2013-04-01 20:26:43 +00:00
|
|
|
/*
|
2013-08-18 14:16:15 +00:00
|
|
|
* Copyright 2011-2013 Blender Foundation
|
2013-04-01 20:26:43 +00:00
|
|
|
*
|
2013-08-18 14:16:15 +00:00
|
|
|
* Licensed under the Apache License, Version 2.0 (the "License");
|
|
|
|
* you may not use this file except in compliance with the License.
|
|
|
|
* You may obtain a copy of the License at
|
2013-04-01 20:26:43 +00:00
|
|
|
*
|
2013-08-18 14:16:15 +00:00
|
|
|
* http://www.apache.org/licenses/LICENSE-2.0
|
2013-04-01 20:26:43 +00:00
|
|
|
*
|
2013-08-18 14:16:15 +00:00
|
|
|
* Unless required by applicable law or agreed to in writing, software
|
|
|
|
* distributed under the License is distributed on an "AS IS" BASIS,
|
|
|
|
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
|
* See the License for the specific language governing permissions and
|
2014-12-25 02:50:24 +01:00
|
|
|
* limitations under the License.
|
2013-04-01 20:26:43 +00:00
|
|
|
*/
|
|
|
|
|
2020-03-19 09:33:03 +01:00
|
|
|
#include "render/tables.h"
|
Cycles: Make all #include statements relative to cycles source directory
The idea is to make include statements more explicit and obvious where the
file is coming from, additionally reducing chance of wrong header being
picked up.
For example, it was not obvious whether bvh.h was refferring to builder
or traversal, whenter node.h is a generic graph node or a shader node
and cases like that.
Surely this might look obvious for the active developers, but after some
time of not touching the code it becomes less obvious where file is coming
from.
This was briefly mentioned in T50824 and seems @brecht is fine with such
explicitness, but need to agree with all active developers before committing
this.
Please note that this patch is lacking changes related on GPU/OpenCL
support. This will be solved if/when we all agree this is a good idea to move
forward.
Reviewers: brecht, lukasstockner97, maiself, nirved, dingto, juicyfruit, swerner
Reviewed By: lukasstockner97, maiself, nirved, dingto
Subscribers: brecht
Differential Revision: https://developer.blender.org/D2586
2017-03-28 20:39:14 +02:00
|
|
|
#include "device/device.h"
|
|
|
|
#include "render/scene.h"
|
2020-10-01 23:16:01 +02:00
|
|
|
#include "render/stats.h"
|
2013-04-01 20:26:43 +00:00
|
|
|
|
Cycles: Make all #include statements relative to cycles source directory
The idea is to make include statements more explicit and obvious where the
file is coming from, additionally reducing chance of wrong header being
picked up.
For example, it was not obvious whether bvh.h was refferring to builder
or traversal, whenter node.h is a generic graph node or a shader node
and cases like that.
Surely this might look obvious for the active developers, but after some
time of not touching the code it becomes less obvious where file is coming
from.
This was briefly mentioned in T50824 and seems @brecht is fine with such
explicitness, but need to agree with all active developers before committing
this.
Please note that this patch is lacking changes related on GPU/OpenCL
support. This will be solved if/when we all agree this is a good idea to move
forward.
Reviewers: brecht, lukasstockner97, maiself, nirved, dingto, juicyfruit, swerner
Reviewed By: lukasstockner97, maiself, nirved, dingto
Subscribers: brecht
Differential Revision: https://developer.blender.org/D2586
2017-03-28 20:39:14 +02:00
|
|
|
#include "util/util_logging.h"
|
2020-10-01 23:16:01 +02:00
|
|
|
#include "util/util_time.h"
|
2013-04-01 20:26:43 +00:00
|
|
|
|
|
|
|
CCL_NAMESPACE_BEGIN
|
|
|
|
|
2013-04-01 20:26:52 +00:00
|
|
|
/* Lookup Tables */
|
|
|
|
|
2013-04-01 20:26:43 +00:00
|
|
|
LookupTables::LookupTables()
|
|
|
|
{
|
Cycles: optimize device updates
This optimizes device updates (during user edits or frame changes in
the viewport) by avoiding unnecessary computations. To achieve this,
we use a combination of the sockets' update flags as well as some new
flags passed to the various managers when tagging for an update to tell
exactly what the tagging is for (e.g. shader was modified, object was
removed, etc.).
Besides avoiding recomputations, we also avoid resending to the devices
unmodified data arrays, thus reducing bandwidth usage. For OptiX and
Embree, BVH packing was also multithreaded.
The performance improvements may vary depending on the used device (CPU
or GPU), and the content of the scene. Simple scenes (e.g. with no adaptive
subdivision or volumes) rendered using OptiX will benefit from this work
the most.
On average, for a variety of animated scenes, this gives a 3x speedup.
Reviewed By: #cycles, brecht
Maniphest Tasks: T79174
Differential Revision: https://developer.blender.org/D9555
2021-01-22 15:01:26 +01:00
|
|
|
need_update_ = true;
|
2013-04-01 20:26:43 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
LookupTables::~LookupTables()
|
|
|
|
{
|
2019-04-17 06:17:24 +02:00
|
|
|
assert(lookup_tables.size() == 0);
|
2013-04-01 20:26:43 +00:00
|
|
|
}
|
|
|
|
|
2020-10-01 23:16:01 +02:00
|
|
|
void LookupTables::device_update(Device *, DeviceScene *dscene, Scene *scene)
|
2013-04-01 20:26:43 +00:00
|
|
|
{
|
Cycles: optimize device updates
This optimizes device updates (during user edits or frame changes in
the viewport) by avoiding unnecessary computations. To achieve this,
we use a combination of the sockets' update flags as well as some new
flags passed to the various managers when tagging for an update to tell
exactly what the tagging is for (e.g. shader was modified, object was
removed, etc.).
Besides avoiding recomputations, we also avoid resending to the devices
unmodified data arrays, thus reducing bandwidth usage. For OptiX and
Embree, BVH packing was also multithreaded.
The performance improvements may vary depending on the used device (CPU
or GPU), and the content of the scene. Simple scenes (e.g. with no adaptive
subdivision or volumes) rendered using OptiX will benefit from this work
the most.
On average, for a variety of animated scenes, this gives a 3x speedup.
Reviewed By: #cycles, brecht
Maniphest Tasks: T79174
Differential Revision: https://developer.blender.org/D9555
2021-01-22 15:01:26 +01:00
|
|
|
if (!need_update())
|
2019-04-17 06:17:24 +02:00
|
|
|
return;
|
2013-04-01 20:26:43 +00:00
|
|
|
|
2020-10-01 23:16:01 +02:00
|
|
|
scoped_callback_timer timer([scene](double time) {
|
|
|
|
if (scene->update_stats) {
|
|
|
|
scene->update_stats->tables.times.add_entry({"device_update", time});
|
|
|
|
}
|
|
|
|
});
|
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
VLOG(1) << "Total " << lookup_tables.size() << " lookup tables.";
|
2016-04-22 10:55:26 +02:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
if (lookup_tables.size() > 0)
|
|
|
|
dscene->lookup_table.copy_to_device();
|
2013-04-01 20:26:43 +00:00
|
|
|
|
Cycles: optimize device updates
This optimizes device updates (during user edits or frame changes in
the viewport) by avoiding unnecessary computations. To achieve this,
we use a combination of the sockets' update flags as well as some new
flags passed to the various managers when tagging for an update to tell
exactly what the tagging is for (e.g. shader was modified, object was
removed, etc.).
Besides avoiding recomputations, we also avoid resending to the devices
unmodified data arrays, thus reducing bandwidth usage. For OptiX and
Embree, BVH packing was also multithreaded.
The performance improvements may vary depending on the used device (CPU
or GPU), and the content of the scene. Simple scenes (e.g. with no adaptive
subdivision or volumes) rendered using OptiX will benefit from this work
the most.
On average, for a variety of animated scenes, this gives a 3x speedup.
Reviewed By: #cycles, brecht
Maniphest Tasks: T79174
Differential Revision: https://developer.blender.org/D9555
2021-01-22 15:01:26 +01:00
|
|
|
need_update_ = false;
|
2013-04-01 20:26:43 +00:00
|
|
|
}
|
|
|
|
|
2017-10-21 01:09:59 +02:00
|
|
|
void LookupTables::device_free(Device *, DeviceScene *dscene)
|
2013-04-01 20:26:43 +00:00
|
|
|
{
|
2019-04-17 06:17:24 +02:00
|
|
|
dscene->lookup_table.free();
|
2013-04-01 20:26:43 +00:00
|
|
|
}
|
|
|
|
|
Cycles: optimize device updates
This optimizes device updates (during user edits or frame changes in
the viewport) by avoiding unnecessary computations. To achieve this,
we use a combination of the sockets' update flags as well as some new
flags passed to the various managers when tagging for an update to tell
exactly what the tagging is for (e.g. shader was modified, object was
removed, etc.).
Besides avoiding recomputations, we also avoid resending to the devices
unmodified data arrays, thus reducing bandwidth usage. For OptiX and
Embree, BVH packing was also multithreaded.
The performance improvements may vary depending on the used device (CPU
or GPU), and the content of the scene. Simple scenes (e.g. with no adaptive
subdivision or volumes) rendered using OptiX will benefit from this work
the most.
On average, for a variety of animated scenes, this gives a 3x speedup.
Reviewed By: #cycles, brecht
Maniphest Tasks: T79174
Differential Revision: https://developer.blender.org/D9555
2021-01-22 15:01:26 +01:00
|
|
|
bool LookupTables::need_update() const
|
|
|
|
{
|
|
|
|
return need_update_;
|
|
|
|
}
|
|
|
|
|
2013-04-01 20:26:43 +00:00
|
|
|
static size_t round_up_to_multiple(size_t size, size_t chunk)
|
|
|
|
{
|
2019-04-17 06:17:24 +02:00
|
|
|
return ((size + chunk - 1) / chunk) * chunk;
|
2013-04-01 20:26:43 +00:00
|
|
|
}
|
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
size_t LookupTables::add_table(DeviceScene *dscene, vector<float> &data)
|
2013-04-01 20:26:43 +00:00
|
|
|
{
|
2019-04-17 06:17:24 +02:00
|
|
|
assert(data.size() > 0);
|
2013-04-01 20:26:43 +00:00
|
|
|
|
Cycles: optimize device updates
This optimizes device updates (during user edits or frame changes in
the viewport) by avoiding unnecessary computations. To achieve this,
we use a combination of the sockets' update flags as well as some new
flags passed to the various managers when tagging for an update to tell
exactly what the tagging is for (e.g. shader was modified, object was
removed, etc.).
Besides avoiding recomputations, we also avoid resending to the devices
unmodified data arrays, thus reducing bandwidth usage. For OptiX and
Embree, BVH packing was also multithreaded.
The performance improvements may vary depending on the used device (CPU
or GPU), and the content of the scene. Simple scenes (e.g. with no adaptive
subdivision or volumes) rendered using OptiX will benefit from this work
the most.
On average, for a variety of animated scenes, this gives a 3x speedup.
Reviewed By: #cycles, brecht
Maniphest Tasks: T79174
Differential Revision: https://developer.blender.org/D9555
2021-01-22 15:01:26 +01:00
|
|
|
need_update_ = true;
|
2013-04-01 20:26:43 +00:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
Table new_table;
|
|
|
|
new_table.offset = 0;
|
|
|
|
new_table.size = round_up_to_multiple(data.size(), TABLE_CHUNK_SIZE);
|
2013-04-01 20:26:43 +00:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
/* find space to put lookup table */
|
|
|
|
list<Table>::iterator table;
|
2013-04-01 20:26:43 +00:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
for (table = lookup_tables.begin(); table != lookup_tables.end(); table++) {
|
|
|
|
if (new_table.offset + new_table.size <= table->offset) {
|
|
|
|
lookup_tables.insert(table, new_table);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
new_table.offset = table->offset + table->size;
|
|
|
|
}
|
2013-04-01 20:26:43 +00:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
if (table == lookup_tables.end()) {
|
|
|
|
/* add at the end */
|
|
|
|
lookup_tables.push_back(new_table);
|
|
|
|
dscene->lookup_table.resize(new_table.offset + new_table.size);
|
|
|
|
}
|
2013-04-01 20:26:43 +00:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
/* copy table data and return offset */
|
|
|
|
float *dtable = dscene->lookup_table.data();
|
|
|
|
memcpy(dtable + new_table.offset, &data[0], sizeof(float) * data.size());
|
2017-10-20 04:32:29 +02:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
return new_table.offset;
|
2013-04-01 20:26:43 +00:00
|
|
|
}
|
|
|
|
|
2016-05-08 01:44:01 +02:00
|
|
|
void LookupTables::remove_table(size_t *offset)
|
2013-04-01 20:26:43 +00:00
|
|
|
{
|
2019-04-17 06:17:24 +02:00
|
|
|
if (*offset == TABLE_OFFSET_INVALID) {
|
|
|
|
/* The table isn't even allocated, so just return here. */
|
|
|
|
return;
|
|
|
|
}
|
2016-05-08 01:44:01 +02:00
|
|
|
|
Cycles: optimize device updates
This optimizes device updates (during user edits or frame changes in
the viewport) by avoiding unnecessary computations. To achieve this,
we use a combination of the sockets' update flags as well as some new
flags passed to the various managers when tagging for an update to tell
exactly what the tagging is for (e.g. shader was modified, object was
removed, etc.).
Besides avoiding recomputations, we also avoid resending to the devices
unmodified data arrays, thus reducing bandwidth usage. For OptiX and
Embree, BVH packing was also multithreaded.
The performance improvements may vary depending on the used device (CPU
or GPU), and the content of the scene. Simple scenes (e.g. with no adaptive
subdivision or volumes) rendered using OptiX will benefit from this work
the most.
On average, for a variety of animated scenes, this gives a 3x speedup.
Reviewed By: #cycles, brecht
Maniphest Tasks: T79174
Differential Revision: https://developer.blender.org/D9555
2021-01-22 15:01:26 +01:00
|
|
|
need_update_ = true;
|
2013-04-01 20:26:43 +00:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
list<Table>::iterator table;
|
2013-04-01 20:26:43 +00:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
for (table = lookup_tables.begin(); table != lookup_tables.end(); table++) {
|
|
|
|
if (table->offset == *offset) {
|
|
|
|
lookup_tables.erase(table);
|
|
|
|
*offset = TABLE_OFFSET_INVALID;
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
}
|
2013-04-01 20:26:43 +00:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
assert(table != lookup_tables.end());
|
2013-04-01 20:26:43 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
CCL_NAMESPACE_END
|