int i915_gem_render_state_init(struct drm_i915_gem_request *req)
{
struct render_state so;
int ret;
ret = i915_gem_render_state_prepare(req->ring, &so);
if (ret)
return ret;
if (so.rodata == NULL)
return 0;
ret = req->ring->dispatch_execbuffer(req, so.ggtt_offset,
so.rodata->batch_items * 4,
I915_DISPATCH_SECURE);
if (ret)
goto out;
if (so.aux_batch_size > 8) {
ret = req->ring->dispatch_execbuffer(req,
(so.ggtt_offset +
so.aux_batch_offset),
so.aux_batch_size,
I915_DISPATCH_SECURE);
if (ret)
goto out;
}
i915_vma_move_to_active(i915_gem_obj_to_ggtt(so.obj), req);
out:
i915_gem_render_state_fini(&so);
return ret;
}
/**
* i915_gem_set_tiling - IOCTL handler to set tiling mode
* @dev: DRM device
* @data: data pointer for the ioctl
* @file: DRM file for the ioctl call
*
* Sets the tiling mode of an object, returning the required swizzling of
* bit 6 of addresses in the object.
*
* Called by the user via ioctl.
*
* Returns:
* Zero on success, negative errno on failure.
*/
int
i915_gem_set_tiling(struct drm_device *dev, void *data,
struct drm_file *file)
{
struct drm_i915_gem_set_tiling *args = data;
struct drm_i915_private *dev_priv = to_i915(dev);
struct drm_i915_gem_object *obj;
int ret = 0;
obj = to_intel_bo(drm_gem_object_lookup(file, args->handle));
if (&obj->base == NULL)
return -ENOENT;
if (!i915_tiling_ok(dev,
args->stride, obj->base.size, args->tiling_mode)) {
drm_gem_object_unreference_unlocked(&obj->base);
return -EINVAL;
}
intel_runtime_pm_get(dev_priv);
mutex_lock(&dev->struct_mutex);
if (obj->pin_display || obj->framebuffer_references) {
ret = -EBUSY;
goto err;
}
if (args->tiling_mode == I915_TILING_NONE) {
args->swizzle_mode = I915_BIT_6_SWIZZLE_NONE;
args->stride = 0;
} else {
if (args->tiling_mode == I915_TILING_X)
args->swizzle_mode = dev_priv->mm.bit_6_swizzle_x;
else
args->swizzle_mode = dev_priv->mm.bit_6_swizzle_y;
/* Hide bit 17 swizzling from the user. This prevents old Mesa
* from aborting the application on sw fallbacks to bit 17,
* and we use the pread/pwrite bit17 paths to swizzle for it.
* If there was a user that was relying on the swizzle
* information for drm_intel_bo_map()ed reads/writes this would
* break it, but we don't have any of those.
*/
if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_17)
args->swizzle_mode = I915_BIT_6_SWIZZLE_9;
if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_10_17)
args->swizzle_mode = I915_BIT_6_SWIZZLE_9_10;
/* If we can't handle the swizzling, make it untiled. */
if (args->swizzle_mode == I915_BIT_6_SWIZZLE_UNKNOWN) {
args->tiling_mode = I915_TILING_NONE;
args->swizzle_mode = I915_BIT_6_SWIZZLE_NONE;
args->stride = 0;
}
}
if (args->tiling_mode != obj->tiling_mode ||
args->stride != obj->stride) {
/* We need to rebind the object if its current allocation
* no longer meets the alignment restrictions for its new
* tiling mode. Otherwise we can just leave it alone, but
* need to ensure that any fence register is updated before
* the next fenced (either through the GTT or by the BLT unit
* on older GPUs) access.
*
* After updating the tiling parameters, we then flag whether
* we need to update an associated fence register. Note this
* has to also include the unfenced register the GPU uses
* whilst executing a fenced command for an untiled object.
*/
if (obj->map_and_fenceable &&
!i915_gem_object_fence_ok(obj, args->tiling_mode))
ret = i915_vma_unbind(i915_gem_obj_to_ggtt(obj));
if (ret == 0) {
if (obj->pages &&
obj->madv == I915_MADV_WILLNEED &&
dev_priv->quirks & QUIRK_PIN_SWIZZLED_PAGES) {
if (args->tiling_mode == I915_TILING_NONE)
i915_gem_object_unpin_pages(obj);
if (obj->tiling_mode == I915_TILING_NONE)
i915_gem_object_pin_pages(obj);
}
obj->fence_dirty =
obj->last_fenced_req ||
obj->fence_reg != I915_FENCE_REG_NONE;
obj->tiling_mode = args->tiling_mode;
obj->stride = args->stride;
/* Force the fence to be reacquired for GTT access */
i915_gem_release_mmap(obj);
}
}
/* we have to maintain this existing ABI... */
args->stride = obj->stride;
args->tiling_mode = obj->tiling_mode;
/* Try to preallocate memory required to save swizzling on put-pages */
if (i915_gem_object_needs_bit17_swizzle(obj)) {
if (obj->bit_17 == NULL) {
obj->bit_17 = kcalloc(BITS_TO_LONGS(obj->base.size >> PAGE_SHIFT),
sizeof(long), GFP_KERNEL);
}
} else {
static int do_switch(struct i915_hw_context *to)
{
struct intel_ring_buffer *ring = to->ring;
struct i915_hw_context *from = ring->last_context;
u32 hw_flags = 0;
int ret, i;
BUG_ON(from != NULL && from->obj != NULL && from->obj->pin_count == 0);
if (from == to && !to->remap_slice)
return 0;
ret = i915_gem_obj_ggtt_pin(to->obj, CONTEXT_ALIGN, false, false);
if (ret)
return ret;
/*
* Pin can switch back to the default context if we end up calling into
* evict_everything - as a last ditch gtt defrag effort that also
* switches to the default context. Hence we need to reload from here.
*/
from = ring->last_context;
/*
* Clear this page out of any CPU caches for coherent swap-in/out. Note
* that thanks to write = false in this call and us not setting any gpu
* write domains when putting a context object onto the active list
* (when switching away from it), this won't block.
*
* XXX: We need a real interface to do this instead of trickery.
*/
ret = i915_gem_object_set_to_gtt_domain(to->obj, false);
if (ret) {
i915_gem_object_unpin(to->obj);
return ret;
}
if (!to->obj->has_global_gtt_mapping)
i915_gem_gtt_bind_object(to->obj, to->obj->cache_level);
if (!to->is_initialized || is_default_context(to))
hw_flags |= MI_RESTORE_INHIBIT;
ret = mi_set_context(ring, to, hw_flags);
if (ret) {
i915_gem_object_unpin(to->obj);
return ret;
}
for (i = 0; i < MAX_L3_SLICES; i++) {
if (!(to->remap_slice & (1<<i)))
continue;
ret = i915_gem_l3_remap(ring, i);
/* If it failed, try again next round */
if (ret)
DRM_DEBUG_DRIVER("L3 remapping failed\n");
else
to->remap_slice &= ~(1<<i);
}
/* The backing object for the context is done after switching to the
* *next* context. Therefore we cannot retire the previous context until
* the next context has already started running. In fact, the below code
* is a bit suboptimal because the retiring can occur simply after the
* MI_SET_CONTEXT instead of when the next seqno has completed.
*/
if (from != NULL) {
from->obj->base.read_domains = I915_GEM_DOMAIN_INSTRUCTION;
i915_vma_move_to_active(i915_gem_obj_to_ggtt(from->obj), ring);
/* As long as MI_SET_CONTEXT is serializing, ie. it flushes the
* whole damn pipeline, we don't need to explicitly mark the
* object dirty. The only exception is that the context must be
* correct in case the object gets swapped out. Ideally we'd be
* able to defer doing this until we know the object would be
* swapped, but there is no way to do that yet.
*/
from->obj->dirty = 1;
BUG_ON(from->obj->ring != ring);
/* obj is kept alive until the next request by its active ref */
i915_gem_object_unpin(from->obj);
i915_gem_context_unreference(from);
}
i915_gem_context_reference(to);
ring->last_context = to;
to->is_initialized = true;
return 0;
}
