drm/i915: add render state initialization
HW guys say that it is not a cool idea to let device go into rc6 without proper 3d pipeline state. For each new uninitialized context, generate a valid null render state to be run on context creation. This patch introduces a skeleton with empty states. v2: - No need to vmap (Chris Wilson) - use .c files for state (Daniel Vetter) - no need to flush as i915_add_request does it - remove parameter for batch alloc size - don't wait for the init (Ben Widawsky) v3: - move to cpu/gpu (Chris Wilson) Tested-by: Kristen Carlson Accardi <kristen@linux.intel.com> (v1) Tested-by: Oscar Mateo <oscar.mateo@intel.com> Reviewed-by: Damien Lespiau <damien.lespiau@intel.com> Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This commit is contained in:
Родитель
8b1bc9b4f1
Коммит
9d0a6fa6c5
|
@ -18,6 +18,7 @@ i915-$(CONFIG_DEBUG_FS) += i915_debugfs.o
|
|||
# GEM code
|
||||
i915-y += i915_cmd_parser.o \
|
||||
i915_gem_context.o \
|
||||
i915_gem_render_state.o \
|
||||
i915_gem_debug.o \
|
||||
i915_gem_dmabuf.o \
|
||||
i915_gem_evict.o \
|
||||
|
@ -32,6 +33,11 @@ i915-y += i915_cmd_parser.o \
|
|||
intel_ringbuffer.o \
|
||||
intel_uncore.o
|
||||
|
||||
# autogenerated null render state
|
||||
i915-y += intel_renderstate_gen6.o \
|
||||
intel_renderstate_gen7.o \
|
||||
intel_renderstate_gen8.o
|
||||
|
||||
# modesetting core code
|
||||
i915-y += intel_bios.o \
|
||||
intel_display.o \
|
||||
|
|
|
@ -2388,6 +2388,8 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
|
|||
int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
|
||||
struct drm_file *file);
|
||||
|
||||
/* i915_gem_render_state.c */
|
||||
int i915_gem_render_state_init(struct intel_ring_buffer *ring);
|
||||
/* i915_gem_evict.c */
|
||||
int __must_check i915_gem_evict_something(struct drm_device *dev,
|
||||
struct i915_address_space *vm,
|
||||
|
|
|
@ -699,6 +699,12 @@ static int do_switch(struct intel_ring_buffer *ring,
|
|||
/* obj is kept alive until the next request by its active ref */
|
||||
i915_gem_object_ggtt_unpin(from->obj);
|
||||
i915_gem_context_unreference(from);
|
||||
} else {
|
||||
if (to->is_initialized == false) {
|
||||
ret = i915_gem_render_state_init(ring);
|
||||
if (ret)
|
||||
DRM_ERROR("init render state: %d\n", ret);
|
||||
}
|
||||
}
|
||||
|
||||
to->is_initialized = true;
|
||||
|
|
|
@ -0,0 +1,194 @@
|
|||
/*
|
||||
* Copyright © 2014 Intel Corporation
|
||||
*
|
||||
* Permission is hereby granted, free of charge, to any person obtaining a
|
||||
* copy of this software and associated documentation files (the "Software"),
|
||||
* to deal in the Software without restriction, including without limitation
|
||||
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
|
||||
* and/or sell copies of the Software, and to permit persons to whom the
|
||||
* Software is furnished to do so, subject to the following conditions:
|
||||
*
|
||||
* The above copyright notice and this permission notice (including the next
|
||||
* paragraph) shall be included in all copies or substantial portions of the
|
||||
* Software.
|
||||
*
|
||||
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
||||
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
||||
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
|
||||
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
||||
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
|
||||
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
|
||||
* IN THE SOFTWARE.
|
||||
*
|
||||
* Authors:
|
||||
* Mika Kuoppala <mika.kuoppala@intel.com>
|
||||
*
|
||||
*/
|
||||
|
||||
#include "i915_drv.h"
|
||||
#include "intel_renderstate.h"
|
||||
|
||||
struct i915_render_state {
|
||||
struct drm_i915_gem_object *obj;
|
||||
unsigned long ggtt_offset;
|
||||
void *batch;
|
||||
u32 size;
|
||||
u32 len;
|
||||
};
|
||||
|
||||
static struct i915_render_state *render_state_alloc(struct drm_device *dev)
|
||||
{
|
||||
struct i915_render_state *so;
|
||||
struct page *page;
|
||||
int ret;
|
||||
|
||||
so = kzalloc(sizeof(*so), GFP_KERNEL);
|
||||
if (!so)
|
||||
return ERR_PTR(-ENOMEM);
|
||||
|
||||
so->obj = i915_gem_alloc_object(dev, 4096);
|
||||
if (so->obj == NULL) {
|
||||
ret = -ENOMEM;
|
||||
goto free;
|
||||
}
|
||||
so->size = 4096;
|
||||
|
||||
ret = i915_gem_obj_ggtt_pin(so->obj, 4096, 0);
|
||||
if (ret)
|
||||
goto free_gem;
|
||||
|
||||
BUG_ON(so->obj->pages->nents != 1);
|
||||
page = sg_page(so->obj->pages->sgl);
|
||||
|
||||
so->batch = kmap(page);
|
||||
if (!so->batch) {
|
||||
ret = -ENOMEM;
|
||||
goto unpin;
|
||||
}
|
||||
|
||||
so->ggtt_offset = i915_gem_obj_ggtt_offset(so->obj);
|
||||
|
||||
return so;
|
||||
unpin:
|
||||
i915_gem_object_ggtt_unpin(so->obj);
|
||||
free_gem:
|
||||
drm_gem_object_unreference(&so->obj->base);
|
||||
free:
|
||||
kfree(so);
|
||||
return ERR_PTR(ret);
|
||||
}
|
||||
|
||||
static void render_state_free(struct i915_render_state *so)
|
||||
{
|
||||
kunmap(so->batch);
|
||||
i915_gem_object_ggtt_unpin(so->obj);
|
||||
drm_gem_object_unreference(&so->obj->base);
|
||||
kfree(so);
|
||||
}
|
||||
|
||||
static const struct intel_renderstate_rodata *
|
||||
render_state_get_rodata(struct drm_device *dev, const int gen)
|
||||
{
|
||||
switch (gen) {
|
||||
case 6:
|
||||
return &gen6_null_state;
|
||||
case 7:
|
||||
return &gen7_null_state;
|
||||
case 8:
|
||||
return &gen8_null_state;
|
||||
}
|
||||
|
||||
return NULL;
|
||||
}
|
||||
|
||||
static int render_state_setup(const int gen,
|
||||
const struct intel_renderstate_rodata *rodata,
|
||||
struct i915_render_state *so)
|
||||
{
|
||||
const u64 goffset = i915_gem_obj_ggtt_offset(so->obj);
|
||||
u32 reloc_index = 0;
|
||||
u32 * const d = so->batch;
|
||||
unsigned int i = 0;
|
||||
int ret;
|
||||
|
||||
if (!rodata || rodata->batch_items * 4 > so->size)
|
||||
return -EINVAL;
|
||||
|
||||
ret = i915_gem_object_set_to_cpu_domain(so->obj, true);
|
||||
if (ret)
|
||||
return ret;
|
||||
|
||||
while (i < rodata->batch_items) {
|
||||
u32 s = rodata->batch[i];
|
||||
|
||||
if (reloc_index < rodata->reloc_items &&
|
||||
i * 4 == rodata->reloc[reloc_index]) {
|
||||
|
||||
s += goffset & 0xffffffff;
|
||||
|
||||
/* We keep batch offsets max 32bit */
|
||||
if (gen >= 8) {
|
||||
if (i + 1 >= rodata->batch_items ||
|
||||
rodata->batch[i + 1] != 0)
|
||||
return -EINVAL;
|
||||
|
||||
d[i] = s;
|
||||
i++;
|
||||
s = (goffset & 0xffffffff00000000ull) >> 32;
|
||||
}
|
||||
|
||||
reloc_index++;
|
||||
}
|
||||
|
||||
d[i] = s;
|
||||
i++;
|
||||
}
|
||||
|
||||
ret = i915_gem_object_set_to_gtt_domain(so->obj, false);
|
||||
if (ret)
|
||||
return ret;
|
||||
|
||||
if (rodata->reloc_items != reloc_index) {
|
||||
DRM_ERROR("not all relocs resolved, %d out of %d\n",
|
||||
reloc_index, rodata->reloc_items);
|
||||
return -EINVAL;
|
||||
}
|
||||
|
||||
so->len = rodata->batch_items * 4;
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
int i915_gem_render_state_init(struct intel_ring_buffer *ring)
|
||||
{
|
||||
const int gen = INTEL_INFO(ring->dev)->gen;
|
||||
struct i915_render_state *so;
|
||||
const struct intel_renderstate_rodata *rodata;
|
||||
u32 seqno;
|
||||
int ret;
|
||||
|
||||
rodata = render_state_get_rodata(ring->dev, gen);
|
||||
if (rodata == NULL)
|
||||
return 0;
|
||||
|
||||
so = render_state_alloc(ring->dev);
|
||||
if (IS_ERR(so))
|
||||
return PTR_ERR(so);
|
||||
|
||||
ret = render_state_setup(gen, rodata, so);
|
||||
if (ret)
|
||||
goto out;
|
||||
|
||||
ret = ring->dispatch_execbuffer(ring,
|
||||
i915_gem_obj_ggtt_offset(so->obj),
|
||||
so->len,
|
||||
I915_DISPATCH_SECURE);
|
||||
if (ret)
|
||||
goto out;
|
||||
|
||||
ret = i915_add_request(ring, &seqno);
|
||||
|
||||
out:
|
||||
render_state_free(so);
|
||||
return ret;
|
||||
}
|
|
@ -0,0 +1,48 @@
|
|||
/*
|
||||
* Copyright © 2014 Intel Corporation
|
||||
*
|
||||
* Permission is hereby granted, free of charge, to any person obtaining a
|
||||
* copy of this software and associated documentation files (the "Software"),
|
||||
* to deal in the Software without restriction, including without limitation
|
||||
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
|
||||
* and/or sell copies of the Software, and to permit persons to whom the
|
||||
* Software is furnished to do so, subject to the following conditions:
|
||||
*
|
||||
* The above copyright notice and this permission notice (including the next
|
||||
* paragraph) shall be included in all copies or substantial portions of the
|
||||
* Software.
|
||||
*
|
||||
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
||||
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
||||
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
|
||||
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
||||
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
|
||||
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
|
||||
* DEALINGS IN THE SOFTWARE.
|
||||
*/
|
||||
|
||||
#ifndef _INTEL_RENDERSTATE_H
|
||||
#define _INTEL_RENDERSTATE_H
|
||||
|
||||
#include <linux/types.h>
|
||||
|
||||
struct intel_renderstate_rodata {
|
||||
const u32 *reloc;
|
||||
const u32 reloc_items;
|
||||
const u32 *batch;
|
||||
const u32 batch_items;
|
||||
};
|
||||
|
||||
extern const struct intel_renderstate_rodata gen6_null_state;
|
||||
extern const struct intel_renderstate_rodata gen7_null_state;
|
||||
extern const struct intel_renderstate_rodata gen8_null_state;
|
||||
|
||||
#define RO_RENDERSTATE(_g) \
|
||||
const struct intel_renderstate_rodata gen ## _g ## _null_state = { \
|
||||
.reloc = gen ## _g ## _null_state_relocs, \
|
||||
.reloc_items = sizeof(gen ## _g ## _null_state_relocs)/4, \
|
||||
.batch = gen ## _g ## _null_state_batch, \
|
||||
.batch_items = sizeof(gen ## _g ## _null_state_batch)/4, \
|
||||
}
|
||||
|
||||
#endif /* INTEL_RENDERSTATE_H */
|
|
@ -0,0 +1,10 @@
|
|||
#include "intel_renderstate.h"
|
||||
|
||||
static const u32 gen6_null_state_relocs[] = {
|
||||
};
|
||||
|
||||
static const u32 gen6_null_state_batch[] = {
|
||||
0x0a << 23, /* MI_BATCH_BUFFER_END */
|
||||
};
|
||||
|
||||
RO_RENDERSTATE(6);
|
|
@ -0,0 +1,10 @@
|
|||
#include "intel_renderstate.h"
|
||||
|
||||
static const u32 gen7_null_state_relocs[] = {
|
||||
};
|
||||
|
||||
static const u32 gen7_null_state_batch[] = {
|
||||
0x0a << 23, /* MI_BATCH_BUFFER_END */
|
||||
};
|
||||
|
||||
RO_RENDERSTATE(7);
|
|
@ -0,0 +1,10 @@
|
|||
#include "intel_renderstate.h"
|
||||
|
||||
static const u32 gen8_null_state_relocs[] = {
|
||||
};
|
||||
|
||||
static const u32 gen8_null_state_batch[] = {
|
||||
0x0a << 23, /* MI_BATCH_BUFFER_END */
|
||||
};
|
||||
|
||||
RO_RENDERSTATE(8);
|
Загрузка…
Ссылка в новой задаче