From patchwork Mon Oct 21 10:48:07 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Abdiel Janulgue X-Patchwork-Id: 11201877 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 81DD413B1 for ; Mon, 21 Oct 2019 10:48:25 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 68E3D2067B for ; Mon, 21 Oct 2019 10:48:25 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 68E3D2067B Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=linux.intel.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=intel-gfx-bounces@lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 3296289E5F; Mon, 21 Oct 2019 10:48:23 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by gabe.freedesktop.org (Postfix) with ESMTPS id 96D4789E47 for ; Mon, 21 Oct 2019 10:48:22 +0000 (UTC) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga002.jf.intel.com ([10.7.209.21]) by orsmga102.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 21 Oct 2019 03:48:21 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.67,323,1566889200"; d="scan'208";a="209306837" Received: from ahedstro-mobl1.ger.corp.intel.com (HELO skylake-nuc.ger.corp.intel.com) ([10.249.254.249]) by orsmga002.jf.intel.com with ESMTP; 21 Oct 2019 03:48:19 -0700 From: Abdiel Janulgue To: intel-gfx@lists.freedesktop.org Date: Mon, 21 Oct 2019 13:48:07 +0300 Message-Id: <20191021104811.25663-1-abdiel.janulgue@linux.intel.com> X-Mailer: git-send-email 2.23.0 MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH 1/5] drm/i915: Allow i915 to manage the vma offset nodes instead of drm core X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Matthew Auld Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" Have i915 replace the core drm_gem_mmap implementation to overcome its limitation in having only a single mmap offset node per gem object. This change allows us to have multiple mmap offsets per object and enables a mmapping instance to use unique fault-handlers per user vma. This allows i915 to store extra data within vma->vm_private_data and assign the pagefault ops for each mmap instance allowing objects to use multiple fault handlers depending on its backing storage. v2: - Fix race condition exposed by gem_mmap_gtt@close-race. Simplify lifetime management of the mmap offset objects be ensuring it is owned by the parent gem object instead of refcounting. - Track mmo used by fencing to Avoid locking when revoking mmaps during GPU reset. - Rebase. v3: - Simplify mmo tracking v4: - use vma->mmo in __i915_gem_object_release_mmap_gtt Signed-off-by: Abdiel Janulgue Cc: Matthew Auld Cc: Joonas Lahtinen Cc: Chris Wilson --- drivers/gpu/drm/i915/gem/i915_gem_domain.c | 2 +- drivers/gpu/drm/i915/gem/i915_gem_mman.c | 229 ++++++++++++++++-- drivers/gpu/drm/i915/gem/i915_gem_object.c | 13 + drivers/gpu/drm/i915/gem/i915_gem_object.h | 9 +- .../gpu/drm/i915/gem/i915_gem_object_types.h | 19 ++ .../drm/i915/gem/selftests/i915_gem_mman.c | 12 +- drivers/gpu/drm/i915/gt/intel_reset.c | 7 +- drivers/gpu/drm/i915/i915_drv.c | 10 +- drivers/gpu/drm/i915/i915_drv.h | 3 +- drivers/gpu/drm/i915/i915_gem.c | 2 +- drivers/gpu/drm/i915/i915_vma.c | 15 +- drivers/gpu/drm/i915/i915_vma.h | 4 + 12 files changed, 275 insertions(+), 50 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c index 9937b4c341f1..40792d2017a7 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c @@ -254,7 +254,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj, } if (obj->userfault_count) - __i915_gem_object_release_mmap(obj); + __i915_gem_object_release_mmap_gtt(obj); /* * As we no longer need a fence for GTT access, diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index fd4122d8c0a9..3491bb06606b 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -219,7 +219,8 @@ vm_fault_t i915_gem_fault(struct vm_fault *vmf) { #define MIN_CHUNK_PAGES (SZ_1M >> PAGE_SHIFT) struct vm_area_struct *area = vmf->vma; - struct drm_i915_gem_object *obj = to_intel_bo(area->vm_private_data); + struct i915_mmap_offset *priv = area->vm_private_data; + struct drm_i915_gem_object *obj = priv->obj; struct drm_device *dev = obj->base.dev; struct drm_i915_private *i915 = to_i915(dev); struct intel_runtime_pm *rpm = &i915->runtime_pm; @@ -312,6 +313,9 @@ vm_fault_t i915_gem_fault(struct vm_fault *vmf) list_add(&obj->userfault_link, &i915->ggtt.userfault_list); mutex_unlock(&i915->ggtt.vm.mutex); + /* Track the mmo associated with the fenced vma */ + vma->mmo = priv; + if (CONFIG_DRM_I915_USERFAULT_AUTOSUSPEND) intel_wakeref_auto(&i915->ggtt.userfault_wakeref, msecs_to_jiffies_timeout(CONFIG_DRM_I915_USERFAULT_AUTOSUSPEND)); @@ -358,7 +362,7 @@ vm_fault_t i915_gem_fault(struct vm_fault *vmf) } } -void __i915_gem_object_release_mmap(struct drm_i915_gem_object *obj) +void __i915_gem_object_release_mmap_gtt(struct drm_i915_gem_object *obj) { struct i915_vma *vma; @@ -366,20 +370,16 @@ void __i915_gem_object_release_mmap(struct drm_i915_gem_object *obj) obj->userfault_count = 0; list_del(&obj->userfault_link); - drm_vma_node_unmap(&obj->base.vma_node, - obj->base.dev->anon_inode->i_mapping); - for_each_ggtt_vma(vma, obj) + for_each_ggtt_vma(vma, obj) { + if (vma->mmo) + drm_vma_node_unmap(&vma->mmo->vma_node, + obj->base.dev->anon_inode->i_mapping); i915_vma_unset_userfault(vma); + } } /** - * i915_gem_object_release_mmap - remove physical page mappings - * @obj: obj in question - * - * Preserve the reservation of the mmapping with the DRM core code, but - * relinquish ownership of the pages back to the system. - * * It is vital that we remove the page mapping if we have mapped a tiled * object through the GTT and then lose the fence register due to * resource pressure. Similarly if the object has been moved out of the @@ -387,7 +387,7 @@ void __i915_gem_object_release_mmap(struct drm_i915_gem_object *obj) * mapping will then trigger a page fault on the next user access, allowing * fixup by i915_gem_fault(). */ -void i915_gem_object_release_mmap(struct drm_i915_gem_object *obj) +static void i915_gem_object_release_mmap_gtt(struct drm_i915_gem_object *obj) { struct drm_i915_private *i915 = to_i915(obj->base.dev); intel_wakeref_t wakeref; @@ -406,7 +406,7 @@ void i915_gem_object_release_mmap(struct drm_i915_gem_object *obj) if (!obj->userfault_count) goto out; - __i915_gem_object_release_mmap(obj); + __i915_gem_object_release_mmap_gtt(obj); /* Ensure that the CPU's PTE are revoked and there are not outstanding * memory transactions from userspace before we return. The TLB @@ -422,15 +422,63 @@ void i915_gem_object_release_mmap(struct drm_i915_gem_object *obj) intel_runtime_pm_put(&i915->runtime_pm, wakeref); } -static int create_mmap_offset(struct drm_i915_gem_object *obj) +static void i915_gem_object_release_mmap_offset(struct drm_i915_gem_object *obj) +{ + struct i915_mmap_offset *mmo; + + mutex_lock(&obj->mmo_lock); + list_for_each_entry(mmo, &obj->mmap_offsets, offset) { + /* vma_node_unmap for GTT mmaps handled already in + * __i915_gem_object_release_mmap_gtt + */ + if (mmo->mmap_type != I915_MMAP_TYPE_GTT) + drm_vma_node_unmap(&mmo->vma_node, + obj->base.dev->anon_inode->i_mapping); + } + mutex_unlock(&obj->mmo_lock); +} + +/** + * i915_gem_object_release_mmap - remove physical page mappings + * @obj: obj in question + * + * Preserve the reservation of the mmapping with the DRM core code, but + * relinquish ownership of the pages back to the system. + */ +void i915_gem_object_release_mmap(struct drm_i915_gem_object *obj) +{ + i915_gem_object_release_mmap_gtt(obj); + i915_gem_object_release_mmap_offset(obj); +} + +static void init_mmap_offset(struct drm_i915_gem_object *obj, + struct i915_mmap_offset *mmo) +{ + mutex_lock(&obj->mmo_lock); + list_add(&mmo->offset, &obj->mmap_offsets); + mutex_unlock(&obj->mmo_lock); + + mmo->obj = obj; + mmo->dev = obj->base.dev; +} + +static int create_mmap_offset(struct drm_i915_gem_object *obj, + struct i915_mmap_offset *mmo) { struct drm_i915_private *i915 = to_i915(obj->base.dev); struct intel_gt *gt = &i915->gt; + struct drm_device *dev = obj->base.dev; int err; - err = drm_gem_create_mmap_offset(&obj->base); - if (likely(!err)) + drm_vma_node_reset(&mmo->vma_node); + if (mmo->file) + drm_vma_node_allow(&mmo->vma_node, mmo->file); + err = drm_vma_offset_add(dev->vma_offset_manager, &mmo->vma_node, + obj->base.size / PAGE_SIZE); + if (likely(!err)) { + init_mmap_offset(obj, mmo); return 0; + } /* Attempt to reap some mmap space from dead objects */ err = intel_gt_retire_requests_timeout(gt, MAX_SCHEDULE_TIMEOUT); @@ -438,16 +486,23 @@ static int create_mmap_offset(struct drm_i915_gem_object *obj) return err; i915_gem_drain_freed_objects(i915); - return drm_gem_create_mmap_offset(&obj->base); + err = drm_vma_offset_add(dev->vma_offset_manager, &mmo->vma_node, + obj->base.size / PAGE_SIZE); + if (err) + return err; + + init_mmap_offset(obj, mmo); + return 0; } -int -i915_gem_mmap_gtt(struct drm_file *file, - struct drm_device *dev, - u32 handle, - u64 *offset) +static int +__assign_gem_object_mmap_data(struct drm_file *file, + u32 handle, + enum i915_mmap_type mmap_type, + u64 *offset) { struct drm_i915_gem_object *obj; + struct i915_mmap_offset *mmo; int ret; obj = i915_gem_object_lookup(file, handle); @@ -459,10 +514,21 @@ i915_gem_mmap_gtt(struct drm_file *file, goto out; } - ret = create_mmap_offset(obj); - if (ret == 0) - *offset = drm_vma_node_offset_addr(&obj->base.vma_node); + mmo = kzalloc(sizeof(*mmo), GFP_KERNEL); + if (!mmo) { + ret = -ENOMEM; + goto out; + } + + mmo->file = file; + ret = create_mmap_offset(obj, mmo); + if (ret) { + kfree(mmo); + goto out; + } + mmo->mmap_type = mmap_type; + *offset = drm_vma_node_offset_addr(&mmo->vma_node); out: i915_gem_object_put(obj); return ret; @@ -489,7 +555,118 @@ i915_gem_mmap_gtt_ioctl(struct drm_device *dev, void *data, { struct drm_i915_gem_mmap_gtt *args = data; - return i915_gem_mmap_gtt(file, dev, args->handle, &args->offset); + return __assign_gem_object_mmap_data(file, args->handle, + I915_MMAP_TYPE_GTT, + &args->offset); +} + +void i915_mmap_offset_destroy(struct i915_mmap_offset *mmo, struct mutex *mutex) +{ + if (mmo->file) + drm_vma_node_revoke(&mmo->vma_node, mmo->file); + drm_vma_offset_remove(mmo->dev->vma_offset_manager, &mmo->vma_node); + + mutex_lock(mutex); + list_del(&mmo->offset); + mutex_unlock(mutex); + + kfree(mmo); +} + +static void i915_gem_vm_open(struct vm_area_struct *vma) +{ + struct i915_mmap_offset *mmo = vma->vm_private_data; + struct drm_i915_gem_object *obj = mmo->obj; + + GEM_BUG_ON(!obj); + i915_gem_object_get(obj); +} + +static void i915_gem_vm_close(struct vm_area_struct *vma) +{ + struct i915_mmap_offset *mmo = vma->vm_private_data; + struct drm_i915_gem_object *obj = mmo->obj; + + GEM_BUG_ON(!obj); + i915_gem_object_put(obj); +} + +static const struct vm_operations_struct i915_gem_gtt_vm_ops = { + .fault = i915_gem_fault, + .open = i915_gem_vm_open, + .close = i915_gem_vm_close, +}; + +/* This overcomes the limitation in drm_gem_mmap's assignment of a + * drm_gem_object as the vma->vm_private_data. Since we need to + * be able to resolve multiple mmap offsets which could be tied + * to a single gem object. + */ +int i915_gem_mmap(struct file *filp, struct vm_area_struct *vma) +{ + struct drm_vma_offset_node *node; + struct drm_file *priv = filp->private_data; + struct drm_device *dev = priv->minor->dev; + struct i915_mmap_offset *mmo = NULL; + struct drm_gem_object *obj = NULL; + + if (drm_dev_is_unplugged(dev)) + return -ENODEV; + + drm_vma_offset_lock_lookup(dev->vma_offset_manager); + node = drm_vma_offset_exact_lookup_locked(dev->vma_offset_manager, + vma->vm_pgoff, + vma_pages(vma)); + if (likely(node)) { + mmo = container_of(node, struct i915_mmap_offset, + vma_node); + /* + * In our dependency chain, the drm_vma_offset_node + * depends on the validity of the mmo, which depends on + * the gem object. However the only reference we have + * at this point is the mmo (as the parent of the node). + * Try to check if the gem object was at least cleared. + */ + if (!mmo || !mmo->obj) { + drm_vma_offset_unlock_lookup(dev->vma_offset_manager); + return -EINVAL; + } + /* + * Skip 0-refcnted objects as it is in the process of being + * destroyed and will be invalid when the vma manager lock + * is released. + */ + obj = &mmo->obj->base; + if (!kref_get_unless_zero(&obj->refcount)) + obj = NULL; + + } + drm_vma_offset_unlock_lookup(dev->vma_offset_manager); + + if (!obj) + return -EINVAL; + + if (!drm_vma_node_is_allowed(node, priv)) { + drm_gem_object_put_unlocked(obj); + return -EACCES; + } + + if (to_intel_bo(obj)->readonly) { + if (vma->vm_flags & VM_WRITE) { + drm_gem_object_put_unlocked(obj); + return -EINVAL; + } + vma->vm_flags &= ~VM_MAYWRITE; + } + + vma->vm_flags |= VM_PFNMAP | VM_DONTEXPAND | VM_DONTDUMP; + vma->vm_page_prot = pgprot_writecombine(vm_get_page_prot(vma->vm_flags)); + vma->vm_page_prot = pgprot_decrypted(vma->vm_page_prot); + vma->vm_private_data = mmo; + + vma->vm_ops = &i915_gem_gtt_vm_ops; + + return 0; } #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c index dbf9be9a79f4..8abcd5202374 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c @@ -58,6 +58,9 @@ void i915_gem_object_init(struct drm_i915_gem_object *obj, INIT_LIST_HEAD(&obj->lut_list); + mutex_init(&obj->mmo_lock); + INIT_LIST_HEAD(&obj->mmap_offsets); + init_rcu_head(&obj->rcu); obj->ops = ops; @@ -155,6 +158,8 @@ static void __i915_gem_free_objects(struct drm_i915_private *i915, wakeref = intel_runtime_pm_get(&i915->runtime_pm); llist_for_each_entry_safe(obj, on, freed, freed) { + struct i915_mmap_offset *mmo, *mn; + trace_i915_gem_object_destroy(obj); if (!list_empty(&obj->vma.list)) { @@ -180,6 +185,14 @@ static void __i915_gem_free_objects(struct drm_i915_private *i915, spin_unlock(&obj->vma.lock); } + i915_gem_object_release_mmap(obj); + + list_for_each_entry_safe(mmo, mn, &obj->mmap_offsets, offset) { + mmo->obj = NULL; + i915_mmap_offset_destroy(mmo, &obj->mmo_lock); + } + + GEM_BUG_ON(!list_empty(&obj->mmap_offsets)); GEM_BUG_ON(atomic_read(&obj->bind_count)); GEM_BUG_ON(obj->userfault_count); GEM_BUG_ON(!list_empty(&obj->lut_list)); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h index 85921796851f..d529d011e85c 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h @@ -131,13 +131,13 @@ void i915_gem_object_unlock_fence(struct drm_i915_gem_object *obj, static inline void i915_gem_object_set_readonly(struct drm_i915_gem_object *obj) { - obj->base.vma_node.readonly = true; + obj->readonly = true; } static inline bool i915_gem_object_is_readonly(const struct drm_i915_gem_object *obj) { - return obj->base.vma_node.readonly; + return obj->readonly; } static inline bool @@ -375,7 +375,7 @@ static inline void i915_gem_object_unpin_map(struct drm_i915_gem_object *obj) i915_gem_object_unpin_pages(obj); } -void __i915_gem_object_release_mmap(struct drm_i915_gem_object *obj); +void __i915_gem_object_release_mmap_gtt(struct drm_i915_gem_object *obj); void i915_gem_object_release_mmap(struct drm_i915_gem_object *obj); void @@ -461,6 +461,9 @@ int i915_gem_object_wait(struct drm_i915_gem_object *obj, int i915_gem_object_wait_priority(struct drm_i915_gem_object *obj, unsigned int flags, const struct i915_sched_attr *attr); + +void i915_mmap_offset_destroy(struct i915_mmap_offset *mmo, struct mutex *mutex); + #define I915_PRIORITY_DISPLAY I915_USER_PRIORITY(I915_PRIORITY_MAX) #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 a387e3ee728b..c5c305bd9927 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h @@ -62,6 +62,20 @@ struct drm_i915_gem_object_ops { void (*release)(struct drm_i915_gem_object *obj); }; +enum i915_mmap_type { + I915_MMAP_TYPE_GTT = 0, +}; + +struct i915_mmap_offset { + struct drm_device *dev; + struct drm_vma_offset_node vma_node; + struct drm_i915_gem_object *obj; + struct drm_file *file; + enum i915_mmap_type mmap_type; + + struct list_head offset; +}; + struct drm_i915_gem_object { struct drm_gem_object base; @@ -117,6 +131,11 @@ struct drm_i915_gem_object { unsigned int userfault_count; struct list_head userfault_link; + /* Protects access to mmap offsets */ + struct mutex mmo_lock; + struct list_head mmap_offsets; + bool readonly:1; + I915_SELFTEST_DECLARE(struct list_head st_link); unsigned long flags; diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c index 65d4dbf91999..a4b14a2fb77a 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c @@ -557,15 +557,20 @@ static bool assert_mmap_offset(struct drm_i915_private *i915, int expected) { struct drm_i915_gem_object *obj; + /* Ownership transferred to parent gem object in create_mmap_offset */ + struct i915_mmap_offset *mmo = kzalloc(sizeof(*mmo), GFP_KERNEL); int err; obj = i915_gem_object_create_internal(i915, size); if (IS_ERR(obj)) return PTR_ERR(obj); - err = create_mmap_offset(obj); + err = create_mmap_offset(obj, mmo); + if (err) + kfree(mmo); i915_gem_object_put(obj); + return err == expected; } @@ -601,6 +606,8 @@ static int igt_mmap_offset_exhaustion(void *arg) struct drm_mm *mm = &i915->drm.vma_offset_manager->vm_addr_space_mm; struct drm_i915_gem_object *obj; struct drm_mm_node resv, *hole; + /* Ownership transferred to parent gem object in create_mmap_offset */ + struct i915_mmap_offset *mmo = kzalloc(sizeof(*mmo), GFP_KERNEL); u64 hole_start, hole_end; int loop, err; @@ -644,9 +651,10 @@ static int igt_mmap_offset_exhaustion(void *arg) goto out; } - err = create_mmap_offset(obj); + err = create_mmap_offset(obj, mmo); if (err) { pr_err("Unable to insert object into reclaimed hole\n"); + kfree(mmo); goto err_obj; } diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c index bf8d1ed4b1d8..5e985c0cb949 100644 --- a/drivers/gpu/drm/i915/gt/intel_reset.c +++ b/drivers/gpu/drm/i915/gt/intel_reset.c @@ -667,8 +667,13 @@ static void revoke_mmaps(struct intel_gt *gt) continue; GEM_BUG_ON(vma->fence != >->ggtt->fence_regs[i]); - node = &vma->obj->base.vma_node; + + if (!vma->mmo) + continue; + + node = &vma->mmo->vma_node; vma_offset = vma->ggtt_view.partial.offset << PAGE_SHIFT; + unmap_mapping_range(gt->i915->drm.anon_inode->i_mapping, drm_vma_node_offset_addr(node) + vma_offset, vma->size, diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 21b36a6143b9..ead2e6ca7907 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -2648,18 +2648,12 @@ const struct dev_pm_ops i915_pm_ops = { .runtime_resume = intel_runtime_resume, }; -static const struct vm_operations_struct i915_gem_vm_ops = { - .fault = i915_gem_fault, - .open = drm_gem_vm_open, - .close = drm_gem_vm_close, -}; - static const struct file_operations i915_driver_fops = { .owner = THIS_MODULE, .open = drm_open, .release = drm_release, .unlocked_ioctl = drm_ioctl, - .mmap = drm_gem_mmap, + .mmap = i915_gem_mmap, .poll = drm_poll, .read = drm_read, .compat_ioctl = i915_compat_ioctl, @@ -2748,7 +2742,6 @@ static struct drm_driver driver = { .gem_close_object = i915_gem_close_object, .gem_free_object_unlocked = i915_gem_free_object, - .gem_vm_ops = &i915_gem_vm_ops, .prime_handle_to_fd = drm_gem_prime_handle_to_fd, .prime_fd_to_handle = drm_gem_prime_fd_to_handle, @@ -2759,7 +2752,6 @@ static struct drm_driver driver = { .get_scanout_position = i915_get_crtc_scanoutpos, .dumb_create = i915_gem_dumb_create, - .dumb_map_offset = i915_gem_mmap_gtt, .ioctls = i915_ioctls, .num_ioctls = ARRAY_SIZE(i915_ioctls), .fops = &i915_driver_fops, diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 8882c0908c3b..ed0fc1d6dfab 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -1932,8 +1932,6 @@ i915_mutex_lock_interruptible(struct drm_device *dev) int i915_gem_dumb_create(struct drm_file *file_priv, struct drm_device *dev, struct drm_mode_create_dumb *args); -int i915_gem_mmap_gtt(struct drm_file *file_priv, struct drm_device *dev, - u32 handle, u64 *offset); int i915_gem_mmap_gtt_version(void); int __must_check i915_gem_set_global_seqno(struct drm_device *dev, u32 seqno); @@ -1958,6 +1956,7 @@ void i915_gem_driver_release(struct drm_i915_private *dev_priv); void i915_gem_suspend(struct drm_i915_private *dev_priv); void i915_gem_suspend_late(struct drm_i915_private *dev_priv); void i915_gem_resume(struct drm_i915_private *dev_priv); +int i915_gem_mmap(struct file *filp, struct vm_area_struct *vma); vm_fault_t i915_gem_fault(struct vm_fault *vmf); int i915_gem_open(struct drm_i915_private *i915, struct drm_file *file); diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index a2ba123f7b31..2e20c6020334 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -852,7 +852,7 @@ void i915_gem_runtime_suspend(struct drm_i915_private *i915) list_for_each_entry_safe(obj, on, &i915->ggtt.userfault_list, userfault_link) - __i915_gem_object_release_mmap(obj); + __i915_gem_object_release_mmap_gtt(obj); /* * The fence will be lost when the device powers down. If any were diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index e90c4d0af8fd..c89e7b7a797b 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -1058,7 +1058,7 @@ static void __i915_vma_iounmap(struct i915_vma *vma) void i915_vma_revoke_mmap(struct i915_vma *vma) { - struct drm_vma_offset_node *node = &vma->obj->base.vma_node; + struct drm_vma_offset_node *node; u64 vma_offset; lockdep_assert_held(&vma->vm->mutex); @@ -1070,10 +1070,15 @@ void i915_vma_revoke_mmap(struct i915_vma *vma) GEM_BUG_ON(!vma->obj->userfault_count); vma_offset = vma->ggtt_view.partial.offset << PAGE_SHIFT; - unmap_mapping_range(vma->vm->i915->drm.anon_inode->i_mapping, - drm_vma_node_offset_addr(node) + vma_offset, - vma->size, - 1); + + if (vma->mmo) { + node = &vma->mmo->vma_node; + + unmap_mapping_range(vma->vm->i915->drm.anon_inode->i_mapping, + drm_vma_node_offset_addr(node) + vma_offset, + vma->size, + 1); + } i915_vma_unset_userfault(vma); if (!--vma->obj->userfault_count) diff --git a/drivers/gpu/drm/i915/i915_vma.h b/drivers/gpu/drm/i915/i915_vma.h index 858908e3d1cc..34afd1a59ea5 100644 --- a/drivers/gpu/drm/i915/i915_vma.h +++ b/drivers/gpu/drm/i915/i915_vma.h @@ -63,6 +63,10 @@ struct i915_vma { u64 display_alignment; struct i915_page_sizes page_sizes; + /* mmap-offset associated with fencing for this vma */ + struct i915_mmap_offset *mmo; + struct list_head mmo_link; + u32 fence_size; u32 fence_alignment; From patchwork Mon Oct 21 10:48:08 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Abdiel Janulgue X-Patchwork-Id: 11201881 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 74DB814E5 for ; Mon, 21 Oct 2019 10:48:30 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 5D0E22067B for ; Mon, 21 Oct 2019 10:48:30 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 5D0E22067B Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=linux.intel.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=intel-gfx-bounces@lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id E9E1989E50; Mon, 21 Oct 2019 10:48:29 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by gabe.freedesktop.org (Postfix) with ESMTPS id 9161289D84 for ; Mon, 21 Oct 2019 10:48:23 +0000 (UTC) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga002.jf.intel.com ([10.7.209.21]) by orsmga102.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 21 Oct 2019 03:48:23 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.67,323,1566889200"; d="scan'208";a="209306841" Received: from ahedstro-mobl1.ger.corp.intel.com (HELO skylake-nuc.ger.corp.intel.com) ([10.249.254.249]) by orsmga002.jf.intel.com with ESMTP; 21 Oct 2019 03:48:22 -0700 From: Abdiel Janulgue To: intel-gfx@lists.freedesktop.org Date: Mon, 21 Oct 2019 13:48:08 +0300 Message-Id: <20191021104811.25663-2-abdiel.janulgue@linux.intel.com> X-Mailer: git-send-email 2.23.0 In-Reply-To: <20191021104811.25663-1-abdiel.janulgue@linux.intel.com> References: <20191021104811.25663-1-abdiel.janulgue@linux.intel.com> MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH 2/5] drm/i915: define HAS_MAPPABLE_APERTURE X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Matthew Auld Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" From: Daniele Ceraolo Spurio The following patches in the series will use it to avoid certain operations when aperture is not available in HW. Signed-off-by: Daniele Ceraolo Spurio Cc: Matthew Auld --- drivers/gpu/drm/i915/i915_drv.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index ed0fc1d6dfab..a407ca62888b 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -1732,6 +1732,8 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define OVERLAY_NEEDS_PHYSICAL(dev_priv) \ (INTEL_INFO(dev_priv)->display.overlay_needs_physical) +#define HAS_MAPPABLE_APERTURE(dev_priv) (dev_priv->ggtt.mappable_end > 0) + /* Early gen2 have a totally busted CS tlb and require pinned batches. */ #define HAS_BROKEN_CS_TLB(dev_priv) (IS_I830(dev_priv) || IS_I845G(dev_priv)) From patchwork Mon Oct 21 10:48:09 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Abdiel Janulgue X-Patchwork-Id: 11201879 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id EC35B1515 for ; Mon, 21 Oct 2019 10:48:27 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id D49AF2067B for ; Mon, 21 Oct 2019 10:48:27 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org D49AF2067B Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=linux.intel.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=intel-gfx-bounces@lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id F07E589E47; Mon, 21 Oct 2019 10:48:26 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by gabe.freedesktop.org (Postfix) with ESMTPS id 88B5F89D84 for ; Mon, 21 Oct 2019 10:48:25 +0000 (UTC) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga002.jf.intel.com ([10.7.209.21]) by orsmga102.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 21 Oct 2019 03:48:25 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.67,323,1566889200"; d="scan'208";a="209306847" Received: from ahedstro-mobl1.ger.corp.intel.com (HELO skylake-nuc.ger.corp.intel.com) ([10.249.254.249]) by orsmga002.jf.intel.com with ESMTP; 21 Oct 2019 03:48:23 -0700 From: Abdiel Janulgue To: intel-gfx@lists.freedesktop.org Date: Mon, 21 Oct 2019 13:48:09 +0300 Message-Id: <20191021104811.25663-3-abdiel.janulgue@linux.intel.com> X-Mailer: git-send-email 2.23.0 In-Reply-To: <20191021104811.25663-1-abdiel.janulgue@linux.intel.com> References: <20191021104811.25663-1-abdiel.janulgue@linux.intel.com> MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH 3/5] drm/i915: Introduce DRM_I915_GEM_MMAP_OFFSET X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Matthew Auld Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" This is really just an alias of mmap_gtt. The 'mmap offset' nomenclature comes from the value returned by this ioctl which is the offset into the device fd which userpace uses with mmap(2). mmap_gtt was our initial mmap_offset implementation, this extends our CPU mmap support to allow additional fault handlers that depends on the object's backing pages. Note that we multiplex mmap_gtt and mmap_offset through the same ioctl, and use the zero extending behaviour of drm to differentiate between them, when we inspect the flags. v2: - Drop the alias, just rename the struct (Chris) - Don't bail out on no PAT when doing WB mmaps - Prepare uAPI for further extensions v3: - drop MMAP_OFFSET_FLAGS Signed-off-by: Abdiel Janulgue Signed-off-by: Matthew Auld Cc: Joonas Lahtinen --- drivers/gpu/drm/i915/gem/i915_gem_mman.c | 44 ++++++++++++++++++- .../gpu/drm/i915/gem/i915_gem_object_types.h | 3 ++ include/uapi/drm/i915_drm.h | 27 +++++++++++- 3 files changed, 71 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index 3491bb06606b..a3eb194fd3b4 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -144,6 +144,9 @@ static unsigned int tile_row_pages(const struct drm_i915_gem_object *obj) * 3 - Remove implicit set-domain(GTT) and synchronisation on initial * pagefault; swapin remains transparent. * + * 4 - Support multiple fault handlers per object depending on object's + * backing storage (a.k.a. MMAP_OFFSET). + * * Restrictions: * * * snoopable objects cannot be accessed via the GTT. It can cause machine @@ -171,7 +174,7 @@ static unsigned int tile_row_pages(const struct drm_i915_gem_object *obj) */ int i915_gem_mmap_gtt_version(void) { - return 3; + return 4; } static inline struct i915_ggtt_view @@ -534,6 +537,35 @@ __assign_gem_object_mmap_data(struct drm_file *file, return ret; } +static int gem_mmap_offset(struct drm_device *dev, void *data, + struct drm_file *file) +{ + struct drm_i915_gem_mmap_offset *args = data; + enum i915_mmap_type type; + + switch (args->flags) { + case I915_MMAP_OFFSET_WC: + if (!boot_cpu_has(X86_FEATURE_PAT)) + return -ENODEV; + type = I915_MMAP_TYPE_WC; + break; + case I915_MMAP_OFFSET_WB: + type = I915_MMAP_TYPE_WB; + break; + case I915_MMAP_OFFSET_UC: + if (!boot_cpu_has(X86_FEATURE_PAT)) + return -ENODEV; + type = I915_MMAP_TYPE_UC; + break; + default: + return -EINVAL; + + }; + + return __assign_gem_object_mmap_data(file, args->handle, type, + &args->offset); +} + /** * i915_gem_mmap_gtt_ioctl - prepare an object for GTT mmap'ing * @dev: DRM device @@ -553,7 +585,15 @@ int i915_gem_mmap_gtt_ioctl(struct drm_device *dev, void *data, struct drm_file *file) { - struct drm_i915_gem_mmap_gtt *args = data; + struct drm_i915_gem_mmap_offset *args = data; + struct drm_i915_private *i915 = to_i915(dev); + + if (args->flags) + return gem_mmap_offset(dev, data, file); + + if (!HAS_MAPPABLE_APERTURE(i915)) + /* No aperture, cannot mmap via legacy GTT */ + return -ENODEV; return __assign_gem_object_mmap_data(file, args->handle, I915_MMAP_TYPE_GTT, 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 c5c305bd9927..c643d45f5c0d 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h @@ -64,6 +64,9 @@ struct drm_i915_gem_object_ops { enum i915_mmap_type { I915_MMAP_TYPE_GTT = 0, + I915_MMAP_TYPE_WC, + I915_MMAP_TYPE_WB, + I915_MMAP_TYPE_UC, }; struct i915_mmap_offset { diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 63d40cba97e0..190659923dea 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -394,7 +394,7 @@ typedef struct _drm_i915_sarea { #define DRM_IOCTL_I915_GEM_PREAD DRM_IOW (DRM_COMMAND_BASE + DRM_I915_GEM_PREAD, struct drm_i915_gem_pread) #define DRM_IOCTL_I915_GEM_PWRITE DRM_IOW (DRM_COMMAND_BASE + DRM_I915_GEM_PWRITE, struct drm_i915_gem_pwrite) #define DRM_IOCTL_I915_GEM_MMAP DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GEM_MMAP, struct drm_i915_gem_mmap) -#define DRM_IOCTL_I915_GEM_MMAP_GTT DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GEM_MMAP_GTT, struct drm_i915_gem_mmap_gtt) +#define DRM_IOCTL_I915_GEM_MMAP_GTT DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GEM_MMAP_GTT, struct drm_i915_gem_mmap_offset) #define DRM_IOCTL_I915_GEM_SET_DOMAIN DRM_IOW (DRM_COMMAND_BASE + DRM_I915_GEM_SET_DOMAIN, struct drm_i915_gem_set_domain) #define DRM_IOCTL_I915_GEM_SW_FINISH DRM_IOW (DRM_COMMAND_BASE + DRM_I915_GEM_SW_FINISH, struct drm_i915_gem_sw_finish) #define DRM_IOCTL_I915_GEM_SET_TILING DRM_IOWR (DRM_COMMAND_BASE + DRM_I915_GEM_SET_TILING, struct drm_i915_gem_set_tiling) @@ -793,6 +793,31 @@ struct drm_i915_gem_mmap_gtt { __u64 offset; }; +struct drm_i915_gem_mmap_offset { + /** Handle for the object being mapped. */ + __u32 handle; + __u32 pad; + /** + * Fake offset to use for subsequent mmap call + * + * This is a fixed-size type for 32/64 compatibility. + */ + __u64 offset; + + /** + * Flags for extended behaviour. + * + * It is mandatory that either one of the MMAP_OFFSET flags + * should be passed here. + */ + __u64 flags; +#define I915_MMAP_OFFSET_WC 1 +#define I915_MMAP_OFFSET_WB 2 +#define I915_MMAP_OFFSET_UC 3 + + __u64 extensions; +}; + struct drm_i915_gem_set_domain { /** Handle for the object */ __u32 handle; From patchwork Mon Oct 21 10:48:10 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Abdiel Janulgue X-Patchwork-Id: 11201885 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id BDF3A14E5 for ; Mon, 21 Oct 2019 10:48:36 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id A5C752067B for ; Mon, 21 Oct 2019 10:48:36 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org A5C752067B Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=linux.intel.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=intel-gfx-bounces@lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 32BF289EAC; Mon, 21 Oct 2019 10:48:36 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by gabe.freedesktop.org (Postfix) with ESMTPS id 7EB5C89AEB for ; Mon, 21 Oct 2019 10:48:27 +0000 (UTC) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga002.jf.intel.com ([10.7.209.21]) by orsmga102.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 21 Oct 2019 03:48:27 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.67,323,1566889200"; d="scan'208";a="209306857" Received: from ahedstro-mobl1.ger.corp.intel.com (HELO skylake-nuc.ger.corp.intel.com) ([10.249.254.249]) by orsmga002.jf.intel.com with ESMTP; 21 Oct 2019 03:48:25 -0700 From: Abdiel Janulgue To: intel-gfx@lists.freedesktop.org Date: Mon, 21 Oct 2019 13:48:10 +0300 Message-Id: <20191021104811.25663-4-abdiel.janulgue@linux.intel.com> X-Mailer: git-send-email 2.23.0 In-Reply-To: <20191021104811.25663-1-abdiel.janulgue@linux.intel.com> References: <20191021104811.25663-1-abdiel.janulgue@linux.intel.com> MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH 4/5] drm/i915: cpu-map based dumb buffers X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Matthew Auld Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" No reason for sticking to GTT mmaps now we have CPU WC mmaps via our new mmap offset plumbing. Signed-off-by: Abdiel Janulgue Cc: Matthew Auld --- drivers/gpu/drm/i915/gem/i915_gem_mman.c | 10 ++++++++++ drivers/gpu/drm/i915/i915_drv.c | 1 + drivers/gpu/drm/i915/i915_drv.h | 2 ++ 3 files changed, 13 insertions(+) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index a3eb194fd3b4..8e2919832483 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -566,6 +566,16 @@ static int gem_mmap_offset(struct drm_device *dev, void *data, &args->offset); } +int +i915_gem_mmap_dumb(struct drm_file *file, + struct drm_device *dev, + u32 handle, + u64 *offset) +{ + return __assign_gem_object_mmap_data(file, handle, I915_MMAP_TYPE_WC, + offset); +} + /** * i915_gem_mmap_gtt_ioctl - prepare an object for GTT mmap'ing * @dev: DRM device diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index ead2e6ca7907..2157bc961e1f 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -2752,6 +2752,7 @@ static struct drm_driver driver = { .get_scanout_position = i915_get_crtc_scanoutpos, .dumb_create = i915_gem_dumb_create, + .dumb_map_offset = i915_gem_mmap_dumb, .ioctls = i915_ioctls, .num_ioctls = ARRAY_SIZE(i915_ioctls), .fops = &i915_driver_fops, diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index a407ca62888b..800674d167a5 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -1934,6 +1934,8 @@ i915_mutex_lock_interruptible(struct drm_device *dev) int i915_gem_dumb_create(struct drm_file *file_priv, struct drm_device *dev, struct drm_mode_create_dumb *args); +int i915_gem_mmap_dumb(struct drm_file *file_priv, struct drm_device *dev, + u32 handle, u64 *offset); int i915_gem_mmap_gtt_version(void); int __must_check i915_gem_set_global_seqno(struct drm_device *dev, u32 seqno); From patchwork Mon Oct 21 10:48:11 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Abdiel Janulgue X-Patchwork-Id: 11201883 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id CFDB013B1 for ; Mon, 21 Oct 2019 10:48:31 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id B80492067B for ; Mon, 21 Oct 2019 10:48:31 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org B80492067B Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=linux.intel.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=intel-gfx-bounces@lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 8FD9B89E86; Mon, 21 Oct 2019 10:48:30 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by gabe.freedesktop.org (Postfix) with ESMTPS id 811A989E50 for ; Mon, 21 Oct 2019 10:48:29 +0000 (UTC) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga002.jf.intel.com ([10.7.209.21]) by orsmga102.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 21 Oct 2019 03:48:29 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.67,323,1566889200"; d="scan'208";a="209306862" Received: from ahedstro-mobl1.ger.corp.intel.com (HELO skylake-nuc.ger.corp.intel.com) ([10.249.254.249]) by orsmga002.jf.intel.com with ESMTP; 21 Oct 2019 03:48:27 -0700 From: Abdiel Janulgue To: intel-gfx@lists.freedesktop.org Date: Mon, 21 Oct 2019 13:48:11 +0300 Message-Id: <20191021104811.25663-5-abdiel.janulgue@linux.intel.com> X-Mailer: git-send-email 2.23.0 In-Reply-To: <20191021104811.25663-1-abdiel.janulgue@linux.intel.com> References: <20191021104811.25663-1-abdiel.janulgue@linux.intel.com> MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH 5/5] drm/i915: Add cpu fault handler for mmap_offset X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Matthew Auld Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" Fault handler to handle missing pages for shmem-backed objects. v2: bail out of inserting PTEs when failing to insert the fault address v3: has struct page check Signed-off-by: Abdiel Janulgue Signed-off-by: Matthew Auld Cc: Joonas Lahtinen --- drivers/gpu/drm/i915/gem/i915_gem_mman.c | 128 ++++++++++++++++++----- 1 file changed, 103 insertions(+), 25 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index 8e2919832483..300b1b8b5a8a 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -5,6 +5,7 @@ */ #include +#include #include #include "gt/intel_gt.h" @@ -200,6 +201,70 @@ compute_partial_view(const struct drm_i915_gem_object *obj, return view; } +static vm_fault_t i915_error_to_vmf_fault(int err) +{ + switch (err) { + default: + WARN_ONCE(err, "unhandled error in %s: %i\n", __func__, err); + /* fallthrough */ + case -EIO: /* shmemfs failure from swap device */ + case -EFAULT: /* purged object */ + case -ENODEV: /* bad object, how did you get here! */ + return VM_FAULT_SIGBUS; + + case -ENOSPC: /* shmemfs allocation failure */ + case -ENOMEM: /* our allocation failure */ + return VM_FAULT_OOM; + + case 0: + case -EAGAIN: + case -ERESTARTSYS: + case -EINTR: + case -EBUSY: + /* + * EBUSY is ok: this just means that another thread + * already did the job. + */ + return VM_FAULT_NOPAGE; + } +} + +static vm_fault_t i915_gem_fault_cpu(struct vm_fault *vmf) +{ + struct vm_area_struct *area = vmf->vma; + struct i915_mmap_offset *priv = area->vm_private_data; + struct drm_i915_gem_object *obj = priv->obj; + vm_fault_t vmf_ret; + unsigned long i, size = area->vm_end - area->vm_start; + bool write = area->vm_flags & VM_WRITE; + int ret; + + if (!i915_gem_object_has_struct_page(obj)) + return VM_FAULT_SIGBUS; + + /* Sanity check that we allow writing into this object */ + if (i915_gem_object_is_readonly(obj) && write) + return VM_FAULT_SIGBUS; + + ret = i915_gem_object_pin_pages(obj); + if (ret) + return i915_error_to_vmf_fault(ret); + + for (i = 0; i < size >> PAGE_SHIFT; i++) { + struct page *page = i915_gem_object_get_page(obj, i); + + vmf_ret = vmf_insert_pfn(area, + (unsigned long)area->vm_start + i * PAGE_SIZE, + page_to_pfn(page)); + if (vmf_ret != VM_FAULT_NOPAGE) + break; + } + + i915_gem_object_unpin_pages(obj); + + return vmf_ret; +} + /** * i915_gem_fault - fault a page into the GTT * @vmf: fault info @@ -339,30 +404,7 @@ vm_fault_t i915_gem_fault(struct vm_fault *vmf) intel_runtime_pm_put(rpm, wakeref); i915_gem_object_unpin_pages(obj); err: - switch (ret) { - default: - WARN_ONCE(ret, "unhandled error in %s: %i\n", __func__, ret); - /* fallthrough */ - case -EIO: /* shmemfs failure from swap device */ - case -EFAULT: /* purged object */ - case -ENODEV: /* bad object, how did you get here! */ - return VM_FAULT_SIGBUS; - - case -ENOSPC: /* shmemfs allocation failure */ - case -ENOMEM: /* our allocation failure */ - return VM_FAULT_OOM; - - case 0: - case -EAGAIN: - case -ERESTARTSYS: - case -EINTR: - case -EBUSY: - /* - * EBUSY is ok: this just means that another thread - * already did the job. - */ - return VM_FAULT_NOPAGE; - } + return i915_error_to_vmf_fault(ret); } void __i915_gem_object_release_mmap_gtt(struct drm_i915_gem_object *obj) @@ -647,6 +689,33 @@ static const struct vm_operations_struct i915_gem_gtt_vm_ops = { .close = i915_gem_vm_close, }; +static const struct vm_operations_struct i915_gem_cpu_vm_ops = { + .fault = i915_gem_fault_cpu, + .open = i915_gem_vm_open, + .close = i915_gem_vm_close, +}; + +static void set_vmdata_mmap_offset(struct i915_mmap_offset *mmo, struct vm_area_struct *vma) +{ + switch (mmo->mmap_type) { + case I915_MMAP_TYPE_WC: + vma->vm_page_prot = + pgprot_writecombine(vm_get_page_prot(vma->vm_flags)); + break; + case I915_MMAP_TYPE_WB: + vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); + break; + case I915_MMAP_TYPE_UC: + vma->vm_page_prot = + pgprot_noncached(vm_get_page_prot(vma->vm_flags)); + break; + default: + break; + } + + vma->vm_ops = &i915_gem_cpu_vm_ops; +} + /* This overcomes the limitation in drm_gem_mmap's assignment of a * drm_gem_object as the vma->vm_private_data. Since we need to * be able to resolve multiple mmap offsets which could be tied @@ -714,7 +783,16 @@ int i915_gem_mmap(struct file *filp, struct vm_area_struct *vma) vma->vm_page_prot = pgprot_decrypted(vma->vm_page_prot); vma->vm_private_data = mmo; - vma->vm_ops = &i915_gem_gtt_vm_ops; + switch (mmo->mmap_type) { + case I915_MMAP_TYPE_WC: + case I915_MMAP_TYPE_WB: + case I915_MMAP_TYPE_UC: + set_vmdata_mmap_offset(mmo, vma); + break; + case I915_MMAP_TYPE_GTT: + vma->vm_ops = &i915_gem_gtt_vm_ops; + break; + } return 0; }