivers/gpu/drm/i915/gt/intel_gt_pm.c index a459a42ad5c2..7e64b7d7d330 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_pm.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_pm.c @@ -94,8 +94,9 @@ static int __gt_park(struct intel_wakeref *wf) intel_uncore_forcewake_put(&i915->uncore, FORCEWAKE_ALL); } + /* Defer dropping the display power well for 100ms, it's slow! */ GEM_BUG_ON(!wakeref); - intel_display_power_put(i915, POWER_DOMAIN_GT_IRQ, wakeref); + intel_display_power_put_async(i915, POWER_DOMAIN_GT_IRQ, wakeref); i915_globals_park(); diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c index 9fdefbdc3546..75dd0e0367b7 100644 --- a/drivers/gpu/drm/i915/gt/intel_lrc.c +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c @@ -845,12 +845,6 @@ static const u8 *reg_offsets(const struct intel_engine_cs *engine) } } -static void unwind_wa_tail(struct i915_request *rq) -{ - rq->tail = intel_ring_wrap(rq->ring, rq->wa_tail - WA_TAIL_BYTES); - assert_ring_tail_valid(rq->ring, rq->tail); -} - static struct i915_request * __unwind_incomplete_requests(struct intel_engine_cs *engine) { @@ -863,12 +857,10 @@ __unwind_incomplete_requests(struct intel_engine_cs *engine) list_for_each_entry_safe_reverse(rq, rn, &engine->active.requests, sched.link) { - if (i915_request_completed(rq)) continue; /* XXX */ __i915_request_unsubmit(rq); - unwind_wa_tail(rq); /* * Push the request back into the queue for later resubmission. @@ -1161,13 +1153,29 @@ execlists_schedule_out(struct i915_request *rq) i915_request_put(rq); } -static u64 execlists_update_context(const struct i915_request *rq) +static u64 execlists_update_context(struct i915_request *rq) { struct intel_context *ce = rq->hw_context; - u64 desc; + u64 desc = ce->lrc_desc; + u32 tail; - ce->lrc_reg_state[CTX_RING_TAIL] = - intel_ring_set_tail(rq->ring, rq->tail); + /* + * WaIdleLiteRestore:bdw,skl + * + * We should never submit the context with the same RING_TAIL twice + * just in case we submit an empty ring, which confuses the HW. + * + * We append a couple of NOOPs (gen8_emit_wa_tail) after the end of + * the normal request to be able to always advance the RING_TAIL on + * subsequent resubmissions (for lite restore). Should that fail us, + * and we try and submit the same tail again, force the context + * reload. + */ + tail = intel_ring_set_tail(rq->ring, rq->tail); + if (unlikely(ce->lrc_reg_state[CTX_RING_TAIL] == tail)) + desc |= CTX_DESC_FORCE_RESTORE; + ce->lrc_reg_state[CTX_RING_TAIL] = tail; + rq->tail = rq->wa_tail; /* * Make sure the context image is complete before we submit it to HW. @@ -1186,13 +1194,11 @@ static u64 execlists_update_context(const struct i915_request *rq) */ mb(); - desc = ce->lrc_desc; - ce->lrc_desc &= ~CTX_DESC_FORCE_RESTORE; - /* Wa_1607138340:tgl */ if (IS_TGL_REVID(rq->i915, TGL_REVID_A0, TGL_REVID_A0)) desc |= CTX_DESC_FORCE_RESTORE; + ce->lrc_desc &= ~CTX_DESC_FORCE_RESTORE; return desc; } @@ -1703,16 +1709,6 @@ static void execlists_dequeue(struct intel_engine_cs *engine) return; } - - /* - * WaIdleLiteRestore:bdw,skl - * Apply the wa NOOPs to prevent - * ring:HEAD == rq:TAIL as we resubmit the - * request. See gen8_emit_fini_breadcrumb() for - * where we prepare the padding after the - * end of the request. - */ - last->tail = last->wa_tail; } } @@ -4120,17 +4116,18 @@ static void virtual_context_destroy(struct kref *kref) for (n = 0; n < ve->num_siblings; n++) { struct intel_engine_cs *sibling = ve->siblings[n]; struct rb_node *node = &ve->nodes[sibling->id].rb; + unsigned long flags; if (RB_EMPTY_NODE(node)) continue; - spin_lock_irq(&sibling->active.lock); + spin_lock_irqsave(&sibling->active.lock, flags); /* Detachment is lazily performed in the execlists tasklet */ if (!RB_EMPTY_NODE(node)) rb_erase_cached(node, &sibling->execlists.virtual); - spin_unlock_irq(&sibling->active.lock); + spin_unlock_irqrestore(&sibling->active.lock, flags); } GEM_BUG_ON(__tasklet_is_scheduled(&ve->base.execlists.tasklet)); diff --git a/drivers/gpu/drm/i915/gvt/dmabuf.c b/drivers/gpu/drm/i915/gvt/dmabuf.c index e451298d11c3..2477a1e5a166 100644 --- a/drivers/gpu/drm/i915/gvt/dmabuf.c +++ b/drivers/gpu/drm/i915/gvt/dmabuf.c @@ -36,13 +36,32 @@ #define GEN8_DECODE_PTE(pte) (pte & GENMASK_ULL(63, 12)) +static int vgpu_pin_dma_address(struct intel_vgpu *vgpu, + unsigned long size, + dma_addr_t dma_addr) +{ + int ret = 0; + + if (intel_gvt_hypervisor_dma_pin_guest_page(vgpu, dma_addr)) + ret = -EINVAL; + + return ret; +} + +static void vgpu_unpin_dma_address(struct intel_vgpu *vgpu, + dma_addr_t dma_addr) +{ + intel_gvt_hypervisor_dma_unmap_guest_page(vgpu, dma_addr); +} + static int vgpu_gem_get_pages( struct drm_i915_gem_object *obj) { struct drm_i915_private *dev_priv = to_i915(obj->base.dev); + struct intel_vgpu *vgpu; struct sg_table *st; struct scatterlist *sg; - int i, ret; + int i, j, ret; gen8_pte_t __iomem *gtt_entries; struct intel_vgpu_fb_info *fb_info; u32 page_num; @@ -51,6 +70,10 @@ static int vgpu_gem_get_pages( if (WARN_ON(!fb_info)) return -ENODEV; + vgpu = fb_info->obj->vgpu; + if (WARN_ON(!vgpu)) + return -ENODEV; + st = kmalloc(sizeof(*st), GFP_KERNEL); if (unlikely(!st)) return -ENOMEM; @@ -64,21 +87,53 @@ static int vgpu_gem_get_pages( gtt_entries = (gen8_pte_t __iomem *)dev_priv->ggtt.gsm + (fb_info->start >> PAGE_SHIFT); for_each_sg(st->sgl, sg, page_num, i) { + dma_addr_t dma_addr = + GEN8_DECODE_PTE(readq(>t_entries[i])); + if (vgpu_pin_dma_address(vgpu, PAGE_SIZE, dma_addr)) { + ret = -EINVAL; + goto out; + } + sg->offset = 0; sg->length = PAGE_SIZE; - sg_dma_address(sg) = - GEN8_DECODE_PTE(readq(>t_entries[i])); sg_dma_len(sg) = PAGE_SIZE; + sg_dma_address(sg) = dma_addr; } __i915_gem_object_set_pages(obj, st, PAGE_SIZE); +out: + if (ret) { + dma_addr_t dma_addr; + + for_each_sg(st->sgl, sg, i, j) { + dma_addr = sg_dma_address(sg); + if (dma_addr) + vgpu_unpin_dma_address(vgpu, dma_addr); + } + sg_free_table(st); + kfree(st); + } + + return ret; - return 0; } static void vgpu_gem_put_pages(struct drm_i915_gem_object *obj, struct sg_table *pages) { + struct scatterlist *sg; + + if (obj->base.dma_buf) { + struct intel_vgpu_fb_info *fb_info = obj->gvt_info; + struct intel_vgpu_dmabuf_obj *obj = fb_info->obj; + struct intel_vgpu *vgpu = obj->vgpu; + int i; + + for_each_sg(pages->sgl, sg, fb_info->size, i) + vgpu_unpin_dma_address(vgpu, + sg_dma_address(sg)); + } + sg_free_table(pages); kfree(pages); } @@ -163,6 +218,7 @@ static struct drm_i915_gem_object *vgpu_create_gem(struct drm_device *dev, drm_gem_private_object_init(dev, &obj->base, roundup(info->size, PAGE_SIZE)); i915_gem_object_init(obj, &intel_vgpu_gem_ops, &lock_class); + i915_gem_object_set_readonly(obj); obj->read_domains = I915_GEM_DOMAIN_GTT; obj->write_domain = 0; diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c index bb9fe6bf5275..1043e6d564df 100644 --- a/drivers/gpu/drm/i915/gvt/handlers.c +++ b/drivers/gpu/drm/i915/gvt/handlers.c @@ -341,6 +341,10 @@ static int gdrst_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, gvt_dbg_mmio("vgpu%d: request VCS2 Reset\n", vgpu->id); engine_mask |= BIT(VCS1); } + if (data & GEN9_GRDOM_GUC) { + gvt_dbg_mmio("vgpu%d: request GUC Reset\n", vgpu->id); + vgpu_vreg_t(vgpu, GUC_STATUS) |= GS_MIA_IN_RESET; + } engine_mask &= INTEL_INFO(vgpu->gvt->dev_priv)->engine_mask; } @@ -1636,6 +1640,16 @@ static int edp_psr_imr_iir_write(struct intel_vgpu *vgpu, return 0; } +static int guc_status_read(struct intel_vgpu *vgpu, + unsigned int offset, void *p_data, + unsigned int bytes) +{ + /* keep MIA_IN_RESET before clearing */ + read_vreg(vgpu, offset, p_data, bytes); + vgpu_vreg(vgpu, offset) &= ~GS_MIA_IN_RESET; + return 0; +} + static int mmio_read_from_hw(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { @@ -2672,6 +2686,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt) MMIO_DH(EDP_PSR_IMR, D_BDW_PLUS, NULL, edp_psr_imr_iir_write); MMIO_DH(EDP_PSR_IIR, D_BDW_PLUS, NULL, edp_psr_imr_iir_write); + MMIO_DH(GUC_STATUS, D_ALL, guc_status_read, NULL); + return 0; } diff --git a/drivers/gpu/drm/i915/gvt/hypercall.h b/drivers/gpu/drm/i915/gvt/hypercall.h index 4862fb12778e..b19a3b1ea4c1 100644 --- a/drivers/gpu/drm/i915/gvt/hypercall.h +++ b/drivers/gpu/drm/i915/gvt/hypercall.h @@ -62,6 +62,8 @@ struct intel_gvt_mpt { unsigned long size, dma_addr_t *dma_addr); void (*dma_unmap_guest_page)(unsigned long handle, dma_addr_t dma_addr); + int (*dma_pin_guest_page)(unsigned long handle, dma_addr_t dma_addr); + int (*map_gfn_to_mfn)(unsigned long handle, unsigned long gfn, unsigned long mfn, unsigned int nr, bool map); int (*set_trap_area)(unsigned long handle, u64 start, u64 end, diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c index 04a5a0d90823..3259a1fa69e1 100644 --- a/drivers/gpu/drm/i915/gvt/kvmgt.c +++ b/drivers/gpu/drm/i915/gvt/kvmgt.c @@ -1916,6 +1916,28 @@ err_unlock: return ret; } +static int kvmgt_dma_pin_guest_page(unsigned long handle, dma_addr_t dma_addr) +{ + struct kvmgt_guest_info *info; + struct gvt_dma *entry; + int ret = 0; + + if (!handle_valid(handle)) + return -ENODEV; + + info = (struct kvmgt_guest_info *)handle; + + mutex_lock(&info->vgpu->vdev.cache_lock); + entry = __gvt_cache_find_dma_addr(info->vgpu, dma_addr); + if (entry) + kref_get(&entry->ref); + else + ret = -ENOMEM; + mutex_unlock(&info->vgpu->vdev.cache_lock); + + return ret; +} + static void __gvt_dma_release(struct kref *ref) { struct gvt_dma *entry = container_of(ref, typeof(*entry), ref); @@ -2027,6 +2049,7 @@ static struct intel_gvt_mpt kvmgt_mpt = { .gfn_to_mfn = kvmgt_gfn_to_pfn, .dma_map_guest_page = kvmgt_dma_map_guest_page, .dma_unmap_guest_page = kvmgt_dma_unmap_guest_page, + .dma_pin_guest_page = kvmgt_dma_pin_guest_page, .set_opregion = kvmgt_set_opregion, .set_edid = kvmgt_set_edid, .get_vfio_device = kvmgt_get_vfio_device, diff --git a/drivers/gpu/drm/i915/gvt/mpt.h b/drivers/gpu/drm/i915/gvt/mpt.h index 0f9440128123..9ad224df9c68 100644 --- a/drivers/gpu/drm/i915/gvt/mpt.h +++ b/drivers/gpu/drm/i915/gvt/mpt.h @@ -255,6 +255,21 @@ static inline void intel_gvt_hypervisor_dma_unmap_guest_page( } /** + * intel_gvt_hypervisor_dma_pin_guest_page - pin guest dma buf + * @vgpu: a vGPU + * @dma_addr: guest dma addr + * + * Returns: + * 0 on success, negative error code if failed. + */ +static inline int +intel_gvt_hypervisor_dma_pin_guest_page(struct intel_vgpu *vgpu, + dma_addr_t dma_addr) +{ + return intel_gvt_host.mpt->dma_pin_guest_page(vgpu->handle, dma_addr); +} + +/** * intel_gvt_hypervisor_map_gfn_to_mfn - map a GFN region to MFN * @vgpu: a vGPU * @gfn: guest PFN diff --git a/drivers/gpu/drm/i915/gvt/vgpu.c b/drivers/gpu/drm/i915/gvt/vgpu.c index d5a6e4e3d0fd..85bd9bf4f6ee 100644 --- a/drivers/gpu/drm/i915/gvt/vgpu.c +++ b/drivers/gpu/drm/i915/gvt/vgpu.c @@ -212,9 +212,9 @@ static void intel_gvt_update_vgpu_types(struct intel_gvt *gvt) */ void intel_gvt_activate_vgpu(struct intel_vgpu *vgpu) { - mutex_lock(&vgpu->gvt->lock); + mutex_lock(&vgpu->vgpu_lock); vgpu->active = true; - mutex_unlock(&vgpu->gvt->lock); + mutex_unlock(&vgpu->vgpu_lock); } /** diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index b9eb6b3149b7..905890e3ac24 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -45,6 +45,7 @@ #include "gem/i915_gem_context.h" #include "gem/i915_gem_ioctls.h" #include "gem/i915_gem_pm.h" +#include "gt/intel_context.h" #include "gt/intel_engine_user.h" #include "gt/intel_gt.h" #include "gt/intel_gt_pm.h" @@ -160,7 +161,7 @@ i915_gem_phys_pwrite(struct drm_i915_gem_object *obj, * We manually control the domain here and pretend that it * remains coherent i.e. in the GTT domain, like shmem_pwrite. */ - intel_frontbuffer_invalidate(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_invalidate_frontbuffer(obj, ORIGIN_CPU); if (copy_from_user(vaddr, user_data, args->size)) return -EFAULT; @@ -168,7 +169,7 @@ i915_gem_phys_pwrite(struct drm_i915_gem_object *obj, drm_clflush_virt_range(vaddr, args->size); intel_gt_chipset_flush(&to_i915(obj->base.dev)->gt); - intel_frontbuffer_flush(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU); return 0; } @@ -588,7 +589,7 @@ i915_gem_gtt_pwrite_fast(struct drm_i915_gem_object *obj, goto out_unpin; } - intel_frontbuffer_invalidate(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_invalidate_frontbuffer(obj, ORIGIN_CPU); user_data = u64_to_user_ptr(args->data_ptr); offset = args->offset; @@ -630,7 +631,7 @@ i915_gem_gtt_pwrite_fast(struct drm_i915_gem_object *obj, user_data += page_length; offset += page_length; } - intel_frontbuffer_flush(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU); i915_gem_object_unlock_fence(obj, fence); out_unpin: @@ -720,7 +721,7 @@ i915_gem_shmem_pwrite(struct drm_i915_gem_object *obj, offset = 0; } - intel_frontbuffer_flush(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU); i915_gem_object_unlock_fence(obj, fence); return ret; @@ -1053,6 +1054,18 @@ out: return err; } +static int __intel_context_flush_retire(struct intel_context *ce) +{ + struct intel_timeline *tl; + + tl = intel_context_timeline_lock(ce); + if (IS_ERR(tl)) + return PTR_ERR(tl); + + intel_context_timeline_unlock(tl); + return 0; +} + static int __intel_engines_record_defaults(struct intel_gt *gt) { struct i915_request *requests[I915_NUM_ENGINES] = {}; @@ -1121,13 +1134,20 @@ err_rq: if (!rq) continue; - /* We want to be able to unbind the state from the GGTT */ - GEM_BUG_ON(intel_context_is_pinned(rq->hw_context)); - + GEM_BUG_ON(!test_bit(CONTEXT_ALLOC_BIT, + &rq->hw_context->flags)); state = rq->hw_context->state; if (!state) continue; + /* Serialise with retirement on another CPU */ + err = __intel_context_flush_retire(rq->hw_context); + if (err) + goto out; + + /* We want to be able to unbind the state from the GGTT */ + GEM_BUG_ON(intel_context_is_pinned(rq->hw_context)); + /* * As we will hold a reference to the logical state, it will * not be torn down with the context, and importantly the diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c index 65d7c2e599de..2ae14bc14931 100644 --- a/drivers/gpu/drm/i915/i915_perf.c +++ b/drivers/gpu/drm/i915/i915_perf.c @@ -2078,20 +2078,12 @@ gen8_update_reg_state_unlocked(const struct intel_context *ce, u32 *reg_state = ce->lrc_reg_state; int i; - if (IS_GEN(stream->perf->i915, 12)) { - u32 format = stream->oa_buffer.format; + reg_state[ctx_oactxctrl + 1] = + (stream->period_exponent << GEN8_OA_TIMER_PERIOD_SHIFT) | + (stream->periodic ? GEN8_OA_TIMER_ENABLE : 0) | + GEN8_OA_COUNTER_RESUME; - reg_state[ctx_oactxctrl + 1] = - (format << GEN12_OAR_OACONTROL_COUNTER_FORMAT_SHIFT) | - (stream->oa_config ? GEN12_OAR_OACONTROL_COUNTER_ENABLE : 0); - } else { - reg_state[ctx_oactxctrl + 1] = - (stream->period_exponent << GEN8_OA_TIMER_PERIOD_SHIFT) | - (stream->periodic ? GEN8_OA_TIMER_ENABLE : 0) | - GEN8_OA_COUNTER_RESUME; - } - - for (i = 0; !!ctx_flexeu0 && i < ARRAY_SIZE(flex_regs); i++) + for (i = 0; i < ARRAY_SIZE(flex_regs); i++) reg_state[ctx_flexeu0 + i * 2 + 1] = oa_config_flex_reg(stream->oa_config, flex_regs[i]); @@ -2224,34 +2216,51 @@ static int gen8_configure_context(struct i915_gem_context *ctx, return err; } -static int gen12_emit_oar_config(struct intel_context *ce, bool enable) +static int gen12_configure_oar_context(struct i915_perf_stream *stream, bool enable) { - struct i915_request *rq; - u32 *cs; - int err = 0; - - rq = i915_request_create(ce); - if (IS_ERR(rq)) - return PTR_ERR(rq); - - cs = intel_ring_begin(rq, 4); - if (IS_ERR(cs)) { - err = PTR_ERR(cs); - goto out; - } - - *cs++ = MI_LOAD_REGISTER_IMM(1); - *cs++ = i915_mmio_reg_offset(RING_CONTEXT_CONTROL(ce->engine->mmio_base)); - *cs++ = _MASKED_FIELD(GEN12_CTX_CTRL_OAR_CONTEXT_ENABLE, - enable ? GEN12_CTX_CTRL_OAR_CONTEXT_ENABLE : 0); - *cs++ = MI_NOOP; + int err; + struct intel_context *ce = stream->pinned_ctx; + u32 format = stream->oa_buffer.format; + struct flex regs_context[] = { + { + GEN8_OACTXCONTROL, + stream->perf->ctx_oactxctrl_offset + 1, + enable ? GEN8_OA_COUNTER_RESUME : 0, + }, + }; + /* Offsets in regs_lri are not used since this configuration is only + * applied using LRI. Initialize the correct offsets for posterity. + */ +#define GEN12_OAR_OACONTROL_OFFSET 0x5B0 + struct flex regs_lri[] = { + { + GEN12_OAR_OACONTROL, + GEN12_OAR_OACONTROL_OFFSET + 1, + (format << GEN12_OAR_OACONTROL_COUNTER_FORMAT_SHIFT) | + (enable ? GEN12_OAR_OACONTROL_COUNTER_ENABLE : 0) + }, + { + RING_CONTEXT_CONTROL(ce->engine->mmio_base), + CTX_CONTEXT_CONTROL, + _MASKED_FIELD(GEN12_CTX_CTRL_OAR_CONTEXT_ENABLE, + enable ? + GEN12_CTX_CTRL_OAR_CONTEXT_ENABLE : + 0) + }, + }; - intel_ring_advance(rq, cs); + /* Modify the context image of pinned context with regs_context*/ + err = intel_context_lock_pinned(ce); + if (err) + return err; -out: - i915_request_add(rq); + err = gen8_modify_context(ce, regs_context, ARRAY_SIZE(regs_context)); + intel_context_unlock_pinned(ce); + if (err) + return err; - return err; + /* Apply regs_lri using LRI with pinned context */ + return gen8_modify_self(ce, regs_lri, ARRAY_SIZE(regs_lri)); } /* @@ -2277,53 +2286,16 @@ out: * per-context OA state. * * Note: it's only the RCS/Render context that has any OA state. + * Note: the first flex register passed must always be R_PWR_CLK_STATE */ -static int lrc_configure_all_contexts(struct i915_perf_stream *stream, - const struct i915_oa_config *oa_config) +static int oa_configure_all_contexts(struct i915_perf_stream *stream, + struct flex *regs, + size_t num_regs) { struct drm_i915_private *i915 = stream->perf->i915; - /* The MMIO offsets for Flex EU registers aren't contiguous */ - const u32 ctx_flexeu0 = stream->perf->ctx_flexeu0_offset; -#define ctx_flexeuN(N) (ctx_flexeu0 + 2 * (N) + 1) - struct flex regs[] = { - { - GEN8_R_PWR_CLK_STATE, - CTX_R_PWR_CLK_STATE, - }, - { - IS_GEN(i915, 12) ? - GEN12_OAR_OACONTROL : GEN8_OACTXCONTROL, - stream->perf->ctx_oactxctrl_offset + 1, - }, - { EU_PERF_CNTL0, ctx_flexeuN(0) }, - { EU_PERF_CNTL1, ctx_flexeuN(1) }, - { EU_PERF_CNTL2, ctx_flexeuN(2) }, - { EU_PERF_CNTL3, ctx_flexeuN(3) }, - { EU_PERF_CNTL4, ctx_flexeuN(4) }, - { EU_PERF_CNTL5, ctx_flexeuN(5) }, - { EU_PERF_CNTL6, ctx_flexeuN(6) }, - }; -#undef ctx_flexeuN struct intel_engine_cs *engine; struct i915_gem_context *ctx, *cn; - size_t array_size = IS_GEN(i915, 12) ? 2 : ARRAY_SIZE(regs); - int i, err; - - if (IS_GEN(i915, 12)) { - u32 format = stream->oa_buffer.format; - - regs[1].value = - (format << GEN12_OAR_OACONTROL_COUNTER_FORMAT_SHIFT) | - (oa_config ? GEN12_OAR_OACONTROL_COUNTER_ENABLE : 0); - } else { - regs[1].value = - (stream->period_exponent << GEN8_OA_TIMER_PERIOD_SHIFT) | - (stream->periodic ? GEN8_OA_TIMER_ENABLE : 0) | - GEN8_OA_COUNTER_RESUME; - } - - for (i = 2; !!ctx_flexeu0 && i < array_size; i++) - regs[i].value = oa_config_flex_reg(oa_config, regs[i].reg); + int err; lockdep_assert_held(&stream->perf->lock); @@ -2353,7 +2325,7 @@ static int lrc_configure_all_contexts(struct i915_perf_stream *stream, spin_unlock(&i915->gem.contexts.lock); - err = gen8_configure_context(ctx, regs, array_size); + err = gen8_configure_context(ctx, regs, num_regs); if (err) { i915_gem_context_put(ctx); return err; @@ -2378,7 +2350,7 @@ static int lrc_configure_all_contexts(struct i915_perf_stream *stream, regs[0].value = intel_sseu_make_rpcs(i915, &ce->sseu); - err = gen8_modify_self(ce, regs, array_size); + err = gen8_modify_self(ce, regs, num_regs); if (err) return err; } @@ -2386,6 +2358,56 @@ static int lrc_configure_all_contexts(struct i915_perf_stream *stream, return 0; } +static int gen12_configure_all_contexts(struct i915_perf_stream *stream, + const struct i915_oa_config *oa_config) +{ + struct flex regs[] = { + { + GEN8_R_PWR_CLK_STATE, + CTX_R_PWR_CLK_STATE, + }, + }; + + return oa_configure_all_contexts(stream, regs, ARRAY_SIZE(regs)); +} + +static int lrc_configure_all_contexts(struct i915_perf_stream *stream, + const struct i915_oa_config *oa_config) +{ + /* The MMIO offsets for Flex EU registers aren't contiguous */ + const u32 ctx_flexeu0 = stream->perf->ctx_flexeu0_offset; +#define ctx_flexeuN(N) (ctx_flexeu0 + 2 * (N) + 1) + struct flex regs[] = { + { + GEN8_R_PWR_CLK_STATE, + CTX_R_PWR_CLK_STATE, + }, + { + GEN8_OACTXCONTROL, + stream->perf->ctx_oactxctrl_offset + 1, + }, + { EU_PERF_CNTL0, ctx_flexeuN(0) }, + { EU_PERF_CNTL1, ctx_flexeuN(1) }, + { EU_PERF_CNTL2, ctx_flexeuN(2) }, + { EU_PERF_CNTL3, ctx_flexeuN(3) }, + { EU_PERF_CNTL4, ctx_flexeuN(4) }, + { EU_PERF_CNTL5, ctx_flexeuN(5) }, + { EU_PERF_CNTL6, ctx_flexeuN(6) }, + }; +#undef ctx_flexeuN + int i; + + regs[1].value = + (stream->period_exponent << GEN8_OA_TIMER_PERIOD_SHIFT) | + (stream->periodic ? GEN8_OA_TIMER_ENABLE : 0) | + GEN8_OA_COUNTER_RESUME; + + for (i = 2; i < ARRAY_SIZE(regs); i++) + regs[i].value = oa_config_flex_reg(oa_config, regs[i].reg); + + return oa_configure_all_contexts(stream, regs, ARRAY_SIZE(regs)); +} + static int gen8_enable_metric_set(struct i915_perf_stream *stream) { struct intel_uncore *uncore = stream->uncore; @@ -2464,7 +2486,7 @@ static int gen12_enable_metric_set(struct i915_perf_stream *stream) * to make sure all slices/subslices are ON before writing to NOA * registers. */ - ret = lrc_configure_all_contexts(stream, oa_config); + ret = gen12_configure_all_contexts(stream, oa_config); if (ret) return ret; @@ -2474,8 +2496,7 @@ static int gen12_enable_metric_set(struct i915_perf_stream *stream) * requested this. */ if (stream->ctx) { - ret = gen12_emit_oar_config(stream->pinned_ctx, - oa_config != NULL); + ret = gen12_configure_oar_context(stream, true); if (ret) return ret; } @@ -2509,11 +2530,11 @@ static void gen12_disable_metric_set(struct i915_perf_stream *stream) struct intel_uncore *uncore = stream->uncore; /* Reset all contexts' slices/subslices configurations. */ - lrc_configure_all_contexts(stream, NULL); + gen12_configure_all_contexts(stream, NULL); /* disable the context save/restore or OAR counters */ if (stream->ctx) - gen12_emit_oar_config(stream->pinned_ctx, false); + gen12_configure_oar_context(stream, false); /* Make sure we disable noa to save power. */ intel_uncore_rmw(uncore, RPM_CONFIG1, GEN10_GT_NOA_ENABLE, 0); @@ -2713,7 +2734,8 @@ static int i915_oa_stream_init(struct i915_perf_stream *stream, return -EINVAL; } - if (!(props->sample_flags & SAMPLE_OA_REPORT)) { + if (!(props->sample_flags & SAMPLE_OA_REPORT) && + (INTEL_GEN(perf->i915) < 12 || !stream->ctx)) { DRM_DEBUG("Only OA report sampling supported\n"); return -EINVAL; } @@ -2745,7 +2767,7 @@ static int i915_oa_stream_init(struct i915_perf_stream *stream, format_size = perf->oa_formats[props->oa_format].size; - stream->sample_flags |= SAMPLE_OA_REPORT; + stream->sample_flags = props->sample_flags; stream->sample_size += format_size; stream->oa_buffer.format_size = format_size; @@ -2854,7 +2876,11 @@ void i915_oa_init_reg_state(const struct intel_context *ce, return; stream = engine->i915->perf.exclusive_stream; - if (stream) + /* + * For gen12, only CTX_R_PWR_CLK_STATE needs update, but the caller + * is already doing that, so nothing to be done for gen12 here. + */ + if (stream && INTEL_GEN(stream->perf->i915) < 12) gen8_update_reg_state_unlocked(ce, stream); } diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c index 2814218c5ba1..6f09aa0be80a 100644 --- a/drivers/gpu/drm/i915/i915_pmu.c +++ b/drivers/gpu/drm/i915/i915_pmu.c @@ -144,61 +144,40 @@ static inline s64 ktime_since(const ktime_t kt) return ktime_to_ns(ktime_sub(ktime_get(), kt)); } -static u64 __pmu_estimate_rc6(struct i915_pmu *pmu) -{ - u64 val; - - /* - * We think we are runtime suspended. - * - * Report the delta from when the device was suspended to now, - * on top of the last known real value, as the approximated RC6 - * counter value. - */ - val = ktime_since(pmu->sleep_last); - val += pmu->sample[__I915_SAMPLE_RC6].cur; - - pmu->sample[__I915_SAMPLE_RC6_ESTIMATED].cur = val; - - return val; -} - -static u64 __pmu_update_rc6(struct i915_pmu *pmu, u64 val) -{ - /* - * If we are coming back from being runtime suspended we must - * be careful not to report a larger value than returned - * previously. - */ - if (val >= pmu->sample[__I915_SAMPLE_RC6_ESTIMATED].cur) { - pmu->sample[__I915_SAMPLE_RC6_ESTIMATED].cur = 0; - pmu->sample[__I915_SAMPLE_RC6].cur = val; - } else { - val = pmu->sample[__I915_SAMPLE_RC6_ESTIMATED].cur; - } - - return val; -} - static u64 get_rc6(struct intel_gt *gt) { struct drm_i915_private *i915 = gt->i915; struct i915_pmu *pmu = &i915->pmu; unsigned long flags; + bool awake = false; u64 val; - val = 0; if (intel_gt_pm_get_if_awake(gt)) { val = __get_rc6(gt); intel_gt_pm_put_async(gt); + awake = true; } spin_lock_irqsave(&pmu->lock, flags); - if (val) - val = __pmu_update_rc6(pmu, val); + if (awake) { + pmu->sample[__I915_SAMPLE_RC6].cur = val; + } else { + /* + * We think we are runtime suspended. + * + * Report the delta from when the device was suspended to now, + * on top of the last known real value, as the approximated RC6 + * counter value. + */ + val = ktime_since(pmu->sleep_last); + val += pmu->sample[__I915_SAMPLE_RC6].cur; + } + + if (val < pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur) + val = pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur; else - val = __pmu_estimate_rc6(pmu); + pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur = val; spin_unlock_irqrestore(&pmu->lock, flags); @@ -210,20 +189,11 @@ static void park_rc6(struct drm_i915_private *i915) struct i915_pmu *pmu = &i915->pmu; if (pmu->enable & config_enabled_mask(I915_PMU_RC6_RESIDENCY)) - __pmu_update_rc6(pmu, __get_rc6(&i915->gt)); + pmu->sample[__I915_SAMPLE_RC6].cur = __get_rc6(&i915->gt); pmu->sleep_last = ktime_get(); } -static void unpark_rc6(struct drm_i915_private *i915) -{ - struct i915_pmu *pmu = &i915->pmu; - - /* Estimate how long we slept and accumulate that into rc6 counters */ - if (pmu->enable & config_enabled_mask(I915_PMU_RC6_RESIDENCY)) - __pmu_estimate_rc6(pmu); -} - #else static u64 get_rc6(struct intel_gt *gt) @@ -232,7 +202,6 @@ static u64 get_rc6(struct intel_gt *gt) } static void park_rc6(struct drm_i915_private *i915) {} -static void unpark_rc6(struct drm_i915_private *i915) {} #endif @@ -281,8 +250,6 @@ void i915_pmu_gt_unparked(struct drm_i915_private *i915) */ __i915_pmu_maybe_start_timer(pmu); - unpark_rc6(i915); - spin_unlock_irq(&pmu->lock); } diff --git a/drivers/gpu/drm/i915/i915_pmu.h b/drivers/gpu/drm/i915/i915_pmu.h index bf52e3983631..6c1647c5daf2 100644 --- a/drivers/gpu/drm/i915/i915_pmu.h +++ b/drivers/gpu/drm/i915/i915_pmu.h @@ -18,7 +18,7 @@ enum { __I915_SAMPLE_FREQ_ACT = 0, __I915_SAMPLE_FREQ_REQ, __I915_SAMPLE_RC6, - __I915_SAMPLE_RC6_ESTIMATED, + __I915_SAMPLE_RC6_LAST_REPORTED, __I915_NUM_PMU_SAMPLERS }; diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 73079b503724..4fd3d76db346 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -9405,11 +9405,9 @@ enum skl_power_gate { #define _ICL_AUX_REG_IDX(pw_idx) ((pw_idx) - ICL_PW_CTL_IDX_AUX_A) #define _ICL_AUX_ANAOVRD1_A 0x162398 #define _ICL_AUX_ANAOVRD1_B 0x6C398 -#define _TGL_AUX_ANAOVRD1_C 0x160398 #define ICL_AUX_ANAOVRD1(pw_idx) _MMIO(_PICK(_ICL_AUX_REG_IDX(pw_idx), \ _ICL_AUX_ANAOVRD1_A, \ - _ICL_AUX_ANAOVRD1_B, \ - _TGL_AUX_ANAOVRD1_C)) + _ICL_AUX_ANAOVRD1_B)) #define ICL_AUX_ANAOVRD1_LDO_BYPASS (1 << 7) #define ICL_AUX_ANAOVRD1_ENABLE (1 << 0) @@ -11994,7 +11992,7 @@ enum skl_power_gate { /* This register controls the Display State Buffer (DSB) engines. */ #define _DSBSL_INSTANCE_BASE 0x70B00 #define DSBSL_INSTANCE(pipe, id) (_DSBSL_INSTANCE_BASE + \ - (pipe) * 0x1000 + (id) * 100) + (pipe) * 0x1000 + (id) * 0x100) #define DSB_HEAD(pipe, id) _MMIO(DSBSL_INSTANCE(pipe, id) + 0x0) #define DSB_TAIL(pipe, id) _MMIO(DSBSL_INSTANCE(pipe, id) + 0x4) #define DSB_CTRL(pipe, id) _MMIO(DSBSL_INSTANCE(pipe, id) + 0x8) diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c index bbd71af00a91..765bec89fc0d 100644 --- a/drivers/gpu/drm/i915/i915_request.c +++ b/drivers/gpu/drm/i915/i915_request.c @@ -300,11 +300,11 @@ void i915_request_retire_upto(struct i915_request *rq) } static int -__i915_request_await_execution(struct i915_request *rq, - struct i915_request *signal, - void (*hook)(struct i915_request *rq, - struct dma_fence *signal), - gfp_t gfp) +__await_execution(struct i915_request *rq, + struct i915_request *signal, + void (*hook)(struct i915_request *rq, + struct dma_fence *signal), + gfp_t gfp) { struct execute_cb *cb; @@ -341,6 +341,8 @@ __i915_request_await_execution(struct i915_request *rq, } spin_unlock_irq(&signal->lock); + /* Copy across semaphore status as we need the same behaviour */ + rq->sched.flags |= signal->sched.flags; return 0; } @@ -811,31 +813,21 @@ already_busywaiting(struct i915_request *rq) } static int -emit_semaphore_wait(struct i915_request *to, - struct i915_request *from, - gfp_t gfp) +__emit_semaphore_wait(struct i915_request *to, + struct i915_request *from, + u32 seqno) { const int has_token = INTEL_GEN(to->i915) >= 12; u32 hwsp_offset; - int len; + int len, err; u32 *cs; GEM_BUG_ON(INTEL_GEN(to->i915) < 8); - /* Just emit the first semaphore we see as request space is limited. */ - if (already_busywaiting(to) & from->engine->mask) - goto await_fence; - - if (i915_request_await_start(to, from) < 0) - goto await_fence; - - /* Only submit our spinner after the signaler is running! */ - if (__i915_request_await_execution(to, from, NULL, gfp)) - goto await_fence; - /* We need to pin the signaler's HWSP until we are finished reading. */ - if (intel_timeline_read_hwsp(from, to, &hwsp_offset)) - goto await_fence; + err = intel_timeline_read_hwsp(from, to, &hwsp_offset); + if (err) + return err; len = 4; if (has_token) @@ -858,7 +850,7 @@ emit_semaphore_wait(struct i915_request *to, MI_SEMAPHORE_POLL | MI_SEMAPHORE_SAD_GTE_SDD) + has_token; - *cs++ = from->fence.seqno; + *cs++ = seqno; *cs++ = hwsp_offset; *cs++ = 0; if (has_token) { @@ -867,6 +859,28 @@ emit_semaphore_wait(struct i915_request *to, } intel_ring_advance(to, cs); + return 0; +} + +static int +emit_semaphore_wait(struct i915_request *to, + struct i915_request *from, + gfp_t gfp) +{ + /* Just emit the first semaphore we see as request space is limited. */ + if (already_busywaiting(to) & from->engine->mask) + goto await_fence; + + if (i915_request_await_start(to, from) < 0) + goto await_fence; + + /* Only submit our spinner after the signaler is running! */ + if (__await_execution(to, from, NULL, gfp)) + goto await_fence; + + if (__emit_semaphore_wait(to, from, from->fence.seqno)) + goto await_fence; + to->sched.semaphores |= from->engine->mask; to->sched.flags |= I915_SCHED_HAS_SEMAPHORE_CHAIN; return 0; @@ -980,6 +994,57 @@ i915_request_await_dma_fence(struct i915_request *rq, struct dma_fence *fence) return 0; } +static bool intel_timeline_sync_has_start(struct intel_timeline *tl, + struct dma_fence *fence) +{ + return __intel_timeline_sync_is_later(tl, + fence->context, + fence->seqno - 1); +} + +static int intel_timeline_sync_set_start(struct intel_timeline *tl, + const struct dma_fence *fence) +{ + return __intel_timeline_sync_set(tl, fence->context, fence->seqno - 1); +} + +static int +__i915_request_await_execution(struct i915_request *to, + struct i915_request *from, + void (*hook)(struct i915_request *rq, + struct dma_fence *signal)) +{ + int err; + + /* Submit both requests at the same time */ + err = __await_execution(to, from, hook, I915_FENCE_GFP); + if (err) + return err; + + /* Squash repeated depenendices to the same timelines */ + if (intel_timeline_sync_has_start(i915_request_timeline(to), + &from->fence)) + return 0; + + /* Ensure both start together [after all semaphores in signal] */ + if (intel_engine_has_semaphores(to->engine)) + err = __emit_semaphore_wait(to, from, from->fence.seqno - 1); + else + err = i915_request_await_start(to, from); + if (err < 0) + return err; + + /* Couple the dependency tree for PI on this exposed to->fence */ + if (to->engine->schedule) { + err = i915_sched_node_add_dependency(&to->sched, &from->sched); + if (err < 0) + return err; + } + + return intel_timeline_sync_set_start(i915_request_timeline(to), + &from->fence); +} + int i915_request_await_execution(struct i915_request *rq, struct dma_fence *fence, @@ -1013,8 +1078,7 @@ i915_request_await_execution(struct i915_request *rq, if (dma_fence_is_i915(fence)) ret = __i915_request_await_execution(rq, to_request(fence), - hook, - I915_FENCE_GFP); + hook); else ret = i915_sw_fence_await_dma_fence(&rq->submit, fence, I915_FENCE_TIMEOUT, diff --git a/drivers/gpu/drm/i915/i915_scheduler.c b/drivers/gpu/drm/i915/i915_scheduler.c index 010d67f48ad9..247a9671bca5 100644 --- a/drivers/gpu/drm/i915/i915_scheduler.c +++ b/drivers/gpu/drm/i915/i915_scheduler.c @@ -474,7 +474,6 @@ void i915_sched_node_fini(struct i915_sched_node *node) * so we may be called out-of-order. */ list_for_each_entry_safe(dep, tmp, &node->signalers_list, signal_link) { - GEM_BUG_ON(!node_signaled(dep->signaler)); GEM_BUG_ON(!list_empty(&dep->dfs_link)); list_del(&dep->wait_link); diff --git a/drivers/gpu/drm/i915/i915_sw_fence_work.c b/drivers/gpu/drm/i915/i915_sw_fence_work.c index 07552cd544f2..8538ee7a521d 100644 --- a/drivers/gpu/drm/i915/i915_sw_fence_work.c +++ b/drivers/gpu/drm/i915/i915_sw_fence_work.c @@ -78,12 +78,11 @@ static const struct dma_fence_ops fence_ops = { void dma_fence_work_init(struct dma_fence_work *f, const struct dma_fence_work_ops *ops) { + f->ops = ops; spin_lock_init(&f->lock); dma_fence_init(&f->dma, &fence_ops, &f->lock, 0, 0); i915_sw_fence_init(&f->chain, fence_notify); INIT_WORK(&f->work, fence_work); - - f->ops = ops; } int dma_fence_work_chain(struct dma_fence_work *f, struct dma_fence *signal) diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index e5512f26e20a..01c822256b39 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -1104,8 +1104,14 @@ int i915_vma_move_to_active(struct i915_vma *vma, return err; if (flags & EXEC_OBJECT_WRITE) { - if (intel_frontbuffer_invalidate(obj->frontbuffer, ORIGIN_CS)) - i915_active_add_request(&obj->frontbuffer->write, rq); + struct intel_frontbuffer *front; + + front = __intel_frontbuffer_get(obj); + if (unlikely(front)) { + if (intel_frontbuffer_invalidate(front, ORIGIN_CS)) + i915_active_add_request(&front->write, rq); + intel_frontbuffer_put(front); + } dma_resv_add_excl_fence(vma->resv, &rq->fence); obj->write_domain = I915_GEM_DOMAIN_RENDER; diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 809bff955b5a..75ae6f495161 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -4291,8 +4291,8 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *crtc_state, &crtc_state->wm.skl.optimal.planes[plane_id]; if (plane_id == PLANE_CURSOR) { - if (WARN_ON(wm->wm[level].min_ddb_alloc > - total[PLANE_CURSOR])) { + if (wm->wm[level].min_ddb_alloc > total[PLANE_CURSOR]) { + WARN_ON(wm->wm[level].min_ddb_alloc != U16_MAX); blocks = U32_MAX; break; } |