diff options
Diffstat (limited to 'drivers/gpu/drm/i915')
38 files changed, 516 insertions, 357 deletions
diff --git a/drivers/gpu/drm/i915/gvt/aperture_gm.c b/drivers/gpu/drm/i915/gvt/aperture_gm.c index 7311aeab16f7..3b6caaca9751 100644 --- a/drivers/gpu/drm/i915/gvt/aperture_gm.c +++ b/drivers/gpu/drm/i915/gvt/aperture_gm.c @@ -49,20 +49,21 @@ static int alloc_gm(struct intel_vgpu *vgpu, bool high_gm) if (high_gm) { node = &vgpu->gm.high_gm_node; size = vgpu_hidden_sz(vgpu); - start = gvt_hidden_gmadr_base(gvt); - end = gvt_hidden_gmadr_end(gvt); + start = ALIGN(gvt_hidden_gmadr_base(gvt), I915_GTT_PAGE_SIZE); + end = ALIGN(gvt_hidden_gmadr_end(gvt), I915_GTT_PAGE_SIZE); flags = PIN_HIGH; } else { node = &vgpu->gm.low_gm_node; size = vgpu_aperture_sz(vgpu); - start = gvt_aperture_gmadr_base(gvt); - end = gvt_aperture_gmadr_end(gvt); + start = ALIGN(gvt_aperture_gmadr_base(gvt), I915_GTT_PAGE_SIZE); + end = ALIGN(gvt_aperture_gmadr_end(gvt), I915_GTT_PAGE_SIZE); flags = PIN_MAPPABLE; } mutex_lock(&dev_priv->drm.struct_mutex); ret = i915_gem_gtt_insert(&dev_priv->ggtt.base, node, - size, 4096, I915_COLOR_UNEVICTABLE, + size, I915_GTT_PAGE_SIZE, + I915_COLOR_UNEVICTABLE, start, end, flags); mutex_unlock(&dev_priv->drm.struct_mutex); if (ret) @@ -254,7 +255,7 @@ static int alloc_resource(struct intel_vgpu *vgpu, if (request > avail) goto no_enough_resource; - vgpu_aperture_sz(vgpu) = request; + vgpu_aperture_sz(vgpu) = ALIGN(request, I915_GTT_PAGE_SIZE); item = "high GM space"; max = gvt_hidden_sz(gvt) - HOST_HIGH_GM_SIZE; @@ -265,7 +266,7 @@ static int alloc_resource(struct intel_vgpu *vgpu, if (request > avail) goto no_enough_resource; - vgpu_hidden_sz(vgpu) = request; + vgpu_hidden_sz(vgpu) = ALIGN(request, I915_GTT_PAGE_SIZE); item = "fence"; max = gvt_fence_sz(gvt) - HOST_FENCE; diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c index 9a4b23c3ee97..b9c8e2407682 100644 --- a/drivers/gpu/drm/i915/gvt/cmd_parser.c +++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c @@ -481,7 +481,6 @@ struct parser_exec_state { (s->vgpu->gvt->device_info.gmadr_bytes_in_cmd >> 2) static unsigned long bypass_scan_mask = 0; -static bool bypass_batch_buffer_scan = true; /* ring ALL, type = 0 */ static struct sub_op_bits sub_op_mi[] = { @@ -1135,6 +1134,8 @@ static int skl_decode_mi_display_flip(struct parser_exec_state *s, u32 dword2 = cmd_val(s, 2); u32 plane = (dword0 & GENMASK(12, 8)) >> 8; + info->plane = PRIMARY_PLANE; + switch (plane) { case MI_DISPLAY_FLIP_SKL_PLANE_1_A: info->pipe = PIPE_A; @@ -1148,12 +1149,28 @@ static int skl_decode_mi_display_flip(struct parser_exec_state *s, info->pipe = PIPE_C; info->event = PRIMARY_C_FLIP_DONE; break; + + case MI_DISPLAY_FLIP_SKL_PLANE_2_A: + info->pipe = PIPE_A; + info->event = SPRITE_A_FLIP_DONE; + info->plane = SPRITE_PLANE; + break; + case MI_DISPLAY_FLIP_SKL_PLANE_2_B: + info->pipe = PIPE_B; + info->event = SPRITE_B_FLIP_DONE; + info->plane = SPRITE_PLANE; + break; + case MI_DISPLAY_FLIP_SKL_PLANE_2_C: + info->pipe = PIPE_C; + info->event = SPRITE_C_FLIP_DONE; + info->plane = SPRITE_PLANE; + break; + default: gvt_err("unknown plane code %d\n", plane); return -EINVAL; } - info->pipe = PRIMARY_PLANE; info->stride_val = (dword1 & GENMASK(15, 6)) >> 6; info->tile_val = (dword1 & GENMASK(2, 0)); info->surf_val = (dword2 & GENMASK(31, 12)) >> 12; @@ -1525,9 +1542,6 @@ static int batch_buffer_needs_scan(struct parser_exec_state *s) { struct intel_gvt *gvt = s->vgpu->gvt; - if (bypass_batch_buffer_scan) - return 0; - if (IS_BROADWELL(gvt->dev_priv) || IS_SKYLAKE(gvt->dev_priv)) { /* BDW decides privilege based on address space */ if (cmd_val(s, 0) & (1 << 8)) diff --git a/drivers/gpu/drm/i915/gvt/display.c b/drivers/gpu/drm/i915/gvt/display.c index c0c884aeb30e..6d8fde880c39 100644 --- a/drivers/gpu/drm/i915/gvt/display.c +++ b/drivers/gpu/drm/i915/gvt/display.c @@ -83,7 +83,7 @@ static int pipe_is_enabled(struct intel_vgpu *vgpu, int pipe) return 0; } -/* EDID with 1024x768 as its resolution */ +/* EDID with 1920x1200 as its resolution */ static unsigned char virtual_dp_monitor_edid[] = { /*Header*/ 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x00, @@ -97,11 +97,16 @@ static unsigned char virtual_dp_monitor_edid[] = { 0xfc, 0x81, 0xa4, 0x55, 0x4d, 0x9d, 0x25, 0x12, 0x50, 0x54, /* Established Timings: maximum resolution is 1024x768 */ 0x21, 0x08, 0x00, - /* Standard Timings. All invalid */ - 0x00, 0xc0, 0x00, 0xc0, 0x00, 0x40, 0x00, 0x80, 0x00, 0x00, - 0x00, 0x40, 0x00, 0x00, 0x00, 0x01, - /* 18 Byte Data Blocks 1: invalid */ - 0x00, 0x00, 0x80, 0xa0, 0x70, 0xb0, + /* + * Standard Timings. + * below new resolutions can be supported: + * 1920x1080, 1280x720, 1280x960, 1280x1024, + * 1440x900, 1600x1200, 1680x1050 + */ + 0xd1, 0xc0, 0x81, 0xc0, 0x81, 0x40, 0x81, 0x80, 0x95, 0x00, + 0xa9, 0x40, 0xb3, 0x00, 0x01, 0x01, + /* 18 Byte Data Blocks 1: max resolution is 1920x1200 */ + 0x28, 0x3c, 0x80, 0xa0, 0x70, 0xb0, 0x23, 0x40, 0x30, 0x20, 0x36, 0x00, 0x06, 0x44, 0x21, 0x00, 0x00, 0x1a, /* 18 Byte Data Blocks 2: invalid */ 0x00, 0x00, 0x00, 0xfd, 0x00, 0x18, 0x3c, 0x18, 0x50, 0x11, 0x00, 0x0a, @@ -115,7 +120,7 @@ static unsigned char virtual_dp_monitor_edid[] = { /* Extension Block Count */ 0x00, /* Checksum */ - 0xef, + 0x45, }; #define DPCD_HEADER_SIZE 0xb @@ -328,3 +333,15 @@ int intel_vgpu_init_display(struct intel_vgpu *vgpu) else return setup_virtual_dp_monitor(vgpu, PORT_B, GVT_DP_B); } + +/** + * intel_vgpu_reset_display- reset vGPU virtual display emulation + * @vgpu: a vGPU + * + * This function is used to reset vGPU virtual display emulation stuffs + * + */ +void intel_vgpu_reset_display(struct intel_vgpu *vgpu) +{ + emulate_monitor_status_change(vgpu); +} diff --git a/drivers/gpu/drm/i915/gvt/display.h b/drivers/gpu/drm/i915/gvt/display.h index 7a60cb848268..8b234ea961f6 100644 --- a/drivers/gpu/drm/i915/gvt/display.h +++ b/drivers/gpu/drm/i915/gvt/display.h @@ -158,6 +158,7 @@ void intel_gvt_emulate_vblank(struct intel_gvt *gvt); void intel_gvt_check_vblank_emulation(struct intel_gvt *gvt); int intel_vgpu_init_display(struct intel_vgpu *vgpu); +void intel_vgpu_reset_display(struct intel_vgpu *vgpu); void intel_vgpu_clean_display(struct intel_vgpu *vgpu); #endif diff --git a/drivers/gpu/drm/i915/gvt/execlist.c b/drivers/gpu/drm/i915/gvt/execlist.c index f32bb6f6495c..46eb9fd3c03f 100644 --- a/drivers/gpu/drm/i915/gvt/execlist.c +++ b/drivers/gpu/drm/i915/gvt/execlist.c @@ -364,58 +364,30 @@ static void free_workload(struct intel_vgpu_workload *workload) #define get_desc_from_elsp_dwords(ed, i) \ ((struct execlist_ctx_descriptor_format *)&((ed)->data[i * 2])) - -#define BATCH_BUFFER_ADDR_MASK ((1UL << 32) - (1U << 2)) -#define BATCH_BUFFER_ADDR_HIGH_MASK ((1UL << 16) - (1U)) -static int set_gma_to_bb_cmd(struct intel_shadow_bb_entry *entry_obj, - unsigned long add, int gmadr_bytes) -{ - if (WARN_ON(gmadr_bytes != 4 && gmadr_bytes != 8)) - return -1; - - *((u32 *)(entry_obj->bb_start_cmd_va + (1 << 2))) = add & - BATCH_BUFFER_ADDR_MASK; - if (gmadr_bytes == 8) { - *((u32 *)(entry_obj->bb_start_cmd_va + (2 << 2))) = - add & BATCH_BUFFER_ADDR_HIGH_MASK; - } - - return 0; -} - static void prepare_shadow_batch_buffer(struct intel_vgpu_workload *workload) { - int gmadr_bytes = workload->vgpu->gvt->device_info.gmadr_bytes_in_cmd; + const int gmadr_bytes = workload->vgpu->gvt->device_info.gmadr_bytes_in_cmd; + struct intel_shadow_bb_entry *entry_obj; /* pin the gem object to ggtt */ - if (!list_empty(&workload->shadow_bb)) { - struct intel_shadow_bb_entry *entry_obj = - list_first_entry(&workload->shadow_bb, - struct intel_shadow_bb_entry, - list); - struct intel_shadow_bb_entry *temp; + list_for_each_entry(entry_obj, &workload->shadow_bb, list) { + struct i915_vma *vma; - list_for_each_entry_safe(entry_obj, temp, &workload->shadow_bb, - list) { - struct i915_vma *vma; - - vma = i915_gem_object_ggtt_pin(entry_obj->obj, NULL, 0, - 4, 0); - if (IS_ERR(vma)) { - gvt_err("Cannot pin\n"); - return; - } - - /* FIXME: we are not tracking our pinned VMA leaving it - * up to the core to fix up the stray pin_count upon - * free. - */ - - /* update the relocate gma with shadow batch buffer*/ - set_gma_to_bb_cmd(entry_obj, - i915_ggtt_offset(vma), - gmadr_bytes); + vma = i915_gem_object_ggtt_pin(entry_obj->obj, NULL, 0, 4, 0); + if (IS_ERR(vma)) { + gvt_err("Cannot pin\n"); + return; } + + /* FIXME: we are not tracking our pinned VMA leaving it + * up to the core to fix up the stray pin_count upon + * free. + */ + + /* update the relocate gma with shadow batch buffer*/ + entry_obj->bb_start_cmd_va[1] = i915_ggtt_offset(vma); + if (gmadr_bytes == 8) + entry_obj->bb_start_cmd_va[2] = 0; } } @@ -515,7 +487,7 @@ static void release_shadow_batch_buffer(struct intel_vgpu_workload *workload) static void release_shadow_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx) { - if (wa_ctx->indirect_ctx.size == 0) + if (!wa_ctx->indirect_ctx.obj) return; i915_gem_object_unpin_map(wa_ctx->indirect_ctx.obj); @@ -826,7 +798,7 @@ int intel_vgpu_init_execlist(struct intel_vgpu *vgpu) INIT_LIST_HEAD(&vgpu->workload_q_head[i]); } - vgpu->workloads = kmem_cache_create("gvt-g vgpu workload", + vgpu->workloads = kmem_cache_create("gvt-g_vgpu_workload", sizeof(struct intel_vgpu_workload), 0, SLAB_HWCACHE_ALIGN, NULL); diff --git a/drivers/gpu/drm/i915/gvt/firmware.c b/drivers/gpu/drm/i915/gvt/firmware.c index 2fae2a2ca96f..1cb29b2d7dc6 100644 --- a/drivers/gpu/drm/i915/gvt/firmware.c +++ b/drivers/gpu/drm/i915/gvt/firmware.c @@ -48,31 +48,6 @@ struct gvt_firmware_header { unsigned char data[1]; }; -#define RD(offset) (readl(mmio + offset.reg)) -#define WR(v, offset) (writel(v, mmio + offset.reg)) - -static void bdw_forcewake_get(void __iomem *mmio) -{ - WR(_MASKED_BIT_DISABLE(0xffff), FORCEWAKE_MT); - - RD(ECOBUS); - - if (wait_for((RD(FORCEWAKE_ACK_HSW) & FORCEWAKE_KERNEL) == 0, 50)) - gvt_err("fail to wait forcewake idle\n"); - - WR(_MASKED_BIT_ENABLE(FORCEWAKE_KERNEL), FORCEWAKE_MT); - - if (wait_for((RD(FORCEWAKE_ACK_HSW) & FORCEWAKE_KERNEL), 50)) - gvt_err("fail to wait forcewake ack\n"); - - if (wait_for((RD(GEN6_GT_THREAD_STATUS_REG) & - GEN6_GT_THREAD_STATUS_CORE_MASK) == 0, 50)) - gvt_err("fail to wait c0 wake up\n"); -} - -#undef RD -#undef WR - #define dev_to_drm_minor(d) dev_get_drvdata((d)) static ssize_t @@ -91,9 +66,9 @@ static struct bin_attribute firmware_attr = { .mmap = NULL, }; -static int expose_firmware_sysfs(struct intel_gvt *gvt, - void __iomem *mmio) +static int expose_firmware_sysfs(struct intel_gvt *gvt) { + struct drm_i915_private *dev_priv = gvt->dev_priv; struct intel_gvt_device_info *info = &gvt->device_info; struct pci_dev *pdev = gvt->dev_priv->drm.pdev; struct intel_gvt_mmio_info *e; @@ -132,7 +107,7 @@ static int expose_firmware_sysfs(struct intel_gvt *gvt, for (j = 0; j < e->length; j += 4) *(u32 *)(p + e->offset + j) = - readl(mmio + e->offset + j); + I915_READ_NOTRACE(_MMIO(e->offset + j)); } memcpy(gvt->firmware.mmio, p, info->mmio_size); @@ -235,7 +210,6 @@ int intel_gvt_load_firmware(struct intel_gvt *gvt) struct gvt_firmware_header *h; const struct firmware *fw; char *path; - void __iomem *mmio; void *mem; int ret; @@ -260,17 +234,6 @@ int intel_gvt_load_firmware(struct intel_gvt *gvt) firmware->mmio = mem; - mmio = pci_iomap(pdev, info->mmio_bar, info->mmio_size); - if (!mmio) { - kfree(path); - kfree(firmware->cfg_space); - kfree(firmware->mmio); - return -EINVAL; - } - - if (IS_BROADWELL(gvt->dev_priv) || IS_SKYLAKE(gvt->dev_priv)) - bdw_forcewake_get(mmio); - sprintf(path, "%s/vid_0x%04x_did_0x%04x_rid_0x%04x.golden_hw_state", GVT_FIRMWARE_PATH, pdev->vendor, pdev->device, pdev->revision); @@ -300,13 +263,11 @@ int intel_gvt_load_firmware(struct intel_gvt *gvt) release_firmware(fw); firmware->firmware_loaded = true; - pci_iounmap(pdev, mmio); return 0; out_free_fw: release_firmware(fw); expose_firmware: - expose_firmware_sysfs(gvt, mmio); - pci_iounmap(pdev, mmio); + expose_firmware_sysfs(gvt); return 0; } diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c index 47dec4acf7ff..28c92346db0e 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.c +++ b/drivers/gpu/drm/i915/gvt/gtt.c @@ -606,21 +606,33 @@ struct intel_vgpu_guest_page *intel_vgpu_find_guest_page( static inline int init_shadow_page(struct intel_vgpu *vgpu, struct intel_vgpu_shadow_page *p, int type) { + struct device *kdev = &vgpu->gvt->dev_priv->drm.pdev->dev; + dma_addr_t daddr; + + daddr = dma_map_page(kdev, p->page, 0, 4096, PCI_DMA_BIDIRECTIONAL); + if (dma_mapping_error(kdev, daddr)) { + gvt_err("fail to map dma addr\n"); + return -EINVAL; + } + p->vaddr = page_address(p->page); p->type = type; INIT_HLIST_NODE(&p->node); - p->mfn = intel_gvt_hypervisor_virt_to_mfn(p->vaddr); - if (p->mfn == INTEL_GVT_INVALID_ADDR) - return -EFAULT; - + p->mfn = daddr >> GTT_PAGE_SHIFT; hash_add(vgpu->gtt.shadow_page_hash_table, &p->node, p->mfn); return 0; } -static inline void clean_shadow_page(struct intel_vgpu_shadow_page *p) +static inline void clean_shadow_page(struct intel_vgpu *vgpu, + struct intel_vgpu_shadow_page *p) { + struct device *kdev = &vgpu->gvt->dev_priv->drm.pdev->dev; + + dma_unmap_page(kdev, p->mfn << GTT_PAGE_SHIFT, 4096, + PCI_DMA_BIDIRECTIONAL); + if (!hlist_unhashed(&p->node)) hash_del(&p->node); } @@ -670,7 +682,7 @@ static void ppgtt_free_shadow_page(struct intel_vgpu_ppgtt_spt *spt) { trace_spt_free(spt->vgpu->id, spt, spt->shadow_page.type); - clean_shadow_page(&spt->shadow_page); + clean_shadow_page(spt->vgpu, &spt->shadow_page); intel_vgpu_clean_guest_page(spt->vgpu, &spt->guest_page); list_del_init(&spt->post_shadow_list); @@ -1875,8 +1887,9 @@ static int alloc_scratch_pages(struct intel_vgpu *vgpu, int page_entry_num = GTT_PAGE_SIZE >> vgpu->gvt->device_info.gtt_entry_size_shift; void *scratch_pt; - unsigned long mfn; int i; + struct device *dev = &vgpu->gvt->dev_priv->drm.pdev->dev; + dma_addr_t daddr; if (WARN_ON(type < GTT_TYPE_PPGTT_PTE_PT || type >= GTT_TYPE_MAX)) return -EINVAL; @@ -1887,16 +1900,18 @@ static int alloc_scratch_pages(struct intel_vgpu *vgpu, return -ENOMEM; } - mfn = intel_gvt_hypervisor_virt_to_mfn(scratch_pt); - if (mfn == INTEL_GVT_INVALID_ADDR) { - gvt_err("fail to translate vaddr:0x%lx\n", (unsigned long)scratch_pt); - free_page((unsigned long)scratch_pt); - return -EFAULT; + daddr = dma_map_page(dev, virt_to_page(scratch_pt), 0, + 4096, PCI_DMA_BIDIRECTIONAL); + if (dma_mapping_error(dev, daddr)) { + gvt_err("fail to dmamap scratch_pt\n"); + __free_page(virt_to_page(scratch_pt)); + return -ENOMEM; } - gtt->scratch_pt[type].page_mfn = mfn; + gtt->scratch_pt[type].page_mfn = + (unsigned long)(daddr >> GTT_PAGE_SHIFT); gtt->scratch_pt[type].page = virt_to_page(scratch_pt); gvt_dbg_mm("vgpu%d create scratch_pt: type %d mfn=0x%lx\n", - vgpu->id, type, mfn); + vgpu->id, type, gtt->scratch_pt[type].page_mfn); /* Build the tree by full filled the scratch pt with the entries which * point to the next level scratch pt or scratch page. The @@ -1930,9 +1945,14 @@ static int alloc_scratch_pages(struct intel_vgpu *vgpu, static int release_scratch_page_tree(struct intel_vgpu *vgpu) { int i; + struct device *dev = &vgpu->gvt->dev_priv->drm.pdev->dev; + dma_addr_t daddr; for (i = GTT_TYPE_PPGTT_PTE_PT; i < GTT_TYPE_MAX; i++) { if (vgpu->gtt.scratch_pt[i].page != NULL) { + daddr = (dma_addr_t)(vgpu->gtt.scratch_pt[i].page_mfn << + GTT_PAGE_SHIFT); + dma_unmap_page(dev, daddr, 4096, PCI_DMA_BIDIRECTIONAL); __free_page(vgpu->gtt.scratch_pt[i].page); vgpu->gtt.scratch_pt[i].page = NULL; vgpu->gtt.scratch_pt[i].page_mfn = 0; @@ -2192,6 +2212,8 @@ int intel_gvt_init_gtt(struct intel_gvt *gvt) { int ret; void *page; + struct device *dev = &gvt->dev_priv->drm.pdev->dev; + dma_addr_t daddr; gvt_dbg_core("init gtt\n"); @@ -2209,14 +2231,16 @@ int intel_gvt_init_gtt(struct intel_gvt *gvt) gvt_err("fail to allocate scratch ggtt page\n"); return -ENOMEM; } - gvt->gtt.scratch_ggtt_page = virt_to_page(page); - gvt->gtt.scratch_ggtt_mfn = intel_gvt_hypervisor_virt_to_mfn(page); - if (gvt->gtt.scratch_ggtt_mfn == INTEL_GVT_INVALID_ADDR) { - gvt_err("fail to translate scratch ggtt page\n"); - __free_page(gvt->gtt.scratch_ggtt_page); - return -EFAULT; + daddr = dma_map_page(dev, virt_to_page(page), 0, + 4096, PCI_DMA_BIDIRECTIONAL); + if (dma_mapping_error(dev, daddr)) { + gvt_err("fail to dmamap scratch ggtt page\n"); + __free_page(virt_to_page(page)); + return -ENOMEM; } + gvt->gtt.scratch_ggtt_page = virt_to_page(page); + gvt->gtt.scratch_ggtt_mfn = (unsigned long)(daddr >> GTT_PAGE_SHIFT); if (enable_out_of_sync) { ret = setup_spt_oos(gvt); @@ -2239,6 +2263,12 @@ int intel_gvt_init_gtt(struct intel_gvt *gvt) */ void intel_gvt_clean_gtt(struct intel_gvt *gvt) { + struct device *dev = &gvt->dev_priv->drm.pdev->dev; + dma_addr_t daddr = (dma_addr_t)(gvt->gtt.scratch_ggtt_mfn << + GTT_PAGE_SHIFT); + + dma_unmap_page(dev, daddr, 4096, PCI_DMA_BIDIRECTIONAL); + __free_page(gvt->gtt.scratch_ggtt_page); if (enable_out_of_sync) diff --git a/drivers/gpu/drm/i915/gvt/gvt.c b/drivers/gpu/drm/i915/gvt/gvt.c index e6bf5c533fbe..3b9d59e457ba 100644 --- a/drivers/gpu/drm/i915/gvt/gvt.c +++ b/drivers/gpu/drm/i915/gvt/gvt.c @@ -68,8 +68,6 @@ static const struct intel_gvt_ops intel_gvt_ops = { */ int intel_gvt_init_host(void) { - int ret; - if (intel_gvt_host.initialized) return 0; @@ -96,11 +94,6 @@ int intel_gvt_init_host(void) if (!intel_gvt_host.mpt) return -EINVAL; - /* Try to detect if we're running in host instead of VM. */ - ret = intel_gvt_hypervisor_detect_host(); - if (ret) - return -ENODEV; - gvt_dbg_core("Running with hypervisor %s in host mode\n", supported_hypervisors[intel_gvt_host.hypervisor_type]); diff --git a/drivers/gpu/drm/i915/gvt/hypercall.h b/drivers/gpu/drm/i915/gvt/hypercall.h index 30e543f5a703..df7f33abd393 100644 --- a/drivers/gpu/drm/i915/gvt/hypercall.h +++ b/drivers/gpu/drm/i915/gvt/hypercall.h @@ -38,7 +38,6 @@ * both Xen and KVM by providing dedicated hypervisor-related MPT modules. */ struct intel_gvt_mpt { - int (*detect_host)(void); int (*host_init)(struct device *dev, void *gvt, const void *ops); void (*host_exit)(struct device *dev, void *gvt); int (*attach_vgpu)(void *vgpu, unsigned long *handle); diff --git a/drivers/gpu/drm/i915/gvt/interrupt.c b/drivers/gpu/drm/i915/gvt/interrupt.c index f7be02ac4be1..92bb247e3478 100644 --- a/drivers/gpu/drm/i915/gvt/interrupt.c +++ b/drivers/gpu/drm/i915/gvt/interrupt.c @@ -176,26 +176,15 @@ int intel_vgpu_reg_imr_handler(struct intel_vgpu *vgpu, { struct intel_gvt *gvt = vgpu->gvt; struct intel_gvt_irq_ops *ops = gvt->irq.ops; - u32 changed, masked, unmasked; u32 imr = *(u32 *)p_data; - gvt_dbg_irq("write IMR %x with val %x\n", - reg, imr); - - gvt_dbg_irq("old vIMR %x\n", vgpu_vreg(vgpu, reg)); - - /* figure out newly masked/unmasked bits */ - changed = vgpu_vreg(vgpu, reg) ^ imr; - masked = (vgpu_vreg(vgpu, reg) & changed) ^ changed; - unmasked = masked ^ changed; - - gvt_dbg_irq("changed %x, masked %x, unmasked %x\n", - changed, masked, unmasked); + gvt_dbg_irq("write IMR %x, new %08x, old %08x, changed %08x\n", + reg, imr, vgpu_vreg(vgpu, reg), vgpu_vreg(vgpu, reg) ^ imr); vgpu_vreg(vgpu, reg) = imr; ops->check_pending_irq(vgpu); - gvt_dbg_irq("IRQ: new vIMR %x\n", vgpu_vreg(vgpu, reg)); + return 0; } @@ -217,14 +206,11 @@ int intel_vgpu_reg_master_irq_handler(struct intel_vgpu *vgpu, { struct intel_gvt *gvt = vgpu->gvt; struct intel_gvt_irq_ops *ops = gvt->irq.ops; - u32 changed, enabled, disabled; u32 ier = *(u32 *)p_data; u32 virtual_ier = vgpu_vreg(vgpu, reg); - gvt_dbg_irq("write master irq reg %x with val %x\n", - reg, ier); - - gvt_dbg_irq("old vreg %x\n", vgpu_vreg(vgpu, reg)); + gvt_dbg_irq("write MASTER_IRQ %x, new %08x, old %08x, changed %08x\n", + reg, ier, virtual_ier, virtual_ier ^ ier); /* * GEN8_MASTER_IRQ is a special irq register, @@ -236,16 +222,8 @@ int intel_vgpu_reg_master_irq_handler(struct intel_vgpu *vgpu, vgpu_vreg(vgpu, reg) &= ~GEN8_MASTER_IRQ_CONTROL; vgpu_vreg(vgpu, reg) |= ier; - /* figure out newly enabled/disable bits */ - changed = virtual_ier ^ ier; - enabled = (virtual_ier & changed) ^ changed; - disabled = enabled ^ changed; - - gvt_dbg_irq("changed %x, enabled %x, disabled %x\n", - changed, enabled, disabled); - ops->check_pending_irq(vgpu); - gvt_dbg_irq("new vreg %x\n", vgpu_vreg(vgpu, reg)); + return 0; } @@ -268,21 +246,11 @@ int intel_vgpu_reg_ier_handler(struct intel_vgpu *vgpu, struct intel_gvt *gvt = vgpu->gvt; struct intel_gvt_irq_ops *ops = gvt->irq.ops; struct intel_gvt_irq_info *info; - u32 changed, enabled, disabled; u32 ier = *(u32 *)p_data; - gvt_dbg_irq("write IER %x with val %x\n", - reg, ier); - - gvt_dbg_irq("old vIER %x\n", vgpu_vreg(vgpu, reg)); + gvt_dbg_irq("write IER %x, new %08x, old %08x, changed %08x\n", + reg, ier, vgpu_vreg(vgpu, reg), vgpu_vreg(vgpu, reg) ^ ier); - /* figure out newly enabled/disable bits */ - changed = vgpu_vreg(vgpu, reg) ^ ier; - enabled = (vgpu_vreg(vgpu, reg) & changed) ^ changed; - disabled = enabled ^ changed; - - gvt_dbg_irq("changed %x, enabled %x, disabled %x\n", - changed, enabled, disabled); vgpu_vreg(vgpu, reg) = ier; info = regbase_to_irq_info(gvt, ier_to_regbase(reg)); @@ -293,7 +261,7 @@ int intel_vgpu_reg_ier_handler(struct intel_vgpu *vgpu, update_upstream_irq(vgpu, info); ops->check_pending_irq(vgpu); - gvt_dbg_irq("new vIER %x\n", vgpu_vreg(vgpu, reg)); + return 0; } @@ -317,7 +285,8 @@ int intel_vgpu_reg_iir_handler(struct intel_vgpu *vgpu, unsigned int reg, iir_to_regbase(reg)); u32 iir = *(u32 *)p_data; - gvt_dbg_irq("write IIR %x with val %x\n", reg, iir); + gvt_dbg_irq("write IIR %x, new %08x, old %08x, changed %08x\n", + reg, iir, vgpu_vreg(vgpu, reg), vgpu_vreg(vgpu, reg) ^ iir); if (WARN_ON(!info)) return -EINVAL; @@ -619,6 +588,10 @@ static void gen8_init_irq( SET_BIT_INFO(irq, 3, PRIMARY_A_FLIP_DONE, INTEL_GVT_IRQ_INFO_DE_PIPE_A); SET_BIT_INFO(irq, 3, PRIMARY_B_FLIP_DONE, INTEL_GVT_IRQ_INFO_DE_PIPE_B); SET_BIT_INFO(irq, 3, PRIMARY_C_FLIP_DONE, INTEL_GVT_IRQ_INFO_DE_PIPE_C); + + SET_BIT_INFO(irq, 4, SPRITE_A_FLIP_DONE, INTEL_GVT_IRQ_INFO_DE_PIPE_A); + SET_BIT_INFO(irq, 4, SPRITE_B_FLIP_DONE, INTEL_GVT_IRQ_INFO_DE_PIPE_B); + SET_BIT_INFO(irq, 4, SPRITE_C_FLIP_DONE, INTEL_GVT_IRQ_INFO_DE_PIPE_C); } /* GEN8 interrupt PCU events */ diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c index 0c9234a87a20..0f7f5d97f582 100644 --- a/drivers/gpu/drm/i915/gvt/kvmgt.c +++ b/drivers/gpu/drm/i915/gvt/kvmgt.c @@ -77,7 +77,7 @@ struct kvmgt_guest_info { struct gvt_dma { struct rb_node node; gfn_t gfn; - kvm_pfn_t pfn; + unsigned long iova; }; static inline bool handle_valid(unsigned long handle) @@ -89,6 +89,35 @@ static int kvmgt_guest_init(struct mdev_device *mdev); static void intel_vgpu_release_work(struct work_struct *work); static bool kvmgt_guest_exit(struct kvmgt_guest_info *info); +static int gvt_dma_map_iova(struct intel_vgpu *vgpu, kvm_pfn_t pfn, + unsigned long *iova) +{ + struct page *page; + struct device *dev = &vgpu->gvt->dev_priv->drm.pdev->dev; + dma_addr_t daddr; + + page = pfn_to_page(pfn); + if (is_error_page(page)) + return -EFAULT; + + daddr = dma_map_page(dev, page, 0, PAGE_SIZE, + PCI_DMA_BIDIRECTIONAL); + if (dma_mapping_error(dev, daddr)) + return -ENOMEM; + + *iova = (unsigned long)(daddr >> PAGE_SHIFT); + return 0; +} + +static void gvt_dma_unmap_iova(struct intel_vgpu *vgpu, unsigned long iova) +{ + struct device *dev = &vgpu->gvt->dev_priv->drm.pdev->dev; + dma_addr_t daddr; + + daddr = (dma_addr_t)(iova << PAGE_SHIFT); + dma_unmap_page(dev, daddr, PAGE_SIZE, PCI_DMA_BIDIRECTIONAL); +} + static struct gvt_dma *__gvt_cache_find(struct intel_vgpu *vgpu, gfn_t gfn) { struct rb_node *node = vgpu->vdev.cache.rb_node; @@ -111,21 +140,22 @@ out: return ret; } -static kvm_pfn_t gvt_cache_find(struct intel_vgpu *vgpu, gfn_t gfn) +static unsigned long gvt_cache_find(struct intel_vgpu *vgpu, gfn_t gfn) { struct gvt_dma *entry; - kvm_pfn_t pfn; + unsigned long iova; mutex_lock(&vgpu->vdev.cache_lock); entry = __gvt_cache_find(vgpu, gfn); - pfn = (entry == NULL) ? 0 : entry->pfn; + iova = (entry == NULL) ? INTEL_GVT_INVALID_ADDR : entry->iova; mutex_unlock(&vgpu->vdev.cache_lock); - return pfn; + return iova; } -static void gvt_cache_add(struct intel_vgpu *vgpu, gfn_t gfn, kvm_pfn_t pfn) +static void gvt_cache_add(struct intel_vgpu *vgpu, gfn_t gfn, + unsigned long iova) { struct gvt_dma *new, *itr; struct rb_node **link = &vgpu->vdev.cache.rb_node, *parent = NULL; @@ -135,7 +165,7 @@ static void gvt_cache_add(struct intel_vgpu *vgpu, gfn_t gfn, kvm_pfn_t pfn) return; new->gfn = gfn; - new->pfn = pfn; + new->iova = iova; mutex_lock(&vgpu->vdev.cache_lock); while (*link) { @@ -182,6 +212,7 @@ static void gvt_cache_remove(struct intel_vgpu *vgpu, gfn_t gfn) } g1 = gfn; + gvt_dma_unmap_iova(vgpu, this->iova); rc = vfio_unpin_pages(dev, &g1, 1); WARN_ON(rc != 1); __gvt_cache_remove_entry(vgpu, this); @@ -204,6 +235,7 @@ static void gvt_cache_destroy(struct intel_vgpu *vgpu) mutex_lock(&vgpu->vdev.cache_lock); while ((node = rb_first(&vgpu->vdev.cache))) { dma = rb_entry(node, struct gvt_dma, node); + gvt_dma_unmap_iova(vgpu, dma->iova); gfn = dma->gfn; vfio_unpin_pages(dev, &gfn, 1); @@ -230,8 +262,8 @@ static struct intel_vgpu_type *intel_gvt_find_vgpu_type(struct intel_gvt *gvt, return NULL; } -static ssize_t available_instance_show(struct kobject *kobj, struct device *dev, - char *buf) +static ssize_t available_instances_show(struct kobject *kobj, + struct device *dev, char *buf) { struct intel_vgpu_type *type; unsigned int num = 0; @@ -269,12 +301,12 @@ static ssize_t description_show(struct kobject *kobj, struct device *dev, type->fence); } -static MDEV_TYPE_ATTR_RO(available_instance); +static MDEV_TYPE_ATTR_RO(available_instances); static MDEV_TYPE_ATTR_RO(device_api); static MDEV_TYPE_ATTR_RO(description); static struct attribute *type_attrs[] = { - &mdev_type_attr_available_instance.attr, + &mdev_type_attr_available_instances.attr, &mdev_type_attr_device_api.attr, &mdev_type_attr_description.attr, NULL, @@ -965,11 +997,6 @@ static long intel_vgpu_ioctl(struct mdev_device *mdev, unsigned int cmd, sparse->areas[0].offset = PAGE_ALIGN(vgpu_aperture_offset(vgpu)); sparse->areas[0].size = vgpu_aperture_sz(vgpu); - if (!caps.buf) { - kfree(caps.buf); - caps.buf = NULL; - caps.size = 0; - } break; case VFIO_PCI_BAR3_REGION_INDEX ... VFIO_PCI_BAR5_REGION_INDEX: @@ -1248,43 +1275,6 @@ static void kvmgt_page_track_flush_slot(struct kvm *kvm, spin_unlock(&kvm->mmu_lock); } -static bool kvmgt_check_guest(void) -{ - unsigned int eax, ebx, ecx, edx; - char s[12]; - unsigned int *i; - - eax = KVM_CPUID_SIGNATURE; - ebx = ecx = edx = 0; - - asm volatile ("cpuid" - : "+a"(eax), "=b"(ebx), "=c"(ecx), "=d"(edx) - : - : "cc", "memory"); - i = (unsigned int *)s; - i[0] = ebx; - i[1] = ecx; - i[2] = edx; - - return !strncmp(s, "KVMKVMKVM", strlen("KVMKVMKVM")); -} - -/** - * NOTE: - * It's actually impossible to check if we are running in KVM host, - * since the "KVM host" is simply native. So we only dectect guest here. - */ -static int kvmgt_detect_host(void) -{ -#ifdef CONFIG_INTEL_IOMMU - if (intel_iommu_gfx_mapped) { - gvt_err("Hardware IOMMU compatibility not yet supported, try to boot with intel_iommu=igfx_off\n"); - return -ENODEV; - } -#endif - return kvmgt_check_guest() ? -ENODEV : 0; -} - static bool __kvmgt_vgpu_exist(struct intel_vgpu *vgpu, struct kvm *kvm) { struct intel_vgpu *itr; @@ -1390,7 +1380,7 @@ static int kvmgt_inject_msi(unsigned long handle, u32 addr, u16 data) static unsigned long kvmgt_gfn_to_pfn(unsigned long handle, unsigned long gfn) { - unsigned long pfn; + unsigned long iova, pfn; struct kvmgt_guest_info *info; struct device *dev; int rc; @@ -1399,9 +1389,9 @@ static unsigned long kvmgt_gfn_to_pfn(unsigned long handle, unsigned long gfn) return INTEL_GVT_INVALID_ADDR; info = (struct kvmgt_guest_info *)handle; - pfn = gvt_cache_find(info->vgpu, gfn); - if (pfn != 0) - return pfn; + iova = gvt_cache_find(info->vgpu, gfn); + if (iova != INTEL_GVT_INVALID_ADDR) + return iova; pfn = INTEL_GVT_INVALID_ADDR; dev = mdev_dev(info->vgpu->vdev.mdev); @@ -1410,9 +1400,16 @@ static unsigned long kvmgt_gfn_to_pfn(unsigned long handle, unsigned long gfn) gvt_err("vfio_pin_pages failed for gfn 0x%lx: %d\n", gfn, rc); return INTEL_GVT_INVALID_ADDR; } + /* transfer to host iova for GFX to use DMA */ + rc = gvt_dma_map_iova(info->vgpu, pfn, &iova); + if (rc) { + gvt_err("gvt_dma_map_iova failed for gfn: 0x%lx\n", gfn); + vfio_unpin_pages(dev, &gfn, 1); + return INTEL_GVT_INVALID_ADDR; + } - gvt_cache_add(info->vgpu, gfn, pfn); - return pfn; + gvt_cache_add(info->vgpu, gfn, iova); + return iova; } static int kvmgt_rw_gpa(unsigned long handle, unsigned long gpa, @@ -1459,7 +1456,6 @@ static unsigned long kvmgt_virt_to_pfn(void *addr) } struct intel_gvt_mpt kvmgt_mpt = { - .detect_host = kvmgt_detect_host, .host_init = kvmgt_host_init, .host_exit = kvmgt_host_exit, .attach_vgpu = kvmgt_attach_vgpu, diff --git a/drivers/gpu/drm/i915/gvt/mpt.h b/drivers/gpu/drm/i915/gvt/mpt.h index 1af5830c0a56..419353624c5a 100644 --- a/drivers/gpu/drm/i915/gvt/mpt.h +++ b/drivers/gpu/drm/i915/gvt/mpt.h @@ -44,18 +44,6 @@ */ /** - * intel_gvt_hypervisor_detect_host - check if GVT-g is running within - * hypervisor host/privilged domain - * - * Returns: - * Zero on success, -ENODEV if current kernel is running inside a VM - */ -static inline int intel_gvt_hypervisor_detect_host(void) -{ - return intel_gvt_host.mpt->detect_host(); -} - -/** * intel_gvt_hypervisor_host_init - init GVT-g host side * * Returns: diff --git a/drivers/gpu/drm/i915/gvt/render.c b/drivers/gpu/drm/i915/gvt/render.c index 44136b1f3aab..2b3a642284b6 100644 --- a/drivers/gpu/drm/i915/gvt/render.c +++ b/drivers/gpu/drm/i915/gvt/render.c @@ -236,12 +236,18 @@ static void restore_mocs(struct intel_vgpu *vgpu, int ring_id) } } +#define CTX_CONTEXT_CONTROL_VAL 0x03 + void intel_gvt_load_render_mmio(struct intel_vgpu *vgpu, int ring_id) { struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv; struct render_mmio *mmio; u32 v; int i, array_size; + u32 *reg_state = vgpu->shadow_ctx->engine[ring_id].lrc_reg_state; + u32 ctx_ctrl = reg_state[CTX_CONTEXT_CONTROL_VAL]; + u32 inhibit_mask = + _MASKED_BIT_ENABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT); if (IS_SKYLAKE(vgpu->gvt->dev_priv)) { mmio = gen9_render_mmio_list; @@ -257,6 +263,17 @@ void intel_gvt_load_render_mmio(struct intel_vgpu *vgpu, int ring_id) continue; mmio->value = I915_READ(mmio->reg); + + /* + * if it is an inhibit context, load in_context mmio + * into HW by mmio write. If it is not, skip this mmio + * write. + */ + if (mmio->in_context && + ((ctx_ctrl & inhibit_mask) != inhibit_mask) && + i915.enable_execlists) + continue; + if (mmio->mask) v = vgpu_vreg(vgpu, mmio->reg) | (mmio->mask << 16); else diff --git a/drivers/gpu/drm/i915/gvt/sched_policy.c b/drivers/gpu/drm/i915/gvt/sched_policy.c index 678b0be85376..06c9584ac5f0 100644 --- a/drivers/gpu/drm/i915/gvt/sched_policy.c +++ b/drivers/gpu/drm/i915/gvt/sched_policy.c @@ -125,7 +125,6 @@ static void tbs_sched_func(struct work_struct *work) vgpu_data = scheduler->current_vgpu->sched_data; head = &vgpu_data->list; } else { - gvt_dbg_sched("no current vgpu search from q head\n"); head = &sched_data->runq_head; } diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c index 7ea68a75dc46..d6b6d0efdd1a 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.c +++ b/drivers/gpu/drm/i915/gvt/scheduler.c @@ -169,7 +169,8 @@ static int dispatch_workload(struct intel_vgpu_workload *workload) gvt_dbg_sched("ring id %d prepare to dispatch workload %p\n", ring_id, workload); - shadow_ctx->desc_template = workload->ctx_desc.addressing_mode << + shadow_ctx->desc_template &= ~(0x3 << GEN8_CTX_ADDRESSING_MODE_SHIFT); + shadow_ctx->desc_template |= workload->ctx_desc.addressing_mode << GEN8_CTX_ADDRESSING_MODE_SHIFT; mutex_lock(&dev_priv->drm.struct_mutex); @@ -456,7 +457,7 @@ static int workload_thread(void *priv) } complete: - gvt_dbg_sched("will complete workload %p\n, status: %d\n", + gvt_dbg_sched("will complete workload %p, status: %d\n", workload, workload->status); if (workload->req) diff --git a/drivers/gpu/drm/i915/gvt/scheduler.h b/drivers/gpu/drm/i915/gvt/scheduler.h index 3b30c28bff51..2833dfa8c9ae 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.h +++ b/drivers/gpu/drm/i915/gvt/scheduler.h @@ -113,7 +113,7 @@ struct intel_shadow_bb_entry { struct drm_i915_gem_object *obj; void *va; unsigned long len; - void *bb_start_cmd_va; + u32 *bb_start_cmd_va; }; #define workload_q_head(vgpu, ring_id) \ diff --git a/drivers/gpu/drm/i915/gvt/vgpu.c b/drivers/gpu/drm/i915/gvt/vgpu.c index 7295bc8e12fb..95a97aa0051e 100644 --- a/drivers/gpu/drm/i915/gvt/vgpu.c +++ b/drivers/gpu/drm/i915/gvt/vgpu.c @@ -74,7 +74,7 @@ void populate_pvinfo_page(struct intel_vgpu *vgpu) int intel_gvt_init_vgpu_types(struct intel_gvt *gvt) { unsigned int num_types; - unsigned int i, low_avail; + unsigned int i, low_avail, high_avail; unsigned int min_low; /* vGPU type name is defined as GVTg_Vx_y which contains @@ -89,9 +89,9 @@ int intel_gvt_init_vgpu_types(struct intel_gvt *gvt) * to indicate how many vGPU instance can be created for this * type. * - * Currently use static size here as we init type earlier.. */ - low_avail = MB_TO_BYTES(256) - HOST_LOW_GM_SIZE; + low_avail = gvt_aperture_sz(gvt) - HOST_LOW_GM_SIZE; + high_avail = gvt_hidden_sz(gvt) - HOST_HIGH_GM_SIZE; num_types = 4; gvt->types = kzalloc(num_types * sizeof(struct intel_vgpu_type), @@ -106,7 +106,8 @@ int intel_gvt_init_vgpu_types(struct intel_gvt *gvt) gvt->types[i].low_gm_size = min_low; gvt->types[i].high_gm_size = max((min_low<<3), MB_TO_BYTES(384U)); gvt->types[i].fence = 4; - gvt->types[i].max_instance = low_avail / min_low; + gvt->types[i].max_instance = min(low_avail / min_low, + high_avail / gvt->types[i].high_gm_size); gvt->types[i].avail_instance = gvt->types[i].max_instance; if (IS_GEN8(gvt->dev_priv)) @@ -142,9 +143,9 @@ static void intel_gvt_update_vgpu_types(struct intel_gvt *gvt) /* Need to depend on maxium hw resource size but keep on * static config for now. */ - low_gm_avail = MB_TO_BYTES(256) - HOST_LOW_GM_SIZE - + low_gm_avail = gvt_aperture_sz(gvt) - HOST_LOW_GM_SIZE - gvt->gm.vgpu_allocated_low_gm_size; - high_gm_avail = MB_TO_BYTES(256) * 8UL - HOST_HIGH_GM_SIZE - + high_gm_avail = gvt_hidden_sz(gvt) - HOST_HIGH_GM_SIZE - gvt->gm.vgpu_allocated_high_gm_size; fence_avail = gvt_fence_sz(gvt) - HOST_FENCE - gvt->fence.vgpu_allocated_fence_num; @@ -384,6 +385,7 @@ void intel_gvt_reset_vgpu_locked(struct intel_vgpu *vgpu, bool dmlr, intel_vgpu_reset_resource(vgpu); intel_vgpu_reset_mmio(vgpu); populate_pvinfo_page(vgpu); + intel_vgpu_reset_display(vgpu); if (dmlr) intel_vgpu_reset_cfg_space(vgpu); diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 4ae69ebe166e..f6017f2cfb86 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -213,7 +213,8 @@ static void intel_detect_pch(struct drm_i915_private *dev_priv) } else if (id == INTEL_PCH_KBP_DEVICE_ID_TYPE) { dev_priv->pch_type = PCH_KBP; DRM_DEBUG_KMS("Found KabyPoint PCH\n"); - WARN_ON(!IS_KABYLAKE(dev_priv)); + WARN_ON(!IS_SKYLAKE(dev_priv) && + !IS_KABYLAKE(dev_priv)); } else if ((id == INTEL_PCH_P2X_DEVICE_ID_TYPE) || (id == INTEL_PCH_P3X_DEVICE_ID_TYPE) || ((id == INTEL_PCH_QEMU_DEVICE_ID_TYPE) && @@ -824,10 +825,6 @@ static int i915_driver_init_early(struct drm_i915_private *dev_priv, if (ret < 0) return ret; - ret = intel_gvt_init(dev_priv); - if (ret < 0) - goto err_workqueues; - /* This must be called before any calls to HAS_PCH_* */ intel_detect_pch(dev_priv); @@ -841,7 +838,7 @@ static int i915_driver_init_early(struct drm_i915_private *dev_priv, intel_init_audio_hooks(dev_priv); ret = i915_gem_load_init(dev_priv); if (ret < 0) - goto err_gvt; + goto err_workqueues; intel_display_crc_init(dev_priv); @@ -853,8 +850,6 @@ static int i915_driver_init_early(struct drm_i915_private *dev_priv, return 0; -err_gvt: - intel_gvt_cleanup(dev_priv); err_workqueues: i915_workqueues_cleanup(dev_priv); return ret; @@ -1077,6 +1072,10 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv) DRM_DEBUG_DRIVER("can't enable MSI"); } + ret = intel_gvt_init(dev_priv); + if (ret) + goto out_ggtt; + return 0; out_ggtt: @@ -1290,6 +1289,8 @@ void i915_driver_unload(struct drm_device *dev) intel_display_power_get(dev_priv, POWER_DOMAIN_INIT); + intel_gvt_cleanup(dev_priv); + i915_driver_unregister(dev_priv); drm_vblank_cleanup(dev); @@ -2377,7 +2378,7 @@ static int intel_runtime_suspend(struct device *kdev) assert_forcewakes_inactive(dev_priv); - if (!IS_VALLEYVIEW(dev_priv) || !IS_CHERRYVIEW(dev_priv)) + if (!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv)) intel_hpd_poll_init(dev_priv); DRM_DEBUG_KMS("Device suspended\n"); @@ -2426,6 +2427,7 @@ static int intel_runtime_resume(struct device *kdev) * we can do is to hope that things will still work (and disable RPM). */ i915_gem_init_swizzling(dev_priv); + i915_gem_restore_fences(dev_priv); intel_runtime_pm_enable_interrupts(dev_priv); diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 244628065f94..e44c598ecb82 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -2242,6 +2242,11 @@ struct drm_i915_private { struct i915_frontbuffer_tracking fb_tracking; + struct intel_atomic_helper { + struct llist_head free_list; + struct work_struct free_work; + } atomic_helper; + u16 orig_clock; bool mchbar_need_disable; diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index c8689892a89f..88f3628b4e29 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -440,7 +440,7 @@ i915_gem_object_wait_reservation(struct reservation_object *resv, timeout = i915_gem_object_wait_fence(shared[i], flags, timeout, rps); - if (timeout <= 0) + if (timeout < 0) break; dma_fence_put(shared[i]); @@ -453,7 +453,7 @@ i915_gem_object_wait_reservation(struct reservation_object *resv, excl = reservation_object_get_excl_rcu(resv); } - if (excl && timeout > 0) + if (excl && timeout >= 0) timeout = i915_gem_object_wait_fence(excl, flags, timeout, rps); dma_fence_put(excl); @@ -2009,8 +2009,16 @@ void i915_gem_runtime_suspend(struct drm_i915_private *dev_priv) for (i = 0; i < dev_priv->num_fence_regs; i++) { struct drm_i915_fence_reg *reg = &dev_priv->fence_regs[i]; - if (WARN_ON(reg->pin_count)) - continue; + /* Ideally we want to assert that the fence register is not + * live at this point (i.e. that no piece of code will be + * trying to write through fence + GTT, as that both violates + * our tracking of activity and associated locking/barriers, + * but also is illegal given that the hw is powered down). + * + * Previously we used reg->pin_count as a "liveness" indicator. + * That is not sufficient, and we need a more fine-grained + * tool if we want to have a sanity check here. + */ if (!reg->vma) continue; @@ -2735,21 +2743,17 @@ static void i915_gem_reset_engine(struct intel_engine_cs *engine) engine->irq_seqno_barrier(engine); request = i915_gem_find_active_request(engine); - if (!request) - return; + if (request && i915_gem_reset_request(request)) { + DRM_DEBUG_DRIVER("resetting %s to restart from tail of request 0x%x\n", + engine->name, request->global_seqno); - if (!i915_gem_reset_request(request)) - return; - - DRM_DEBUG_DRIVER("resetting %s to restart from tail of request 0x%x\n", - engine->name, request->global_seqno); + /* If this context is now banned, skip all pending requests. */ + if (i915_gem_context_is_banned(request->ctx)) + engine_skip_context(request); + } /* Setup the CS to resume from the breadcrumb of the hung request */ engine->reset_hw(engine, request); - - /* If this context is now banned, skip all of its pending requests. */ - if (i915_gem_context_is_banned(request->ctx)) - engine_skip_context(request); } void i915_gem_reset_finish(struct drm_i915_private *dev_priv) @@ -3517,7 +3521,7 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj, vma->display_alignment = max_t(u64, vma->display_alignment, alignment); /* Treat this as an end-of-frame, like intel_user_framebuffer_dirty() */ - if (obj->cache_dirty) { + if (obj->cache_dirty || obj->base.write_domain == I915_GEM_DOMAIN_CPU) { i915_gem_clflush_object(obj, true); intel_fb_obj_flush(obj, false, ORIGIN_DIRTYFB); } diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index 57bec08e80c5..d02cfaefe1c8 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -1180,14 +1180,14 @@ validate_exec_list(struct drm_device *dev, if (exec[i].offset != gen8_canonical_addr(exec[i].offset & PAGE_MASK)) return -EINVAL; - - /* From drm_mm perspective address space is continuous, - * so from this point we're always using non-canonical - * form internally. - */ - exec[i].offset = gen8_noncanonical_addr(exec[i].offset); } + /* From drm_mm perspective address space is continuous, + * so from this point we're always using non-canonical + * form internally. + */ + exec[i].offset = gen8_noncanonical_addr(exec[i].offset); + if (exec[i].alignment && !is_power_of_2(exec[i].alignment)) return -EINVAL; diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 30d8dbd04f0b..2801a4d56324 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -755,9 +755,10 @@ static bool gen8_ppgtt_clear_pt(struct i915_address_space *vm, GEM_BUG_ON(pte_end > GEN8_PTES); bitmap_clear(pt->used_ptes, pte, num_entries); - - if (bitmap_empty(pt->used_ptes, GEN8_PTES)) - return true; + if (USES_FULL_PPGTT(vm->i915)) { + if (bitmap_empty(pt->used_ptes, GEN8_PTES)) + return true; + } pt_vaddr = kmap_px(pt); diff --git a/drivers/gpu/drm/i915/i915_gem_internal.c b/drivers/gpu/drm/i915/i915_gem_internal.c index 17ce53d0d092..933019e1b206 100644 --- a/drivers/gpu/drm/i915/i915_gem_internal.c +++ b/drivers/gpu/drm/i915/i915_gem_internal.c @@ -46,16 +46,39 @@ static struct sg_table * i915_gem_object_get_pages_internal(struct drm_i915_gem_object *obj) { struct drm_i915_private *i915 = to_i915(obj->base.dev); - unsigned int npages = obj->base.size / PAGE_SIZE; struct sg_table *st; struct scatterlist *sg; + unsigned int npages; int max_order; gfp_t gfp; + max_order = MAX_ORDER; +#ifdef CONFIG_SWIOTLB + if (swiotlb_nr_tbl()) { + unsigned int max_segment; + + max_segment = swiotlb_max_segment(); + if (max_segment) { + max_segment = max_t(unsigned int, max_segment, + PAGE_SIZE) >> PAGE_SHIFT; + max_order = min(max_order, ilog2(max_segment)); + } + } +#endif + + gfp = GFP_KERNEL | __GFP_HIGHMEM | __GFP_RECLAIMABLE; + if (IS_I965GM(i915) || IS_I965G(i915)) { + /* 965gm cannot relocate objects above 4GiB. */ + gfp &= ~__GFP_HIGHMEM; + gfp |= __GFP_DMA32; + } + +create_st: st = kmalloc(sizeof(*st), GFP_KERNEL); if (!st) return ERR_PTR(-ENOMEM); + npages = obj->base.size / PAGE_SIZE; if (sg_alloc_table(st, npages, GFP_KERNEL)) { kfree(st); return ERR_PTR(-ENOMEM); @@ -64,19 +87,6 @@ i915_gem_object_get_pages_internal(struct drm_i915_gem_object *obj) sg = st->sgl; st->nents = 0; - max_order = MAX_ORDER; -#ifdef CONFIG_SWIOTLB - if (swiotlb_nr_tbl()) /* minimum max swiotlb size is IO_TLB_SEGSIZE */ - max_order = min(max_order, ilog2(IO_TLB_SEGPAGES)); -#endif - - gfp = GFP_KERNEL | __GFP_HIGHMEM | __GFP_RECLAIMABLE; - if (IS_I965GM(i915) || IS_I965G(i915)) { - /* 965gm cannot relocate objects above 4GiB. */ - gfp &= ~__GFP_HIGHMEM; - gfp |= __GFP_DMA32; - } - do { int order = min(fls(npages) - 1, max_order); struct page *page; @@ -104,8 +114,15 @@ i915_gem_object_get_pages_internal(struct drm_i915_gem_object *obj) sg = __sg_next(sg); } while (1); - if (i915_gem_gtt_prepare_pages(obj, st)) + if (i915_gem_gtt_prepare_pages(obj, st)) { + /* Failed to dma-map try again with single page sg segments */ + if (get_order(st->sgl->length)) { + internal_free_pages(st); + max_order = 0; + goto create_st; + } goto err; + } /* Mark the pages as dontneed whilst they are still pinned. As soon * as they are unpinned they are allowed to be reaped by the shrinker, diff --git a/drivers/gpu/drm/i915/i915_gem_request.c b/drivers/gpu/drm/i915/i915_gem_request.c index 72b7f7d9461d..f31deeb72703 100644 --- a/drivers/gpu/drm/i915/i915_gem_request.c +++ b/drivers/gpu/drm/i915/i915_gem_request.c @@ -1025,8 +1025,13 @@ __i915_request_wait_for_execute(struct drm_i915_gem_request *request, break; } + if (!timeout) { + timeout = -ETIME; + break; + } + timeout = io_schedule_timeout(timeout); - } while (timeout); + } while (1); finish_wait(&request->execute.wait, &wait); if (flags & I915_WAIT_LOCKED) diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c index ec7c5d80fe4f..9673bcc3b6ad 100644 --- a/drivers/gpu/drm/i915/i915_gem_stolen.c +++ b/drivers/gpu/drm/i915/i915_gem_stolen.c @@ -405,6 +405,11 @@ int i915_gem_init_stolen(struct drm_i915_private *dev_priv) mutex_init(&dev_priv->mm.stolen_lock); + if (intel_vgpu_active(dev_priv)) { + DRM_INFO("iGVT-g active, disabling use of stolen memory\n"); + return 0; + } + #ifdef CONFIG_INTEL_IOMMU if (intel_iommu_gfx_mapped && INTEL_GEN(dev_priv) < 8) { DRM_INFO("DMAR active, disabling use of stolen memory\n"); diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c index b1361cfd4c5c..974ac08df473 100644 --- a/drivers/gpu/drm/i915/i915_gem_tiling.c +++ b/drivers/gpu/drm/i915/i915_gem_tiling.c @@ -173,7 +173,7 @@ i915_tiling_ok(struct drm_i915_gem_object *obj, else tile_width = 512; - if (!IS_ALIGNED(stride, tile_width)) + if (!stride || !IS_ALIGNED(stride, tile_width)) return false; /* 965+ just needs multiples of tile width */ diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index ee313247673b..53bb7de6020d 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -3123,19 +3123,16 @@ static void ibx_hpd_irq_setup(struct drm_i915_private *dev_priv) I915_WRITE(PCH_PORT_HOTPLUG, hotplug); } -static void spt_hpd_irq_setup(struct drm_i915_private *dev_priv) +static void spt_hpd_detection_setup(struct drm_i915_private *dev_priv) { - u32 hotplug_irqs, hotplug, enabled_irqs; - - hotplug_irqs = SDE_HOTPLUG_MASK_SPT; - enabled_irqs = intel_hpd_enabled_irqs(dev_priv, hpd_spt); - - ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs); + u32 hotplug; /* Enable digital hotplug on the PCH */ hotplug = I915_READ(PCH_PORT_HOTPLUG); - hotplug |= PORTD_HOTPLUG_ENABLE | PORTC_HOTPLUG_ENABLE | - PORTB_HOTPLUG_ENABLE | PORTA_HOTPLUG_ENABLE; + hotplug |= PORTA_HOTPLUG_ENABLE | + PORTB_HOTPLUG_ENABLE | + PORTC_HOTPLUG_ENABLE | + PORTD_HOTPLUG_ENABLE; I915_WRITE(PCH_PORT_HOTPLUG, hotplug); hotplug = I915_READ(PCH_PORT_HOTPLUG2); @@ -3143,6 +3140,18 @@ static void spt_hpd_irq_setup(struct drm_i915_private *dev_priv) I915_WRITE(PCH_PORT_HOTPLUG2, hotplug); } +static void spt_hpd_irq_setup(struct drm_i915_private *dev_priv) +{ + u32 hotplug_irqs, enabled_irqs; + + hotplug_irqs = SDE_HOTPLUG_MASK_SPT; + enabled_irqs = intel_hpd_enabled_irqs(dev_priv, hpd_spt); + + ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs); + + spt_hpd_detection_setup(dev_priv); +} + static void ilk_hpd_irq_setup(struct drm_i915_private *dev_priv) { u32 hotplug_irqs, hotplug, enabled_irqs; @@ -3177,18 +3186,15 @@ static void ilk_hpd_irq_setup(struct drm_i915_private *dev_priv) ibx_hpd_irq_setup(dev_priv); } -static void bxt_hpd_irq_setup(struct drm_i915_private *dev_priv) +static void __bxt_hpd_detection_setup(struct drm_i915_private *dev_priv, + u32 enabled_irqs) { - u32 hotplug_irqs, hotplug, enabled_irqs; - - enabled_irqs = intel_hpd_enabled_irqs(dev_priv, hpd_bxt); - hotplug_irqs = BXT_DE_PORT_HOTPLUG_MASK; - - bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs); + u32 hotplug; hotplug = I915_READ(PCH_PORT_HOTPLUG); - hotplug |= PORTC_HOTPLUG_ENABLE | PORTB_HOTPLUG_ENABLE | - PORTA_HOTPLUG_ENABLE; + hotplug |= PORTA_HOTPLUG_ENABLE | + PORTB_HOTPLUG_ENABLE | + PORTC_HOTPLUG_ENABLE; DRM_DEBUG_KMS("Invert bit setting: hp_ctl:%x hp_port:%x\n", hotplug, enabled_irqs); @@ -3198,7 +3204,6 @@ static void bxt_hpd_irq_setup(struct drm_i915_private *dev_priv) * For BXT invert bit has to be set based on AOB design * for HPD detection logic, update it based on VBT fields. */ - if ((enabled_irqs & BXT_DE_PORT_HP_DDIA) && intel_bios_is_port_hpd_inverted(dev_priv, PORT_A)) hotplug |= BXT_DDIA_HPD_INVERT; @@ -3212,6 +3217,23 @@ static void bxt_hpd_irq_setup(struct drm_i915_private *dev_priv) I915_WRITE(PCH_PORT_HOTPLUG, hotplug); } +static void bxt_hpd_detection_setup(struct drm_i915_private *dev_priv) +{ + __bxt_hpd_detection_setup(dev_priv, BXT_DE_PORT_HOTPLUG_MASK); +} + +static void bxt_hpd_irq_setup(struct drm_i915_private *dev_priv) +{ + u32 hotplug_irqs, enabled_irqs; + + enabled_irqs = intel_hpd_enabled_irqs(dev_priv, hpd_bxt); + hotplug_irqs = BXT_DE_PORT_HOTPLUG_MASK; + + bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs); + + __bxt_hpd_detection_setup(dev_priv, enabled_irqs); +} + static void ibx_irq_postinstall(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); @@ -3227,6 +3249,12 @@ static void ibx_irq_postinstall(struct drm_device *dev) gen5_assert_iir_is_zero(dev_priv, SDEIIR); I915_WRITE(SDEIMR, ~mask); + + if (HAS_PCH_IBX(dev_priv) || HAS_PCH_CPT(dev_priv) || + HAS_PCH_LPT(dev_priv)) + ; /* TODO: Enable HPD detection on older PCH platforms too */ + else + spt_hpd_detection_setup(dev_priv); } static void gen5_gt_irq_postinstall(struct drm_device *dev) @@ -3438,6 +3466,9 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv) GEN5_IRQ_INIT(GEN8_DE_PORT_, ~de_port_masked, de_port_enables); GEN5_IRQ_INIT(GEN8_DE_MISC_, ~de_misc_masked, de_misc_masked); + + if (IS_GEN9_LP(dev_priv)) + bxt_hpd_detection_setup(dev_priv); } static int gen8_irq_postinstall(struct drm_device *dev) diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 72f9f36ae5ce..675323189f2c 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -3307,8 +3307,10 @@ enum skl_disp_power_wells { /* * Logical Context regs */ -#define CCID _MMIO(0x2180) -#define CCID_EN (1<<0) +#define CCID _MMIO(0x2180) +#define CCID_EN BIT(0) +#define CCID_EXTENDED_STATE_RESTORE BIT(2) +#define CCID_EXTENDED_STATE_SAVE BIT(3) /* * Notes on SNB/IVB/VLV context size: * - Power context is saved elsewhere (LLC or stolen) diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c index 385e29af8baa..2bf5aca6e37c 100644 --- a/drivers/gpu/drm/i915/intel_crt.c +++ b/drivers/gpu/drm/i915/intel_crt.c @@ -499,6 +499,7 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector) struct drm_i915_private *dev_priv = to_i915(crt->base.base.dev); struct edid *edid; struct i2c_adapter *i2c; + bool ret = false; BUG_ON(crt->base.type != INTEL_OUTPUT_ANALOG); @@ -515,17 +516,17 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector) */ if (!is_digital) { DRM_DEBUG_KMS("CRT detected via DDC:0x50 [EDID]\n"); - return true; + ret = true; + } else { + DRM_DEBUG_KMS("CRT not detected via DDC:0x50 [EDID reports a digital panel]\n"); } - - DRM_DEBUG_KMS("CRT not detected via DDC:0x50 [EDID reports a digital panel]\n"); } else { DRM_DEBUG_KMS("CRT not detected via DDC:0x50 [no valid EDID found]\n"); } kfree(edid); - return false; + return ret; } static enum drm_connector_status diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 36ecc864e711..a2fece5e9fb3 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2578,8 +2578,9 @@ intel_fill_fb_info(struct drm_i915_private *dev_priv, * We only keep the x/y offsets, so push all of the * gtt offset into the x/y offsets. */ - _intel_adjust_tile_offset(&x, &y, tile_size, - tile_width, tile_height, pitch_tiles, + _intel_adjust_tile_offset(&x, &y, + tile_width, tile_height, + tile_size, pitch_tiles, gtt_offset_rotated * tile_size, 0); gtt_offset_rotated += rot_info->plane[i].width * rot_info->plane[i].height; @@ -4253,10 +4254,10 @@ static void page_flip_completed(struct intel_crtc *intel_crtc) drm_crtc_vblank_put(&intel_crtc->base); wake_up_all(&dev_priv->pending_flip_queue); - queue_work(dev_priv->wq, &work->unpin_work); - trace_i915_flip_complete(intel_crtc->plane, work->pending_flip_obj); + + queue_work(dev_priv->wq, &work->unpin_work); } static int intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc) @@ -6882,6 +6883,12 @@ static void intel_crtc_disable_noatomic(struct drm_crtc *crtc) } state = drm_atomic_state_alloc(crtc->dev); + if (!state) { + DRM_DEBUG_KMS("failed to disable [CRTC:%d:%s], out of memory", + crtc->base.id, crtc->name); + return; + } + state->acquire_ctx = crtc->dev->mode_config.acquire_ctx; /* Everything's already locked, -EDEADLK can't happen. */ @@ -14563,8 +14570,14 @@ intel_atomic_commit_ready(struct i915_sw_fence *fence, break; case FENCE_FREE: - drm_atomic_state_put(&state->base); - break; + { + struct intel_atomic_helper *helper = + &to_i915(state->base.dev)->atomic_helper; + + if (llist_add(&state->freed, &helper->free_list)) + schedule_work(&helper->free_work); + break; + } } return NOTIFY_DONE; @@ -16587,6 +16600,18 @@ fail: drm_modeset_acquire_fini(&ctx); } +static void intel_atomic_helper_free_state(struct work_struct *work) +{ + struct drm_i915_private *dev_priv = + container_of(work, typeof(*dev_priv), atomic_helper.free_work); + struct intel_atomic_state *state, *next; + struct llist_node *freed; + + freed = llist_del_all(&dev_priv->atomic_helper.free_list); + llist_for_each_entry_safe(state, next, freed, freed) + drm_atomic_state_put(&state->base); +} + int intel_modeset_init(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); @@ -16606,6 +16631,9 @@ int intel_modeset_init(struct drm_device *dev) dev->mode_config.funcs = &intel_mode_funcs; + INIT_WORK(&dev_priv->atomic_helper.free_work, + intel_atomic_helper_free_state); + intel_init_quirks(dev); intel_init_pm(dev_priv); @@ -17263,6 +17291,9 @@ void intel_modeset_cleanup(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); + flush_work(&dev_priv->atomic_helper.free_work); + WARN_ON(!llist_empty(&dev_priv->atomic_helper.free_list)); + intel_disable_gt_powersave(dev_priv); /* diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 3d8ac8aa7214..d1670b8afbf5 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -2887,6 +2887,9 @@ static void vlv_detach_power_sequencer(struct intel_dp *intel_dp) WARN_ON(intel_dp->active_pipe != INVALID_PIPE); + if (WARN_ON(pipe != PIPE_A && pipe != PIPE_B)) + return; + edp_panel_vdd_off_sync(intel_dp); /* @@ -2914,9 +2917,6 @@ static void vlv_steal_power_sequencer(struct drm_device *dev, lockdep_assert_held(&dev_priv->pps_mutex); - if (WARN_ON(pipe != PIPE_A && pipe != PIPE_B)) - return; - for_each_intel_encoder(dev, encoder) { struct intel_dp *intel_dp; enum port port; @@ -4406,8 +4406,8 @@ static bool bxt_digital_port_connected(struct drm_i915_private *dev_priv, * * Return %true if @port is connected, %false otherwise. */ -static bool intel_digital_port_connected(struct drm_i915_private *dev_priv, - struct intel_digital_port *port) +bool intel_digital_port_connected(struct drm_i915_private *dev_priv, + struct intel_digital_port *port) { if (HAS_PCH_IBX(dev_priv)) return ibx_digital_port_connected(dev_priv, port); diff --git a/drivers/gpu/drm/i915/intel_dpll_mgr.c b/drivers/gpu/drm/i915/intel_dpll_mgr.c index c92a2558beb4..e59e43a9f3a6 100644 --- a/drivers/gpu/drm/i915/intel_dpll_mgr.c +++ b/drivers/gpu/drm/i915/intel_dpll_mgr.c @@ -1855,7 +1855,8 @@ bxt_get_dpll(struct intel_crtc *crtc, return NULL; if ((encoder->type == INTEL_OUTPUT_DP || - encoder->type == INTEL_OUTPUT_EDP) && + encoder->type == INTEL_OUTPUT_EDP || + encoder->type == INTEL_OUTPUT_DP_MST) && !bxt_ddi_dp_set_dpll_hw_state(clock, &dpll_hw_state)) return NULL; diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 0cec0013ace0..40fed65a791d 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -371,6 +371,8 @@ struct intel_atomic_state { struct skl_wm_values wm_results; struct i915_sw_fence commit_ready; + + struct llist_node freed; }; struct intel_plane_state { @@ -1485,6 +1487,8 @@ bool __intel_dp_read_desc(struct intel_dp *intel_dp, bool intel_dp_read_desc(struct intel_dp *intel_dp); int intel_dp_link_required(int pixel_clock, int bpp); int intel_dp_max_data_rate(int max_link_clock, int max_lanes); +bool intel_digital_port_connected(struct drm_i915_private *dev_priv, + struct intel_digital_port *port); /* intel_dp_aux_backlight.c */ int intel_dp_aux_init_backlight_funcs(struct intel_connector *intel_connector); diff --git a/drivers/gpu/drm/i915/intel_gvt.c b/drivers/gpu/drm/i915/intel_gvt.c index 290384e86c63..d23c0fcff751 100644 --- a/drivers/gpu/drm/i915/intel_gvt.c +++ b/drivers/gpu/drm/i915/intel_gvt.c @@ -67,6 +67,11 @@ int intel_gvt_init(struct drm_i915_private *dev_priv) return 0; } + if (intel_vgpu_active(dev_priv)) { + DRM_DEBUG_DRIVER("GVT-g is disabled for guest\n"); + goto bail; + } + if (!is_supported_device(dev_priv)) { DRM_DEBUG_DRIVER("Unsupported device. GVT-g is disabled\n"); goto bail; diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 432ee495dec2..ebf8023d21e6 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -360,7 +360,8 @@ execlists_update_context_pdps(struct i915_hw_ppgtt *ppgtt, u32 *reg_state) static u64 execlists_update_context(struct drm_i915_gem_request *rq) { struct intel_context *ce = &rq->ctx->engine[rq->engine->id]; - struct i915_hw_ppgtt *ppgtt = rq->ctx->ppgtt; + struct i915_hw_ppgtt *ppgtt = + rq->ctx->ppgtt ?: rq->i915->mm.aliasing_ppgtt; u32 *reg_state = ce->lrc_reg_state; reg_state[CTX_RING_TAIL+1] = rq->tail; @@ -1389,7 +1390,20 @@ static void reset_common_ring(struct intel_engine_cs *engine, { struct drm_i915_private *dev_priv = engine->i915; struct execlist_port *port = engine->execlist_port; - struct intel_context *ce = &request->ctx->engine[engine->id]; + struct intel_context *ce; + + /* If the request was innocent, we leave the request in the ELSP + * and will try to replay it on restarting. The context image may + * have been corrupted by the reset, in which case we may have + * to service a new GPU hang, but more likely we can continue on + * without impact. + * + * If the request was guilty, we presume the context is corrupt + * and have to at least restore the RING register in the context + * image back to the expected values to skip over the guilty request. + */ + if (!request || request->fence.error != -EIO) + return; /* We want a simple context + ring to execute the breadcrumb update. * We cannot rely on the context being intact across the GPU hang, @@ -1398,6 +1412,7 @@ static void reset_common_ring(struct intel_engine_cs *engine, * future request will be after userspace has had the opportunity * to recreate its own state. */ + ce = &request->ctx->engine[engine->id]; execlists_init_reg_state(ce->lrc_reg_state, request->ctx, engine, ce->ring); diff --git a/drivers/gpu/drm/i915/intel_lspcon.c b/drivers/gpu/drm/i915/intel_lspcon.c index f6d4e6940257..c300647ef604 100644 --- a/drivers/gpu/drm/i915/intel_lspcon.c +++ b/drivers/gpu/drm/i915/intel_lspcon.c @@ -158,6 +158,8 @@ static bool lspcon_probe(struct intel_lspcon *lspcon) static void lspcon_resume_in_pcon_wa(struct intel_lspcon *lspcon) { struct intel_dp *intel_dp = lspcon_to_intel_dp(lspcon); + struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); + struct drm_i915_private *dev_priv = to_i915(dig_port->base.base.dev); unsigned long start = jiffies; if (!lspcon->desc_valid) @@ -173,7 +175,8 @@ static void lspcon_resume_in_pcon_wa(struct intel_lspcon *lspcon) if (!__intel_dp_read_desc(intel_dp, &desc)) return; - if (!memcmp(&intel_dp->desc, &desc, sizeof(desc))) { + if (intel_digital_port_connected(dev_priv, dig_port) && + !memcmp(&intel_dp->desc, &desc, sizeof(desc))) { DRM_DEBUG_KMS("LSPCON recovering in PCON mode after %u ms\n", jiffies_to_msecs(jiffies - start)); return; diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c index f4429f67a4e3..4a862a358c70 100644 --- a/drivers/gpu/drm/i915/intel_opregion.c +++ b/drivers/gpu/drm/i915/intel_opregion.c @@ -982,7 +982,18 @@ int intel_opregion_setup(struct drm_i915_private *dev_priv) opregion->vbt_size = vbt_size; } else { vbt = base + OPREGION_VBT_OFFSET; - vbt_size = OPREGION_ASLE_EXT_OFFSET - OPREGION_VBT_OFFSET; + /* + * The VBT specification says that if the ASLE ext + * mailbox is not used its area is reserved, but + * on some CHT boards the VBT extends into the + * ASLE ext area. Allow this even though it is + * against the spec, so we do not end up rejecting + * the VBT on those boards (and end up not finding the + * LCD panel because of this). + */ + vbt_size = (mboxes & MBOX_ASLE_EXT) ? + OPREGION_ASLE_EXT_OFFSET : OPREGION_SIZE; + vbt_size -= OPREGION_VBT_OFFSET; if (intel_bios_is_valid_vbt(vbt, vbt_size)) { DRM_DEBUG_KMS("Found valid VBT in ACPI OpRegion (Mailbox #4)\n"); opregion->vbt = vbt; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 69035e4f9b3b..91bc4abf5d3e 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -599,10 +599,62 @@ out: static void reset_ring_common(struct intel_engine_cs *engine, struct drm_i915_gem_request *request) { - struct intel_ring *ring = request->ring; + /* Try to restore the logical GPU state to match the continuation + * of the request queue. If we skip the context/PD restore, then + * the next request may try to execute assuming that its context + * is valid and loaded on the GPU and so may try to access invalid + * memory, prompting repeated GPU hangs. + * + * If the request was guilty, we still restore the logical state + * in case the next request requires it (e.g. the aliasing ppgtt), + * but skip over the hung batch. + * + * If the request was innocent, we try to replay the request with + * the restored context. + */ + if (request) { + struct drm_i915_private *dev_priv = request->i915; + struct intel_context *ce = &request->ctx->engine[engine->id]; + struct i915_hw_ppgtt *ppgtt; + + /* FIXME consider gen8 reset */ + + if (ce->state) { + I915_WRITE(CCID, + i915_ggtt_offset(ce->state) | + BIT(8) /* must be set! */ | + CCID_EXTENDED_STATE_SAVE | + CCID_EXTENDED_STATE_RESTORE | + CCID_EN); + } - ring->head = request->postfix; - ring->last_retired_head = -1; + ppgtt = request->ctx->ppgtt ?: engine->i915->mm.aliasing_ppgtt; + if (ppgtt) { + u32 pd_offset = ppgtt->pd.base.ggtt_offset << 10; + + I915_WRITE(RING_PP_DIR_DCLV(engine), PP_DIR_DCLV_2G); + I915_WRITE(RING_PP_DIR_BASE(engine), pd_offset); + + /* Wait for the PD reload to complete */ + if (intel_wait_for_register(dev_priv, + RING_PP_DIR_BASE(engine), + BIT(0), 0, + 10)) + DRM_ERROR("Wait for reload of ppgtt page-directory timed out\n"); + + ppgtt->pd_dirty_rings &= ~intel_engine_flag(engine); + } + + /* If the rq hung, jump to its breadcrumb and skip the batch */ + if (request->fence.error == -EIO) { + struct intel_ring *ring = request->ring; + + ring->head = request->postfix; + ring->last_retired_head = -1; + } + } else { + engine->legacy_active_context = NULL; + } } static int intel_ring_workarounds_emit(struct drm_i915_gem_request *req) |