mirror of
https://gitlab.freedesktop.org/emersion/libliftoff.git
synced 2024-12-27 21:59:24 +01:00
aa2478cda8
This allows to unclutter output_choose_layers a little.
665 lines
18 KiB
C
665 lines
18 KiB
C
#include <assert.h>
|
|
#include <errno.h>
|
|
#include <inttypes.h>
|
|
#include <limits.h>
|
|
#include <stdlib.h>
|
|
#include <stdio.h>
|
|
#include <string.h>
|
|
#include <unistd.h>
|
|
#include "private.h"
|
|
|
|
static int guess_plane_zpos_from_type(struct liftoff_display *display,
|
|
uint32_t plane_id, uint32_t type)
|
|
{
|
|
struct liftoff_plane *primary;
|
|
|
|
/* From far to close to the eye: primary, overlay, cursor. Unless
|
|
* the overlay ID < primary ID. */
|
|
switch (type) {
|
|
case DRM_PLANE_TYPE_PRIMARY:
|
|
return 0;
|
|
case DRM_PLANE_TYPE_CURSOR:
|
|
return 2;
|
|
case DRM_PLANE_TYPE_OVERLAY:
|
|
if (liftoff_list_empty(&display->planes)) {
|
|
return 0; /* No primary plane, shouldn't happen */
|
|
}
|
|
primary = liftoff_container_of(display->planes.next,
|
|
primary, link);
|
|
if (plane_id < primary->id) {
|
|
return -1;
|
|
} else {
|
|
return 1;
|
|
}
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
static struct liftoff_plane *plane_create(struct liftoff_display *display,
|
|
uint32_t id)
|
|
{
|
|
struct liftoff_plane *plane, *cur;
|
|
drmModePlane *drm_plane;
|
|
drmModeObjectProperties *drm_props;
|
|
uint32_t i;
|
|
drmModePropertyRes *drm_prop;
|
|
struct liftoff_plane_property *prop;
|
|
uint64_t value;
|
|
bool has_type = false, has_zpos = false;
|
|
|
|
plane = calloc(1, sizeof(*plane));
|
|
if (plane == NULL) {
|
|
return NULL;
|
|
}
|
|
|
|
drm_plane = drmModeGetPlane(display->drm_fd, id);
|
|
if (drm_plane == NULL) {
|
|
return NULL;
|
|
}
|
|
plane->id = drm_plane->plane_id;
|
|
plane->possible_crtcs = drm_plane->possible_crtcs;
|
|
drmModeFreePlane(drm_plane);
|
|
|
|
drm_props = drmModeObjectGetProperties(display->drm_fd, id,
|
|
DRM_MODE_OBJECT_PLANE);
|
|
if (drm_props == NULL) {
|
|
return NULL;
|
|
}
|
|
plane->props = calloc(drm_props->count_props,
|
|
sizeof(struct liftoff_plane_property));
|
|
if (plane->props == NULL) {
|
|
drmModeFreeObjectProperties(drm_props);
|
|
return NULL;
|
|
}
|
|
for (i = 0; i < drm_props->count_props; i++) {
|
|
drm_prop = drmModeGetProperty(display->drm_fd,
|
|
drm_props->props[i]);
|
|
if (drm_prop == NULL) {
|
|
drmModeFreeObjectProperties(drm_props);
|
|
return NULL;
|
|
}
|
|
prop = &plane->props[i];
|
|
memcpy(prop->name, drm_prop->name, sizeof(prop->name));
|
|
prop->id = drm_prop->prop_id;
|
|
drmModeFreeProperty(drm_prop);
|
|
plane->props_len++;
|
|
|
|
value = drm_props->prop_values[i];
|
|
if (strcmp(prop->name, "type") == 0) {
|
|
plane->type = value;
|
|
has_type = true;
|
|
} else if (strcmp(prop->name, "zpos") == 0) {
|
|
plane->zpos = value;
|
|
has_zpos = true;
|
|
}
|
|
}
|
|
drmModeFreeObjectProperties(drm_props);
|
|
|
|
if (!has_type) {
|
|
fprintf(stderr,
|
|
"plane %"PRIu32" is missing the 'type' property\n",
|
|
plane->id);
|
|
free(plane);
|
|
return NULL;
|
|
} else if (!has_zpos) {
|
|
plane->zpos = guess_plane_zpos_from_type(display, plane->id,
|
|
plane->type);
|
|
}
|
|
|
|
/* During plane allocation, we will use the plane list order to fill
|
|
* planes with FBs. Primary planes need to be filled first, then planes
|
|
* far from the primary planes, then planes closer and closer to the
|
|
* primary plane. */
|
|
if (plane->type == DRM_PLANE_TYPE_PRIMARY) {
|
|
liftoff_list_insert(&display->planes, &plane->link);
|
|
} else {
|
|
liftoff_list_for_each(cur, &display->planes, link) {
|
|
if (cur->type != DRM_PLANE_TYPE_PRIMARY &&
|
|
plane->zpos >= cur->zpos) {
|
|
liftoff_list_insert(cur->link.prev, &plane->link);
|
|
break;
|
|
}
|
|
}
|
|
|
|
if (plane->link.next == NULL) { /* not inserted */
|
|
liftoff_list_insert(display->planes.prev, &plane->link);
|
|
}
|
|
}
|
|
|
|
return plane;
|
|
}
|
|
|
|
static void plane_destroy(struct liftoff_plane *plane)
|
|
{
|
|
liftoff_list_remove(&plane->link);
|
|
free(plane->props);
|
|
free(plane);
|
|
}
|
|
|
|
struct liftoff_display *liftoff_display_create(int drm_fd)
|
|
{
|
|
struct liftoff_display *display;
|
|
drmModeRes *drm_res;
|
|
drmModePlaneRes *drm_plane_res;
|
|
uint32_t i;
|
|
|
|
display = calloc(1, sizeof(*display));
|
|
if (display == NULL) {
|
|
return NULL;
|
|
}
|
|
|
|
liftoff_list_init(&display->planes);
|
|
liftoff_list_init(&display->outputs);
|
|
|
|
display->drm_fd = dup(drm_fd);
|
|
if (display->drm_fd < 0) {
|
|
liftoff_display_destroy(display);
|
|
return NULL;
|
|
}
|
|
|
|
drm_res = drmModeGetResources(drm_fd);
|
|
if (drm_res == NULL) {
|
|
liftoff_display_destroy(display);
|
|
return NULL;
|
|
}
|
|
|
|
display->crtcs = malloc(drm_res->count_crtcs * sizeof(uint32_t));
|
|
if (display->crtcs == NULL) {
|
|
drmModeFreeResources(drm_res);
|
|
liftoff_display_destroy(display);
|
|
return NULL;
|
|
}
|
|
display->crtcs_len = drm_res->count_crtcs;
|
|
memcpy(display->crtcs, drm_res->crtcs,
|
|
drm_res->count_crtcs * sizeof(uint32_t));
|
|
|
|
drmModeFreeResources(drm_res);
|
|
|
|
/* TODO: allow users to choose which layers to hand over */
|
|
drm_plane_res = drmModeGetPlaneResources(drm_fd);
|
|
if (drm_plane_res == NULL) {
|
|
liftoff_display_destroy(display);
|
|
return NULL;
|
|
}
|
|
|
|
for (i = 0; i < drm_plane_res->count_planes; i++) {
|
|
if (plane_create(display, drm_plane_res->planes[i]) == NULL) {
|
|
liftoff_display_destroy(display);
|
|
return NULL;
|
|
}
|
|
}
|
|
drmModeFreePlaneResources(drm_plane_res);
|
|
|
|
return display;
|
|
}
|
|
|
|
void liftoff_display_destroy(struct liftoff_display *display)
|
|
{
|
|
struct liftoff_plane *plane, *tmp;
|
|
|
|
close(display->drm_fd);
|
|
liftoff_list_for_each_safe(plane, tmp, &display->planes, link) {
|
|
plane_destroy(plane);
|
|
}
|
|
free(display->crtcs);
|
|
free(display);
|
|
}
|
|
|
|
static struct liftoff_plane_property *plane_get_property(struct liftoff_plane *plane,
|
|
const char *name)
|
|
{
|
|
size_t i;
|
|
|
|
for (i = 0; i < plane->props_len; i++) {
|
|
if (strcmp(plane->props[i].name, name) == 0) {
|
|
return &plane->props[i];
|
|
}
|
|
}
|
|
return NULL;
|
|
}
|
|
|
|
static bool plane_set_prop(struct liftoff_plane *plane, drmModeAtomicReq *req,
|
|
struct liftoff_plane_property *prop, uint64_t value)
|
|
{
|
|
int ret;
|
|
|
|
fprintf(stderr, " Setting %s = %"PRIu64"\n", prop->name, value);
|
|
ret = drmModeAtomicAddProperty(req, plane->id, prop->id, value);
|
|
if (ret < 0) {
|
|
perror("drmModeAtomicAddProperty");
|
|
return false;
|
|
}
|
|
|
|
return true;
|
|
}
|
|
|
|
static bool plane_apply(struct liftoff_plane *plane, struct liftoff_layer *layer,
|
|
drmModeAtomicReq *req, bool *compatible)
|
|
{
|
|
int cursor;
|
|
size_t i;
|
|
struct liftoff_layer_property *layer_prop;
|
|
struct liftoff_plane_property *plane_prop;
|
|
|
|
*compatible = true;
|
|
cursor = drmModeAtomicGetCursor(req);
|
|
|
|
if (layer == NULL) {
|
|
plane_prop = plane_get_property(plane, "FB_ID");
|
|
if (plane_prop == NULL) {
|
|
fprintf(stderr, "plane %"PRIu32" is missing the FB_ID "
|
|
"property\n", plane->id);
|
|
return false;
|
|
}
|
|
return plane_set_prop(plane, req, plane_prop, 0);
|
|
}
|
|
|
|
plane_prop = plane_get_property(plane, "CRTC_ID");
|
|
if (plane_prop == NULL) {
|
|
fprintf(stderr, "plane %"PRIu32" is missing the CRTC_ID "
|
|
"property\n", plane->id);
|
|
return false;
|
|
}
|
|
if (!plane_set_prop(plane, req, plane_prop, layer->output->crtc_id)) {
|
|
return false;
|
|
}
|
|
|
|
for (i = 0; i < layer->props_len; i++) {
|
|
layer_prop = &layer->props[i];
|
|
if (strcmp(layer_prop->name, "zpos") == 0) {
|
|
/* We don't yet support setting the zpos property. We
|
|
* only use it (read-only) during plane allocation. */
|
|
continue;
|
|
}
|
|
|
|
plane_prop = plane_get_property(plane, layer_prop->name);
|
|
if (plane_prop == NULL) {
|
|
*compatible = false;
|
|
drmModeAtomicSetCursor(req, cursor);
|
|
return true;
|
|
}
|
|
|
|
if (!plane_set_prop(plane, req, plane_prop, layer_prop->value)) {
|
|
drmModeAtomicSetCursor(req, cursor);
|
|
return false;
|
|
}
|
|
}
|
|
|
|
return true;
|
|
}
|
|
|
|
/* Plane allocation algorithm
|
|
*
|
|
* Goal: KMS exposes a set of hardware planes, user submitted a set of layers.
|
|
* We want to map as many layers as possible to planes.
|
|
*
|
|
* However, all layers can't be mapped to any plane. There are constraints,
|
|
* sometimes depending on driver-specific limitations or the configuration of
|
|
* other planes.
|
|
*
|
|
* The only way to discover driver-specific limitations is via an atomic test
|
|
* commit: we submit a plane configuration, and KMS replies whether it's
|
|
* supported or not. Thus we need to incrementally build a valid configuration.
|
|
*
|
|
* Let's take an example with 2 planes and 3 layers. Plane 1 is only compatible
|
|
* with layer 2 and plane 2 is only compatible with layer 3. Our algorithm will
|
|
* discover the solution by building the mapping one plane at a time. It first
|
|
* starts with plane 1: an atomic commit assigning layer 1 to plane 1 is
|
|
* submitted. It fails, because this isn't supported by the driver. Then layer
|
|
* 2 is assigned to plane 1 and the atomic test succeeds. We can go on and
|
|
* repeat the operation with plane 2. After exploring the whole tree, we end up
|
|
* with a valid allocation.
|
|
*
|
|
*
|
|
* layer 1 layer 1
|
|
* +---------> failure +---------> failure
|
|
* | |
|
|
* | |
|
|
* | |
|
|
* +---------+ | +---------+ |
|
|
* | | | layer 2 | | | layer 3 final allocation:
|
|
* | plane 1 +------------>+ plane 2 +--+---------> plane 1 → layer 2
|
|
* | | | | | plane 2 → layer 3
|
|
* +---------+ | +---------+
|
|
* |
|
|
* |
|
|
* | layer 3
|
|
* +---------> failure
|
|
*
|
|
*
|
|
* Note how layer 2 isn't considered for plane 2: it's already mapped to plane
|
|
* 1. Also note that branches are pruned as soon as an atomic test fails.
|
|
*
|
|
* In practice, the primary plane is treated separately. This is where layers
|
|
* that can't be mapped to any plane (e.g. layer 1 in our example) will be
|
|
* composited. The primary plane is the first that will be allocated. Then all
|
|
* other planes will be allocated, from the topmost one to the bottommost one.
|
|
*
|
|
* The "zpos" property (which defines ordering between layers/planes) is handled
|
|
* as a special case. If it's set on layers, it adds additional constraints on
|
|
* their relative ordering. If two layers intersect, their relative zpos needs
|
|
* to be preserved during plane allocation.
|
|
*
|
|
* Implementation-wise, the output_choose_layers function is called at each node
|
|
* of the tree. It iterates over layers, check constraints, performs an atomic
|
|
* test commit and calls itself recursively on the next plane.
|
|
*/
|
|
|
|
/* Global data for the allocation algorithm */
|
|
struct plane_alloc {
|
|
drmModeAtomicReq *req;
|
|
size_t planes_len;
|
|
|
|
struct liftoff_layer **best;
|
|
int best_score;
|
|
};
|
|
|
|
/* Transient data, arguments for each step */
|
|
struct plane_data {
|
|
struct liftoff_list *plane_link; /* liftoff_plane.link */
|
|
size_t plane_idx;
|
|
|
|
struct liftoff_layer **alloc; /* only items up to plane_idx are valid */
|
|
int score;
|
|
int last_plane_zpos, last_layer_zpos;
|
|
};
|
|
|
|
static void plane_data_init_next(struct plane_data *data,
|
|
struct plane_data *prev,
|
|
struct liftoff_layer *layer)
|
|
{
|
|
struct liftoff_plane *plane;
|
|
struct liftoff_layer_property *zpos_prop;
|
|
|
|
plane = liftoff_container_of(prev->plane_link, plane, link);
|
|
|
|
data->plane_link = prev->plane_link->next;
|
|
data->plane_idx = prev->plane_idx + 1;
|
|
data->alloc = prev->alloc;
|
|
data->alloc[prev->plane_idx] = layer;
|
|
|
|
if (layer != NULL) {
|
|
data->score = prev->score + 1;
|
|
} else {
|
|
data->score = prev->score;
|
|
}
|
|
if (layer != NULL && plane->type != DRM_PLANE_TYPE_PRIMARY) {
|
|
data->last_plane_zpos = plane->zpos;
|
|
} else {
|
|
data->last_plane_zpos = prev->last_plane_zpos;
|
|
}
|
|
|
|
zpos_prop = NULL;
|
|
if (layer != NULL) {
|
|
zpos_prop = layer_get_property(layer, "zpos");
|
|
}
|
|
if (zpos_prop != NULL && plane->type != DRM_PLANE_TYPE_PRIMARY) {
|
|
data->last_layer_zpos = zpos_prop->value;
|
|
} else {
|
|
data->last_layer_zpos = prev->last_layer_zpos;
|
|
}
|
|
}
|
|
|
|
static bool is_layer_allocated(struct plane_data *data,
|
|
struct liftoff_layer *layer)
|
|
{
|
|
size_t i;
|
|
|
|
/* TODO: speed this up with an array of bools indicating whether a layer
|
|
* has been allocated */
|
|
for (i = 0; i < data->plane_idx; i++) {
|
|
if (data->alloc[i] == layer) {
|
|
return true;
|
|
}
|
|
}
|
|
return false;
|
|
}
|
|
|
|
static bool has_composited_layer_on_top(struct liftoff_output *output,
|
|
struct plane_data *data, uint64_t zpos)
|
|
{
|
|
struct liftoff_layer *other_layer;
|
|
struct liftoff_layer_property *other_zpos_prop;
|
|
|
|
liftoff_list_for_each(other_layer, &output->layers, link) {
|
|
if (is_layer_allocated(data, other_layer)) {
|
|
continue;
|
|
}
|
|
|
|
other_zpos_prop = layer_get_property(other_layer, "zpos");
|
|
if (other_zpos_prop == NULL) {
|
|
continue;
|
|
}
|
|
|
|
if (other_zpos_prop->value > zpos) {
|
|
return true;
|
|
}
|
|
}
|
|
|
|
return false;
|
|
}
|
|
|
|
bool output_choose_layers(struct liftoff_output *output,
|
|
struct plane_alloc *alloc, struct plane_data *data)
|
|
{
|
|
struct liftoff_display *display;
|
|
struct liftoff_plane *plane;
|
|
struct liftoff_layer *layer;
|
|
int cursor, ret;
|
|
size_t remaining_planes;
|
|
bool compatible;
|
|
struct plane_data next_data;
|
|
struct liftoff_layer_property *zpos_prop;
|
|
|
|
display = output->display;
|
|
|
|
if (data->plane_link == &display->planes) { /* Allocation finished */
|
|
if (data->score > alloc->best_score) {
|
|
/* We found a better allocation */
|
|
alloc->best_score = data->score;
|
|
memcpy(alloc->best, data->alloc,
|
|
alloc->planes_len * sizeof(struct liftoff_layer *));
|
|
}
|
|
return true;
|
|
}
|
|
plane = liftoff_container_of(data->plane_link, plane, link);
|
|
|
|
remaining_planes = alloc->planes_len - data->plane_idx;
|
|
if (alloc->best_score >= data->score + (int)remaining_planes) {
|
|
/* Even if we find a layer for all remaining planes, we won't
|
|
* find a better allocation. Give up. */
|
|
return true;
|
|
}
|
|
|
|
cursor = drmModeAtomicGetCursor(alloc->req);
|
|
|
|
if (plane->layer != NULL) {
|
|
goto skip;
|
|
}
|
|
if ((plane->possible_crtcs & (1 << output->crtc_index)) == 0) {
|
|
goto skip;
|
|
}
|
|
|
|
fprintf(stderr, "Performing allocation for plane %"PRIu32" (%zu/%zu)\n",
|
|
plane->id, data->plane_idx + 1, alloc->planes_len);
|
|
|
|
liftoff_list_for_each(layer, &output->layers, link) {
|
|
if (layer->plane != NULL) {
|
|
continue;
|
|
}
|
|
|
|
/* Skip this layer if already allocated */
|
|
if (is_layer_allocated(data, layer)) {
|
|
continue;
|
|
}
|
|
|
|
zpos_prop = layer_get_property(layer, "zpos");
|
|
if (zpos_prop != NULL) {
|
|
if ((int)zpos_prop->value > data->last_layer_zpos) {
|
|
/* This layer needs to be on top of the last
|
|
* allocated one */
|
|
/* TODO: don't skip if they don't intersect? */
|
|
fprintf(stderr, "Layer %p -> plane %"PRIu32": "
|
|
"layer zpos invalid\n",
|
|
(void *)layer, plane->id);
|
|
continue;
|
|
}
|
|
if ((int)zpos_prop->value < data->last_layer_zpos &&
|
|
plane->zpos >= data->last_plane_zpos) {
|
|
/* This layer needs to be under the last
|
|
* allocated one, but this plane isn't under the
|
|
* last one */
|
|
/* TODO: don't skip if they don't intersect? */
|
|
fprintf(stderr, "Layer %p -> plane %"PRIu32": "
|
|
"plane zpos invalid\n",
|
|
(void *)layer, plane->id);
|
|
continue;
|
|
}
|
|
}
|
|
|
|
if (plane->type != DRM_PLANE_TYPE_PRIMARY &&
|
|
zpos_prop != NULL &&
|
|
has_composited_layer_on_top(output, data,
|
|
zpos_prop->value)) {
|
|
fprintf(stderr, "Layer %p -> plane %"PRIu32": "
|
|
"has composited layer on top\n",
|
|
(void *)layer, plane->id);
|
|
continue;
|
|
}
|
|
|
|
/* Try to use this layer for the current plane */
|
|
fprintf(stderr, "Layer %p -> plane %"PRIu32": "
|
|
"applying properties...\n", (void *)layer, plane->id);
|
|
if (!plane_apply(plane, layer, alloc->req, &compatible)) {
|
|
return false;
|
|
}
|
|
if (!compatible) {
|
|
fprintf(stderr, "Layer %p -> plane %"PRIu32": "
|
|
"incompatible properties\n",
|
|
(void *)layer, plane->id);
|
|
continue;
|
|
}
|
|
|
|
ret = drmModeAtomicCommit(display->drm_fd, alloc->req,
|
|
DRM_MODE_ATOMIC_TEST_ONLY, NULL);
|
|
if (ret == 0) {
|
|
fprintf(stderr, "Layer %p -> plane %"PRIu32": success\n",
|
|
(void *)layer, plane->id);
|
|
/* Continue with the next plane */
|
|
plane_data_init_next(&next_data, data, layer);
|
|
if (!output_choose_layers(output, alloc, &next_data)) {
|
|
return false;
|
|
}
|
|
} else if (-ret != EINVAL && -ret != ERANGE) {
|
|
perror("drmModeAtomicCommit");
|
|
return false;
|
|
}
|
|
|
|
drmModeAtomicSetCursor(alloc->req, cursor);
|
|
}
|
|
|
|
skip:
|
|
/* Try not to use the current plane */
|
|
plane_data_init_next(&next_data, data, NULL);
|
|
if (!output_choose_layers(output, alloc, &next_data)) {
|
|
return false;
|
|
}
|
|
drmModeAtomicSetCursor(alloc->req, cursor);
|
|
|
|
return true;
|
|
}
|
|
|
|
bool liftoff_display_apply(struct liftoff_display *display, drmModeAtomicReq *req)
|
|
{
|
|
struct liftoff_output *output;
|
|
struct liftoff_plane *plane;
|
|
struct liftoff_layer *layer;
|
|
struct plane_alloc alloc;
|
|
struct plane_data data;
|
|
size_t i;
|
|
bool compatible;
|
|
|
|
/* Unset all existing plane and layer mappings.
|
|
TODO: incremental updates keeping old configuration if possible */
|
|
liftoff_list_for_each(plane, &display->planes, link) {
|
|
if (plane->layer != NULL) {
|
|
plane->layer->plane = NULL;
|
|
plane->layer = NULL;
|
|
}
|
|
}
|
|
|
|
/* Disable all planes. Do it before building mappings to make sure not
|
|
to hit bandwidth limits because too many planes are enabled. */
|
|
liftoff_list_for_each(plane, &display->planes, link) {
|
|
if (plane->layer == NULL) {
|
|
fprintf(stderr, "Disabling plane %d\n", plane->id);
|
|
if (!plane_apply(plane, NULL, req, &compatible)) {
|
|
return false;
|
|
}
|
|
assert(compatible);
|
|
}
|
|
}
|
|
|
|
alloc.req = req;
|
|
alloc.planes_len = liftoff_list_length(&display->planes);
|
|
|
|
data.alloc = malloc(alloc.planes_len * sizeof(*data.alloc));
|
|
alloc.best = malloc(alloc.planes_len * sizeof(*alloc.best));
|
|
if (data.alloc == NULL || alloc.best == NULL) {
|
|
perror("malloc");
|
|
return false;
|
|
}
|
|
|
|
/* TODO: maybe start by allocating the primary plane on each output to
|
|
* make sure we can display at least something without hitting bandwidth
|
|
* issues? Also: be fair when mapping planes to outputs, don't give all
|
|
* planes to a single output. Also: don't treat each output separately,
|
|
* allocate planes for all outputs at once. */
|
|
liftoff_list_for_each(output, &display->outputs, link) {
|
|
/* For each plane, try to find a layer. Don't do it the other
|
|
* way around (ie. for each layer, try to find a plane) because
|
|
* some drivers want user-space to enable the primary plane
|
|
* before any other plane. */
|
|
|
|
alloc.best_score = 0;
|
|
memset(alloc.best, 0, alloc.planes_len * sizeof(*alloc.best));
|
|
data.plane_link = display->planes.next;
|
|
data.plane_idx = 0;
|
|
data.score = 0;
|
|
data.last_layer_zpos = INT_MAX;
|
|
data.last_plane_zpos = INT_MAX;
|
|
if (!output_choose_layers(output, &alloc, &data)) {
|
|
return false;
|
|
}
|
|
|
|
fprintf(stderr, "Found plane allocation for output %p "
|
|
"with score=%d\n", (void *)output, alloc.best_score);
|
|
|
|
/* Apply the best allocation */
|
|
i = 0;
|
|
liftoff_list_for_each(plane, &display->planes, link) {
|
|
layer = alloc.best[i];
|
|
i++;
|
|
if (layer == NULL) {
|
|
continue;
|
|
}
|
|
|
|
fprintf(stderr, "Assigning layer %p to plane %d\n",
|
|
(void *)layer, plane->id);
|
|
if (!plane_apply(plane, layer, req, &compatible)) {
|
|
return false;
|
|
}
|
|
assert(compatible);
|
|
|
|
assert(plane->layer == NULL);
|
|
assert(layer->plane == NULL);
|
|
plane->layer = layer;
|
|
layer->plane = plane;
|
|
}
|
|
}
|
|
|
|
free(data.alloc);
|
|
free(alloc.best);
|
|
|
|
return true;
|
|
}
|