#include <drm/drm_vma_manager.h>
#include <drm/i915_drm.h>
#include "i915_drv.h"
+#include "i915_gem_dmabuf.h"
#include "i915_vgpu.h"
#include "i915_trace.h"
#include "intel_drv.h"
+#include "intel_frontbuffer.h"
#include "intel_mocs.h"
+#include <linux/reservation.h>
#include <linux/shmem_fs.h>
#include <linux/slab.h>
#include <linux/swap.h>
static void i915_gem_object_flush_gtt_write_domain(struct drm_i915_gem_object *obj);
static void i915_gem_object_flush_cpu_write_domain(struct drm_i915_gem_object *obj);
-static void
-i915_gem_object_retire__write(struct drm_i915_gem_object *obj);
-static void
-i915_gem_object_retire__read(struct drm_i915_gem_object *obj, int ring);
static bool cpu_cache_is_coherent(struct drm_device *dev,
enum i915_cache_level level)
static bool cpu_write_needs_clflush(struct drm_i915_gem_object *obj)
{
+ if (obj->base.write_domain == I915_GEM_DOMAIN_CPU)
+ return false;
+
if (!cpu_cache_is_coherent(obj->base.dev, obj->cache_level))
return true;
return obj->pin_display;
}
+static int
+insert_mappable_node(struct drm_i915_private *i915,
+ struct drm_mm_node *node, u32 size)
+{
+ memset(node, 0, sizeof(*node));
+ return drm_mm_insert_node_in_range_generic(&i915->ggtt.base.mm, node,
+ size, 0, 0, 0,
+ i915->ggtt.mappable_end,
+ DRM_MM_SEARCH_DEFAULT,
+ DRM_MM_CREATE_DEFAULT);
+}
+
+static void
+remove_mappable_node(struct drm_mm_node *node)
+{
+ drm_mm_remove_node(node);
+}
+
/* some bookkeeping */
static void i915_gem_info_add_obj(struct drm_i915_private *dev_priv,
size_t size)
int i915_mutex_lock_interruptible(struct drm_device *dev)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
int ret;
ret = i915_gem_wait_for_error(&dev_priv->gpu_error);
if (ret)
return ret;
- WARN_ON(i915_verify_lists(dev));
return 0;
}
pinned = 0;
mutex_lock(&dev->struct_mutex);
list_for_each_entry(vma, &ggtt->base.active_list, vm_link)
- if (vma->pin_count)
+ if (i915_vma_is_pinned(vma))
pinned += vma->node.size;
list_for_each_entry(vma, &ggtt->base.inactive_list, vm_link)
- if (vma->pin_count)
+ if (i915_vma_is_pinned(vma))
pinned += vma->node.size;
mutex_unlock(&dev->struct_mutex);
vaddr += PAGE_SIZE;
}
- i915_gem_chipset_flush(obj->base.dev);
+ i915_gem_chipset_flush(to_i915(obj->base.dev));
st = kmalloc(sizeof(*st), GFP_KERNEL);
if (st == NULL)
.release = i915_gem_object_release_phys,
};
-static int
-drop_pages(struct drm_i915_gem_object *obj)
+int
+i915_gem_object_unbind(struct drm_i915_gem_object *obj)
{
- struct i915_vma *vma, *next;
+ struct i915_vma *vma;
+ LIST_HEAD(still_in_list);
int ret;
- drm_gem_object_reference(&obj->base);
- list_for_each_entry_safe(vma, next, &obj->vma_list, obj_link)
- if (i915_vma_unbind(vma))
+ /* The vma will only be freed if it is marked as closed, and if we wait
+ * upon rendering to the vma, we may unbind anything in the list.
+ */
+ while ((vma = list_first_entry_or_null(&obj->vma_list,
+ struct i915_vma,
+ obj_link))) {
+ list_move_tail(&vma->obj_link, &still_in_list);
+ ret = i915_vma_unbind(vma);
+ if (ret)
break;
-
- ret = i915_gem_object_put_pages(obj);
- drm_gem_object_unreference(&obj->base);
+ }
+ list_splice(&still_in_list, &obj->vma_list);
return ret;
}
+/**
+ * Ensures that all rendering to the object has completed and the object is
+ * safe to unbind from the GTT or access from the CPU.
+ * @obj: i915 gem object
+ * @readonly: waiting for just read access or read-write access
+ */
+int
+i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
+ bool readonly)
+{
+ struct reservation_object *resv;
+ struct i915_gem_active *active;
+ unsigned long active_mask;
+ int idx;
+
+ lockdep_assert_held(&obj->base.dev->struct_mutex);
+
+ if (!readonly) {
+ active = obj->last_read;
+ active_mask = i915_gem_object_get_active(obj);
+ } else {
+ active_mask = 1;
+ active = &obj->last_write;
+ }
+
+ for_each_active(active_mask, idx) {
+ int ret;
+
+ ret = i915_gem_active_wait(&active[idx],
+ &obj->base.dev->struct_mutex);
+ if (ret)
+ return ret;
+ }
+
+ resv = i915_gem_object_get_dmabuf_resv(obj);
+ if (resv) {
+ long err;
+
+ err = reservation_object_wait_timeout_rcu(resv, !readonly, true,
+ MAX_SCHEDULE_TIMEOUT);
+ if (err < 0)
+ return err;
+ }
+
+ return 0;
+}
+
+/* A nonblocking variant of the above wait. Must be called prior to
+ * acquiring the mutex for the object, as the object state may change
+ * during this call. A reference must be held by the caller for the object.
+ */
+static __must_check int
+__unsafe_wait_rendering(struct drm_i915_gem_object *obj,
+ struct intel_rps_client *rps,
+ bool readonly)
+{
+ struct i915_gem_active *active;
+ unsigned long active_mask;
+ int idx;
+
+ active_mask = __I915_BO_ACTIVE(obj);
+ if (!active_mask)
+ return 0;
+
+ if (!readonly) {
+ active = obj->last_read;
+ } else {
+ active_mask = 1;
+ active = &obj->last_write;
+ }
+
+ for_each_active(active_mask, idx) {
+ int ret;
+
+ ret = i915_gem_active_wait_unlocked(&active[idx],
+ true, NULL, rps);
+ if (ret)
+ return ret;
+ }
+
+ return 0;
+}
+
+static struct intel_rps_client *to_rps_client(struct drm_file *file)
+{
+ struct drm_i915_file_private *fpriv = file->driver_priv;
+
+ return &fpriv->rps;
+}
+
int
i915_gem_object_attach_phys(struct drm_i915_gem_object *obj,
int align)
if (obj->base.filp == NULL)
return -EINVAL;
- ret = drop_pages(obj);
+ ret = i915_gem_object_unbind(obj);
+ if (ret)
+ return ret;
+
+ ret = i915_gem_object_put_pages(obj);
if (ret)
return ret;
}
drm_clflush_virt_range(vaddr, args->size);
- i915_gem_chipset_flush(dev);
+ i915_gem_chipset_flush(to_i915(dev));
out:
intel_fb_obj_flush(obj, false, ORIGIN_CPU);
void *i915_gem_object_alloc(struct drm_device *dev)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
return kmem_cache_zalloc(dev_priv->objects, GFP_KERNEL);
}
void i915_gem_object_free(struct drm_i915_gem_object *obj)
{
- struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
kmem_cache_free(dev_priv->objects, obj);
}
return -EINVAL;
/* Allocate the new object */
- obj = i915_gem_alloc_object(dev, size);
- if (obj == NULL)
- return -ENOMEM;
+ obj = i915_gem_object_create(dev, size);
+ if (IS_ERR(obj))
+ return PTR_ERR(obj);
ret = drm_gem_handle_create(file, &obj->base, &handle);
/* drop reference from allocate - handle holds it now */
- drm_gem_object_unreference_unlocked(&obj->base);
+ i915_gem_object_put_unlocked(obj);
if (ret)
return ret;
/**
* Creates a new mm object and returns a handle to it.
+ * @dev: drm device pointer
+ * @data: ioctl data blob
+ * @file: drm file pointer
*/
int
i915_gem_create_ioctl(struct drm_device *dev, void *data,
*needs_clflush = 0;
- if (WARN_ON((obj->ops->flags & I915_GEM_OBJECT_HAS_STRUCT_PAGE) == 0))
+ if (WARN_ON(!i915_gem_object_has_struct_page(obj)))
return -EINVAL;
+ ret = i915_gem_object_wait_rendering(obj, true);
+ if (ret)
+ return ret;
+
if (!(obj->base.read_domains & I915_GEM_DOMAIN_CPU)) {
/* If we're not in the cpu read domain, set ourself into the gtt
* read domain and manually flush cachelines (if required). This
* anyway again before the next pread happens. */
*needs_clflush = !cpu_cache_is_coherent(obj->base.dev,
obj->cache_level);
- ret = i915_gem_object_wait_rendering(obj, true);
- if (ret)
- return ret;
}
ret = i915_gem_object_get_pages(obj);
return ret ? - EFAULT : 0;
}
+static inline unsigned long
+slow_user_access(struct io_mapping *mapping,
+ uint64_t page_base, int page_offset,
+ char __user *user_data,
+ unsigned long length, bool pwrite)
+{
+ void __iomem *ioaddr;
+ void *vaddr;
+ uint64_t unwritten;
+
+ ioaddr = io_mapping_map_wc(mapping, page_base, PAGE_SIZE);
+ /* We can use the cpu mem copy function because this is X86. */
+ vaddr = (void __force *)ioaddr + page_offset;
+ if (pwrite)
+ unwritten = __copy_from_user(vaddr, user_data, length);
+ else
+ unwritten = __copy_to_user(user_data, vaddr, length);
+
+ io_mapping_unmap(ioaddr);
+ return unwritten;
+}
+
+static int
+i915_gem_gtt_pread(struct drm_device *dev,
+ struct drm_i915_gem_object *obj, uint64_t size,
+ uint64_t data_offset, uint64_t data_ptr)
+{
+ struct drm_i915_private *dev_priv = to_i915(dev);
+ struct i915_ggtt *ggtt = &dev_priv->ggtt;
+ struct drm_mm_node node;
+ char __user *user_data;
+ uint64_t remain;
+ uint64_t offset;
+ int ret;
+
+ ret = i915_gem_object_ggtt_pin(obj, NULL, 0, 0, PIN_MAPPABLE);
+ if (ret) {
+ ret = insert_mappable_node(dev_priv, &node, PAGE_SIZE);
+ if (ret)
+ goto out;
+
+ ret = i915_gem_object_get_pages(obj);
+ if (ret) {
+ remove_mappable_node(&node);
+ goto out;
+ }
+
+ i915_gem_object_pin_pages(obj);
+ } else {
+ node.start = i915_gem_obj_ggtt_offset(obj);
+ node.allocated = false;
+ ret = i915_gem_object_put_fence(obj);
+ if (ret)
+ goto out_unpin;
+ }
+
+ ret = i915_gem_object_set_to_gtt_domain(obj, false);
+ if (ret)
+ goto out_unpin;
+
+ user_data = u64_to_user_ptr(data_ptr);
+ remain = size;
+ offset = data_offset;
+
+ mutex_unlock(&dev->struct_mutex);
+ if (likely(!i915.prefault_disable)) {
+ ret = fault_in_multipages_writeable(user_data, remain);
+ if (ret) {
+ mutex_lock(&dev->struct_mutex);
+ goto out_unpin;
+ }
+ }
+
+ while (remain > 0) {
+ /* Operation in this page
+ *
+ * page_base = page offset within aperture
+ * page_offset = offset within page
+ * page_length = bytes to copy for this page
+ */
+ u32 page_base = node.start;
+ unsigned page_offset = offset_in_page(offset);
+ unsigned page_length = PAGE_SIZE - page_offset;
+ page_length = remain < page_length ? remain : page_length;
+ if (node.allocated) {
+ wmb();
+ ggtt->base.insert_page(&ggtt->base,
+ i915_gem_object_get_dma_address(obj, offset >> PAGE_SHIFT),
+ node.start,
+ I915_CACHE_NONE, 0);
+ wmb();
+ } else {
+ page_base += offset & PAGE_MASK;
+ }
+ /* This is a slow read/write as it tries to read from
+ * and write to user memory which may result into page
+ * faults, and so we cannot perform this under struct_mutex.
+ */
+ if (slow_user_access(ggtt->mappable, page_base,
+ page_offset, user_data,
+ page_length, false)) {
+ ret = -EFAULT;
+ break;
+ }
+
+ remain -= page_length;
+ user_data += page_length;
+ offset += page_length;
+ }
+
+ mutex_lock(&dev->struct_mutex);
+ if (ret == 0 && (obj->base.read_domains & I915_GEM_DOMAIN_GTT) == 0) {
+ /* The user has modified the object whilst we tried
+ * reading from it, and we now have no idea what domain
+ * the pages should be in. As we have just been touching
+ * them directly, flush everything back to the GTT
+ * domain.
+ */
+ ret = i915_gem_object_set_to_gtt_domain(obj, false);
+ }
+
+out_unpin:
+ if (node.allocated) {
+ wmb();
+ ggtt->base.clear_range(&ggtt->base,
+ node.start, node.size,
+ true);
+ i915_gem_object_unpin_pages(obj);
+ remove_mappable_node(&node);
+ } else {
+ i915_gem_object_ggtt_unpin(obj);
+ }
+out:
+ return ret;
+}
+
static int
i915_gem_shmem_pread(struct drm_device *dev,
struct drm_i915_gem_object *obj,
int needs_clflush = 0;
struct sg_page_iter sg_iter;
+ if (!i915_gem_object_has_struct_page(obj))
+ return -ENODEV;
+
user_data = u64_to_user_ptr(args->data_ptr);
remain = args->size;
/**
* Reads data from the object referenced by handle.
+ * @dev: drm device pointer
+ * @data: ioctl data blob
+ * @file: drm file pointer
*
* On error, the contents of *data are undefined.
*/
if (ret)
return ret;
- obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
- if (&obj->base == NULL) {
+ obj = i915_gem_object_lookup(file, args->handle);
+ if (!obj) {
ret = -ENOENT;
goto unlock;
}
goto out;
}
- /* prime objects have no backing filp to GEM pread/pwrite
- * pages from.
- */
- if (!obj->base.filp) {
- ret = -EINVAL;
- goto out;
- }
-
trace_i915_gem_object_pread(obj, args->offset, args->size);
ret = i915_gem_shmem_pread(dev, obj, args, file);
+ /* pread for non shmem backed objects */
+ if (ret == -EFAULT || ret == -ENODEV) {
+ intel_runtime_pm_get(to_i915(dev));
+ ret = i915_gem_gtt_pread(dev, obj, args->size,
+ args->offset, args->data_ptr);
+ intel_runtime_pm_put(to_i915(dev));
+ }
+
out:
- drm_gem_object_unreference(&obj->base);
+ i915_gem_object_put(obj);
unlock:
mutex_unlock(&dev->struct_mutex);
return ret;
/**
* This is the fast pwrite path, where we copy the data directly from the
* user into the GTT, uncached.
+ * @i915: i915 device private data
+ * @obj: i915 gem object
+ * @args: pwrite arguments structure
+ * @file: drm file pointer
*/
static int
-i915_gem_gtt_pwrite_fast(struct drm_device *dev,
+i915_gem_gtt_pwrite_fast(struct drm_i915_private *i915,
struct drm_i915_gem_object *obj,
struct drm_i915_gem_pwrite *args,
struct drm_file *file)
{
- struct drm_i915_private *dev_priv = to_i915(dev);
- struct i915_ggtt *ggtt = &dev_priv->ggtt;
- ssize_t remain;
- loff_t offset, page_base;
+ struct i915_ggtt *ggtt = &i915->ggtt;
+ struct drm_device *dev = obj->base.dev;
+ struct drm_mm_node node;
+ uint64_t remain, offset;
char __user *user_data;
- int page_offset, page_length, ret;
+ int ret;
+ bool hit_slow_path = false;
- ret = i915_gem_obj_ggtt_pin(obj, 0, PIN_MAPPABLE | PIN_NONBLOCK);
- if (ret)
- goto out;
+ if (obj->tiling_mode != I915_TILING_NONE)
+ return -EFAULT;
+
+ ret = i915_gem_object_ggtt_pin(obj, NULL, 0, 0,
+ PIN_MAPPABLE | PIN_NONBLOCK);
+ if (ret) {
+ ret = insert_mappable_node(i915, &node, PAGE_SIZE);
+ if (ret)
+ goto out;
+
+ ret = i915_gem_object_get_pages(obj);
+ if (ret) {
+ remove_mappable_node(&node);
+ goto out;
+ }
+
+ i915_gem_object_pin_pages(obj);
+ } else {
+ node.start = i915_gem_obj_ggtt_offset(obj);
+ node.allocated = false;
+ ret = i915_gem_object_put_fence(obj);
+ if (ret)
+ goto out_unpin;
+ }
ret = i915_gem_object_set_to_gtt_domain(obj, true);
if (ret)
goto out_unpin;
- ret = i915_gem_object_put_fence(obj);
- if (ret)
- goto out_unpin;
+ intel_fb_obj_invalidate(obj, ORIGIN_GTT);
+ obj->dirty = true;
user_data = u64_to_user_ptr(args->data_ptr);
+ offset = args->offset;
remain = args->size;
-
- offset = i915_gem_obj_ggtt_offset(obj) + args->offset;
-
- intel_fb_obj_invalidate(obj, ORIGIN_GTT);
-
- while (remain > 0) {
+ while (remain) {
/* Operation in this page
*
* page_base = page offset within aperture
* page_offset = offset within page
* page_length = bytes to copy for this page
*/
- page_base = offset & PAGE_MASK;
- page_offset = offset_in_page(offset);
- page_length = remain;
- if ((page_offset + remain) > PAGE_SIZE)
- page_length = PAGE_SIZE - page_offset;
-
+ u32 page_base = node.start;
+ unsigned page_offset = offset_in_page(offset);
+ unsigned page_length = PAGE_SIZE - page_offset;
+ page_length = remain < page_length ? remain : page_length;
+ if (node.allocated) {
+ wmb(); /* flush the write before we modify the GGTT */
+ ggtt->base.insert_page(&ggtt->base,
+ i915_gem_object_get_dma_address(obj, offset >> PAGE_SHIFT),
+ node.start, I915_CACHE_NONE, 0);
+ wmb(); /* flush modifications to the GGTT (insert_page) */
+ } else {
+ page_base += offset & PAGE_MASK;
+ }
/* If we get a fault while copying data, then (presumably) our
* source page isn't available. Return the error and we'll
* retry in the slow path.
+ * If the object is non-shmem backed, we retry again with the
+ * path that handles page fault.
*/
if (fast_user_write(ggtt->mappable, page_base,
page_offset, user_data, page_length)) {
- ret = -EFAULT;
- goto out_flush;
+ hit_slow_path = true;
+ mutex_unlock(&dev->struct_mutex);
+ if (slow_user_access(ggtt->mappable,
+ page_base,
+ page_offset, user_data,
+ page_length, true)) {
+ ret = -EFAULT;
+ mutex_lock(&dev->struct_mutex);
+ goto out_flush;
+ }
+
+ mutex_lock(&dev->struct_mutex);
}
remain -= page_length;
}
out_flush:
+ if (hit_slow_path) {
+ if (ret == 0 &&
+ (obj->base.read_domains & I915_GEM_DOMAIN_GTT) == 0) {
+ /* The user has modified the object whilst we tried
+ * reading from it, and we now have no idea what domain
+ * the pages should be in. As we have just been touching
+ * them directly, flush everything back to the GTT
+ * domain.
+ */
+ ret = i915_gem_object_set_to_gtt_domain(obj, false);
+ }
+ }
+
intel_fb_obj_flush(obj, false, ORIGIN_GTT);
out_unpin:
- i915_gem_object_ggtt_unpin(obj);
+ if (node.allocated) {
+ wmb();
+ ggtt->base.clear_range(&ggtt->base,
+ node.start, node.size,
+ true);
+ i915_gem_object_unpin_pages(obj);
+ remove_mappable_node(&node);
+ } else {
+ i915_gem_object_ggtt_unpin(obj);
+ }
out:
return ret;
}
obj_do_bit17_swizzling = i915_gem_object_needs_bit17_swizzle(obj);
+ ret = i915_gem_object_wait_rendering(obj, false);
+ if (ret)
+ return ret;
+
if (obj->base.write_domain != I915_GEM_DOMAIN_CPU) {
/* If we're not in the cpu write domain, set ourself into the gtt
* write domain and manually flush cachelines (if required). This
* optimizes for the case when the gpu will use the data
* right away and we therefore have to clflush anyway. */
needs_clflush_after = cpu_write_needs_clflush(obj);
- ret = i915_gem_object_wait_rendering(obj, false);
- if (ret)
- return ret;
}
/* Same trick applies to invalidate partially written cachelines read
* before writing. */
}
if (needs_clflush_after)
- i915_gem_chipset_flush(dev);
+ i915_gem_chipset_flush(to_i915(dev));
else
obj->cache_dirty = true;
/**
* Writes data to the object referenced by handle.
+ * @dev: drm device
+ * @data: ioctl data blob
+ * @file: drm file
*
* On error, the contents of the buffer that were to be modified are undefined.
*/
i915_gem_pwrite_ioctl(struct drm_device *dev, void *data,
struct drm_file *file)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
struct drm_i915_gem_pwrite *args = data;
struct drm_i915_gem_object *obj;
int ret;
if (ret)
goto put_rpm;
- obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
- if (&obj->base == NULL) {
+ obj = i915_gem_object_lookup(file, args->handle);
+ if (!obj) {
ret = -ENOENT;
goto unlock;
}
goto out;
}
- /* prime objects have no backing filp to GEM pread/pwrite
- * pages from.
- */
- if (!obj->base.filp) {
- ret = -EINVAL;
- goto out;
- }
-
trace_i915_gem_object_pwrite(obj, args->offset, args->size);
ret = -EFAULT;
* pread/pwrite currently are reading and writing from the CPU
* perspective, requiring manual detiling by the client.
*/
- if (obj->tiling_mode == I915_TILING_NONE &&
- obj->base.write_domain != I915_GEM_DOMAIN_CPU &&
+ if (!i915_gem_object_has_struct_page(obj) ||
cpu_write_needs_clflush(obj)) {
- ret = i915_gem_gtt_pwrite_fast(dev, obj, args, file);
+ ret = i915_gem_gtt_pwrite_fast(dev_priv, obj, args, file);
/* Note that the gtt paths might fail with non-page-backed user
* pointers (e.g. gtt mappings when moving data between
* textures). Fallback to the shmem path in that case. */
if (ret == -EFAULT || ret == -ENOSPC) {
if (obj->phys_handle)
ret = i915_gem_phys_pwrite(obj, args, file);
- else
+ else if (i915_gem_object_has_struct_page(obj))
ret = i915_gem_shmem_pwrite(dev, obj, args, file);
+ else
+ ret = -ENODEV;
}
out:
- drm_gem_object_unreference(&obj->base);
+ i915_gem_object_put(obj);
unlock:
mutex_unlock(&dev->struct_mutex);
put_rpm:
return ret;
}
-static int
-i915_gem_check_wedge(unsigned reset_counter, bool interruptible)
+static enum fb_op_origin
+write_origin(struct drm_i915_gem_object *obj, unsigned domain)
{
- if (__i915_terminally_wedged(reset_counter))
- return -EIO;
-
- if (__i915_reset_in_progress(reset_counter)) {
- /* Non-interruptible callers can't handle -EAGAIN, hence return
- * -EIO unconditionally for these. */
- if (!interruptible)
- return -EIO;
-
- return -EAGAIN;
- }
-
- return 0;
+ return domain == I915_GEM_DOMAIN_GTT && !obj->has_wc_mmap ?
+ ORIGIN_GTT : ORIGIN_CPU;
}
-static void fake_irq(unsigned long data)
+/**
+ * Called when user space prepares to use an object with the CPU, either
+ * through the mmap ioctl's mapping or a GTT mapping.
+ * @dev: drm device
+ * @data: ioctl data blob
+ * @file: drm file
+ */
+int
+i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
+ struct drm_file *file)
{
- wake_up_process((struct task_struct *)data);
-}
+ struct drm_i915_gem_set_domain *args = data;
+ struct drm_i915_gem_object *obj;
+ uint32_t read_domains = args->read_domains;
+ uint32_t write_domain = args->write_domain;
+ int ret;
-static bool missed_irq(struct drm_i915_private *dev_priv,
- struct intel_engine_cs *engine)
-{
- return test_bit(engine->id, &dev_priv->gpu_error.missed_irq_rings);
-}
-
-static unsigned long local_clock_us(unsigned *cpu)
-{
- unsigned long t;
-
- /* Cheaply and approximately convert from nanoseconds to microseconds.
- * The result and subsequent calculations are also defined in the same
- * approximate microseconds units. The principal source of timing
- * error here is from the simple truncation.
- *
- * Note that local_clock() is only defined wrt to the current CPU;
- * the comparisons are no longer valid if we switch CPUs. Instead of
- * blocking preemption for the entire busywait, we can detect the CPU
- * switch and use that as indicator of system load and a reason to
- * stop busywaiting, see busywait_stop().
- */
- *cpu = get_cpu();
- t = local_clock() >> 10;
- put_cpu();
-
- return t;
-}
-
-static bool busywait_stop(unsigned long timeout, unsigned cpu)
-{
- unsigned this_cpu;
-
- if (time_after(local_clock_us(&this_cpu), timeout))
- return true;
-
- return this_cpu != cpu;
-}
-
-static int __i915_spin_request(struct drm_i915_gem_request *req, int state)
-{
- unsigned long timeout;
- unsigned cpu;
-
- /* When waiting for high frequency requests, e.g. during synchronous
- * rendering split between the CPU and GPU, the finite amount of time
- * required to set up the irq and wait upon it limits the response
- * rate. By busywaiting on the request completion for a short while we
- * can service the high frequency waits as quick as possible. However,
- * if it is a slow request, we want to sleep as quickly as possible.
- * The tradeoff between waiting and sleeping is roughly the time it
- * takes to sleep on a request, on the order of a microsecond.
- */
-
- if (req->engine->irq_refcount)
- return -EBUSY;
-
- /* Only spin if we know the GPU is processing this request */
- if (!i915_gem_request_started(req, true))
- return -EAGAIN;
-
- timeout = local_clock_us(&cpu) + 5;
- while (!need_resched()) {
- if (i915_gem_request_completed(req, true))
- return 0;
-
- if (signal_pending_state(state, current))
- break;
-
- if (busywait_stop(timeout, cpu))
- break;
-
- cpu_relax_lowlatency();
- }
-
- if (i915_gem_request_completed(req, false))
- return 0;
-
- return -EAGAIN;
-}
-
-/**
- * __i915_wait_request - wait until execution of request has finished
- * @req: duh!
- * @interruptible: do an interruptible wait (normally yes)
- * @timeout: in - how long to wait (NULL forever); out - how much time remaining
- *
- * Note: It is of utmost importance that the passed in seqno and reset_counter
- * values have been read by the caller in an smp safe manner. Where read-side
- * locks are involved, it is sufficient to read the reset_counter before
- * unlocking the lock that protects the seqno. For lockless tricks, the
- * reset_counter _must_ be read before, and an appropriate smp_rmb must be
- * inserted.
- *
- * Returns 0 if the request was found within the alloted time. Else returns the
- * errno with remaining time filled in timeout argument.
- */
-int __i915_wait_request(struct drm_i915_gem_request *req,
- bool interruptible,
- s64 *timeout,
- struct intel_rps_client *rps)
-{
- struct intel_engine_cs *engine = i915_gem_request_get_engine(req);
- struct drm_device *dev = engine->dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
- const bool irq_test_in_progress =
- ACCESS_ONCE(dev_priv->gpu_error.test_irq_rings) & intel_engine_flag(engine);
- int state = interruptible ? TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE;
- DEFINE_WAIT(wait);
- unsigned long timeout_expire;
- s64 before = 0; /* Only to silence a compiler warning. */
- int ret;
-
- WARN(!intel_irqs_enabled(dev_priv), "IRQs disabled");
-
- if (list_empty(&req->list))
- return 0;
-
- if (i915_gem_request_completed(req, true))
- return 0;
-
- timeout_expire = 0;
- if (timeout) {
- if (WARN_ON(*timeout < 0))
- return -EINVAL;
-
- if (*timeout == 0)
- return -ETIME;
-
- timeout_expire = jiffies + nsecs_to_jiffies_timeout(*timeout);
-
- /*
- * Record current time in case interrupted by signal, or wedged.
- */
- before = ktime_get_raw_ns();
- }
-
- if (INTEL_INFO(dev_priv)->gen >= 6)
- gen6_rps_boost(dev_priv, rps, req->emitted_jiffies);
-
- trace_i915_gem_request_wait_begin(req);
-
- /* Optimistic spin for the next jiffie before touching IRQs */
- ret = __i915_spin_request(req, state);
- if (ret == 0)
- goto out;
-
- if (!irq_test_in_progress && WARN_ON(!engine->irq_get(engine))) {
- ret = -ENODEV;
- goto out;
- }
-
- for (;;) {
- struct timer_list timer;
-
- prepare_to_wait(&engine->irq_queue, &wait, state);
-
- /* We need to check whether any gpu reset happened in between
- * the request being submitted and now. If a reset has occurred,
- * the request is effectively complete (we either are in the
- * process of or have discarded the rendering and completely
- * reset the GPU. The results of the request are lost and we
- * are free to continue on with the original operation.
- */
- if (req->reset_counter != i915_reset_counter(&dev_priv->gpu_error)) {
- ret = 0;
- break;
- }
-
- if (i915_gem_request_completed(req, false)) {
- ret = 0;
- break;
- }
-
- if (signal_pending_state(state, current)) {
- ret = -ERESTARTSYS;
- break;
- }
-
- if (timeout && time_after_eq(jiffies, timeout_expire)) {
- ret = -ETIME;
- break;
- }
-
- timer.function = NULL;
- if (timeout || missed_irq(dev_priv, engine)) {
- unsigned long expire;
-
- setup_timer_on_stack(&timer, fake_irq, (unsigned long)current);
- expire = missed_irq(dev_priv, engine) ? jiffies + 1 : timeout_expire;
- mod_timer(&timer, expire);
- }
-
- io_schedule();
-
- if (timer.function) {
- del_singleshot_timer_sync(&timer);
- destroy_timer_on_stack(&timer);
- }
- }
- if (!irq_test_in_progress)
- engine->irq_put(engine);
-
- finish_wait(&engine->irq_queue, &wait);
-
-out:
- trace_i915_gem_request_wait_end(req);
-
- if (timeout) {
- s64 tres = *timeout - (ktime_get_raw_ns() - before);
-
- *timeout = tres < 0 ? 0 : tres;
-
- /*
- * Apparently ktime isn't accurate enough and occasionally has a
- * bit of mismatch in the jiffies<->nsecs<->ktime loop. So patch
- * things up to make the test happy. We allow up to 1 jiffy.
- *
- * This is a regrssion from the timespec->ktime conversion.
- */
- if (ret == -ETIME && *timeout < jiffies_to_usecs(1)*1000)
- *timeout = 0;
- }
-
- return ret;
-}
-
-int i915_gem_request_add_to_client(struct drm_i915_gem_request *req,
- struct drm_file *file)
-{
- struct drm_i915_file_private *file_priv;
-
- WARN_ON(!req || !file || req->file_priv);
-
- if (!req || !file)
- return -EINVAL;
-
- if (req->file_priv)
- return -EINVAL;
-
- file_priv = file->driver_priv;
-
- spin_lock(&file_priv->mm.lock);
- req->file_priv = file_priv;
- list_add_tail(&req->client_list, &file_priv->mm.request_list);
- spin_unlock(&file_priv->mm.lock);
-
- req->pid = get_pid(task_pid(current));
-
- return 0;
-}
-
-static inline void
-i915_gem_request_remove_from_client(struct drm_i915_gem_request *request)
-{
- struct drm_i915_file_private *file_priv = request->file_priv;
-
- if (!file_priv)
- return;
-
- spin_lock(&file_priv->mm.lock);
- list_del(&request->client_list);
- request->file_priv = NULL;
- spin_unlock(&file_priv->mm.lock);
-
- put_pid(request->pid);
- request->pid = NULL;
-}
-
-static void i915_gem_request_retire(struct drm_i915_gem_request *request)
-{
- trace_i915_gem_request_retire(request);
-
- /* We know the GPU must have read the request to have
- * sent us the seqno + interrupt, so use the position
- * of tail of the request to update the last known position
- * of the GPU head.
- *
- * Note this requires that we are always called in request
- * completion order.
- */
- request->ringbuf->last_retired_head = request->postfix;
-
- list_del_init(&request->list);
- i915_gem_request_remove_from_client(request);
-
- i915_gem_request_unreference(request);
-}
-
-static void
-__i915_gem_request_retire__upto(struct drm_i915_gem_request *req)
-{
- struct intel_engine_cs *engine = req->engine;
- struct drm_i915_gem_request *tmp;
-
- lockdep_assert_held(&engine->dev->struct_mutex);
-
- if (list_empty(&req->list))
- return;
-
- do {
- tmp = list_first_entry(&engine->request_list,
- typeof(*tmp), list);
-
- i915_gem_request_retire(tmp);
- } while (tmp != req);
-
- WARN_ON(i915_verify_lists(engine->dev));
-}
-
-/**
- * Waits for a request to be signaled, and cleans up the
- * request and object lists appropriately for that event.
- */
-int
-i915_wait_request(struct drm_i915_gem_request *req)
-{
- struct drm_i915_private *dev_priv = req->i915;
- bool interruptible;
- int ret;
-
- interruptible = dev_priv->mm.interruptible;
-
- BUG_ON(!mutex_is_locked(&dev_priv->dev->struct_mutex));
-
- ret = __i915_wait_request(req, interruptible, NULL, NULL);
- if (ret)
- return ret;
-
- /* If the GPU hung, we want to keep the requests to find the guilty. */
- if (req->reset_counter == i915_reset_counter(&dev_priv->gpu_error))
- __i915_gem_request_retire__upto(req);
-
- return 0;
-}
-
-/**
- * Ensures that all rendering to the object has completed and the object is
- * safe to unbind from the GTT or access from the CPU.
- */
-int
-i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
- bool readonly)
-{
- int ret, i;
-
- if (!obj->active)
- return 0;
-
- if (readonly) {
- if (obj->last_write_req != NULL) {
- ret = i915_wait_request(obj->last_write_req);
- if (ret)
- return ret;
-
- i = obj->last_write_req->engine->id;
- if (obj->last_read_req[i] == obj->last_write_req)
- i915_gem_object_retire__read(obj, i);
- else
- i915_gem_object_retire__write(obj);
- }
- } else {
- for (i = 0; i < I915_NUM_ENGINES; i++) {
- if (obj->last_read_req[i] == NULL)
- continue;
-
- ret = i915_wait_request(obj->last_read_req[i]);
- if (ret)
- return ret;
-
- i915_gem_object_retire__read(obj, i);
- }
- GEM_BUG_ON(obj->active);
- }
-
- return 0;
-}
-
-static void
-i915_gem_object_retire_request(struct drm_i915_gem_object *obj,
- struct drm_i915_gem_request *req)
-{
- int ring = req->engine->id;
-
- if (obj->last_read_req[ring] == req)
- i915_gem_object_retire__read(obj, ring);
- else if (obj->last_write_req == req)
- i915_gem_object_retire__write(obj);
-
- if (req->reset_counter == i915_reset_counter(&req->i915->gpu_error))
- __i915_gem_request_retire__upto(req);
-}
-
-/* A nonblocking variant of the above wait. This is a highly dangerous routine
- * as the object state may change during this call.
- */
-static __must_check int
-i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
- struct intel_rps_client *rps,
- bool readonly)
-{
- struct drm_device *dev = obj->base.dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
- struct drm_i915_gem_request *requests[I915_NUM_ENGINES];
- int ret, i, n = 0;
-
- BUG_ON(!mutex_is_locked(&dev->struct_mutex));
- BUG_ON(!dev_priv->mm.interruptible);
-
- if (!obj->active)
- return 0;
-
- if (readonly) {
- struct drm_i915_gem_request *req;
-
- req = obj->last_write_req;
- if (req == NULL)
- return 0;
-
- requests[n++] = i915_gem_request_reference(req);
- } else {
- for (i = 0; i < I915_NUM_ENGINES; i++) {
- struct drm_i915_gem_request *req;
-
- req = obj->last_read_req[i];
- if (req == NULL)
- continue;
-
- requests[n++] = i915_gem_request_reference(req);
- }
- }
-
- mutex_unlock(&dev->struct_mutex);
- ret = 0;
- for (i = 0; ret == 0 && i < n; i++)
- ret = __i915_wait_request(requests[i], true, NULL, rps);
- mutex_lock(&dev->struct_mutex);
-
- for (i = 0; i < n; i++) {
- if (ret == 0)
- i915_gem_object_retire_request(obj, requests[i]);
- i915_gem_request_unreference(requests[i]);
- }
-
- return ret;
-}
-
-static struct intel_rps_client *to_rps_client(struct drm_file *file)
-{
- struct drm_i915_file_private *fpriv = file->driver_priv;
- return &fpriv->rps;
-}
-
-/**
- * Called when user space prepares to use an object with the CPU, either
- * through the mmap ioctl's mapping or a GTT mapping.
- */
-int
-i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
- struct drm_file *file)
-{
- struct drm_i915_gem_set_domain *args = data;
- struct drm_i915_gem_object *obj;
- uint32_t read_domains = args->read_domains;
- uint32_t write_domain = args->write_domain;
- int ret;
-
- /* Only handle setting domains to types used by the CPU. */
- if (write_domain & I915_GEM_GPU_DOMAINS)
- return -EINVAL;
-
- if (read_domains & I915_GEM_GPU_DOMAINS)
- return -EINVAL;
+ /* Only handle setting domains to types used by the CPU. */
+ if ((write_domain | read_domains) & I915_GEM_GPU_DOMAINS)
+ return -EINVAL;
/* Having something in the write domain implies it's in the read
* domain, and only that read domain. Enforce that in the request.
if (write_domain != 0 && read_domains != write_domain)
return -EINVAL;
- ret = i915_mutex_lock_interruptible(dev);
- if (ret)
- return ret;
-
- obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
- if (&obj->base == NULL) {
- ret = -ENOENT;
- goto unlock;
- }
+ obj = i915_gem_object_lookup(file, args->handle);
+ if (!obj)
+ return -ENOENT;
/* Try to flush the object off the GPU without holding the lock.
* We will repeat the flush holding the lock in the normal manner
* to catch cases where we are gazumped.
*/
- ret = i915_gem_object_wait_rendering__nonblocking(obj,
- to_rps_client(file),
- !write_domain);
+ ret = __unsafe_wait_rendering(obj, to_rps_client(file), !write_domain);
+ if (ret)
+ goto err;
+
+ ret = i915_mutex_lock_interruptible(dev);
if (ret)
- goto unref;
+ goto err;
if (read_domains & I915_GEM_DOMAIN_GTT)
ret = i915_gem_object_set_to_gtt_domain(obj, write_domain != 0);
ret = i915_gem_object_set_to_cpu_domain(obj, write_domain != 0);
if (write_domain != 0)
- intel_fb_obj_invalidate(obj,
- write_domain == I915_GEM_DOMAIN_GTT ?
- ORIGIN_GTT : ORIGIN_CPU);
+ intel_fb_obj_invalidate(obj, write_origin(obj, write_domain));
-unref:
- drm_gem_object_unreference(&obj->base);
-unlock:
+ i915_gem_object_put(obj);
mutex_unlock(&dev->struct_mutex);
return ret;
+
+err:
+ i915_gem_object_put_unlocked(obj);
+ return ret;
}
/**
* Called when user space has done writes to this buffer
+ * @dev: drm device
+ * @data: ioctl data blob
+ * @file: drm file
*/
int
i915_gem_sw_finish_ioctl(struct drm_device *dev, void *data,
if (ret)
return ret;
- obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
- if (&obj->base == NULL) {
+ obj = i915_gem_object_lookup(file, args->handle);
+ if (!obj) {
ret = -ENOENT;
goto unlock;
}
if (obj->pin_display)
i915_gem_object_flush_cpu_write_domain(obj);
- drm_gem_object_unreference(&obj->base);
+ i915_gem_object_put(obj);
unlock:
mutex_unlock(&dev->struct_mutex);
return ret;
}
/**
- * Maps the contents of an object, returning the address it is mapped
- * into.
+ * i915_gem_mmap_ioctl - Maps the contents of an object, returning the address
+ * it is mapped to.
+ * @dev: drm device
+ * @data: ioctl data blob
+ * @file: drm file
*
* While the mapping holds a reference on the contents of the object, it doesn't
* imply a ref on the object itself.
struct drm_file *file)
{
struct drm_i915_gem_mmap *args = data;
- struct drm_gem_object *obj;
+ struct drm_i915_gem_object *obj;
unsigned long addr;
if (args->flags & ~(I915_MMAP_WC))
if (args->flags & I915_MMAP_WC && !boot_cpu_has(X86_FEATURE_PAT))
return -ENODEV;
- obj = drm_gem_object_lookup(file, args->handle);
- if (obj == NULL)
+ obj = i915_gem_object_lookup(file, args->handle);
+ if (!obj)
return -ENOENT;
/* prime objects have no backing filp to GEM mmap
* pages from.
*/
- if (!obj->filp) {
- drm_gem_object_unreference_unlocked(obj);
+ if (!obj->base.filp) {
+ i915_gem_object_put_unlocked(obj);
return -EINVAL;
}
- addr = vm_mmap(obj->filp, 0, args->size,
+ addr = vm_mmap(obj->base.filp, 0, args->size,
PROT_READ | PROT_WRITE, MAP_SHARED,
args->offset);
if (args->flags & I915_MMAP_WC) {
struct vm_area_struct *vma;
if (down_write_killable(&mm->mmap_sem)) {
- drm_gem_object_unreference_unlocked(obj);
+ i915_gem_object_put_unlocked(obj);
return -EINTR;
}
vma = find_vma(mm, addr);
else
addr = -ENOMEM;
up_write(&mm->mmap_sem);
+
+ /* This may race, but that's ok, it only gets set */
+ WRITE_ONCE(obj->has_wc_mmap, true);
}
- drm_gem_object_unreference_unlocked(obj);
+ i915_gem_object_put_unlocked(obj);
if (IS_ERR((void *)addr))
return addr;
struct drm_i915_private *dev_priv = to_i915(dev);
struct i915_ggtt *ggtt = &dev_priv->ggtt;
struct i915_ggtt_view view = i915_ggtt_view_normal;
+ bool write = !!(vmf->flags & FAULT_FLAG_WRITE);
pgoff_t page_offset;
unsigned long pfn;
- int ret = 0;
- bool write = !!(vmf->flags & FAULT_FLAG_WRITE);
-
- intel_runtime_pm_get(dev_priv);
+ int ret;
/* We don't use vmf->pgoff since that has the fake offset */
page_offset = ((unsigned long)vmf->virtual_address - vma->vm_start) >>
PAGE_SHIFT;
- ret = i915_mutex_lock_interruptible(dev);
- if (ret)
- goto out;
-
trace_i915_gem_object_fault(obj, page_offset, true, write);
/* Try to flush the object off the GPU first without holding the lock.
- * Upon reacquiring the lock, we will perform our sanity checks and then
+ * Upon acquiring the lock, we will perform our sanity checks and then
* repeat the flush holding the lock in the normal manner to catch cases
* where we are gazumped.
*/
- ret = i915_gem_object_wait_rendering__nonblocking(obj, NULL, !write);
+ ret = __unsafe_wait_rendering(obj, NULL, !write);
if (ret)
- goto unlock;
+ goto err;
+
+ intel_runtime_pm_get(dev_priv);
+
+ ret = i915_mutex_lock_interruptible(dev);
+ if (ret)
+ goto err_rpm;
/* Access to snoopable pages through the GTT is incoherent. */
if (obj->cache_level != I915_CACHE_NONE && !HAS_LLC(dev)) {
ret = -EFAULT;
- goto unlock;
+ goto err_unlock;
}
/* Use a partial view if the object is bigger than the aperture. */
}
/* Now pin it into the GTT if needed */
- ret = i915_gem_object_ggtt_pin(obj, &view, 0, PIN_MAPPABLE);
+ ret = i915_gem_object_ggtt_pin(obj, &view, 0, 0, PIN_MAPPABLE);
if (ret)
- goto unlock;
+ goto err_unlock;
ret = i915_gem_object_set_to_gtt_domain(obj, write);
if (ret)
- goto unpin;
+ goto err_unpin;
ret = i915_gem_object_get_fence(obj);
if (ret)
- goto unpin;
+ goto err_unpin;
/* Finally, remap it using the new GTT offset */
pfn = ggtt->mappable_base +
(unsigned long)vmf->virtual_address,
pfn + page_offset);
}
-unpin:
+err_unpin:
i915_gem_object_ggtt_unpin_view(obj, &view);
-unlock:
+err_unlock:
mutex_unlock(&dev->struct_mutex);
-out:
+err_rpm:
+ intel_runtime_pm_put(dev_priv);
+err:
switch (ret) {
case -EIO:
/*
ret = VM_FAULT_SIGBUS;
break;
}
-
- intel_runtime_pm_put(dev_priv);
return ret;
}
i915_gem_release_mmap(obj);
}
-uint32_t
-i915_gem_get_gtt_size(struct drm_device *dev, uint32_t size, int tiling_mode)
+/**
+ * i915_gem_get_ggtt_size - return required global GTT size for an object
+ * @dev_priv: i915 device
+ * @size: object size
+ * @tiling_mode: tiling mode
+ *
+ * Return the required global GTT size for an object, taking into account
+ * potential fence register mapping.
+ */
+u64 i915_gem_get_ggtt_size(struct drm_i915_private *dev_priv,
+ u64 size, int tiling_mode)
{
- uint32_t gtt_size;
+ u64 ggtt_size;
+
+ GEM_BUG_ON(size == 0);
- if (INTEL_INFO(dev)->gen >= 4 ||
+ if (INTEL_GEN(dev_priv) >= 4 ||
tiling_mode == I915_TILING_NONE)
return size;
/* Previous chips need a power-of-two fence region when tiling */
- if (INTEL_INFO(dev)->gen == 3)
- gtt_size = 1024*1024;
+ if (IS_GEN3(dev_priv))
+ ggtt_size = 1024*1024;
else
- gtt_size = 512*1024;
+ ggtt_size = 512*1024;
- while (gtt_size < size)
- gtt_size <<= 1;
+ while (ggtt_size < size)
+ ggtt_size <<= 1;
- return gtt_size;
+ return ggtt_size;
}
/**
- * i915_gem_get_gtt_alignment - return required GTT alignment for an object
- * @obj: object to check
+ * i915_gem_get_ggtt_alignment - return required global GTT alignment
+ * @dev_priv: i915 device
+ * @size: object size
+ * @tiling_mode: tiling mode
+ * @fenced: is fenced alignment required or not
*
- * Return the required GTT alignment for an object, taking into account
+ * Return the required global GTT alignment for an object, taking into account
* potential fence register mapping.
*/
-uint32_t
-i915_gem_get_gtt_alignment(struct drm_device *dev, uint32_t size,
- int tiling_mode, bool fenced)
+u64 i915_gem_get_ggtt_alignment(struct drm_i915_private *dev_priv, u64 size,
+ int tiling_mode, bool fenced)
{
+ GEM_BUG_ON(size == 0);
+
/*
* Minimum alignment is 4k (GTT page size), but might be greater
* if a fence register is needed for the object.
*/
- if (INTEL_INFO(dev)->gen >= 4 || (!fenced && IS_G33(dev)) ||
+ if (INTEL_GEN(dev_priv) >= 4 || (!fenced && IS_G33(dev_priv)) ||
tiling_mode == I915_TILING_NONE)
return 4096;
* Previous chips need to be aligned to the size of the smallest
* fence register that can contain the object.
*/
- return i915_gem_get_gtt_size(dev, size, tiling_mode);
+ return i915_gem_get_ggtt_size(dev_priv, size, tiling_mode);
}
static int i915_gem_object_create_mmap_offset(struct drm_i915_gem_object *obj)
{
- struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
int ret;
dev_priv->mm.shrinker_no_lock_stealing = true;
if (ret)
return ret;
- obj = to_intel_bo(drm_gem_object_lookup(file, handle));
- if (&obj->base == NULL) {
+ obj = i915_gem_object_lookup(file, handle);
+ if (!obj) {
ret = -ENOENT;
goto unlock;
}
*offset = drm_vma_node_offset_addr(&obj->base.vma_node);
out:
- drm_gem_object_unreference(&obj->base);
+ i915_gem_object_put(obj);
unlock:
mutex_unlock(&dev->struct_mutex);
return ret;
static void
i915_gem_object_put_pages_gtt(struct drm_i915_gem_object *obj)
{
- struct sg_page_iter sg_iter;
+ struct sgt_iter sgt_iter;
+ struct page *page;
int ret;
BUG_ON(obj->madv == __I915_MADV_PURGED);
if (obj->madv == I915_MADV_DONTNEED)
obj->dirty = 0;
- for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0) {
- struct page *page = sg_page_iter_page(&sg_iter);
-
+ for_each_sgt_page(page, sgt_iter, obj->pages) {
if (obj->dirty)
set_page_dirty(page);
if (obj->pages_pin_count)
return -EBUSY;
- BUG_ON(i915_gem_obj_bound_any(obj));
+ GEM_BUG_ON(obj->bind_count);
/* ->put_pages might need to allocate memory for the bit17 swizzle
* array, hence protect them from being reaped by removing them from gtt
static int
i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj)
{
- struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
int page_count, i;
struct address_space *mapping;
struct sg_table *st;
struct scatterlist *sg;
- struct sg_page_iter sg_iter;
+ struct sgt_iter sgt_iter;
struct page *page;
unsigned long last_pfn = 0; /* suppress gcc warning */
int ret;
err_pages:
sg_mark_end(sg);
- for_each_sg_page(st->sgl, &sg_iter, st->nents, 0)
- put_page(sg_page_iter_page(&sg_iter));
+ for_each_sgt_page(page, sgt_iter, st)
+ put_page(page);
sg_free_table(st);
kfree(st);
int
i915_gem_object_get_pages(struct drm_i915_gem_object *obj)
{
- struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
const struct drm_i915_gem_object_ops *ops = obj->ops;
int ret;
return 0;
}
-void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj)
+/* The 'mapping' part of i915_gem_object_pin_map() below */
+static void *i915_gem_object_map(const struct drm_i915_gem_object *obj)
{
- int ret;
-
- lockdep_assert_held(&obj->base.dev->struct_mutex);
-
- ret = i915_gem_object_get_pages(obj);
- if (ret)
- return ERR_PTR(ret);
-
- i915_gem_object_pin_pages(obj);
-
- if (obj->mapping == NULL) {
- struct page **pages;
-
- pages = NULL;
- if (obj->base.size == PAGE_SIZE)
- obj->mapping = kmap(sg_page(obj->pages->sgl));
- else
- pages = drm_malloc_gfp(obj->base.size >> PAGE_SHIFT,
- sizeof(*pages),
- GFP_TEMPORARY);
- if (pages != NULL) {
- struct sg_page_iter sg_iter;
- int n;
-
- n = 0;
- for_each_sg_page(obj->pages->sgl, &sg_iter,
- obj->pages->nents, 0)
- pages[n++] = sg_page_iter_page(&sg_iter);
-
- obj->mapping = vmap(pages, n, 0, PAGE_KERNEL);
- drm_free_large(pages);
- }
- if (obj->mapping == NULL) {
- i915_gem_object_unpin_pages(obj);
- return ERR_PTR(-ENOMEM);
- }
- }
-
- return obj->mapping;
-}
-
-void i915_vma_move_to_active(struct i915_vma *vma,
- struct drm_i915_gem_request *req)
-{
- struct drm_i915_gem_object *obj = vma->obj;
- struct intel_engine_cs *engine;
-
- engine = i915_gem_request_get_engine(req);
-
- /* Add a reference if we're newly entering the active list. */
- if (obj->active == 0)
- drm_gem_object_reference(&obj->base);
- obj->active |= intel_engine_flag(engine);
-
- list_move_tail(&obj->engine_list[engine->id], &engine->active_list);
- i915_gem_request_assign(&obj->last_read_req[engine->id], req);
-
- list_move_tail(&vma->vm_link, &vma->vm->active_list);
-}
-
-static void
-i915_gem_object_retire__write(struct drm_i915_gem_object *obj)
-{
- GEM_BUG_ON(obj->last_write_req == NULL);
- GEM_BUG_ON(!(obj->active & intel_engine_flag(obj->last_write_req->engine)));
-
- i915_gem_request_assign(&obj->last_write_req, NULL);
- intel_fb_obj_flush(obj, true, ORIGIN_CS);
-}
-
-static void
-i915_gem_object_retire__read(struct drm_i915_gem_object *obj, int ring)
-{
- struct i915_vma *vma;
-
- GEM_BUG_ON(obj->last_read_req[ring] == NULL);
- GEM_BUG_ON(!(obj->active & (1 << ring)));
-
- list_del_init(&obj->engine_list[ring]);
- i915_gem_request_assign(&obj->last_read_req[ring], NULL);
-
- if (obj->last_write_req && obj->last_write_req->engine->id == ring)
- i915_gem_object_retire__write(obj);
+ unsigned long n_pages = obj->base.size >> PAGE_SHIFT;
+ struct sg_table *sgt = obj->pages;
+ struct sgt_iter sgt_iter;
+ struct page *page;
+ struct page *stack_pages[32];
+ struct page **pages = stack_pages;
+ unsigned long i = 0;
+ void *addr;
- obj->active &= ~(1 << ring);
- if (obj->active)
- return;
+ /* A single page can always be kmapped */
+ if (n_pages == 1)
+ return kmap(sg_page(sgt->sgl));
- /* Bump our place on the bound list to keep it roughly in LRU order
- * so that we don't steal from recently used but inactive objects
- * (unless we are forced to ofc!)
- */
- list_move_tail(&obj->global_list,
- &to_i915(obj->base.dev)->mm.bound_list);
-
- list_for_each_entry(vma, &obj->vma_list, obj_link) {
- if (!list_empty(&vma->vm_link))
- list_move_tail(&vma->vm_link, &vma->vm->inactive_list);
+ if (n_pages > ARRAY_SIZE(stack_pages)) {
+ /* Too big for stack -- allocate temporary array instead */
+ pages = drm_malloc_gfp(n_pages, sizeof(*pages), GFP_TEMPORARY);
+ if (!pages)
+ return NULL;
}
- i915_gem_request_assign(&obj->last_fenced_req, NULL);
- drm_gem_object_unreference(&obj->base);
-}
+ for_each_sgt_page(page, sgt_iter, sgt)
+ pages[i++] = page;
-static int
-i915_gem_init_seqno(struct drm_device *dev, u32 seqno)
-{
- struct drm_i915_private *dev_priv = dev->dev_private;
- struct intel_engine_cs *engine;
- int ret;
+ /* Check that we have the expected number of pages */
+ GEM_BUG_ON(i != n_pages);
- /* Carefully retire all requests without writing to the rings */
- for_each_engine(engine, dev_priv) {
- ret = intel_engine_idle(engine);
- if (ret)
- return ret;
- }
- i915_gem_retire_requests(dev);
+ addr = vmap(pages, n_pages, 0, PAGE_KERNEL);
- /* Finally reset hw state */
- for_each_engine(engine, dev_priv)
- intel_ring_init_seqno(engine, seqno);
+ if (pages != stack_pages)
+ drm_free_large(pages);
- return 0;
+ return addr;
}
-int i915_gem_set_seqno(struct drm_device *dev, u32 seqno)
+/* get, pin, and map the pages of the object into kernel space */
+void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
int ret;
- if (seqno == 0)
- return -EINVAL;
+ lockdep_assert_held(&obj->base.dev->struct_mutex);
- /* HWS page needs to be set less than what we
- * will inject to ring
- */
- ret = i915_gem_init_seqno(dev, seqno - 1);
+ ret = i915_gem_object_get_pages(obj);
if (ret)
- return ret;
-
- /* Carefully set the last_seqno value so that wrap
- * detection still works
- */
- dev_priv->next_seqno = seqno;
- dev_priv->last_seqno = seqno - 1;
- if (dev_priv->last_seqno == 0)
- dev_priv->last_seqno--;
-
- return 0;
-}
-
-int
-i915_gem_get_seqno(struct drm_device *dev, u32 *seqno)
-{
- struct drm_i915_private *dev_priv = dev->dev_private;
-
- /* reserve 0 for non-seqno */
- if (dev_priv->next_seqno == 0) {
- int ret = i915_gem_init_seqno(dev, 0);
- if (ret)
- return ret;
-
- dev_priv->next_seqno = 1;
- }
-
- *seqno = dev_priv->last_seqno = dev_priv->next_seqno++;
- return 0;
-}
-
-/*
- * NB: This function is not allowed to fail. Doing so would mean the the
- * request is not being tracked for completion but the work itself is
- * going to happen on the hardware. This would be a Bad Thing(tm).
- */
-void __i915_add_request(struct drm_i915_gem_request *request,
- struct drm_i915_gem_object *obj,
- bool flush_caches)
-{
- struct intel_engine_cs *engine;
- struct drm_i915_private *dev_priv;
- struct intel_ringbuffer *ringbuf;
- u32 request_start;
- int ret;
-
- if (WARN_ON(request == NULL))
- return;
-
- engine = request->engine;
- dev_priv = request->i915;
- ringbuf = request->ringbuf;
-
- /*
- * To ensure that this call will not fail, space for its emissions
- * should already have been reserved in the ring buffer. Let the ring
- * know that it is time to use that space up.
- */
- intel_ring_reserved_space_use(ringbuf);
-
- request_start = intel_ring_get_tail(ringbuf);
- /*
- * Emit any outstanding flushes - execbuf can fail to emit the flush
- * after having emitted the batchbuffer command. Hence we need to fix
- * things up similar to emitting the lazy request. The difference here
- * is that the flush _must_ happen before the next request, no matter
- * what.
- */
- if (flush_caches) {
- if (i915.enable_execlists)
- ret = logical_ring_flush_all_caches(request);
- else
- ret = intel_ring_flush_all_caches(request);
- /* Not allowed to fail! */
- WARN(ret, "*_ring_flush_all_caches failed: %d!\n", ret);
- }
-
- trace_i915_gem_request_add(request);
-
- request->head = request_start;
-
- /* Whilst this request exists, batch_obj will be on the
- * active_list, and so will hold the active reference. Only when this
- * request is retired will the the batch_obj be moved onto the
- * inactive_list and lose its active reference. Hence we do not need
- * to explicitly hold another reference here.
- */
- request->batch_obj = obj;
-
- /* Seal the request and mark it as pending execution. Note that
- * we may inspect this state, without holding any locks, during
- * hangcheck. Hence we apply the barrier to ensure that we do not
- * see a more recent value in the hws than we are tracking.
- */
- request->emitted_jiffies = jiffies;
- request->previous_seqno = engine->last_submitted_seqno;
- smp_store_mb(engine->last_submitted_seqno, request->seqno);
- list_add_tail(&request->list, &engine->request_list);
-
- /* Record the position of the start of the request so that
- * should we detect the updated seqno part-way through the
- * GPU processing the request, we never over-estimate the
- * position of the head.
- */
- request->postfix = intel_ring_get_tail(ringbuf);
-
- if (i915.enable_execlists)
- ret = engine->emit_request(request);
- else {
- ret = engine->add_request(request);
-
- request->tail = intel_ring_get_tail(ringbuf);
- }
- /* Not allowed to fail! */
- WARN(ret, "emit|add_request failed: %d!\n", ret);
-
- i915_queue_hangcheck(engine->dev);
-
- queue_delayed_work(dev_priv->wq,
- &dev_priv->mm.retire_work,
- round_jiffies_up_relative(HZ));
- intel_mark_busy(dev_priv->dev);
-
- /* Sanity check that the reserved size was large enough. */
- intel_ring_reserved_space_end(ringbuf);
-}
-
-static bool i915_context_is_banned(struct drm_i915_private *dev_priv,
- const struct intel_context *ctx)
-{
- unsigned long elapsed;
-
- elapsed = get_seconds() - ctx->hang_stats.guilty_ts;
+ return ERR_PTR(ret);
- if (ctx->hang_stats.banned)
- return true;
+ i915_gem_object_pin_pages(obj);
- if (ctx->hang_stats.ban_period_seconds &&
- elapsed <= ctx->hang_stats.ban_period_seconds) {
- if (!i915_gem_context_is_default(ctx)) {
- DRM_DEBUG("context hanging too fast, banning!\n");
- return true;
- } else if (i915_stop_ring_allow_ban(dev_priv)) {
- if (i915_stop_ring_allow_warn(dev_priv))
- DRM_ERROR("gpu hanging too fast, banning!\n");
- return true;
+ if (!obj->mapping) {
+ obj->mapping = i915_gem_object_map(obj);
+ if (!obj->mapping) {
+ i915_gem_object_unpin_pages(obj);
+ return ERR_PTR(-ENOMEM);
}
}
- return false;
-}
-
-static void i915_set_reset_status(struct drm_i915_private *dev_priv,
- struct intel_context *ctx,
- const bool guilty)
-{
- struct i915_ctx_hang_stats *hs;
-
- if (WARN_ON(!ctx))
- return;
-
- hs = &ctx->hang_stats;
-
- if (guilty) {
- hs->banned = i915_context_is_banned(dev_priv, ctx);
- hs->batch_active++;
- hs->guilty_ts = get_seconds();
- } else {
- hs->batch_pending++;
- }
+ return obj->mapping;
}
-void i915_gem_request_free(struct kref *req_ref)
+static void
+i915_gem_object_retire__write(struct i915_gem_active *active,
+ struct drm_i915_gem_request *request)
{
- struct drm_i915_gem_request *req = container_of(req_ref,
- typeof(*req), ref);
- struct intel_context *ctx = req->ctx;
-
- if (req->file_priv)
- i915_gem_request_remove_from_client(req);
-
- if (ctx) {
- if (i915.enable_execlists && ctx != req->i915->kernel_context)
- intel_lr_context_unpin(ctx, req->engine);
+ struct drm_i915_gem_object *obj =
+ container_of(active, struct drm_i915_gem_object, last_write);
- i915_gem_context_unreference(ctx);
- }
-
- kmem_cache_free(req->i915->requests, req);
+ intel_fb_obj_flush(obj, true, ORIGIN_CS);
}
-static inline int
-__i915_gem_request_alloc(struct intel_engine_cs *engine,
- struct intel_context *ctx,
- struct drm_i915_gem_request **req_out)
+static void
+i915_gem_object_retire__read(struct i915_gem_active *active,
+ struct drm_i915_gem_request *request)
{
- struct drm_i915_private *dev_priv = to_i915(engine->dev);
- unsigned reset_counter = i915_reset_counter(&dev_priv->gpu_error);
- struct drm_i915_gem_request *req;
- int ret;
+ int idx = request->engine->id;
+ struct drm_i915_gem_object *obj =
+ container_of(active, struct drm_i915_gem_object, last_read[idx]);
- if (!req_out)
- return -EINVAL;
+ GEM_BUG_ON(!i915_gem_object_has_active_engine(obj, idx));
- *req_out = NULL;
+ i915_gem_object_clear_active(obj, idx);
+ if (i915_gem_object_is_active(obj))
+ return;
- /* ABI: Before userspace accesses the GPU (e.g. execbuffer), report
- * EIO if the GPU is already wedged, or EAGAIN to drop the struct_mutex
- * and restart.
+ /* Bump our place on the bound list to keep it roughly in LRU order
+ * so that we don't steal from recently used but inactive objects
+ * (unless we are forced to ofc!)
*/
- ret = i915_gem_check_wedge(reset_counter, dev_priv->mm.interruptible);
- if (ret)
- return ret;
-
- req = kmem_cache_zalloc(dev_priv->requests, GFP_KERNEL);
- if (req == NULL)
- return -ENOMEM;
+ if (obj->bind_count)
+ list_move_tail(&obj->global_list,
+ &request->i915->mm.bound_list);
- ret = i915_gem_get_seqno(engine->dev, &req->seqno);
- if (ret)
- goto err;
+ i915_gem_object_put(obj);
+}
- kref_init(&req->ref);
- req->i915 = dev_priv;
- req->engine = engine;
- req->reset_counter = reset_counter;
- req->ctx = ctx;
- i915_gem_context_reference(req->ctx);
+static bool i915_context_is_banned(const struct i915_gem_context *ctx)
+{
+ unsigned long elapsed;
- if (i915.enable_execlists)
- ret = intel_logical_ring_alloc_request_extras(req);
- else
- ret = intel_ring_alloc_request_extras(req);
- if (ret) {
- i915_gem_context_unreference(req->ctx);
- goto err;
- }
+ if (ctx->hang_stats.banned)
+ return true;
- /*
- * Reserve space in the ring buffer for all the commands required to
- * eventually emit this request. This is to guarantee that the
- * i915_add_request() call can't fail. Note that the reserve may need
- * to be redone if the request is not actually submitted straight
- * away, e.g. because a GPU scheduler has deferred it.
- */
- if (i915.enable_execlists)
- ret = intel_logical_ring_reserve_space(req);
- else
- ret = intel_ring_reserve_space(req);
- if (ret) {
- /*
- * At this point, the request is fully allocated even if not
- * fully prepared. Thus it can be cleaned up using the proper
- * free code.
- */
- intel_ring_reserved_space_cancel(req->ringbuf);
- i915_gem_request_unreference(req);
- return ret;
+ elapsed = get_seconds() - ctx->hang_stats.guilty_ts;
+ if (ctx->hang_stats.ban_period_seconds &&
+ elapsed <= ctx->hang_stats.ban_period_seconds) {
+ DRM_DEBUG("context hanging too fast, banning!\n");
+ return true;
}
- *req_out = req;
- return 0;
-
-err:
- kmem_cache_free(dev_priv->requests, req);
- return ret;
+ return false;
}
-/**
- * i915_gem_request_alloc - allocate a request structure
- *
- * @engine: engine that we wish to issue the request on.
- * @ctx: context that the request will be associated with.
- * This can be NULL if the request is not directly related to
- * any specific user context, in which case this function will
- * choose an appropriate context to use.
- *
- * Returns a pointer to the allocated request if successful,
- * or an error code if not.
- */
-struct drm_i915_gem_request *
-i915_gem_request_alloc(struct intel_engine_cs *engine,
- struct intel_context *ctx)
+static void i915_set_reset_status(struct i915_gem_context *ctx,
+ const bool guilty)
{
- struct drm_i915_gem_request *req;
- int err;
+ struct i915_ctx_hang_stats *hs = &ctx->hang_stats;
- if (ctx == NULL)
- ctx = to_i915(engine->dev)->kernel_context;
- err = __i915_gem_request_alloc(engine, ctx, &req);
- return err ? ERR_PTR(err) : req;
+ if (guilty) {
+ hs->banned = i915_context_is_banned(ctx);
+ hs->batch_active++;
+ hs->guilty_ts = get_seconds();
+ } else {
+ hs->batch_pending++;
+ }
}
struct drm_i915_gem_request *
{
struct drm_i915_gem_request *request;
- list_for_each_entry(request, &engine->request_list, list) {
- if (i915_gem_request_completed(request, false))
+ /* We are called by the error capture and reset at a random
+ * point in time. In particular, note that neither is crucially
+ * ordered with an interrupt. After a hang, the GPU is dead and we
+ * assume that no more writes can happen (we waited long enough for
+ * all writes that were in transaction to be flushed) - adding an
+ * extra delay for a recent interrupt is pointless. Hence, we do
+ * not need an engine->irq_seqno_barrier() before the seqno reads.
+ */
+ list_for_each_entry(request, &engine->request_list, link) {
+ if (i915_gem_request_completed(request))
continue;
return request;
return NULL;
}
-static void i915_gem_reset_engine_status(struct drm_i915_private *dev_priv,
- struct intel_engine_cs *engine)
+static void i915_gem_reset_engine_status(struct intel_engine_cs *engine)
{
struct drm_i915_gem_request *request;
bool ring_hung;
request = i915_gem_find_active_request(engine);
-
if (request == NULL)
return;
ring_hung = engine->hangcheck.score >= HANGCHECK_SCORE_RING_HUNG;
- i915_set_reset_status(dev_priv, request->ctx, ring_hung);
-
- list_for_each_entry_continue(request, &engine->request_list, list)
- i915_set_reset_status(dev_priv, request->ctx, false);
+ i915_set_reset_status(request->ctx, ring_hung);
+ list_for_each_entry_continue(request, &engine->request_list, link)
+ i915_set_reset_status(request->ctx, false);
}
-static void i915_gem_reset_engine_cleanup(struct drm_i915_private *dev_priv,
- struct intel_engine_cs *engine)
+static void i915_gem_reset_engine_cleanup(struct intel_engine_cs *engine)
{
- struct intel_ringbuffer *buffer;
+ struct intel_ring *ring;
- while (!list_empty(&engine->active_list)) {
- struct drm_i915_gem_object *obj;
-
- obj = list_first_entry(&engine->active_list,
- struct drm_i915_gem_object,
- engine_list[engine->id]);
-
- i915_gem_object_retire__read(obj, engine->id);
- }
+ /* Mark all pending requests as complete so that any concurrent
+ * (lockless) lookup doesn't try and wait upon the request as we
+ * reset it.
+ */
+ intel_engine_init_seqno(engine, engine->last_submitted_seqno);
/*
* Clear the execlists queue up before freeing the requests, as those
/* Ensure irq handler finishes or is cancelled. */
tasklet_kill(&engine->irq_tasklet);
- spin_lock_bh(&engine->execlist_lock);
- /* list_splice_tail_init checks for empty lists */
- list_splice_tail_init(&engine->execlist_queue,
- &engine->execlist_retired_req_list);
- spin_unlock_bh(&engine->execlist_lock);
-
- intel_execlists_retire_requests(engine);
+ intel_execlists_cancel_requests(engine);
}
/*
* implicit references on things like e.g. ppgtt address spaces through
* the request.
*/
- while (!list_empty(&engine->request_list)) {
+ if (!list_empty(&engine->request_list)) {
struct drm_i915_gem_request *request;
- request = list_first_entry(&engine->request_list,
- struct drm_i915_gem_request,
- list);
+ request = list_last_entry(&engine->request_list,
+ struct drm_i915_gem_request,
+ link);
- i915_gem_request_retire(request);
+ i915_gem_request_retire_upto(request);
}
/* Having flushed all requests from all queues, we know that all
* upon reset is less than when we start. Do one more pass over
* all the ringbuffers to reset last_retired_head.
*/
- list_for_each_entry(buffer, &engine->buffers, link) {
- buffer->last_retired_head = buffer->tail;
- intel_ring_update_space(buffer);
+ list_for_each_entry(ring, &engine->buffers, link) {
+ ring->last_retired_head = ring->tail;
+ intel_ring_update_space(ring);
}
- intel_ring_init_seqno(engine, engine->last_submitted_seqno);
+ engine->i915->gt.active_engines &= ~intel_engine_flag(engine);
}
void i915_gem_reset(struct drm_device *dev)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
struct intel_engine_cs *engine;
/*
* their reference to the objects, the inspection must be done first.
*/
for_each_engine(engine, dev_priv)
- i915_gem_reset_engine_status(dev_priv, engine);
+ i915_gem_reset_engine_status(engine);
for_each_engine(engine, dev_priv)
- i915_gem_reset_engine_cleanup(dev_priv, engine);
+ i915_gem_reset_engine_cleanup(engine);
+ mod_delayed_work(dev_priv->wq, &dev_priv->gt.idle_work, 0);
i915_gem_context_reset(dev);
i915_gem_restore_fences(dev);
-
- WARN_ON(i915_verify_lists(dev));
-}
-
-/**
- * This function clears the request list as sequence numbers are passed.
- */
-void
-i915_gem_retire_requests_ring(struct intel_engine_cs *engine)
-{
- WARN_ON(i915_verify_lists(engine->dev));
-
- /* Retire requests first as we use it above for the early return.
- * If we retire requests last, we may use a later seqno and so clear
- * the requests lists without clearing the active list, leading to
- * confusion.
- */
- while (!list_empty(&engine->request_list)) {
- struct drm_i915_gem_request *request;
-
- request = list_first_entry(&engine->request_list,
- struct drm_i915_gem_request,
- list);
-
- if (!i915_gem_request_completed(request, true))
- break;
-
- i915_gem_request_retire(request);
- }
-
- /* Move any buffers on the active list that are no longer referenced
- * by the ringbuffer to the flushing/inactive lists as appropriate,
- * before we free the context associated with the requests.
- */
- while (!list_empty(&engine->active_list)) {
- struct drm_i915_gem_object *obj;
-
- obj = list_first_entry(&engine->active_list,
- struct drm_i915_gem_object,
- engine_list[engine->id]);
-
- if (!list_empty(&obj->last_read_req[engine->id]->list))
- break;
-
- i915_gem_object_retire__read(obj, engine->id);
- }
-
- if (unlikely(engine->trace_irq_req &&
- i915_gem_request_completed(engine->trace_irq_req, true))) {
- engine->irq_put(engine);
- i915_gem_request_assign(&engine->trace_irq_req, NULL);
- }
-
- WARN_ON(i915_verify_lists(engine->dev));
-}
-
-bool
-i915_gem_retire_requests(struct drm_device *dev)
-{
- struct drm_i915_private *dev_priv = dev->dev_private;
- struct intel_engine_cs *engine;
- bool idle = true;
-
- for_each_engine(engine, dev_priv) {
- i915_gem_retire_requests_ring(engine);
- idle &= list_empty(&engine->request_list);
- if (i915.enable_execlists) {
- spin_lock_bh(&engine->execlist_lock);
- idle &= list_empty(&engine->execlist_queue);
- spin_unlock_bh(&engine->execlist_lock);
-
- intel_execlists_retire_requests(engine);
- }
- }
-
- if (idle)
- mod_delayed_work(dev_priv->wq,
- &dev_priv->mm.idle_work,
- msecs_to_jiffies(100));
-
- return idle;
}
static void
i915_gem_retire_work_handler(struct work_struct *work)
{
struct drm_i915_private *dev_priv =
- container_of(work, typeof(*dev_priv), mm.retire_work.work);
- struct drm_device *dev = dev_priv->dev;
- bool idle;
+ container_of(work, typeof(*dev_priv), gt.retire_work.work);
+ struct drm_device *dev = &dev_priv->drm;
/* Come back later if the device is busy... */
- idle = false;
if (mutex_trylock(&dev->struct_mutex)) {
- idle = i915_gem_retire_requests(dev);
+ i915_gem_retire_requests(dev_priv);
mutex_unlock(&dev->struct_mutex);
}
- if (!idle)
- queue_delayed_work(dev_priv->wq, &dev_priv->mm.retire_work,
+
+ /* Keep the retire handler running until we are finally idle.
+ * We do not need to do this test under locking as in the worst-case
+ * we queue the retire worker once too often.
+ */
+ if (READ_ONCE(dev_priv->gt.awake)) {
+ i915_queue_hangcheck(dev_priv);
+ queue_delayed_work(dev_priv->wq,
+ &dev_priv->gt.retire_work,
round_jiffies_up_relative(HZ));
+ }
}
static void
i915_gem_idle_work_handler(struct work_struct *work)
{
struct drm_i915_private *dev_priv =
- container_of(work, typeof(*dev_priv), mm.idle_work.work);
- struct drm_device *dev = dev_priv->dev;
+ container_of(work, typeof(*dev_priv), gt.idle_work.work);
+ struct drm_device *dev = &dev_priv->drm;
struct intel_engine_cs *engine;
+ unsigned int stuck_engines;
+ bool rearm_hangcheck;
- for_each_engine(engine, dev_priv)
- if (!list_empty(&engine->request_list))
- return;
-
- /* we probably should sync with hangcheck here, using cancel_work_sync.
- * Also locking seems to be fubar here, engine->request_list is protected
- * by dev->struct_mutex. */
+ if (!READ_ONCE(dev_priv->gt.awake))
+ return;
- intel_mark_idle(dev);
+ if (READ_ONCE(dev_priv->gt.active_engines))
+ return;
- if (mutex_trylock(&dev->struct_mutex)) {
- for_each_engine(engine, dev_priv)
- i915_gem_batch_pool_fini(&engine->batch_pool);
+ rearm_hangcheck =
+ cancel_delayed_work_sync(&dev_priv->gpu_error.hangcheck_work);
- mutex_unlock(&dev->struct_mutex);
+ if (!mutex_trylock(&dev->struct_mutex)) {
+ /* Currently busy, come back later */
+ mod_delayed_work(dev_priv->wq,
+ &dev_priv->gt.idle_work,
+ msecs_to_jiffies(50));
+ goto out_rearm;
}
-}
-/**
- * Ensures that an object will eventually get non-busy by flushing any required
- * write domains, emitting any outstanding lazy request and retiring and
- * completed requests.
- */
-static int
-i915_gem_object_flush_active(struct drm_i915_gem_object *obj)
-{
- int i;
+ if (dev_priv->gt.active_engines)
+ goto out_unlock;
- if (!obj->active)
- return 0;
+ for_each_engine(engine, dev_priv)
+ i915_gem_batch_pool_fini(&engine->batch_pool);
- for (i = 0; i < I915_NUM_ENGINES; i++) {
- struct drm_i915_gem_request *req;
+ GEM_BUG_ON(!dev_priv->gt.awake);
+ dev_priv->gt.awake = false;
+ rearm_hangcheck = false;
- req = obj->last_read_req[i];
- if (req == NULL)
- continue;
+ /* As we have disabled hangcheck, we need to unstick any waiters still
+ * hanging around. However, as we may be racing against the interrupt
+ * handler or the waiters themselves, we skip enabling the fake-irq.
+ */
+ stuck_engines = intel_kick_waiters(dev_priv);
+ if (unlikely(stuck_engines))
+ DRM_DEBUG_DRIVER("kicked stuck waiters (%x)...missed irq?\n",
+ stuck_engines);
- if (list_empty(&req->list))
- goto retire;
+ if (INTEL_GEN(dev_priv) >= 6)
+ gen6_rps_idle(dev_priv);
+ intel_runtime_pm_put(dev_priv);
+out_unlock:
+ mutex_unlock(&dev->struct_mutex);
- if (i915_gem_request_completed(req, true)) {
- __i915_gem_request_retire__upto(req);
-retire:
- i915_gem_object_retire__read(obj, i);
- }
+out_rearm:
+ if (rearm_hangcheck) {
+ GEM_BUG_ON(!dev_priv->gt.awake);
+ i915_queue_hangcheck(dev_priv);
}
+}
- return 0;
+void i915_gem_close_object(struct drm_gem_object *gem, struct drm_file *file)
+{
+ struct drm_i915_gem_object *obj = to_intel_bo(gem);
+ struct drm_i915_file_private *fpriv = file->driver_priv;
+ struct i915_vma *vma, *vn;
+
+ mutex_lock(&obj->base.dev->struct_mutex);
+ list_for_each_entry_safe(vma, vn, &obj->vma_list, obj_link)
+ if (vma->vm->file == fpriv)
+ i915_vma_close(vma);
+ mutex_unlock(&obj->base.dev->struct_mutex);
}
/**
* i915_gem_wait_ioctl - implements DRM_IOCTL_I915_GEM_WAIT
- * @DRM_IOCTL_ARGS: standard ioctl arguments
+ * @dev: drm device pointer
+ * @data: ioctl data blob
+ * @file: drm file pointer
*
* Returns 0 if successful, else an error is returned with the remaining time in
* the timeout parameter.
{
struct drm_i915_gem_wait *args = data;
struct drm_i915_gem_object *obj;
- struct drm_i915_gem_request *req[I915_NUM_ENGINES];
+ struct drm_i915_gem_request *requests[I915_NUM_ENGINES];
int i, n = 0;
int ret;
if (ret)
return ret;
- obj = to_intel_bo(drm_gem_object_lookup(file, args->bo_handle));
- if (&obj->base == NULL) {
+ obj = i915_gem_object_lookup(file, args->bo_handle);
+ if (!obj) {
mutex_unlock(&dev->struct_mutex);
return -ENOENT;
}
- /* Need to make sure the object gets inactive eventually. */
- ret = i915_gem_object_flush_active(obj);
- if (ret)
- goto out;
-
- if (!obj->active)
- goto out;
-
- /* Do this after OLR check to make sure we make forward progress polling
- * on this IOCTL with a timeout == 0 (like busy ioctl)
- */
- if (args->timeout_ns == 0) {
- ret = -ETIME;
+ if (!i915_gem_object_is_active(obj))
goto out;
- }
-
- drm_gem_object_unreference(&obj->base);
for (i = 0; i < I915_NUM_ENGINES; i++) {
- if (obj->last_read_req[i] == NULL)
- continue;
+ struct drm_i915_gem_request *req;
- req[n++] = i915_gem_request_reference(obj->last_read_req[i]);
+ req = i915_gem_active_get(&obj->last_read[i],
+ &obj->base.dev->struct_mutex);
+ if (req)
+ requests[n++] = req;
}
+out:
+ i915_gem_object_put(obj);
mutex_unlock(&dev->struct_mutex);
for (i = 0; i < n; i++) {
if (ret == 0)
- ret = __i915_wait_request(req[i], true,
- args->timeout_ns > 0 ? &args->timeout_ns : NULL,
- to_rps_client(file));
- i915_gem_request_unreference__unlocked(req[i]);
+ ret = i915_wait_request(requests[i], true,
+ args->timeout_ns > 0 ? &args->timeout_ns : NULL,
+ to_rps_client(file));
+ i915_gem_request_put(requests[i]);
}
return ret;
-
-out:
- drm_gem_object_unreference(&obj->base);
- mutex_unlock(&dev->struct_mutex);
- return ret;
}
static int
-__i915_gem_object_sync(struct drm_i915_gem_object *obj,
- struct intel_engine_cs *to,
- struct drm_i915_gem_request *from_req,
- struct drm_i915_gem_request **to_req)
+__i915_gem_object_sync(struct drm_i915_gem_request *to,
+ struct drm_i915_gem_request *from)
{
- struct intel_engine_cs *from;
int ret;
- from = i915_gem_request_get_engine(from_req);
- if (to == from)
- return 0;
-
- if (i915_gem_request_completed(from_req, true))
+ if (to->engine == from->engine)
return 0;
- if (!i915_semaphore_is_enabled(obj->base.dev)) {
- struct drm_i915_private *i915 = to_i915(obj->base.dev);
- ret = __i915_wait_request(from_req,
- i915->mm.interruptible,
- NULL,
- &i915->rps.semaphores);
+ if (!i915.semaphores) {
+ ret = i915_wait_request(from,
+ from->i915->mm.interruptible,
+ NULL,
+ NO_WAITBOOST);
if (ret)
return ret;
-
- i915_gem_object_retire_request(obj, from_req);
} else {
- int idx = intel_ring_sync_index(from, to);
- u32 seqno = i915_gem_request_get_seqno(from_req);
-
- WARN_ON(!to_req);
-
- if (seqno <= from->semaphore.sync_seqno[idx])
+ int idx = intel_engine_sync_index(from->engine, to->engine);
+ if (from->fence.seqno <= from->engine->semaphore.sync_seqno[idx])
return 0;
- if (*to_req == NULL) {
- struct drm_i915_gem_request *req;
-
- req = i915_gem_request_alloc(to, NULL);
- if (IS_ERR(req))
- return PTR_ERR(req);
-
- *to_req = req;
- }
-
- trace_i915_gem_ring_sync_to(*to_req, from, from_req);
- ret = to->semaphore.sync_to(*to_req, from, seqno);
+ trace_i915_gem_ring_sync_to(to, from);
+ ret = to->engine->semaphore.sync_to(to, from);
if (ret)
return ret;
- /* We use last_read_req because sync_to()
- * might have just caused seqno wrap under
- * the radar.
- */
- from->semaphore.sync_seqno[idx] =
- i915_gem_request_get_seqno(obj->last_read_req[from->id]);
+ from->engine->semaphore.sync_seqno[idx] = from->fence.seqno;
}
return 0;
* i915_gem_object_sync - sync an object to a ring.
*
* @obj: object which may be in use on another ring.
- * @to: ring we wish to use the object on. May be NULL.
- * @to_req: request we wish to use the object for. See below.
- * This will be allocated and returned if a request is
- * required but not passed in.
+ * @to: request we are wishing to use
*
* This code is meant to abstract object synchronization with the GPU.
- * Calling with NULL implies synchronizing the object with the CPU
- * rather than a particular GPU ring. Conceptually we serialise writes
- * between engines inside the GPU. We only allow one engine to write
- * into a buffer at any time, but multiple readers. To ensure each has
- * a coherent view of memory, we must:
+ * Conceptually we serialise writes between engines inside the GPU.
+ * We only allow one engine to write into a buffer at any time, but
+ * multiple readers. To ensure each has a coherent view of memory, we must:
*
* - If there is an outstanding write request to the object, the new
* request must wait for it to complete (either CPU or in hw, requests
* - If we are a write request (pending_write_domain is set), the new
* request must wait for outstanding read requests to complete.
*
- * For CPU synchronisation (NULL to) no request is required. For syncing with
- * rings to_req must be non-NULL. However, a request does not have to be
- * pre-allocated. If *to_req is NULL and sync commands will be emitted then a
- * request will be allocated automatically and returned through *to_req. Note
- * that it is not guaranteed that commands will be emitted (because the system
- * might already be idle). Hence there is no need to create a request that
- * might never have any work submitted. Note further that if a request is
- * returned in *to_req, it is the responsibility of the caller to submit
- * that request (after potentially adding more work to it).
- *
* Returns 0 if successful, else propagates up the lower layer error.
*/
int
i915_gem_object_sync(struct drm_i915_gem_object *obj,
- struct intel_engine_cs *to,
- struct drm_i915_gem_request **to_req)
+ struct drm_i915_gem_request *to)
{
- const bool readonly = obj->base.pending_write_domain == 0;
- struct drm_i915_gem_request *req[I915_NUM_ENGINES];
- int ret, i, n;
+ struct i915_gem_active *active;
+ unsigned long active_mask;
+ int idx;
- if (!obj->active)
- return 0;
+ lockdep_assert_held(&obj->base.dev->struct_mutex);
- if (to == NULL)
- return i915_gem_object_wait_rendering(obj, readonly);
+ active_mask = i915_gem_object_get_active(obj);
+ if (!active_mask)
+ return 0;
- n = 0;
- if (readonly) {
- if (obj->last_write_req)
- req[n++] = obj->last_write_req;
+ if (obj->base.pending_write_domain) {
+ active = obj->last_read;
} else {
- for (i = 0; i < I915_NUM_ENGINES; i++)
- if (obj->last_read_req[i])
- req[n++] = obj->last_read_req[i];
+ active_mask = 1;
+ active = &obj->last_write;
}
- for (i = 0; i < n; i++) {
- ret = __i915_gem_object_sync(obj, to, req[i], to_req);
+
+ for_each_active(active_mask, idx) {
+ struct drm_i915_gem_request *request;
+ int ret;
+
+ request = i915_gem_active_peek(&active[idx],
+ &obj->base.dev->struct_mutex);
+ if (!request)
+ continue;
+
+ ret = __i915_gem_object_sync(to, request);
if (ret)
return ret;
}
old_write_domain);
}
-static int __i915_vma_unbind(struct i915_vma *vma, bool wait)
+static void __i915_vma_iounmap(struct i915_vma *vma)
+{
+ GEM_BUG_ON(i915_vma_is_pinned(vma));
+
+ if (vma->iomap == NULL)
+ return;
+
+ io_mapping_unmap(vma->iomap);
+ vma->iomap = NULL;
+}
+
+int i915_vma_unbind(struct i915_vma *vma)
{
struct drm_i915_gem_object *obj = vma->obj;
- struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
+ unsigned long active;
int ret;
- if (list_empty(&vma->obj_link))
- return 0;
+ /* First wait upon any activity as retiring the request may
+ * have side-effects such as unpinning or even unbinding this vma.
+ */
+ active = i915_vma_get_active(vma);
+ if (active) {
+ int idx;
+
+ /* When a closed VMA is retired, it is unbound - eek.
+ * In order to prevent it from being recursively closed,
+ * take a pin on the vma so that the second unbind is
+ * aborted.
+ */
+ __i915_vma_pin(vma);
- if (!drm_mm_node_allocated(&vma->node)) {
- i915_gem_vma_destroy(vma);
- return 0;
+ for_each_active(active, idx) {
+ ret = i915_gem_active_retire(&vma->last_read[idx],
+ &vma->vm->dev->struct_mutex);
+ if (ret)
+ break;
+ }
+
+ __i915_vma_unpin(vma);
+ if (ret)
+ return ret;
+
+ GEM_BUG_ON(i915_vma_is_active(vma));
}
- if (vma->pin_count)
+ if (i915_vma_is_pinned(vma))
return -EBUSY;
- BUG_ON(obj->pages == NULL);
+ if (!drm_mm_node_allocated(&vma->node))
+ goto destroy;
- if (wait) {
- ret = i915_gem_object_wait_rendering(obj, false);
- if (ret)
- return ret;
- }
+ GEM_BUG_ON(obj->bind_count == 0);
+ GEM_BUG_ON(!obj->pages);
- if (vma->is_ggtt && vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL) {
+ if (i915_vma_is_ggtt(vma) &&
+ vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL) {
i915_gem_object_finish_gtt(obj);
/* release the fence reg _after_ flushing */
ret = i915_gem_object_put_fence(obj);
if (ret)
return ret;
+
+ __i915_vma_iounmap(vma);
}
- trace_i915_vma_unbind(vma);
+ if (likely(!vma->vm->closed)) {
+ trace_i915_vma_unbind(vma);
+ vma->vm->unbind_vma(vma);
+ }
+ vma->flags &= ~(I915_VMA_GLOBAL_BIND | I915_VMA_LOCAL_BIND);
- vma->vm->unbind_vma(vma);
- vma->bound = 0;
+ drm_mm_remove_node(&vma->node);
+ list_move_tail(&vma->vm_link, &vma->vm->unbound_list);
- list_del_init(&vma->vm_link);
- if (vma->is_ggtt) {
+ if (i915_vma_is_ggtt(vma)) {
if (vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL) {
obj->map_and_fenceable = false;
} else if (vma->ggtt_view.pages) {
vma->ggtt_view.pages = NULL;
}
- drm_mm_remove_node(&vma->node);
- i915_gem_vma_destroy(vma);
-
/* Since the unbound list is global, only move to that list if
* no more VMAs exist. */
- if (list_empty(&obj->vma_list))
- list_move_tail(&obj->global_list, &dev_priv->mm.unbound_list);
+ if (--obj->bind_count == 0)
+ list_move_tail(&obj->global_list,
+ &to_i915(obj->base.dev)->mm.unbound_list);
/* And finally now the object is completely decoupled from this vma,
* we can drop its hold on the backing storage and allow it to be
*/
i915_gem_object_unpin_pages(obj);
- return 0;
-}
-
-int i915_vma_unbind(struct i915_vma *vma)
-{
- return __i915_vma_unbind(vma, true);
-}
+destroy:
+ if (unlikely(i915_vma_is_closed(vma)))
+ i915_vma_destroy(vma);
-int __i915_vma_unbind_no_wait(struct i915_vma *vma)
-{
- return __i915_vma_unbind(vma, false);
+ return 0;
}
-int i915_gpu_idle(struct drm_device *dev)
+int i915_gem_wait_for_idle(struct drm_i915_private *dev_priv)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
struct intel_engine_cs *engine;
int ret;
- /* Flush everything onto the inactive list. */
- for_each_engine(engine, dev_priv) {
- if (!i915.enable_execlists) {
- struct drm_i915_gem_request *req;
-
- req = i915_gem_request_alloc(engine, NULL);
- if (IS_ERR(req))
- return PTR_ERR(req);
+ lockdep_assert_held(&dev_priv->drm.struct_mutex);
- ret = i915_switch_context(req);
- i915_add_request_no_flush(req);
- if (ret)
- return ret;
- }
+ for_each_engine(engine, dev_priv) {
+ if (engine->last_context == NULL)
+ continue;
ret = intel_engine_idle(engine);
if (ret)
return ret;
}
- WARN_ON(i915_verify_lists(dev));
return 0;
}
}
/**
- * Finds free space in the GTT aperture and binds the object or a view of it
- * there.
+ * i915_vma_insert - finds a slot for the vma in its address space
+ * @vma: the vma
+ * @size: requested size in bytes (can be larger than the VMA)
+ * @alignment: required alignment
+ * @flags: mask of PIN_* flags to use
+ *
+ * First we try to allocate some free space that meets the requirements for
+ * the VMA. Failiing that, if the flags permit, it will evict an old VMA,
+ * preferrably the oldest idle entry to make room for the new VMA.
+ *
+ * Returns:
+ * 0 on success, negative error code otherwise.
*/
-static struct i915_vma *
-i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
- struct i915_address_space *vm,
- const struct i915_ggtt_view *ggtt_view,
- unsigned alignment,
- uint64_t flags)
+static int
+i915_vma_insert(struct i915_vma *vma, u64 size, u64 alignment, u64 flags)
{
- struct drm_device *dev = obj->base.dev;
- struct drm_i915_private *dev_priv = to_i915(dev);
- struct i915_ggtt *ggtt = &dev_priv->ggtt;
- u32 fence_alignment, unfenced_alignment;
- u32 search_flag, alloc_flag;
+ struct drm_i915_private *dev_priv = to_i915(vma->vm->dev);
+ struct drm_i915_gem_object *obj = vma->obj;
u64 start, end;
- u64 size, fence_size;
- struct i915_vma *vma;
+ u64 min_alignment;
int ret;
- if (i915_is_ggtt(vm)) {
- u32 view_size;
-
- if (WARN_ON(!ggtt_view))
- return ERR_PTR(-EINVAL);
-
- view_size = i915_ggtt_view_size(obj, ggtt_view);
-
- fence_size = i915_gem_get_gtt_size(dev,
- view_size,
- obj->tiling_mode);
- fence_alignment = i915_gem_get_gtt_alignment(dev,
- view_size,
- obj->tiling_mode,
- true);
- unfenced_alignment = i915_gem_get_gtt_alignment(dev,
- view_size,
- obj->tiling_mode,
- false);
- size = flags & PIN_MAPPABLE ? fence_size : view_size;
- } else {
- fence_size = i915_gem_get_gtt_size(dev,
- obj->base.size,
- obj->tiling_mode);
- fence_alignment = i915_gem_get_gtt_alignment(dev,
- obj->base.size,
- obj->tiling_mode,
- true);
- unfenced_alignment =
- i915_gem_get_gtt_alignment(dev,
- obj->base.size,
- obj->tiling_mode,
- false);
- size = flags & PIN_MAPPABLE ? fence_size : obj->base.size;
+ GEM_BUG_ON(vma->flags & (I915_VMA_GLOBAL_BIND | I915_VMA_LOCAL_BIND));
+ GEM_BUG_ON(drm_mm_node_allocated(&vma->node));
+
+ size = max(size, vma->size);
+ if (flags & PIN_MAPPABLE)
+ size = i915_gem_get_ggtt_size(dev_priv, size, obj->tiling_mode);
+
+ min_alignment =
+ i915_gem_get_ggtt_alignment(dev_priv, size, obj->tiling_mode,
+ flags & PIN_MAPPABLE);
+ if (alignment == 0)
+ alignment = min_alignment;
+ if (alignment & (min_alignment - 1)) {
+ DRM_DEBUG("Invalid object alignment requested %llu, minimum %llu\n",
+ alignment, min_alignment);
+ return -EINVAL;
}
start = flags & PIN_OFFSET_BIAS ? flags & PIN_OFFSET_MASK : 0;
- end = vm->total;
+
+ end = vma->vm->total;
if (flags & PIN_MAPPABLE)
- end = min_t(u64, end, ggtt->mappable_end);
+ end = min_t(u64, end, dev_priv->ggtt.mappable_end);
if (flags & PIN_ZONE_4G)
end = min_t(u64, end, (1ULL << 32) - PAGE_SIZE);
- if (alignment == 0)
- alignment = flags & PIN_MAPPABLE ? fence_alignment :
- unfenced_alignment;
- if (flags & PIN_MAPPABLE && alignment & (fence_alignment - 1)) {
- DRM_DEBUG("Invalid object (view type=%u) alignment requested %u\n",
- ggtt_view ? ggtt_view->type : 0,
- alignment);
- return ERR_PTR(-EINVAL);
- }
-
/* If binding the object/GGTT view requires more space than the entire
* aperture has, reject it early before evicting everything in a vain
* attempt to find space.
*/
if (size > end) {
- DRM_DEBUG("Attempting to bind an object (view type=%u) larger than the aperture: size=%llu > %s aperture=%llu\n",
- ggtt_view ? ggtt_view->type : 0,
- size,
+ DRM_DEBUG("Attempting to bind an object larger than the aperture: request=%llu [object=%zd] > %s aperture=%llu\n",
+ size, obj->base.size,
flags & PIN_MAPPABLE ? "mappable" : "total",
end);
- return ERR_PTR(-E2BIG);
+ return -E2BIG;
}
ret = i915_gem_object_get_pages(obj);
if (ret)
- return ERR_PTR(ret);
+ return ret;
i915_gem_object_pin_pages(obj);
- vma = ggtt_view ? i915_gem_obj_lookup_or_create_ggtt_vma(obj, ggtt_view) :
- i915_gem_obj_lookup_or_create_vma(obj, vm);
-
- if (IS_ERR(vma))
- goto err_unpin;
-
if (flags & PIN_OFFSET_FIXED) {
- uint64_t offset = flags & PIN_OFFSET_MASK;
-
- if (offset & (alignment - 1) || offset + size > end) {
+ u64 offset = flags & PIN_OFFSET_MASK;
+ if (offset & (alignment - 1) || offset > end - size) {
ret = -EINVAL;
- goto err_free_vma;
+ goto err_unpin;
}
+
vma->node.start = offset;
vma->node.size = size;
vma->node.color = obj->cache_level;
- ret = drm_mm_reserve_node(&vm->mm, &vma->node);
+ ret = drm_mm_reserve_node(&vma->vm->mm, &vma->node);
if (ret) {
ret = i915_gem_evict_for_vma(vma);
if (ret == 0)
- ret = drm_mm_reserve_node(&vm->mm, &vma->node);
+ ret = drm_mm_reserve_node(&vma->vm->mm, &vma->node);
+ if (ret)
+ goto err_unpin;
}
- if (ret)
- goto err_free_vma;
} else {
+ u32 search_flag, alloc_flag;
+
if (flags & PIN_HIGH) {
search_flag = DRM_MM_SEARCH_BELOW;
alloc_flag = DRM_MM_CREATE_TOP;
alloc_flag = DRM_MM_CREATE_DEFAULT;
}
+ /* We only allocate in PAGE_SIZE/GTT_PAGE_SIZE (4096) chunks,
+ * so we know that we always have a minimum alignment of 4096.
+ * The drm_mm range manager is optimised to return results
+ * with zero alignment, so where possible use the optimal
+ * path.
+ */
+ if (alignment <= 4096)
+ alignment = 0;
+
search_free:
- ret = drm_mm_insert_node_in_range_generic(&vm->mm, &vma->node,
+ ret = drm_mm_insert_node_in_range_generic(&vma->vm->mm,
+ &vma->node,
size, alignment,
obj->cache_level,
start, end,
search_flag,
alloc_flag);
if (ret) {
- ret = i915_gem_evict_something(dev, vm, size, alignment,
+ ret = i915_gem_evict_something(vma->vm, size, alignment,
obj->cache_level,
start, end,
flags);
if (ret == 0)
goto search_free;
- goto err_free_vma;
+ goto err_unpin;
}
}
- if (WARN_ON(!i915_gem_valid_gtt_space(vma, obj->cache_level))) {
- ret = -EINVAL;
- goto err_remove_node;
- }
-
- trace_i915_vma_bind(vma, flags);
- ret = i915_vma_bind(vma, obj->cache_level, flags);
- if (ret)
- goto err_remove_node;
+ GEM_BUG_ON(!i915_gem_valid_gtt_space(vma, obj->cache_level));
list_move_tail(&obj->global_list, &dev_priv->mm.bound_list);
- list_add_tail(&vma->vm_link, &vm->inactive_list);
+ list_move_tail(&vma->vm_link, &vma->vm->inactive_list);
+ obj->bind_count++;
- return vma;
+ return 0;
-err_remove_node:
- drm_mm_remove_node(&vma->node);
-err_free_vma:
- i915_gem_vma_destroy(vma);
- vma = ERR_PTR(ret);
err_unpin:
i915_gem_object_unpin_pages(obj);
- return vma;
+ return ret;
}
bool
return;
if (i915_gem_clflush_object(obj, obj->pin_display))
- i915_gem_chipset_flush(obj->base.dev);
+ i915_gem_chipset_flush(to_i915(obj->base.dev));
old_write_domain = obj->base.write_domain;
obj->base.write_domain = 0;
/**
* Moves a single object to the GTT read, and possibly write domain.
+ * @obj: object to act on
+ * @write: ask for write access or read only
*
* This function returns when the move is complete, including waiting on
* flushes to occur.
int
i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
{
- struct drm_device *dev = obj->base.dev;
- struct drm_i915_private *dev_priv = to_i915(dev);
- struct i915_ggtt *ggtt = &dev_priv->ggtt;
uint32_t old_write_domain, old_read_domains;
struct i915_vma *vma;
int ret;
- if (obj->base.write_domain == I915_GEM_DOMAIN_GTT)
- return 0;
-
ret = i915_gem_object_wait_rendering(obj, !write);
if (ret)
return ret;
+ if (obj->base.write_domain == I915_GEM_DOMAIN_GTT)
+ return 0;
+
/* Flush and acquire obj->pages so that we are coherent through
* direct access in memory with previous cached writes through
* shmemfs and that our cache domain tracking remains valid.
/* And bump the LRU for this access */
vma = i915_gem_obj_to_ggtt(obj);
- if (vma && drm_mm_node_allocated(&vma->node) && !obj->active)
- list_move_tail(&vma->vm_link,
- &ggtt->base.inactive_list);
+ if (vma &&
+ drm_mm_node_allocated(&vma->node) &&
+ !i915_vma_is_active(vma))
+ list_move_tail(&vma->vm_link, &vma->vm->inactive_list);
return 0;
}
/**
* Changes the cache-level of an object across all VMA.
+ * @obj: object to act on
+ * @cache_level: new cache level to set for the object
*
* After this function returns, the object will be in the new cache-level
* across all GTT and the contents of the backing storage will be coherent,
int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
enum i915_cache_level cache_level)
{
- struct drm_device *dev = obj->base.dev;
- struct i915_vma *vma, *next;
- bool bound = false;
+ struct i915_vma *vma;
int ret = 0;
if (obj->cache_level == cache_level)
* catch the issue of the CS prefetch crossing page boundaries and
* reading an invalid PTE on older architectures.
*/
- list_for_each_entry_safe(vma, next, &obj->vma_list, obj_link) {
+restart:
+ list_for_each_entry(vma, &obj->vma_list, obj_link) {
if (!drm_mm_node_allocated(&vma->node))
continue;
- if (vma->pin_count) {
+ if (i915_vma_is_pinned(vma)) {
DRM_DEBUG("can not change the cache level of pinned objects\n");
return -EBUSY;
}
- if (!i915_gem_valid_gtt_space(vma, cache_level)) {
- ret = i915_vma_unbind(vma);
- if (ret)
- return ret;
- } else
- bound = true;
+ if (i915_gem_valid_gtt_space(vma, cache_level))
+ continue;
+
+ ret = i915_vma_unbind(vma);
+ if (ret)
+ return ret;
+
+ /* As unbinding may affect other elements in the
+ * obj->vma_list (due to side-effects from retiring
+ * an active vma), play safe and restart the iterator.
+ */
+ goto restart;
}
/* We can reuse the existing drm_mm nodes but need to change the
* rewrite the PTE in the belief that doing so tramples upon less
* state and so involves less work.
*/
- if (bound) {
+ if (obj->bind_count) {
/* Before we change the PTE, the GPU must not be accessing it.
* If we wait upon the object, we know that all the bound
* VMA are no longer active.
if (ret)
return ret;
- if (!HAS_LLC(dev) && cache_level != I915_CACHE_NONE) {
+ if (!HAS_LLC(obj->base.dev) && cache_level != I915_CACHE_NONE) {
/* Access to snoopable pages through the GTT is
* incoherent and on some machines causes a hard
* lockup. Relinquish the CPU mmaping to force
* object is now coherent at its new cache level (with respect
* to the access domain).
*/
- if (obj->cache_dirty &&
- obj->base.write_domain != I915_GEM_DOMAIN_CPU &&
- cpu_write_needs_clflush(obj)) {
+ if (obj->cache_dirty && cpu_write_needs_clflush(obj)) {
if (i915_gem_clflush_object(obj, true))
- i915_gem_chipset_flush(obj->base.dev);
+ i915_gem_chipset_flush(to_i915(obj->base.dev));
}
return 0;
struct drm_i915_gem_caching *args = data;
struct drm_i915_gem_object *obj;
- obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
- if (&obj->base == NULL)
+ obj = i915_gem_object_lookup(file, args->handle);
+ if (!obj)
return -ENOENT;
switch (obj->cache_level) {
break;
}
- drm_gem_object_unreference_unlocked(&obj->base);
+ i915_gem_object_put_unlocked(obj);
return 0;
}
int i915_gem_set_caching_ioctl(struct drm_device *dev, void *data,
struct drm_file *file)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
struct drm_i915_gem_caching *args = data;
struct drm_i915_gem_object *obj;
enum i915_cache_level level;
if (ret)
goto rpm_put;
- obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
- if (&obj->base == NULL) {
+ obj = i915_gem_object_lookup(file, args->handle);
+ if (!obj) {
ret = -ENOENT;
goto unlock;
}
ret = i915_gem_object_set_cache_level(obj, level);
- drm_gem_object_unreference(&obj->base);
+ i915_gem_object_put(obj);
unlock:
mutex_unlock(&dev->struct_mutex);
rpm_put:
* (e.g. libkms for the bootup splash), we have to ensure that we
* always use map_and_fenceable for all scanout buffers.
*/
- ret = i915_gem_object_ggtt_pin(obj, view, alignment,
+ ret = i915_gem_object_ggtt_pin(obj, view, 0, alignment,
view->type == I915_GGTT_VIEW_NORMAL ?
PIN_MAPPABLE : 0);
if (ret)
/**
* Moves a single object to the CPU read, and possibly write domain.
+ * @obj: object to act on
+ * @write: requesting write or read-only access
*
* This function returns when the move is complete, including waiting on
* flushes to occur.
uint32_t old_write_domain, old_read_domains;
int ret;
- if (obj->base.write_domain == I915_GEM_DOMAIN_CPU)
- return 0;
-
ret = i915_gem_object_wait_rendering(obj, !write);
if (ret)
return ret;
+ if (obj->base.write_domain == I915_GEM_DOMAIN_CPU)
+ return 0;
+
i915_gem_object_flush_gtt_write_domain(obj);
old_write_domain = obj->base.write_domain;
static int
i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
struct drm_i915_file_private *file_priv = file->driver_priv;
unsigned long recent_enough = jiffies - DRM_I915_THROTTLE_JIFFIES;
struct drm_i915_gem_request *request, *target = NULL;
target = request;
}
if (target)
- i915_gem_request_reference(target);
+ i915_gem_request_get(target);
spin_unlock(&file_priv->mm.lock);
if (target == NULL)
return 0;
- ret = __i915_wait_request(target, true, NULL, NULL);
- if (ret == 0)
- queue_delayed_work(dev_priv->wq, &dev_priv->mm.retire_work, 0);
-
- i915_gem_request_unreference__unlocked(target);
+ ret = i915_wait_request(target, true, NULL, NULL);
+ i915_gem_request_put(target);
return ret;
}
static bool
-i915_vma_misplaced(struct i915_vma *vma, uint32_t alignment, uint64_t flags)
+i915_vma_misplaced(struct i915_vma *vma, u64 size, u64 alignment, u64 flags)
{
struct drm_i915_gem_object *obj = vma->obj;
- if (alignment &&
- vma->node.start & (alignment - 1))
+ if (!drm_mm_node_allocated(&vma->node))
+ return false;
+
+ if (vma->node.size < size)
+ return true;
+
+ if (alignment && vma->node.start & (alignment - 1))
return true;
if (flags & PIN_MAPPABLE && !obj->map_and_fenceable)
void __i915_vma_set_map_and_fenceable(struct i915_vma *vma)
{
struct drm_i915_gem_object *obj = vma->obj;
+ struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
bool mappable, fenceable;
u32 fence_size, fence_alignment;
- fence_size = i915_gem_get_gtt_size(obj->base.dev,
- obj->base.size,
- obj->tiling_mode);
- fence_alignment = i915_gem_get_gtt_alignment(obj->base.dev,
- obj->base.size,
- obj->tiling_mode,
- true);
+ fence_size = i915_gem_get_ggtt_size(dev_priv,
+ obj->base.size,
+ obj->tiling_mode);
+ fence_alignment = i915_gem_get_ggtt_alignment(dev_priv,
+ obj->base.size,
+ obj->tiling_mode,
+ true);
fenceable = (vma->node.size == fence_size &&
(vma->node.start & (fence_alignment - 1)) == 0);
mappable = (vma->node.start + fence_size <=
- to_i915(obj->base.dev)->ggtt.mappable_end);
+ dev_priv->ggtt.mappable_end);
obj->map_and_fenceable = mappable && fenceable;
}
-static int
-i915_gem_object_do_pin(struct drm_i915_gem_object *obj,
- struct i915_address_space *vm,
- const struct i915_ggtt_view *ggtt_view,
- uint32_t alignment,
- uint64_t flags)
+int __i915_vma_do_pin(struct i915_vma *vma,
+ u64 size, u64 alignment, u64 flags)
{
- struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
- struct i915_vma *vma;
- unsigned bound;
+ unsigned int bound = vma->flags;
int ret;
- if (WARN_ON(vm == &dev_priv->mm.aliasing_ppgtt->base))
- return -ENODEV;
-
- if (WARN_ON(flags & (PIN_GLOBAL | PIN_MAPPABLE) && !i915_is_ggtt(vm)))
- return -EINVAL;
-
- if (WARN_ON((flags & (PIN_MAPPABLE | PIN_GLOBAL)) == PIN_MAPPABLE))
- return -EINVAL;
-
- if (WARN_ON(i915_is_ggtt(vm) != !!ggtt_view))
- return -EINVAL;
-
- vma = ggtt_view ? i915_gem_obj_to_ggtt_view(obj, ggtt_view) :
- i915_gem_obj_to_vma(obj, vm);
+ GEM_BUG_ON((flags & (PIN_GLOBAL | PIN_USER)) == 0);
+ GEM_BUG_ON((flags & PIN_GLOBAL) && !i915_vma_is_ggtt(vma));
- if (vma) {
- if (WARN_ON(vma->pin_count == DRM_I915_GEM_OBJECT_MAX_PIN_COUNT))
- return -EBUSY;
-
- if (i915_vma_misplaced(vma, alignment, flags)) {
- WARN(vma->pin_count,
- "bo is already pinned in %s with incorrect alignment:"
- " offset=%08x %08x, req.alignment=%x, req.map_and_fenceable=%d,"
- " obj->map_and_fenceable=%d\n",
- ggtt_view ? "ggtt" : "ppgtt",
- upper_32_bits(vma->node.start),
- lower_32_bits(vma->node.start),
- alignment,
- !!(flags & PIN_MAPPABLE),
- obj->map_and_fenceable);
- ret = i915_vma_unbind(vma);
- if (ret)
- return ret;
-
- vma = NULL;
- }
+ if (WARN_ON(bound & I915_VMA_PIN_OVERFLOW)) {
+ ret = -EBUSY;
+ goto err;
}
- bound = vma ? vma->bound : 0;
- if (vma == NULL || !drm_mm_node_allocated(&vma->node)) {
- vma = i915_gem_object_bind_to_vm(obj, vm, ggtt_view, alignment,
- flags);
- if (IS_ERR(vma))
- return PTR_ERR(vma);
- } else {
- ret = i915_vma_bind(vma, obj->cache_level, flags);
+ if ((bound & I915_VMA_BIND_MASK) == 0) {
+ ret = i915_vma_insert(vma, size, alignment, flags);
if (ret)
- return ret;
+ goto err;
}
- if (ggtt_view && ggtt_view->type == I915_GGTT_VIEW_NORMAL &&
- (bound ^ vma->bound) & GLOBAL_BIND) {
+ ret = i915_vma_bind(vma, vma->obj->cache_level, flags);
+ if (ret)
+ goto err;
+
+ if ((bound ^ vma->flags) & I915_VMA_GLOBAL_BIND)
__i915_vma_set_map_and_fenceable(vma);
- WARN_ON(flags & PIN_MAPPABLE && !obj->map_and_fenceable);
- }
- vma->pin_count++;
+ GEM_BUG_ON(i915_vma_misplaced(vma, size, alignment, flags));
return 0;
-}
-int
-i915_gem_object_pin(struct drm_i915_gem_object *obj,
- struct i915_address_space *vm,
- uint32_t alignment,
- uint64_t flags)
-{
- return i915_gem_object_do_pin(obj, vm,
- i915_is_ggtt(vm) ? &i915_ggtt_view_normal : NULL,
- alignment, flags);
+err:
+ __i915_vma_unpin(vma);
+ return ret;
}
int
i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj,
const struct i915_ggtt_view *view,
- uint32_t alignment,
- uint64_t flags)
+ u64 size,
+ u64 alignment,
+ u64 flags)
{
- struct drm_device *dev = obj->base.dev;
- struct drm_i915_private *dev_priv = to_i915(dev);
- struct i915_ggtt *ggtt = &dev_priv->ggtt;
+ struct i915_vma *vma;
+ int ret;
+
+ if (!view)
+ view = &i915_ggtt_view_normal;
- BUG_ON(!view);
+ vma = i915_gem_obj_lookup_or_create_ggtt_vma(obj, view);
+ if (IS_ERR(vma))
+ return PTR_ERR(vma);
+
+ if (i915_vma_misplaced(vma, size, alignment, flags)) {
+ if (flags & PIN_NONBLOCK &&
+ (i915_vma_is_pinned(vma) || i915_vma_is_active(vma)))
+ return -ENOSPC;
+
+ WARN(i915_vma_is_pinned(vma),
+ "bo is already pinned in ggtt with incorrect alignment:"
+ " offset=%08x %08x, req.alignment=%llx, req.map_and_fenceable=%d,"
+ " obj->map_and_fenceable=%d\n",
+ upper_32_bits(vma->node.start),
+ lower_32_bits(vma->node.start),
+ alignment,
+ !!(flags & PIN_MAPPABLE),
+ obj->map_and_fenceable);
+ ret = i915_vma_unbind(vma);
+ if (ret)
+ return ret;
+ }
- return i915_gem_object_do_pin(obj, &ggtt->base, view,
- alignment, flags | PIN_GLOBAL);
+ return i915_vma_pin(vma, size, alignment, flags | PIN_GLOBAL);
}
void
i915_gem_object_ggtt_unpin_view(struct drm_i915_gem_object *obj,
const struct i915_ggtt_view *view)
{
- struct i915_vma *vma = i915_gem_obj_to_ggtt_view(obj, view);
-
- WARN_ON(vma->pin_count == 0);
- WARN_ON(!i915_gem_obj_ggtt_bound_view(obj, view));
-
- --vma->pin_count;
+ i915_vma_unpin(i915_gem_obj_to_ggtt_view(obj, view));
}
int
if (ret)
return ret;
- obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
- if (&obj->base == NULL) {
+ obj = i915_gem_object_lookup(file, args->handle);
+ if (!obj) {
ret = -ENOENT;
goto unlock;
}
/* Count all active objects as busy, even if they are currently not used
* by the gpu. Users of this interface expect objects to eventually
- * become non-busy without any further actions, therefore emit any
- * necessary flushes here.
+ * become non-busy without any further actions.
*/
- ret = i915_gem_object_flush_active(obj);
- if (ret)
- goto unref;
-
args->busy = 0;
- if (obj->active) {
+ if (i915_gem_object_is_active(obj)) {
+ struct drm_i915_gem_request *req;
int i;
for (i = 0; i < I915_NUM_ENGINES; i++) {
- struct drm_i915_gem_request *req;
-
- req = obj->last_read_req[i];
+ req = i915_gem_active_peek(&obj->last_read[i],
+ &obj->base.dev->struct_mutex);
if (req)
args->busy |= 1 << (16 + req->engine->exec_id);
}
- if (obj->last_write_req)
- args->busy |= obj->last_write_req->engine->exec_id;
+ req = i915_gem_active_peek(&obj->last_write,
+ &obj->base.dev->struct_mutex);
+ if (req)
+ args->busy |= req->engine->exec_id;
}
-unref:
- drm_gem_object_unreference(&obj->base);
+ i915_gem_object_put(obj);
unlock:
mutex_unlock(&dev->struct_mutex);
return ret;
i915_gem_madvise_ioctl(struct drm_device *dev, void *data,
struct drm_file *file_priv)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
struct drm_i915_gem_madvise *args = data;
struct drm_i915_gem_object *obj;
int ret;
if (ret)
return ret;
- obj = to_intel_bo(drm_gem_object_lookup(file_priv, args->handle));
- if (&obj->base == NULL) {
+ obj = i915_gem_object_lookup(file_priv, args->handle);
+ if (!obj) {
ret = -ENOENT;
goto unlock;
}
args->retained = obj->madv != __I915_MADV_PURGED;
out:
- drm_gem_object_unreference(&obj->base);
+ i915_gem_object_put(obj);
unlock:
mutex_unlock(&dev->struct_mutex);
return ret;
INIT_LIST_HEAD(&obj->global_list);
for (i = 0; i < I915_NUM_ENGINES; i++)
- INIT_LIST_HEAD(&obj->engine_list[i]);
+ init_request_active(&obj->last_read[i],
+ i915_gem_object_retire__read);
+ init_request_active(&obj->last_write,
+ i915_gem_object_retire__write);
+ init_request_active(&obj->last_fence, NULL);
INIT_LIST_HEAD(&obj->obj_exec_link);
INIT_LIST_HEAD(&obj->vma_list);
INIT_LIST_HEAD(&obj->batch_pool_link);
obj->fence_reg = I915_FENCE_REG_NONE;
obj->madv = I915_MADV_WILLNEED;
- i915_gem_info_add_obj(obj->base.dev->dev_private, obj->base.size);
+ i915_gem_info_add_obj(to_i915(obj->base.dev), obj->base.size);
}
static const struct drm_i915_gem_object_ops i915_gem_object_ops = {
.put_pages = i915_gem_object_put_pages_gtt,
};
-struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev,
+struct drm_i915_gem_object *i915_gem_object_create(struct drm_device *dev,
size_t size)
{
struct drm_i915_gem_object *obj;
struct address_space *mapping;
gfp_t mask;
+ int ret;
obj = i915_gem_object_alloc(dev);
if (obj == NULL)
- return NULL;
+ return ERR_PTR(-ENOMEM);
- if (drm_gem_object_init(dev, &obj->base, size) != 0) {
- i915_gem_object_free(obj);
- return NULL;
- }
+ ret = drm_gem_object_init(dev, &obj->base, size);
+ if (ret)
+ goto fail;
mask = GFP_HIGHUSER | __GFP_RECLAIMABLE;
if (IS_CRESTLINE(dev) || IS_BROADWATER(dev)) {
trace_i915_gem_object_create(obj);
return obj;
+
+fail:
+ i915_gem_object_free(obj);
+
+ return ERR_PTR(ret);
}
static bool discard_backing_storage(struct drm_i915_gem_object *obj)
{
struct drm_i915_gem_object *obj = to_intel_bo(gem_obj);
struct drm_device *dev = obj->base.dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
struct i915_vma *vma, *next;
intel_runtime_pm_get(dev_priv);
trace_i915_gem_object_destroy(obj);
+ /* All file-owned VMA should have been released by this point through
+ * i915_gem_close_object(), or earlier by i915_gem_context_close().
+ * However, the object may also be bound into the global GTT (e.g.
+ * older GPUs without per-process support, or for direct access through
+ * the GTT either for the user or for scanout). Those VMA still need to
+ * unbound now.
+ */
list_for_each_entry_safe(vma, next, &obj->vma_list, obj_link) {
- int ret;
-
- vma->pin_count = 0;
- ret = i915_vma_unbind(vma);
- if (WARN_ON(ret == -ERESTARTSYS)) {
- bool was_interruptible;
-
- was_interruptible = dev_priv->mm.interruptible;
- dev_priv->mm.interruptible = false;
-
- WARN_ON(i915_vma_unbind(vma));
-
- dev_priv->mm.interruptible = was_interruptible;
- }
+ GEM_BUG_ON(!i915_vma_is_ggtt(vma));
+ GEM_BUG_ON(i915_vma_is_active(vma));
+ vma->flags &= ~I915_VMA_PIN_MASK;
+ i915_vma_close(vma);
}
+ GEM_BUG_ON(obj->bind_count);
/* Stolen objects don't hold a ref, but do hold pin count. Fix that up
* before progressing. */
if (obj->stolen)
i915_gem_object_unpin_pages(obj);
- WARN_ON(obj->frontbuffer_bits);
+ WARN_ON(atomic_read(&obj->frontbuffer_bits));
if (obj->pages && obj->madv == I915_MADV_WILLNEED &&
dev_priv->quirks & QUIRK_PIN_SWIZZLED_PAGES &&
if (discard_backing_storage(obj))
obj->madv = I915_MADV_DONTNEED;
i915_gem_object_put_pages(obj);
- i915_gem_object_free_mmap_offset(obj);
BUG_ON(obj->pages);
struct i915_vma *i915_gem_obj_to_ggtt_view(struct drm_i915_gem_object *obj,
const struct i915_ggtt_view *view)
{
- struct drm_device *dev = obj->base.dev;
- struct drm_i915_private *dev_priv = to_i915(dev);
- struct i915_ggtt *ggtt = &dev_priv->ggtt;
struct i915_vma *vma;
- BUG_ON(!view);
+ GEM_BUG_ON(!view);
list_for_each_entry(vma, &obj->vma_list, obj_link)
- if (vma->vm == &ggtt->base &&
+ if (i915_vma_is_ggtt(vma) &&
i915_ggtt_view_equal(&vma->ggtt_view, view))
return vma;
return NULL;
}
-void i915_gem_vma_destroy(struct i915_vma *vma)
-{
- WARN_ON(vma->node.allocated);
-
- /* Keep the vma as a placeholder in the execbuffer reservation lists */
- if (!list_empty(&vma->exec_list))
- return;
-
- if (!vma->is_ggtt)
- i915_ppgtt_put(i915_vm_to_ppgtt(vma->vm));
-
- list_del(&vma->obj_link);
-
- kmem_cache_free(to_i915(vma->obj->base.dev)->vmas, vma);
-}
-
-static void
-i915_gem_stop_engines(struct drm_device *dev)
-{
- struct drm_i915_private *dev_priv = dev->dev_private;
- struct intel_engine_cs *engine;
-
- for_each_engine(engine, dev_priv)
- dev_priv->gt.stop_engine(engine);
-}
-
int
i915_gem_suspend(struct drm_device *dev)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
int ret = 0;
+ intel_suspend_gt_powersave(dev_priv);
+
mutex_lock(&dev->struct_mutex);
- ret = i915_gpu_idle(dev);
+
+ /* We have to flush all the executing contexts to main memory so
+ * that they can saved in the hibernation image. To ensure the last
+ * context image is coherent, we have to switch away from it. That
+ * leaves the dev_priv->kernel_context still active when
+ * we actually suspend, and its image in memory may not match the GPU
+ * state. Fortunately, the kernel_context is disposable and we do
+ * not rely on its state.
+ */
+ ret = i915_gem_switch_to_kernel_context(dev_priv);
+ if (ret)
+ goto err;
+
+ ret = i915_gem_wait_for_idle(dev_priv);
if (ret)
goto err;
- i915_gem_retire_requests(dev);
+ i915_gem_retire_requests(dev_priv);
- i915_gem_stop_engines(dev);
+ i915_gem_context_lost(dev_priv);
mutex_unlock(&dev->struct_mutex);
cancel_delayed_work_sync(&dev_priv->gpu_error.hangcheck_work);
- cancel_delayed_work_sync(&dev_priv->mm.retire_work);
- flush_delayed_work(&dev_priv->mm.idle_work);
+ cancel_delayed_work_sync(&dev_priv->gt.retire_work);
+ flush_delayed_work(&dev_priv->gt.idle_work);
/* Assert that we sucessfully flushed all the work and
* reset the GPU back to its idle, low power state.
*/
- WARN_ON(dev_priv->mm.busy);
+ WARN_ON(dev_priv->gt.awake);
return 0;
return ret;
}
-int i915_gem_l3_remap(struct drm_i915_gem_request *req, int slice)
+void i915_gem_resume(struct drm_device *dev)
{
- struct intel_engine_cs *engine = req->engine;
- struct drm_device *dev = engine->dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
- u32 *remap_info = dev_priv->l3_parity.remap_info[slice];
- int i, ret;
-
- if (!HAS_L3_DPF(dev) || !remap_info)
- return 0;
+ struct drm_i915_private *dev_priv = to_i915(dev);
- ret = intel_ring_begin(req, GEN7_L3LOG_SIZE / 4 * 3);
- if (ret)
- return ret;
+ mutex_lock(&dev->struct_mutex);
+ i915_gem_restore_gtt_mappings(dev);
- /*
- * Note: We do not worry about the concurrent register cacheline hang
- * here because no other code should access these registers other than
- * at initialization time.
+ /* As we didn't flush the kernel context before suspend, we cannot
+ * guarantee that the context image is complete. So let's just reset
+ * it and start again.
*/
- for (i = 0; i < GEN7_L3LOG_SIZE / 4; i++) {
- intel_ring_emit(engine, MI_LOAD_REGISTER_IMM(1));
- intel_ring_emit_reg(engine, GEN7_L3LOG(slice, i));
- intel_ring_emit(engine, remap_info[i]);
- }
-
- intel_ring_advance(engine);
+ if (i915.enable_execlists)
+ intel_lr_context_reset(dev_priv, dev_priv->kernel_context);
- return ret;
+ mutex_unlock(&dev->struct_mutex);
}
void i915_gem_init_swizzling(struct drm_device *dev)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
if (INTEL_INFO(dev)->gen < 5 ||
dev_priv->mm.bit_6_swizzle_x == I915_BIT_6_SWIZZLE_NONE)
static void init_unused_ring(struct drm_device *dev, u32 base)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
I915_WRITE(RING_CTL(base), 0);
I915_WRITE(RING_HEAD(base), 0);
}
}
-int i915_gem_init_engines(struct drm_device *dev)
-{
- struct drm_i915_private *dev_priv = dev->dev_private;
- int ret;
-
- ret = intel_init_render_ring_buffer(dev);
- if (ret)
- return ret;
-
- if (HAS_BSD(dev)) {
- ret = intel_init_bsd_ring_buffer(dev);
- if (ret)
- goto cleanup_render_ring;
- }
-
- if (HAS_BLT(dev)) {
- ret = intel_init_blt_ring_buffer(dev);
- if (ret)
- goto cleanup_bsd_ring;
- }
-
- if (HAS_VEBOX(dev)) {
- ret = intel_init_vebox_ring_buffer(dev);
- if (ret)
- goto cleanup_blt_ring;
- }
-
- if (HAS_BSD2(dev)) {
- ret = intel_init_bsd2_ring_buffer(dev);
- if (ret)
- goto cleanup_vebox_ring;
- }
-
- return 0;
-
-cleanup_vebox_ring:
- intel_cleanup_engine(&dev_priv->engine[VECS]);
-cleanup_blt_ring:
- intel_cleanup_engine(&dev_priv->engine[BCS]);
-cleanup_bsd_ring:
- intel_cleanup_engine(&dev_priv->engine[VCS]);
-cleanup_render_ring:
- intel_cleanup_engine(&dev_priv->engine[RCS]);
-
- return ret;
-}
-
int
i915_gem_init_hw(struct drm_device *dev)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
struct intel_engine_cs *engine;
- int ret, j;
+ int ret;
/* Double layer security blanket, see i915_gem_init() */
intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
intel_mocs_init_l3cc_table(dev);
/* We can't enable contexts until all firmware is loaded */
- if (HAS_GUC_UCODE(dev)) {
- ret = intel_guc_ucode_load(dev);
- if (ret) {
- DRM_ERROR("Failed to initialize GuC, error %d\n", ret);
- ret = -EIO;
- goto out;
- }
- }
-
- /*
- * Increment the next seqno by 0x100 so we have a visible break
- * on re-initialisation
- */
- ret = i915_gem_set_seqno(dev, dev_priv->next_seqno+0x100);
+ ret = intel_guc_setup(dev);
if (ret)
goto out;
- /* Now it is safe to go back round and do everything else: */
- for_each_engine(engine, dev_priv) {
- struct drm_i915_gem_request *req;
-
- req = i915_gem_request_alloc(engine, NULL);
- if (IS_ERR(req)) {
- ret = PTR_ERR(req);
- break;
- }
+out:
+ intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+ return ret;
+}
- if (engine->id == RCS) {
- for (j = 0; j < NUM_L3_SLICES(dev); j++) {
- ret = i915_gem_l3_remap(req, j);
- if (ret)
- goto err_request;
- }
- }
+bool intel_sanitize_semaphores(struct drm_i915_private *dev_priv, int value)
+{
+ if (INTEL_INFO(dev_priv)->gen < 6)
+ return false;
- ret = i915_ppgtt_init_ring(req);
- if (ret)
- goto err_request;
+ /* TODO: make semaphores and Execlists play nicely together */
+ if (i915.enable_execlists)
+ return false;
- ret = i915_gem_context_enable(req);
- if (ret)
- goto err_request;
+ if (value >= 0)
+ return value;
-err_request:
- i915_add_request_no_flush(req);
- if (ret) {
- DRM_ERROR("Failed to enable %s, error=%d\n",
- engine->name, ret);
- i915_gem_cleanup_engines(dev);
- break;
- }
- }
+#ifdef CONFIG_INTEL_IOMMU
+ /* Enable semaphores on SNB when IO remapping is off */
+ if (INTEL_INFO(dev_priv)->gen == 6 && intel_iommu_gfx_mapped)
+ return false;
+#endif
-out:
- intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
- return ret;
+ return true;
}
int i915_gem_init(struct drm_device *dev)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
int ret;
- i915.enable_execlists = intel_sanitize_enable_execlists(dev,
- i915.enable_execlists);
-
mutex_lock(&dev->struct_mutex);
if (!i915.enable_execlists) {
- dev_priv->gt.execbuf_submit = i915_gem_ringbuffer_submission;
- dev_priv->gt.init_engines = i915_gem_init_engines;
- dev_priv->gt.cleanup_engine = intel_cleanup_engine;
- dev_priv->gt.stop_engine = intel_stop_engine;
+ dev_priv->gt.cleanup_engine = intel_engine_cleanup;
} else {
- dev_priv->gt.execbuf_submit = intel_execlists_submission;
- dev_priv->gt.init_engines = intel_logical_rings_init;
dev_priv->gt.cleanup_engine = intel_logical_ring_cleanup;
- dev_priv->gt.stop_engine = intel_logical_ring_stop;
}
/* This is just a security blanket to placate dragons.
*/
intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
- ret = i915_gem_init_userptr(dev);
+ i915_gem_init_userptr(dev_priv);
+
+ ret = i915_gem_init_ggtt(dev_priv);
if (ret)
goto out_unlock;
- i915_gem_init_ggtt(dev);
-
ret = i915_gem_context_init(dev);
if (ret)
goto out_unlock;
- ret = dev_priv->gt.init_engines(dev);
+ ret = intel_engines_init(dev);
if (ret)
goto out_unlock;
ret = i915_gem_init_hw(dev);
if (ret == -EIO) {
- /* Allow ring initialisation to fail by marking the GPU as
+ /* Allow engine initialisation to fail by marking the GPU as
* wedged. But we only want to do this where the GPU is angry,
* for all other failure, such as an allocation failure, bail.
*/
void
i915_gem_cleanup_engines(struct drm_device *dev)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
struct intel_engine_cs *engine;
for_each_engine(engine, dev_priv)
dev_priv->gt.cleanup_engine(engine);
-
- if (i915.enable_execlists)
- /*
- * Neither the BIOS, ourselves or any other kernel
- * expects the system to be in execlists mode on startup,
- * so we need to reset the GPU back to legacy mode.
- */
- intel_gpu_reset(dev, ALL_ENGINES);
}
static void
init_engine_lists(struct intel_engine_cs *engine)
{
- INIT_LIST_HEAD(&engine->active_list);
INIT_LIST_HEAD(&engine->request_list);
}
void
i915_gem_load_init_fences(struct drm_i915_private *dev_priv)
{
- struct drm_device *dev = dev_priv->dev;
+ struct drm_device *dev = &dev_priv->drm;
if (INTEL_INFO(dev_priv)->gen >= 7 && !IS_VALLEYVIEW(dev_priv) &&
!IS_CHERRYVIEW(dev_priv))
else
dev_priv->num_fence_regs = 8;
- if (intel_vgpu_active(dev))
+ if (intel_vgpu_active(dev_priv))
dev_priv->num_fence_regs =
I915_READ(vgtif_reg(avail_rs.fence_num));
void
i915_gem_load_init(struct drm_device *dev)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(dev);
int i;
dev_priv->objects =
dev_priv->requests =
kmem_cache_create("i915_gem_request",
sizeof(struct drm_i915_gem_request), 0,
- SLAB_HWCACHE_ALIGN,
+ SLAB_HWCACHE_ALIGN |
+ SLAB_RECLAIM_ACCOUNT |
+ SLAB_DESTROY_BY_RCU,
NULL);
- INIT_LIST_HEAD(&dev_priv->vm_list);
INIT_LIST_HEAD(&dev_priv->context_list);
INIT_LIST_HEAD(&dev_priv->mm.unbound_list);
INIT_LIST_HEAD(&dev_priv->mm.bound_list);
init_engine_lists(&dev_priv->engine[i]);
for (i = 0; i < I915_MAX_NUM_FENCES; i++)
INIT_LIST_HEAD(&dev_priv->fence_regs[i].lru_list);
- INIT_DELAYED_WORK(&dev_priv->mm.retire_work,
+ INIT_DELAYED_WORK(&dev_priv->gt.retire_work,
i915_gem_retire_work_handler);
- INIT_DELAYED_WORK(&dev_priv->mm.idle_work,
+ INIT_DELAYED_WORK(&dev_priv->gt.idle_work,
i915_gem_idle_work_handler);
+ init_waitqueue_head(&dev_priv->gpu_error.wait_queue);
init_waitqueue_head(&dev_priv->gpu_error.reset_queue);
dev_priv->relative_constants_mode = I915_EXEC_CONSTANTS_REL_GENERAL;
- /*
- * Set initial sequence number for requests.
- * Using this number allows the wraparound to happen early,
- * catching any obvious problems.
- */
- dev_priv->next_seqno = ((u32)~0 - 0x1100);
- dev_priv->last_seqno = ((u32)~0 - 0x1101);
-
INIT_LIST_HEAD(&dev_priv->mm.fence_list);
init_waitqueue_head(&dev_priv->pending_flip_queue);
dev_priv->mm.interruptible = true;
- mutex_init(&dev_priv->fb_tracking.lock);
+ spin_lock_init(&dev_priv->fb_tracking.lock);
}
void i915_gem_load_cleanup(struct drm_device *dev)
kmem_cache_destroy(dev_priv->requests);
kmem_cache_destroy(dev_priv->vmas);
kmem_cache_destroy(dev_priv->objects);
+
+ /* And ensure that our DESTROY_BY_RCU slabs are truly destroyed */
+ rcu_barrier();
+}
+
+int i915_gem_freeze_late(struct drm_i915_private *dev_priv)
+{
+ struct drm_i915_gem_object *obj;
+
+ /* Called just before we write the hibernation image.
+ *
+ * We need to update the domain tracking to reflect that the CPU
+ * will be accessing all the pages to create and restore from the
+ * hibernation, and so upon restoration those pages will be in the
+ * CPU domain.
+ *
+ * To make sure the hibernation image contains the latest state,
+ * we update that state just before writing out the image.
+ */
+
+ list_for_each_entry(obj, &dev_priv->mm.unbound_list, global_list) {
+ obj->base.read_domains = I915_GEM_DOMAIN_CPU;
+ obj->base.write_domain = I915_GEM_DOMAIN_CPU;
+ }
+
+ list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
+ obj->base.read_domains = I915_GEM_DOMAIN_CPU;
+ obj->base.write_domain = I915_GEM_DOMAIN_CPU;
+ }
+
+ return 0;
}
void i915_gem_release(struct drm_device *dev, struct drm_file *file)
{
struct drm_i915_file_private *file_priv = file->driver_priv;
+ struct drm_i915_gem_request *request;
/* Clean up our request list when the client is going away, so that
* later retire_requests won't dereference our soon-to-be-gone
* file_priv.
*/
spin_lock(&file_priv->mm.lock);
- while (!list_empty(&file_priv->mm.request_list)) {
- struct drm_i915_gem_request *request;
-
- request = list_first_entry(&file_priv->mm.request_list,
- struct drm_i915_gem_request,
- client_list);
- list_del(&request->client_list);
+ list_for_each_entry(request, &file_priv->mm.request_list, client_list)
request->file_priv = NULL;
- }
spin_unlock(&file_priv->mm.lock);
if (!list_empty(&file_priv->rps.link)) {
return -ENOMEM;
file->driver_priv = file_priv;
- file_priv->dev_priv = dev->dev_private;
+ file_priv->dev_priv = to_i915(dev);
file_priv->file = file;
INIT_LIST_HEAD(&file_priv->rps.link);
spin_lock_init(&file_priv->mm.lock);
INIT_LIST_HEAD(&file_priv->mm.request_list);
- file_priv->bsd_ring = -1;
+ file_priv->bsd_engine = -1;
ret = i915_gem_context_open(dev, file);
if (ret)
struct drm_i915_gem_object *new,
unsigned frontbuffer_bits)
{
+ /* Control of individual bits within the mask are guarded by
+ * the owning plane->mutex, i.e. we can never see concurrent
+ * manipulation of individual bits. But since the bitfield as a whole
+ * is updated using RMW, we need to use atomics in order to update
+ * the bits.
+ */
+ BUILD_BUG_ON(INTEL_FRONTBUFFER_BITS_PER_PIPE * I915_MAX_PIPES >
+ sizeof(atomic_t) * BITS_PER_BYTE);
+
if (old) {
- WARN_ON(!mutex_is_locked(&old->base.dev->struct_mutex));
- WARN_ON(!(old->frontbuffer_bits & frontbuffer_bits));
- old->frontbuffer_bits &= ~frontbuffer_bits;
+ WARN_ON(!(atomic_read(&old->frontbuffer_bits) & frontbuffer_bits));
+ atomic_andnot(frontbuffer_bits, &old->frontbuffer_bits);
}
if (new) {
- WARN_ON(!mutex_is_locked(&new->base.dev->struct_mutex));
- WARN_ON(new->frontbuffer_bits & frontbuffer_bits);
- new->frontbuffer_bits |= frontbuffer_bits;
+ WARN_ON(atomic_read(&new->frontbuffer_bits) & frontbuffer_bits);
+ atomic_or(frontbuffer_bits, &new->frontbuffer_bits);
}
}
u64 i915_gem_obj_offset(struct drm_i915_gem_object *o,
struct i915_address_space *vm)
{
- struct drm_i915_private *dev_priv = o->base.dev->dev_private;
+ struct drm_i915_private *dev_priv = to_i915(o->base.dev);
struct i915_vma *vma;
WARN_ON(vm == &dev_priv->mm.aliasing_ppgtt->base);
list_for_each_entry(vma, &o->vma_list, obj_link) {
- if (vma->is_ggtt &&
+ if (i915_vma_is_ggtt(vma) &&
vma->ggtt_view.type != I915_GGTT_VIEW_NORMAL)
continue;
if (vma->vm == vm)
u64 i915_gem_obj_ggtt_offset_view(struct drm_i915_gem_object *o,
const struct i915_ggtt_view *view)
{
- struct drm_i915_private *dev_priv = to_i915(o->base.dev);
- struct i915_ggtt *ggtt = &dev_priv->ggtt;
struct i915_vma *vma;
list_for_each_entry(vma, &o->vma_list, obj_link)
- if (vma->vm == &ggtt->base &&
+ if (i915_vma_is_ggtt(vma) &&
i915_ggtt_view_equal(&vma->ggtt_view, view))
return vma->node.start;
struct i915_vma *vma;
list_for_each_entry(vma, &o->vma_list, obj_link) {
- if (vma->is_ggtt &&
+ if (i915_vma_is_ggtt(vma) &&
vma->ggtt_view.type != I915_GGTT_VIEW_NORMAL)
continue;
if (vma->vm == vm && drm_mm_node_allocated(&vma->node))
bool i915_gem_obj_ggtt_bound_view(struct drm_i915_gem_object *o,
const struct i915_ggtt_view *view)
{
- struct drm_i915_private *dev_priv = to_i915(o->base.dev);
- struct i915_ggtt *ggtt = &dev_priv->ggtt;
struct i915_vma *vma;
list_for_each_entry(vma, &o->vma_list, obj_link)
- if (vma->vm == &ggtt->base &&
+ if (i915_vma_is_ggtt(vma) &&
i915_ggtt_view_equal(&vma->ggtt_view, view) &&
drm_mm_node_allocated(&vma->node))
return true;
return false;
}
-bool i915_gem_obj_bound_any(struct drm_i915_gem_object *o)
-{
- struct i915_vma *vma;
-
- list_for_each_entry(vma, &o->vma_list, obj_link)
- if (drm_mm_node_allocated(&vma->node))
- return true;
-
- return false;
-}
-
-unsigned long i915_gem_obj_size(struct drm_i915_gem_object *o,
- struct i915_address_space *vm)
+unsigned long i915_gem_obj_ggtt_size(struct drm_i915_gem_object *o)
{
- struct drm_i915_private *dev_priv = o->base.dev->dev_private;
struct i915_vma *vma;
- WARN_ON(vm == &dev_priv->mm.aliasing_ppgtt->base);
-
- BUG_ON(list_empty(&o->vma_list));
+ GEM_BUG_ON(list_empty(&o->vma_list));
list_for_each_entry(vma, &o->vma_list, obj_link) {
- if (vma->is_ggtt &&
- vma->ggtt_view.type != I915_GGTT_VIEW_NORMAL)
- continue;
- if (vma->vm == vm)
+ if (i915_vma_is_ggtt(vma) &&
+ vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL)
return vma->node.size;
}
+
return 0;
}
{
struct i915_vma *vma;
list_for_each_entry(vma, &obj->vma_list, obj_link)
- if (vma->pin_count > 0)
+ if (i915_vma_is_pinned(vma))
return true;
return false;
struct page *page;
/* Only default objects have per-page dirty tracking */
- if (WARN_ON((obj->ops->flags & I915_GEM_OBJECT_HAS_STRUCT_PAGE) == 0))
+ if (WARN_ON(!i915_gem_object_has_struct_page(obj)))
return NULL;
page = i915_gem_object_get_page(obj, n);
size_t bytes;
int ret;
- obj = i915_gem_alloc_object(dev, round_up(size, PAGE_SIZE));
- if (IS_ERR_OR_NULL(obj))
+ obj = i915_gem_object_create(dev, round_up(size, PAGE_SIZE));
+ if (IS_ERR(obj))
return obj;
ret = i915_gem_object_set_to_cpu_domain(obj, true);
return obj;
fail:
- drm_gem_object_unreference(&obj->base);
+ i915_gem_object_put(obj);
return ERR_PTR(ret);
}
projects / karo-tx-linux.git / blobdiff