You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
weston/libweston/renderer-gl/gl-renderer.c

3842 lines
104 KiB

/*
* Copyright © 2012 Intel Corporation
* Copyright © 2015,2019 Collabora, Ltd.
* Copyright © 2016 NVIDIA 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.
*/
#include "config.h"
#include <GLES2/gl2.h>
#include <GLES2/gl2ext.h>
#include <stdbool.h>
#include <stdint.h>
#include <stdlib.h>
#include <string.h>
#include <ctype.h>
#include <float.h>
#include <assert.h>
#include <linux/input.h>
#include <drm_fourcc.h>
#include <unistd.h>
#include "linux-sync-file.h"
#include "timeline.h"
#include "gl-renderer.h"
#include "gl-renderer-internal.h"
#include "vertex-clipping.h"
#include "linux-dmabuf.h"
#include "linux-dmabuf-unstable-v1-server-protocol.h"
libweston: Support zwp_surface_synchronization_v1.set_acquire_fence Implement the set_acquire_fence request of the zwp_surface_synchronization_v1 interface. The implementation uses the acquire fence in two ways: 1. If the associated buffer is used as GL render source, an EGLSyncKHR is created from the fence and used to synchronize access. 2. If the associated buffer is used as a plane framebuffer, the acquire fence is treated as an in-fence for the atomic commit operation. If in-fences are not supported and the buffer has an acquire fence, we don't consider it for plane placement. If the used compositor/renderer doesn't support explicit synchronization, we don't advertise the protocol at all. Currently only the DRM and X11 backends when using the GL renderer advertise the protocol for production use. Issues for discussion --------------------- a. Currently, a server-side wait of EGLSyncKHR is performed before using the EGLImage/texture during rendering. Unfortunately, it's not clear from the specs whether this is generally safe to do, or we need to sync before glEGLImageTargetTexture2DOES. The exception is TEXTURE_EXTERNAL_OES where the spec mentions it's enough to sync and then glBindTexture for any changes to take effect. Changes in v5: - Meson support. - Make explicit sync server error reporting more generic, supporting all explicit sync related interfaces not just wp_linux_surface_synchronization. - Fix typo in warning for missing EGL_KHR_wait_sync extension. - Support minor version 2 of the explicit sync protocol (i.e., support fences for opaque EGL buffers). Changes in v4: - Introduce and use fd_clear and and fd_move helpers. - Don't check for a valid buffer when updating surface acquire fence fd from state. - Assert that pending state acquire fence fd is always clear after a commit. - Clarify that WESTON_CAP_EXPLICIT_SYNC applies to just the renderer. - Check for EGL_KHR_wait_sync before using eglWaitSyncKHR. - Dup the acquire fence before passing to EGL. Changes in v3: - Keep acquire_fence_fd in surface instead of buffer. - Clarify that WESTON_CAP_EXPLICIT_SYNC applies to both backend and renderer. - Move comment about non-ownership of in_fence_fd to struct drm_plane_state definition. - Assert that we don't try to use planes with in-fences when using the legacy KMS API. - Remove unnecessary info from wayland error messages. - Handle acquire fence for subsurface commits. - Guard against self-update in fd_update. - Disconnect the client if acquire fence EGLSyncKHR creation or wait fails. - Use updated protocol interface names. - User correct format specifier for resource ids. - Advertise protocol for X11 backend with GL renderer. Changes in v2: - Remove sync file wait fallbacks. - Raise UNSUPPORTED_BUFFER error at commit if we have an acquire fence, but the committed buffer is not a valid linux_dmabuf. - Don't put buffers with in-fences on planes that don't support in-fences. - Don't advertise explicit sync protocol if backend does not support explicit sync. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
#include "linux-explicit-synchronization.h"
#include "pixel-formats.h"
libweston: Support zwp_surface_synchronization_v1.get_release Implement the get_release request of the zwp_surface_synchronization_v1 interface. This commit implements the zwp_buffer_release_v1 interface. It supports the zwp_buffer_release_v1.fenced_release event for surfaces rendered by the GL renderer, and the zwp_buffer_release_v1.immediate_release event for other cases. Note that the immediate_release event is safe to be used for surface buffers used as planes in the DRM backend, since the backend releases them only after the next page flip that doesn't use the buffers has finished. Changes in v7: - Remove "partial" from commit title and description. - Fix inverted check when clearing used_in_output_repaint flag. Changes in v5: - Use the new, generic explicit sync server error reporting function. - Introduce and use weston_buffer_release_move. - Introduce internally and use weston_buffer_release_destroy. Changes in v4: - Support the zwp_buffer_release_v1.fenced_release event. - Support release fences in the GL renderer. - Assert that pending state buffer_release is always NULL after a commit. - Simplify weston_buffer_release_reference. - Move removal of destroy listener before resource destruction to avoid concerns about use-after-free in weston_buffer_release_reference - Rename weston_buffer_release_reference.busy_count to ref_count. - Add documentation for weston_buffer_release and ..._reference. Changes in v3: - Raise NO_BUFFER for get_release if no buffer has been committed, don't raise UNSUPPORTED_BUFFER for non-dmabuf buffers, so get_release works for all valid buffers. - Destroy the buffer_release object after sending an event. - Track lifetime of buffer_release objects per commit, independently of any buffers. - Use updated protocol interface names. - Use correct format specifier for resource ids. Changes in v2: - Raise UNSUPPORTED_BUFFER at commit if client has requested a buffer_release, but the committed buffer is not a valid linux_dmabuf. - Remove tests that are not viable anymore due to our inability to create dmabuf buffers and fences in a unit-test environment. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
#include "shared/fd-util.h"
#include "shared/helpers.h"
#include "shared/platform.h"
#include "shared/timespec-util.h"
#include "shared/weston-egl-ext.h"
#define GR_GL_VERSION(major, minor) \
(((uint32_t)(major) << 16) | (uint32_t)(minor))
#define GR_GL_VERSION_INVALID \
GR_GL_VERSION(0, 0)
#define BUFFER_DAMAGE_COUNT 2
enum gl_border_status {
BORDER_STATUS_CLEAN = 0,
BORDER_TOP_DIRTY = 1 << GL_RENDERER_BORDER_TOP,
BORDER_LEFT_DIRTY = 1 << GL_RENDERER_BORDER_LEFT,
BORDER_RIGHT_DIRTY = 1 << GL_RENDERER_BORDER_RIGHT,
BORDER_BOTTOM_DIRTY = 1 << GL_RENDERER_BORDER_BOTTOM,
BORDER_ALL_DIRTY = 0xf,
BORDER_SIZE_CHANGED = 0x10
};
struct gl_border_image {
GLuint tex;
int32_t width, height;
int32_t tex_width;
void *data;
};
struct gl_output_state {
EGLSurface egl_surface;
pixman_region32_t buffer_damage[BUFFER_DAMAGE_COUNT];
int buffer_damage_index;
enum gl_border_status border_damage[BUFFER_DAMAGE_COUNT];
struct gl_border_image borders[4];
enum gl_border_status border_status;
struct weston_matrix output_matrix;
EGLSyncKHR begin_render_sync, end_render_sync;
/* struct timeline_render_point::link */
struct wl_list timeline_render_point_list;
};
enum buffer_type {
BUFFER_TYPE_NULL,
BUFFER_TYPE_SOLID, /* internal solid color surfaces without a buffer */
BUFFER_TYPE_SHM,
BUFFER_TYPE_EGL
};
struct gl_renderer;
struct egl_image {
struct gl_renderer *renderer;
EGLImageKHR image;
int refcount;
};
enum import_type {
IMPORT_TYPE_INVALID,
IMPORT_TYPE_DIRECT,
IMPORT_TYPE_GL_CONVERSION
};
struct dmabuf_image {
struct linux_dmabuf_buffer *dmabuf;
int num_images;
struct egl_image *images[3];
struct wl_list link;
enum import_type import_type;
GLenum target;
struct gl_shader *shader;
};
struct yuv_plane_descriptor {
int width_divisor;
int height_divisor;
uint32_t format;
int plane_index;
};
enum texture_type {
TEXTURE_Y_XUXV_WL,
TEXTURE_Y_UV_WL,
TEXTURE_Y_U_V_WL,
TEXTURE_XYUV_WL
};
struct yuv_format_descriptor {
uint32_t format;
int input_planes;
int output_planes;
enum texture_type texture_type;
struct yuv_plane_descriptor plane[4];
};
struct gl_surface_state {
GLfloat color[4];
struct gl_shader *shader;
GLuint textures[3];
int num_textures;
bool needs_full_upload;
pixman_region32_t texture_damage;
/* These are only used by SHM surfaces to detect when we need
* to do a full upload to specify a new internal texture
* format */
GLenum gl_format[3];
GLenum gl_pixel_type;
struct egl_image* images[3];
GLenum target;
int num_images;
struct weston_buffer_reference buffer_ref;
libweston: Support zwp_surface_synchronization_v1.get_release Implement the get_release request of the zwp_surface_synchronization_v1 interface. This commit implements the zwp_buffer_release_v1 interface. It supports the zwp_buffer_release_v1.fenced_release event for surfaces rendered by the GL renderer, and the zwp_buffer_release_v1.immediate_release event for other cases. Note that the immediate_release event is safe to be used for surface buffers used as planes in the DRM backend, since the backend releases them only after the next page flip that doesn't use the buffers has finished. Changes in v7: - Remove "partial" from commit title and description. - Fix inverted check when clearing used_in_output_repaint flag. Changes in v5: - Use the new, generic explicit sync server error reporting function. - Introduce and use weston_buffer_release_move. - Introduce internally and use weston_buffer_release_destroy. Changes in v4: - Support the zwp_buffer_release_v1.fenced_release event. - Support release fences in the GL renderer. - Assert that pending state buffer_release is always NULL after a commit. - Simplify weston_buffer_release_reference. - Move removal of destroy listener before resource destruction to avoid concerns about use-after-free in weston_buffer_release_reference - Rename weston_buffer_release_reference.busy_count to ref_count. - Add documentation for weston_buffer_release and ..._reference. Changes in v3: - Raise NO_BUFFER for get_release if no buffer has been committed, don't raise UNSUPPORTED_BUFFER for non-dmabuf buffers, so get_release works for all valid buffers. - Destroy the buffer_release object after sending an event. - Track lifetime of buffer_release objects per commit, independently of any buffers. - Use updated protocol interface names. - Use correct format specifier for resource ids. Changes in v2: - Raise UNSUPPORTED_BUFFER at commit if client has requested a buffer_release, but the committed buffer is not a valid linux_dmabuf. - Remove tests that are not viable anymore due to our inability to create dmabuf buffers and fences in a unit-test environment. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
struct weston_buffer_release_reference buffer_release_ref;
enum buffer_type buffer_type;
int pitch; /* in pixels */
int height; /* in pixels */
bool y_inverted;
bool direct_display;
/* Extension needed for SHM YUV texture */
int offset[3]; /* offset per plane */
int hsub[3]; /* horizontal subsampling per plane */
int vsub[3]; /* vertical subsampling per plane */
struct weston_surface *surface;
libweston: Support zwp_surface_synchronization_v1.get_release Implement the get_release request of the zwp_surface_synchronization_v1 interface. This commit implements the zwp_buffer_release_v1 interface. It supports the zwp_buffer_release_v1.fenced_release event for surfaces rendered by the GL renderer, and the zwp_buffer_release_v1.immediate_release event for other cases. Note that the immediate_release event is safe to be used for surface buffers used as planes in the DRM backend, since the backend releases them only after the next page flip that doesn't use the buffers has finished. Changes in v7: - Remove "partial" from commit title and description. - Fix inverted check when clearing used_in_output_repaint flag. Changes in v5: - Use the new, generic explicit sync server error reporting function. - Introduce and use weston_buffer_release_move. - Introduce internally and use weston_buffer_release_destroy. Changes in v4: - Support the zwp_buffer_release_v1.fenced_release event. - Support release fences in the GL renderer. - Assert that pending state buffer_release is always NULL after a commit. - Simplify weston_buffer_release_reference. - Move removal of destroy listener before resource destruction to avoid concerns about use-after-free in weston_buffer_release_reference - Rename weston_buffer_release_reference.busy_count to ref_count. - Add documentation for weston_buffer_release and ..._reference. Changes in v3: - Raise NO_BUFFER for get_release if no buffer has been committed, don't raise UNSUPPORTED_BUFFER for non-dmabuf buffers, so get_release works for all valid buffers. - Destroy the buffer_release object after sending an event. - Track lifetime of buffer_release objects per commit, independently of any buffers. - Use updated protocol interface names. - Use correct format specifier for resource ids. Changes in v2: - Raise UNSUPPORTED_BUFFER at commit if client has requested a buffer_release, but the committed buffer is not a valid linux_dmabuf. - Remove tests that are not viable anymore due to our inability to create dmabuf buffers and fences in a unit-test environment. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
/* Whether this surface was used in the current output repaint.
Used only in the context of a gl_renderer_repaint_output call. */
bool used_in_output_repaint;
struct wl_listener surface_destroy_listener;
struct wl_listener renderer_destroy_listener;
};
enum timeline_render_point_type {
TIMELINE_RENDER_POINT_TYPE_BEGIN,
TIMELINE_RENDER_POINT_TYPE_END
};
struct timeline_render_point {
struct wl_list link; /* gl_output_state::timeline_render_point_list */
enum timeline_render_point_type type;
int fd;
struct weston_output *output;
struct wl_event_source *event_source;
};
static PFNEGLGETPLATFORMDISPLAYEXTPROC get_platform_display = NULL;
static inline const char *
dump_format(uint32_t format, char out[4])
{
#if BYTE_ORDER == BIG_ENDIAN
format = __builtin_bswap32(format);
#endif
memcpy(out, &format, 4);
return out;
}
static inline struct gl_output_state *
get_output_state(struct weston_output *output)
{
return (struct gl_output_state *)output->renderer_state;
}
static int
gl_renderer_create_surface(struct weston_surface *surface);
static inline struct gl_surface_state *
get_surface_state(struct weston_surface *surface)
{
if (!surface->renderer_state)
gl_renderer_create_surface(surface);
return (struct gl_surface_state *)surface->renderer_state;
}
static void
timeline_render_point_destroy(struct timeline_render_point *trp)
{
wl_list_remove(&trp->link);
wl_event_source_remove(trp->event_source);
close(trp->fd);
free(trp);
}
static int
timeline_render_point_handler(int fd, uint32_t mask, void *data)
{
struct timeline_render_point *trp = data;
const char *tp_name = trp->type == TIMELINE_RENDER_POINT_TYPE_BEGIN ?
"renderer_gpu_begin" : "renderer_gpu_end";
if (mask & WL_EVENT_READABLE) {
struct timespec tspec = { 0 };
if (weston_linux_sync_file_read_timestamp(trp->fd,
&tspec) == 0) {
TL_POINT(trp->output->compositor, tp_name, TLP_GPU(&tspec),
TLP_OUTPUT(trp->output), TLP_END);
}
}
timeline_render_point_destroy(trp);
return 0;
}
static EGLSyncKHR
create_render_sync(struct gl_renderer *gr)
{
static const EGLint attribs[] = { EGL_NONE };
if (!gr->has_native_fence_sync)
return EGL_NO_SYNC_KHR;
return gr->create_sync(gr->egl_display, EGL_SYNC_NATIVE_FENCE_ANDROID,
attribs);
}
static void
timeline_submit_render_sync(struct gl_renderer *gr,
struct weston_compositor *ec,
struct weston_output *output,
EGLSyncKHR sync,
enum timeline_render_point_type type)
{
struct gl_output_state *go;
struct wl_event_loop *loop;
int fd;
struct timeline_render_point *trp;
if (!weston_log_scope_is_enabled(ec->timeline) ||
!gr->has_native_fence_sync ||
sync == EGL_NO_SYNC_KHR)
return;
go = get_output_state(output);
loop = wl_display_get_event_loop(ec->wl_display);
fd = gr->dup_native_fence_fd(gr->egl_display, sync);
if (fd == EGL_NO_NATIVE_FENCE_FD_ANDROID)
return;
trp = zalloc(sizeof *trp);
if (trp == NULL) {
close(fd);
return;
}
trp->type = type;
trp->fd = fd;
trp->output = output;
trp->event_source = wl_event_loop_add_fd(loop, fd,
WL_EVENT_READABLE,
timeline_render_point_handler,
trp);
wl_list_insert(&go->timeline_render_point_list, &trp->link);
}
static struct egl_image*
egl_image_create(struct gl_renderer *gr, EGLenum target,
EGLClientBuffer buffer, const EGLint *attribs)
{
struct egl_image *img;
img = zalloc(sizeof *img);
img->renderer = gr;
img->refcount = 1;
img->image = gr->create_image(gr->egl_display, EGL_NO_CONTEXT,
target, buffer, attribs);
if (img->image == EGL_NO_IMAGE_KHR) {
free(img);
return NULL;
}
return img;
}
static struct egl_image*
egl_image_ref(struct egl_image *image)
{
image->refcount++;
return image;
}
static int
egl_image_unref(struct egl_image *image)
{
struct gl_renderer *gr = image->renderer;
assert(image->refcount > 0);
image->refcount--;
if (image->refcount > 0)
return image->refcount;
gr->destroy_image(gr->egl_display, image->image);
free(image);
return 0;
}
static struct dmabuf_image*
dmabuf_image_create(void)
{
struct dmabuf_image *img;
img = zalloc(sizeof *img);
wl_list_init(&img->link);
return img;
}
static void
dmabuf_image_destroy(struct dmabuf_image *image)
{
int i;
for (i = 0; i < image->num_images; ++i)
egl_image_unref(image->images[i]);
if (image->dmabuf)
linux_dmabuf_buffer_set_user_data(image->dmabuf, NULL, NULL);
wl_list_remove(&image->link);
free(image);
}
#define max(a, b) (((a) > (b)) ? (a) : (b))
#define min(a, b) (((a) > (b)) ? (b) : (a))
/*
* Compute the boundary vertices of the intersection of the global coordinate
* aligned rectangle 'rect', and an arbitrary quadrilateral produced from
* 'surf_rect' when transformed from surface coordinates into global coordinates.
* The vertices are written to 'ex' and 'ey', and the return value is the
* number of vertices. Vertices are produced in clockwise winding order.
* Guarantees to produce either zero vertices, or 3-8 vertices with non-zero
* polygon area.
*/
static int
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
calculate_edges(struct weston_view *ev, pixman_box32_t *rect,
pixman_box32_t *surf_rect, GLfloat *ex, GLfloat *ey)
{
struct clip_context ctx;
int i, n;
GLfloat min_x, max_x, min_y, max_y;
struct polygon8 surf = {
{ surf_rect->x1, surf_rect->x2, surf_rect->x2, surf_rect->x1 },
{ surf_rect->y1, surf_rect->y1, surf_rect->y2, surf_rect->y2 },
4
};
ctx.clip.x1 = rect->x1;
ctx.clip.y1 = rect->y1;
ctx.clip.x2 = rect->x2;
ctx.clip.y2 = rect->y2;
/* transform surface to screen space: */
for (i = 0; i < surf.n; i++)
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
weston_view_to_global_float(ev, surf.x[i], surf.y[i],
&surf.x[i], &surf.y[i]);
/* find bounding box: */
min_x = max_x = surf.x[0];
min_y = max_y = surf.y[0];
for (i = 1; i < surf.n; i++) {
min_x = min(min_x, surf.x[i]);
max_x = max(max_x, surf.x[i]);
min_y = min(min_y, surf.y[i]);
max_y = max(max_y, surf.y[i]);
}
/* First, simple bounding box check to discard early transformed
* surface rects that do not intersect with the clip region:
*/
if ((min_x >= ctx.clip.x2) || (max_x <= ctx.clip.x1) ||
(min_y >= ctx.clip.y2) || (max_y <= ctx.clip.y1))
return 0;
/* Simple case, bounding box edges are parallel to surface edges,
* there will be only four edges. We just need to clip the surface
* vertices to the clip rect bounds:
*/
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
if (!ev->transform.enabled)
return clip_simple(&ctx, &surf, ex, ey);
/* Transformed case: use a general polygon clipping algorithm to
* clip the surface rectangle with each side of 'rect'.
* The algorithm is Sutherland-Hodgman, as explained in
* http://www.codeguru.com/cpp/misc/misc/graphics/article.php/c8965/Polygon-Clipping.htm
* but without looking at any of that code.
*/
n = clip_transformed(&ctx, &surf, ex, ey);
if (n < 3)
return 0;
return n;
}
static bool
merge_down(pixman_box32_t *a, pixman_box32_t *b, pixman_box32_t *merge)
{
if (a->x1 == b->x1 && a->x2 == b->x2 && a->y1 == b->y2) {
merge->x1 = a->x1;
merge->x2 = a->x2;
merge->y1 = b->y1;
merge->y2 = a->y2;
return true;
}
return false;
}
static int
compress_bands(pixman_box32_t *inrects, int nrects,
pixman_box32_t **outrects)
{
bool merged = false;
pixman_box32_t *out, merge_rect;
int i, j, nout;
if (!nrects) {
*outrects = NULL;
return 0;
}
/* nrects is an upper bound - we're not too worried about
* allocating a little extra
*/
out = malloc(sizeof(pixman_box32_t) * nrects);
out[0] = inrects[0];
nout = 1;
for (i = 1; i < nrects; i++) {
for (j = 0; j < nout; j++) {
merged = merge_down(&inrects[i], &out[j], &merge_rect);
if (merged) {
out[j] = merge_rect;
break;
}
}
if (!merged) {
out[nout] = inrects[i];
nout++;
}
}
*outrects = out;
return nout;
}
static int
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
texture_region(struct weston_view *ev, pixman_region32_t *region,
pixman_region32_t *surf_region)
{
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
struct gl_surface_state *gs = get_surface_state(ev->surface);
struct weston_compositor *ec = ev->surface->compositor;
struct gl_renderer *gr = get_renderer(ec);
GLfloat *v, inv_width, inv_height;
unsigned int *vtxcnt, nvtx = 0;
pixman_box32_t *rects, *surf_rects;
pixman_box32_t *raw_rects;
int i, j, k, nrects, nsurf, raw_nrects;
bool used_band_compression;
raw_rects = pixman_region32_rectangles(region, &raw_nrects);
surf_rects = pixman_region32_rectangles(surf_region, &nsurf);
if (raw_nrects < 4) {
used_band_compression = false;
nrects = raw_nrects;
rects = raw_rects;
} else {
nrects = compress_bands(raw_rects, raw_nrects, &rects);
used_band_compression = true;
}
/* worst case we can have 8 vertices per rect (ie. clipped into
* an octagon):
*/
v = wl_array_add(&gr->vertices, nrects * nsurf * 8 * 4 * sizeof *v);
vtxcnt = wl_array_add(&gr->vtxcnt, nrects * nsurf * sizeof *vtxcnt);
inv_width = 1.0 / gs->pitch;
inv_height = 1.0 / gs->height;
for (i = 0; i < nrects; i++) {
pixman_box32_t *rect = &rects[i];
for (j = 0; j < nsurf; j++) {
pixman_box32_t *surf_rect = &surf_rects[j];
GLfloat sx, sy, bx, by;
GLfloat ex[8], ey[8]; /* edge points in screen space */
int n;
/* The transformed surface, after clipping to the clip region,
* can have as many as eight sides, emitted as a triangle-fan.
* The first vertex in the triangle fan can be chosen arbitrarily,
* since the area is guaranteed to be convex.
*
* If a corner of the transformed surface falls outside of the
* clip region, instead of emitting one vertex for the corner
* of the surface, up to two are emitted for two corresponding
* intersection point(s) between the surface and the clip region.
*
* To do this, we first calculate the (up to eight) points that
* form the intersection of the clip rect and the transformed
* surface.
*/
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
n = calculate_edges(ev, rect, surf_rect, ex, ey);
if (n < 3)
continue;
/* emit edge points: */
for (k = 0; k < n; k++) {
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
weston_view_from_global_float(ev, ex[k], ey[k],
&sx, &sy);
/* position: */
*(v++) = ex[k];
*(v++) = ey[k];
/* texcoord: */
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
weston_surface_to_buffer_float(ev->surface,
sx, sy,
&bx, &by);
*(v++) = bx * inv_width;
if (gs->y_inverted) {
*(v++) = by * inv_height;
} else {
*(v++) = (gs->height - by) * inv_height;
}
}
vtxcnt[nvtx++] = n;
}
}
if (used_band_compression)
free(rects);
return nvtx;
}
static void
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
triangle_fan_debug(struct weston_view *view, int first, int count)
{
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
struct weston_compositor *compositor = view->surface->compositor;
struct gl_renderer *gr = get_renderer(compositor);
int i;
GLushort *buffer;
GLushort *index;
int nelems;
static int color_idx = 0;
static const GLfloat color[][4] = {
{ 1.0, 0.0, 0.0, 1.0 },
{ 0.0, 1.0, 0.0, 1.0 },
{ 0.0, 0.0, 1.0, 1.0 },
{ 1.0, 1.0, 1.0, 1.0 },
};
nelems = (count - 1 + count - 2) * 2;
buffer = malloc(sizeof(GLushort) * nelems);
index = buffer;
for (i = 1; i < count; i++) {
*index++ = first;
*index++ = first + i;
}
for (i = 2; i < count; i++) {
*index++ = first + i - 1;
*index++ = first + i;
}
glUseProgram(gr->solid_shader.program);
glUniform4fv(gr->solid_shader.color_uniform, 1,
color[color_idx++ % ARRAY_LENGTH(color)]);
glDrawElements(GL_LINES, nelems, GL_UNSIGNED_SHORT, buffer);
glUseProgram(gr->current_shader->program);
free(buffer);
}
static void
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
repaint_region(struct weston_view *ev, pixman_region32_t *region,
pixman_region32_t *surf_region)
{
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
struct weston_compositor *ec = ev->surface->compositor;
struct gl_renderer *gr = get_renderer(ec);
GLfloat *v;
unsigned int *vtxcnt;
int i, first, nfans;
/* The final region to be painted is the intersection of
* 'region' and 'surf_region'. However, 'region' is in the global
* coordinates, and 'surf_region' is in the surface-local
* coordinates. texture_region() will iterate over all pairs of
* rectangles from both regions, compute the intersection
* polygon for each pair, and store it as a triangle fan if
* it has a non-zero area (at least 3 vertices, actually).
*/
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
nfans = texture_region(ev, region, surf_region);
v = gr->vertices.data;
vtxcnt = gr->vtxcnt.data;
/* position: */
glVertexAttribPointer(0, 2, GL_FLOAT, GL_FALSE, 4 * sizeof *v, &v[0]);
glEnableVertexAttribArray(0);
/* texcoord: */
glVertexAttribPointer(1, 2, GL_FLOAT, GL_FALSE, 4 * sizeof *v, &v[2]);
glEnableVertexAttribArray(1);
for (i = 0, first = 0; i < nfans; i++) {
glDrawArrays(GL_TRIANGLE_FAN, first, vtxcnt[i]);
if (gr->fan_debug)
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
triangle_fan_debug(ev, first, vtxcnt[i]);
first += vtxcnt[i];
}
glDisableVertexAttribArray(1);
glDisableVertexAttribArray(0);
gr->vertices.size = 0;
gr->vtxcnt.size = 0;
}
static int
use_output(struct weston_output *output)
{
static int errored;
struct gl_output_state *go = get_output_state(output);
struct gl_renderer *gr = get_renderer(output->compositor);
EGLBoolean ret;
ret = eglMakeCurrent(gr->egl_display, go->egl_surface,
go->egl_surface, gr->egl_context);
if (ret == EGL_FALSE) {
if (errored)
return -1;
errored = 1;
weston_log("Failed to make EGL context current.\n");
gl_renderer_print_egl_error_state();
return -1;
}
return 0;
}
static int
shader_init(struct gl_shader *shader, struct gl_renderer *gr,
const char *vertex_source, const char *fragment_source);
static void
use_shader(struct gl_renderer *gr, struct gl_shader *shader)
{
if (!shader->program) {
int ret;
ret = shader_init(shader, gr,
shader->vertex_source,
shader->fragment_source);
if (ret < 0)
weston_log("warning: failed to compile shader\n");
}
if (gr->current_shader == shader)
return;
glUseProgram(shader->program);
gr->current_shader = shader;
}
static void
shader_uniforms(struct gl_shader *shader,
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
struct weston_view *view,
struct weston_output *output)
{
int i;
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
struct gl_surface_state *gs = get_surface_state(view->surface);
struct gl_output_state *go = get_output_state(output);
glUniformMatrix4fv(shader->proj_uniform,
1, GL_FALSE, go->output_matrix.d);
glUniform4fv(shader->color_uniform, 1, gs->color);
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
glUniform1f(shader->alpha_uniform, view->alpha);
for (i = 0; i < gs->num_textures; i++)
glUniform1i(shader->tex_uniforms[i], i);
}
libweston: Support zwp_surface_synchronization_v1.set_acquire_fence Implement the set_acquire_fence request of the zwp_surface_synchronization_v1 interface. The implementation uses the acquire fence in two ways: 1. If the associated buffer is used as GL render source, an EGLSyncKHR is created from the fence and used to synchronize access. 2. If the associated buffer is used as a plane framebuffer, the acquire fence is treated as an in-fence for the atomic commit operation. If in-fences are not supported and the buffer has an acquire fence, we don't consider it for plane placement. If the used compositor/renderer doesn't support explicit synchronization, we don't advertise the protocol at all. Currently only the DRM and X11 backends when using the GL renderer advertise the protocol for production use. Issues for discussion --------------------- a. Currently, a server-side wait of EGLSyncKHR is performed before using the EGLImage/texture during rendering. Unfortunately, it's not clear from the specs whether this is generally safe to do, or we need to sync before glEGLImageTargetTexture2DOES. The exception is TEXTURE_EXTERNAL_OES where the spec mentions it's enough to sync and then glBindTexture for any changes to take effect. Changes in v5: - Meson support. - Make explicit sync server error reporting more generic, supporting all explicit sync related interfaces not just wp_linux_surface_synchronization. - Fix typo in warning for missing EGL_KHR_wait_sync extension. - Support minor version 2 of the explicit sync protocol (i.e., support fences for opaque EGL buffers). Changes in v4: - Introduce and use fd_clear and and fd_move helpers. - Don't check for a valid buffer when updating surface acquire fence fd from state. - Assert that pending state acquire fence fd is always clear after a commit. - Clarify that WESTON_CAP_EXPLICIT_SYNC applies to just the renderer. - Check for EGL_KHR_wait_sync before using eglWaitSyncKHR. - Dup the acquire fence before passing to EGL. Changes in v3: - Keep acquire_fence_fd in surface instead of buffer. - Clarify that WESTON_CAP_EXPLICIT_SYNC applies to both backend and renderer. - Move comment about non-ownership of in_fence_fd to struct drm_plane_state definition. - Assert that we don't try to use planes with in-fences when using the legacy KMS API. - Remove unnecessary info from wayland error messages. - Handle acquire fence for subsurface commits. - Guard against self-update in fd_update. - Disconnect the client if acquire fence EGLSyncKHR creation or wait fails. - Use updated protocol interface names. - User correct format specifier for resource ids. - Advertise protocol for X11 backend with GL renderer. Changes in v2: - Remove sync file wait fallbacks. - Raise UNSUPPORTED_BUFFER error at commit if we have an acquire fence, but the committed buffer is not a valid linux_dmabuf. - Don't put buffers with in-fences on planes that don't support in-fences. - Don't advertise explicit sync protocol if backend does not support explicit sync. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
static int
ensure_surface_buffer_is_ready(struct gl_renderer *gr,
struct gl_surface_state *gs)
{
EGLint attribs[] = {
EGL_SYNC_NATIVE_FENCE_FD_ANDROID,
-1,
EGL_NONE
};
struct weston_surface *surface = gs->surface;
struct weston_buffer *buffer = gs->buffer_ref.buffer;
EGLSyncKHR sync;
EGLint wait_ret;
EGLint destroy_ret;
if (!buffer)
return 0;
if (surface->acquire_fence_fd < 0)
return 0;
/* We should only get a fence if we support EGLSyncKHR, since
* we don't advertise the explicit sync protocol otherwise. */
assert(gr->has_native_fence_sync);
/* We should only get a fence for non-SHM buffers, since surface
* commit would have failed otherwise. */
assert(wl_shm_buffer_get(buffer->resource) == NULL);
attribs[1] = dup(surface->acquire_fence_fd);
if (attribs[1] == -1) {
linux_explicit_synchronization_send_server_error(
gs->surface->synchronization_resource,
"Failed to dup acquire fence");
return -1;
}
sync = gr->create_sync(gr->egl_display,
EGL_SYNC_NATIVE_FENCE_ANDROID,
attribs);
if (sync == EGL_NO_SYNC_KHR) {
linux_explicit_synchronization_send_server_error(
gs->surface->synchronization_resource,
"Failed to create EGLSyncKHR object");
close(attribs[1]);
return -1;
}
wait_ret = gr->wait_sync(gr->egl_display, sync, 0);
if (wait_ret == EGL_FALSE) {
linux_explicit_synchronization_send_server_error(
gs->surface->synchronization_resource,
"Failed to wait on EGLSyncKHR object");
/* Continue to try to destroy the sync object. */
}
destroy_ret = gr->destroy_sync(gr->egl_display, sync);
if (destroy_ret == EGL_FALSE) {
linux_explicit_synchronization_send_server_error(
gs->surface->synchronization_resource,
"Failed to destroy on EGLSyncKHR object");
}
return (wait_ret == EGL_TRUE && destroy_ret == EGL_TRUE) ? 0 : -1;
}
/* Checks if a view needs to be censored on an output
* Checks for 2 types of censor requirements
* - recording_censor: Censor protected view when a
* protected view is captured.
* - unprotected_censor: Censor regions of protected views
* when displayed on an output which has lower protection capability.
* Returns the originally stored gl_shader if content censoring is required,
* NULL otherwise.
*/
static struct gl_shader *
setup_censor_overrides(struct weston_output *output,
struct weston_view *ev)
{
struct gl_shader *replaced_shader = NULL;
struct weston_compositor *ec = ev->surface->compositor;
struct gl_renderer *gr = get_renderer(ec);
struct gl_surface_state *gs = get_surface_state(ev->surface);
bool recording_censor =
(output->disable_planes > 0) &&
(ev->surface->desired_protection > WESTON_HDCP_DISABLE);
bool unprotected_censor =
(ev->surface->desired_protection > output->current_protection);
if (gs->direct_display) {
gs->color[0] = 0.40;
gs->color[1] = 0.0;
gs->color[2] = 0.0;
gs->color[3] = 1.0;
gs->shader = &gr->solid_shader;
return gs->shader;
}
/* When not in enforced mode, the client is notified of the protection */
/* change, so content censoring is not required */
if (ev->surface->protection_mode !=
WESTON_SURFACE_PROTECTION_MODE_ENFORCED)
return NULL;
if (recording_censor || unprotected_censor) {
replaced_shader = gs->shader;
gs->color[0] = 0.40;
gs->color[1] = 0.0;
gs->color[2] = 0.0;
gs->color[3] = 1.0;
gs->shader = &gr->solid_shader;
}
return replaced_shader;
}
static void
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
draw_view(struct weston_view *ev, struct weston_output *output,
pixman_region32_t *damage) /* in global coordinates */
{
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
struct weston_compositor *ec = ev->surface->compositor;
struct gl_renderer *gr = get_renderer(ec);
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
struct gl_surface_state *gs = get_surface_state(ev->surface);
/* repaint bounding region in global coordinates: */
pixman_region32_t repaint;
/* opaque region in surface coordinates: */
pixman_region32_t surface_opaque;
/* non-opaque region in surface coordinates: */
pixman_region32_t surface_blend;
GLint filter;
int i;
struct gl_shader *replaced_shader = NULL;
/* In case of a runtime switch of renderers, we may not have received
* an attach for this surface since the switch. In that case we don't
* have a valid buffer or a proper shader set up so skip rendering. */
if (!gs->shader && !gs->direct_display)
return;
pixman_region32_init(&repaint);
pixman_region32_intersect(&repaint,
&ev->transform.boundingbox, damage);
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
pixman_region32_subtract(&repaint, &repaint, &ev->clip);
if (!pixman_region32_not_empty(&repaint))
goto out;
libweston: Support zwp_surface_synchronization_v1.set_acquire_fence Implement the set_acquire_fence request of the zwp_surface_synchronization_v1 interface. The implementation uses the acquire fence in two ways: 1. If the associated buffer is used as GL render source, an EGLSyncKHR is created from the fence and used to synchronize access. 2. If the associated buffer is used as a plane framebuffer, the acquire fence is treated as an in-fence for the atomic commit operation. If in-fences are not supported and the buffer has an acquire fence, we don't consider it for plane placement. If the used compositor/renderer doesn't support explicit synchronization, we don't advertise the protocol at all. Currently only the DRM and X11 backends when using the GL renderer advertise the protocol for production use. Issues for discussion --------------------- a. Currently, a server-side wait of EGLSyncKHR is performed before using the EGLImage/texture during rendering. Unfortunately, it's not clear from the specs whether this is generally safe to do, or we need to sync before glEGLImageTargetTexture2DOES. The exception is TEXTURE_EXTERNAL_OES where the spec mentions it's enough to sync and then glBindTexture for any changes to take effect. Changes in v5: - Meson support. - Make explicit sync server error reporting more generic, supporting all explicit sync related interfaces not just wp_linux_surface_synchronization. - Fix typo in warning for missing EGL_KHR_wait_sync extension. - Support minor version 2 of the explicit sync protocol (i.e., support fences for opaque EGL buffers). Changes in v4: - Introduce and use fd_clear and and fd_move helpers. - Don't check for a valid buffer when updating surface acquire fence fd from state. - Assert that pending state acquire fence fd is always clear after a commit. - Clarify that WESTON_CAP_EXPLICIT_SYNC applies to just the renderer. - Check for EGL_KHR_wait_sync before using eglWaitSyncKHR. - Dup the acquire fence before passing to EGL. Changes in v3: - Keep acquire_fence_fd in surface instead of buffer. - Clarify that WESTON_CAP_EXPLICIT_SYNC applies to both backend and renderer. - Move comment about non-ownership of in_fence_fd to struct drm_plane_state definition. - Assert that we don't try to use planes with in-fences when using the legacy KMS API. - Remove unnecessary info from wayland error messages. - Handle acquire fence for subsurface commits. - Guard against self-update in fd_update. - Disconnect the client if acquire fence EGLSyncKHR creation or wait fails. - Use updated protocol interface names. - User correct format specifier for resource ids. - Advertise protocol for X11 backend with GL renderer. Changes in v2: - Remove sync file wait fallbacks. - Raise UNSUPPORTED_BUFFER error at commit if we have an acquire fence, but the committed buffer is not a valid linux_dmabuf. - Don't put buffers with in-fences on planes that don't support in-fences. - Don't advertise explicit sync protocol if backend does not support explicit sync. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
if (ensure_surface_buffer_is_ready(gr, gs) < 0)
goto out;
replaced_shader = setup_censor_overrides(output, ev);
glBlendFunc(GL_ONE, GL_ONE_MINUS_SRC_ALPHA);
if (gr->fan_debug) {
use_shader(gr, &gr->solid_shader);
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
shader_uniforms(&gr->solid_shader, ev, output);
}
use_shader(gr, gs->shader);
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
shader_uniforms(gs->shader, ev, output);
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
if (ev->transform.enabled || output->zoom.active ||
output->current_scale != ev->surface->buffer_viewport.buffer.scale)
filter = GL_LINEAR;
else
filter = GL_NEAREST;
for (i = 0; i < gs->num_textures; i++) {
glActiveTexture(GL_TEXTURE0 + i);
glBindTexture(gs->target, gs->textures[i]);
glTexParameteri(gs->target, GL_TEXTURE_MIN_FILTER, filter);
glTexParameteri(gs->target, GL_TEXTURE_MAG_FILTER, filter);
}
/* blended region is whole surface minus opaque region: */
pixman_region32_init_rect(&surface_blend, 0, 0,
ev->surface->width, ev->surface->height);
if (ev->geometry.scissor_enabled)
pixman_region32_intersect(&surface_blend, &surface_blend,
&ev->geometry.scissor);
pixman_region32_subtract(&surface_blend, &surface_blend,
&ev->surface->opaque);
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
/* XXX: Should we be using ev->transform.opaque here? */
pixman_region32_init(&surface_opaque);
if (ev->geometry.scissor_enabled)
pixman_region32_intersect(&surface_opaque,
&ev->surface->opaque,
&ev->geometry.scissor);
else
pixman_region32_copy(&surface_opaque, &ev->surface->opaque);
if (pixman_region32_not_empty(&surface_opaque)) {
if (gs->shader == &gr->texture_shader_rgba) {
/* Special case for RGBA textures with possibly
* bad data in alpha channel: use the shader
* that forces texture alpha = 1.0.
* Xwayland surfaces need this.
*/
use_shader(gr, &gr->texture_shader_rgbx);
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
shader_uniforms(&gr->texture_shader_rgbx, ev, output);
}
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
if (ev->alpha < 1.0)
glEnable(GL_BLEND);
else
glDisable(GL_BLEND);
repaint_region(ev, &repaint, &surface_opaque);
libweston: Support zwp_surface_synchronization_v1.get_release Implement the get_release request of the zwp_surface_synchronization_v1 interface. This commit implements the zwp_buffer_release_v1 interface. It supports the zwp_buffer_release_v1.fenced_release event for surfaces rendered by the GL renderer, and the zwp_buffer_release_v1.immediate_release event for other cases. Note that the immediate_release event is safe to be used for surface buffers used as planes in the DRM backend, since the backend releases them only after the next page flip that doesn't use the buffers has finished. Changes in v7: - Remove "partial" from commit title and description. - Fix inverted check when clearing used_in_output_repaint flag. Changes in v5: - Use the new, generic explicit sync server error reporting function. - Introduce and use weston_buffer_release_move. - Introduce internally and use weston_buffer_release_destroy. Changes in v4: - Support the zwp_buffer_release_v1.fenced_release event. - Support release fences in the GL renderer. - Assert that pending state buffer_release is always NULL after a commit. - Simplify weston_buffer_release_reference. - Move removal of destroy listener before resource destruction to avoid concerns about use-after-free in weston_buffer_release_reference - Rename weston_buffer_release_reference.busy_count to ref_count. - Add documentation for weston_buffer_release and ..._reference. Changes in v3: - Raise NO_BUFFER for get_release if no buffer has been committed, don't raise UNSUPPORTED_BUFFER for non-dmabuf buffers, so get_release works for all valid buffers. - Destroy the buffer_release object after sending an event. - Track lifetime of buffer_release objects per commit, independently of any buffers. - Use updated protocol interface names. - Use correct format specifier for resource ids. Changes in v2: - Raise UNSUPPORTED_BUFFER at commit if client has requested a buffer_release, but the committed buffer is not a valid linux_dmabuf. - Remove tests that are not viable anymore due to our inability to create dmabuf buffers and fences in a unit-test environment. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
gs->used_in_output_repaint = true;
}
if (pixman_region32_not_empty(&surface_blend)) {
use_shader(gr, gs->shader);
glEnable(GL_BLEND);
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
repaint_region(ev, &repaint, &surface_blend);
libweston: Support zwp_surface_synchronization_v1.get_release Implement the get_release request of the zwp_surface_synchronization_v1 interface. This commit implements the zwp_buffer_release_v1 interface. It supports the zwp_buffer_release_v1.fenced_release event for surfaces rendered by the GL renderer, and the zwp_buffer_release_v1.immediate_release event for other cases. Note that the immediate_release event is safe to be used for surface buffers used as planes in the DRM backend, since the backend releases them only after the next page flip that doesn't use the buffers has finished. Changes in v7: - Remove "partial" from commit title and description. - Fix inverted check when clearing used_in_output_repaint flag. Changes in v5: - Use the new, generic explicit sync server error reporting function. - Introduce and use weston_buffer_release_move. - Introduce internally and use weston_buffer_release_destroy. Changes in v4: - Support the zwp_buffer_release_v1.fenced_release event. - Support release fences in the GL renderer. - Assert that pending state buffer_release is always NULL after a commit. - Simplify weston_buffer_release_reference. - Move removal of destroy listener before resource destruction to avoid concerns about use-after-free in weston_buffer_release_reference - Rename weston_buffer_release_reference.busy_count to ref_count. - Add documentation for weston_buffer_release and ..._reference. Changes in v3: - Raise NO_BUFFER for get_release if no buffer has been committed, don't raise UNSUPPORTED_BUFFER for non-dmabuf buffers, so get_release works for all valid buffers. - Destroy the buffer_release object after sending an event. - Track lifetime of buffer_release objects per commit, independently of any buffers. - Use updated protocol interface names. - Use correct format specifier for resource ids. Changes in v2: - Raise UNSUPPORTED_BUFFER at commit if client has requested a buffer_release, but the committed buffer is not a valid linux_dmabuf. - Remove tests that are not viable anymore due to our inability to create dmabuf buffers and fences in a unit-test environment. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
gs->used_in_output_repaint = true;
}
pixman_region32_fini(&surface_blend);
pixman_region32_fini(&surface_opaque);
out:
pixman_region32_fini(&repaint);
if (replaced_shader)
gs->shader = replaced_shader;
}
static void
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
repaint_views(struct weston_output *output, pixman_region32_t *damage)
{
struct weston_compositor *compositor = output->compositor;
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
struct weston_view *view;
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
wl_list_for_each_reverse(view, &compositor->view_list, link)
if (view->plane == &compositor->primary_plane)
draw_view(view, output, damage);
}
libweston: Support zwp_surface_synchronization_v1.get_release Implement the get_release request of the zwp_surface_synchronization_v1 interface. This commit implements the zwp_buffer_release_v1 interface. It supports the zwp_buffer_release_v1.fenced_release event for surfaces rendered by the GL renderer, and the zwp_buffer_release_v1.immediate_release event for other cases. Note that the immediate_release event is safe to be used for surface buffers used as planes in the DRM backend, since the backend releases them only after the next page flip that doesn't use the buffers has finished. Changes in v7: - Remove "partial" from commit title and description. - Fix inverted check when clearing used_in_output_repaint flag. Changes in v5: - Use the new, generic explicit sync server error reporting function. - Introduce and use weston_buffer_release_move. - Introduce internally and use weston_buffer_release_destroy. Changes in v4: - Support the zwp_buffer_release_v1.fenced_release event. - Support release fences in the GL renderer. - Assert that pending state buffer_release is always NULL after a commit. - Simplify weston_buffer_release_reference. - Move removal of destroy listener before resource destruction to avoid concerns about use-after-free in weston_buffer_release_reference - Rename weston_buffer_release_reference.busy_count to ref_count. - Add documentation for weston_buffer_release and ..._reference. Changes in v3: - Raise NO_BUFFER for get_release if no buffer has been committed, don't raise UNSUPPORTED_BUFFER for non-dmabuf buffers, so get_release works for all valid buffers. - Destroy the buffer_release object after sending an event. - Track lifetime of buffer_release objects per commit, independently of any buffers. - Use updated protocol interface names. - Use correct format specifier for resource ids. Changes in v2: - Raise UNSUPPORTED_BUFFER at commit if client has requested a buffer_release, but the committed buffer is not a valid linux_dmabuf. - Remove tests that are not viable anymore due to our inability to create dmabuf buffers and fences in a unit-test environment. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
static int
gl_renderer_create_fence_fd(struct weston_output *output);
/* Updates the release fences of surfaces that were used in the current output
* repaint. Should only be used from gl_renderer_repaint_output, so that the
* information in gl_surface_state.used_in_output_repaint is accurate.
*/
static void
update_buffer_release_fences(struct weston_compositor *compositor,
struct weston_output *output)
{
struct weston_view *view;
wl_list_for_each_reverse(view, &compositor->view_list, link) {
struct gl_surface_state *gs;
struct weston_buffer_release *buffer_release;
int fence_fd;
if (view->plane != &compositor->primary_plane)
continue;
gs = get_surface_state(view->surface);
buffer_release = gs->buffer_release_ref.buffer_release;
if (!gs->used_in_output_repaint || !buffer_release)
continue;
fence_fd = gl_renderer_create_fence_fd(output);
/* If we have a buffer_release then it means we support fences,
* and we should be able to create the release fence. If we
* can't, something has gone horribly wrong, so disconnect the
* client.
*/
if (fence_fd == -1) {
linux_explicit_synchronization_send_server_error(
buffer_release->resource,
"Failed to create release fence");
fd_clear(&buffer_release->fence_fd);
continue;
}
/* At the moment it is safe to just replace the fence_fd,
* discarding the previous one:
*
* 1. If the previous fence fd represents a sync fence from
* a previous repaint cycle, that fence fd is now not
* sufficient to provide the release guarantee and should
* be replaced.
*
* 2. If the fence fd represents a sync fence from another
* output in the same repaint cycle, it's fine to replace
* it since we are rendering to all outputs using the same
* EGL context, so a fence issued for a later output rendering
* is guaranteed to signal after fences for previous output
* renderings.
*
* Note that the above is only valid if the buffer_release
* fences only originate from the GL renderer, which guarantees
* a total order of operations and fences. If we introduce
* fences from other sources (e.g., plane out-fences), we will
* need to merge fences instead.
*/
fd_update(&buffer_release->fence_fd, fence_fd);
}
}
static void
draw_output_border_texture(struct gl_output_state *go,
enum gl_renderer_border_side side,
int32_t x, int32_t y,
int32_t width, int32_t height)
{
struct gl_border_image *img = &go->borders[side];
static GLushort indices [] = { 0, 1, 3, 3, 1, 2 };
if (!img->data) {
if (img->tex) {
glDeleteTextures(1, &img->tex);
img->tex = 0;
}
return;
}
if (!img->tex) {
glGenTextures(1, &img->tex);
glBindTexture(GL_TEXTURE_2D, img->tex);
glTexParameteri(GL_TEXTURE_2D,
GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D,
GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D,
GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glTexParameteri(GL_TEXTURE_2D,
GL_TEXTURE_MAG_FILTER, GL_NEAREST);
} else {
glBindTexture(GL_TEXTURE_2D, img->tex);
}
if (go->border_status & (1 << side)) {
glPixelStorei(GL_UNPACK_ROW_LENGTH_EXT, 0);
glPixelStorei(GL_UNPACK_SKIP_PIXELS_EXT, 0);
glPixelStorei(GL_UNPACK_SKIP_ROWS_EXT, 0);
glTexImage2D(GL_TEXTURE_2D, 0, GL_BGRA_EXT,
img->tex_width, img->height, 0,
GL_BGRA_EXT, GL_UNSIGNED_BYTE, img->data);
}
GLfloat texcoord[] = {
0.0f, 0.0f,
(GLfloat)img->width / (GLfloat)img->tex_width, 0.0f,
(GLfloat)img->width / (GLfloat)img->tex_width, 1.0f,
0.0f, 1.0f,
};
GLfloat verts[] = {
x, y,
x + width, y,
x + width, y + height,
x, y + height
};
glVertexAttribPointer(0, 2, GL_FLOAT, GL_FALSE, 0, verts);
glVertexAttribPointer(1, 2, GL_FLOAT, GL_FALSE, 0, texcoord);
glEnableVertexAttribArray(0);
glEnableVertexAttribArray(1);
glDrawElements(GL_TRIANGLES, 6, GL_UNSIGNED_SHORT, indices);
glDisableVertexAttribArray(1);
glDisableVertexAttribArray(0);
}
static int
output_has_borders(struct weston_output *output)
{
struct gl_output_state *go = get_output_state(output);
return go->borders[GL_RENDERER_BORDER_TOP].data ||
go->borders[GL_RENDERER_BORDER_RIGHT].data ||
go->borders[GL_RENDERER_BORDER_BOTTOM].data ||
go->borders[GL_RENDERER_BORDER_LEFT].data;
}
static void
draw_output_borders(struct weston_output *output,
enum gl_border_status border_status)
{
struct gl_output_state *go = get_output_state(output);
struct gl_renderer *gr = get_renderer(output->compositor);
struct gl_shader *shader = &gr->texture_shader_rgba;
struct gl_border_image *top, *bottom, *left, *right;
struct weston_matrix matrix;
int full_width, full_height;
if (border_status == BORDER_STATUS_CLEAN)
return; /* Clean. Nothing to do. */
top = &go->borders[GL_RENDERER_BORDER_TOP];
bottom = &go->borders[GL_RENDERER_BORDER_BOTTOM];
left = &go->borders[GL_RENDERER_BORDER_LEFT];
right = &go->borders[GL_RENDERER_BORDER_RIGHT];
full_width = output->current_mode->width + left->width + right->width;
full_height = output->current_mode->height + top->height + bottom->height;
glDisable(GL_BLEND);
use_shader(gr, shader);
glViewport(0, 0, full_width, full_height);
weston_matrix_init(&matrix);
weston_matrix_translate(&matrix, -full_width/2.0, -full_height/2.0, 0);
weston_matrix_scale(&matrix, 2.0/full_width, -2.0/full_height, 1);
glUniformMatrix4fv(shader->proj_uniform, 1, GL_FALSE, matrix.d);
glUniform1i(shader->tex_uniforms[0], 0);
glUniform1f(shader->alpha_uniform, 1);
glActiveTexture(GL_TEXTURE0);
if (border_status & BORDER_TOP_DIRTY)
draw_output_border_texture(go, GL_RENDERER_BORDER_TOP,
0, 0,
full_width, top->height);
if (border_status & BORDER_LEFT_DIRTY)
draw_output_border_texture(go, GL_RENDERER_BORDER_LEFT,
0, top->height,
left->width, output->current_mode->height);
if (border_status & BORDER_RIGHT_DIRTY)
draw_output_border_texture(go, GL_RENDERER_BORDER_RIGHT,
full_width - right->width, top->height,
right->width, output->current_mode->height);
if (border_status & BORDER_BOTTOM_DIRTY)
draw_output_border_texture(go, GL_RENDERER_BORDER_BOTTOM,
0, full_height - bottom->height,
full_width, bottom->height);
}
static void
output_get_border_damage(struct weston_output *output,
enum gl_border_status border_status,
pixman_region32_t *damage)
{
struct gl_output_state *go = get_output_state(output);
struct gl_border_image *top, *bottom, *left, *right;
int full_width, full_height;
if (border_status == BORDER_STATUS_CLEAN)
return; /* Clean. Nothing to do. */
top = &go->borders[GL_RENDERER_BORDER_TOP];
bottom = &go->borders[GL_RENDERER_BORDER_BOTTOM];
left = &go->borders[GL_RENDERER_BORDER_LEFT];
right = &go->borders[GL_RENDERER_BORDER_RIGHT];
full_width = output->current_mode->width + left->width + right->width;
full_height = output->current_mode->height + top->height + bottom->height;
if (border_status & BORDER_TOP_DIRTY)
pixman_region32_union_rect(damage, damage,
0, 0,
full_width, top->height);
if (border_status & BORDER_LEFT_DIRTY)
pixman_region32_union_rect(damage, damage,
0, top->height,
left->width, output->current_mode->height);
if (border_status & BORDER_RIGHT_DIRTY)
pixman_region32_union_rect(damage, damage,
full_width - right->width, top->height,
right->width, output->current_mode->height);
if (border_status & BORDER_BOTTOM_DIRTY)
pixman_region32_union_rect(damage, damage,
0, full_height - bottom->height,
full_width, bottom->height);
}
static void
output_get_damage(struct weston_output *output,
pixman_region32_t *buffer_damage, uint32_t *border_damage)
{
struct gl_output_state *go = get_output_state(output);
struct gl_renderer *gr = get_renderer(output->compositor);
EGLint buffer_age = 0;
EGLBoolean ret;
int i;
if (gr->has_egl_buffer_age) {
ret = eglQuerySurface(gr->egl_display, go->egl_surface,
EGL_BUFFER_AGE_EXT, &buffer_age);
if (ret == EGL_FALSE) {
weston_log("buffer age query failed.\n");
gl_renderer_print_egl_error_state();
}
}
if (buffer_age == 0 || buffer_age - 1 > BUFFER_DAMAGE_COUNT) {
pixman_region32_copy(buffer_damage, &output->region);
*border_damage = BORDER_ALL_DIRTY;
} else {
for (i = 0; i < buffer_age - 1; i++)
*border_damage |= go->border_damage[(go->buffer_damage_index + i) % BUFFER_DAMAGE_COUNT];
if (*border_damage & BORDER_SIZE_CHANGED) {
/* If we've had a resize, we have to do a full
* repaint. */
*border_damage |= BORDER_ALL_DIRTY;
pixman_region32_copy(buffer_damage, &output->region);
} else {
for (i = 0; i < buffer_age - 1; i++)
pixman_region32_union(buffer_damage,
buffer_damage,
&go->buffer_damage[(go->buffer_damage_index + i) % BUFFER_DAMAGE_COUNT]);
}
}
}
static void
output_rotate_damage(struct weston_output *output,
pixman_region32_t *output_damage,
enum gl_border_status border_status)
{
struct gl_output_state *go = get_output_state(output);
struct gl_renderer *gr = get_renderer(output->compositor);
if (!gr->has_egl_buffer_age)
return;
go->buffer_damage_index += BUFFER_DAMAGE_COUNT - 1;
go->buffer_damage_index %= BUFFER_DAMAGE_COUNT;
pixman_region32_copy(&go->buffer_damage[go->buffer_damage_index], output_damage);
go->border_damage[go->buffer_damage_index] = border_status;
}
/**
* Given a region in Weston's (top-left-origin) global co-ordinate space,
* translate it to the co-ordinate space used by GL for our output
* rendering. This requires shifting it into output co-ordinate space:
* translating for output offset within the global co-ordinate space,
* multiplying by output scale to get buffer rather than logical size.
*
* Finally, if borders are drawn around the output, we translate the area
* to account for the border region around the outside, and add any
* damage if the borders have been redrawn.
*
* @param output The output whose co-ordinate space we are after
* @param global_region The affected region in global co-ordinate space
* @param[out] rects Y-inverted quads in {x,y,w,h} order; caller must free
* @param[out] nrects Number of quads (4x number of co-ordinates)
*/
static void
pixman_region_to_egl_y_invert(struct weston_output *output,
struct pixman_region32 *global_region,
EGLint **rects,
EGLint *nrects)
{
struct gl_output_state *go = get_output_state(output);
pixman_region32_t transformed;
struct pixman_box32 *box;
int buffer_height;
EGLint *d;
int i;
/* Translate from global to output co-ordinate space. */
pixman_region32_init(&transformed);
pixman_region32_copy(&transformed, global_region);
pixman_region32_translate(&transformed, -output->x, -output->y);
weston_transformed_region(output->width, output->height,
output->transform,
output->current_scale,
&transformed, &transformed);
/* If we have borders drawn around the output, shift our output damage
* to account for borders being drawn around the outside, adding any
* damage resulting from borders being redrawn. */
if (output_has_borders(output)) {
pixman_region32_translate(&transformed,
go->borders[GL_RENDERER_BORDER_LEFT].width,
go->borders[GL_RENDERER_BORDER_TOP].height);
output_get_border_damage(output, go->border_status,
&transformed);
}
/* Convert from a Pixman region into {x,y,w,h} quads, flipping in the
* Y axis to account for GL's lower-left-origin co-ordinate space. */
box = pixman_region32_rectangles(&transformed, nrects);
*rects = malloc(*nrects * 4 * sizeof(EGLint));
buffer_height = go->borders[GL_RENDERER_BORDER_TOP].height +
output->current_mode->height +
go->borders[GL_RENDERER_BORDER_BOTTOM].height;
d = *rects;
for (i = 0; i < *nrects; ++i) {
*d++ = box[i].x1;
*d++ = buffer_height - box[i].y2;
*d++ = box[i].x2 - box[i].x1;
*d++ = box[i].y2 - box[i].y1;
}
pixman_region32_fini(&transformed);
}
/* NOTE: We now allow falling back to ARGB gl visuals when XRGB is
* unavailable, so we're assuming the background has no transparency
* and that everything with a blend, like drop shadows, will have something
* opaque (like the background) drawn underneath it.
*
* Depending on the underlying hardware, violating that assumption could
* result in seeing through to another display plane.
*/
static void
gl_renderer_repaint_output(struct weston_output *output,
pixman_region32_t *output_damage)
{
struct gl_output_state *go = get_output_state(output);
struct weston_compositor *compositor = output->compositor;
struct gl_renderer *gr = get_renderer(compositor);
EGLBoolean ret;
static int errored;
/* areas we've damaged since we last used this buffer */
pixman_region32_t previous_damage;
/* total area we need to repaint this time */
pixman_region32_t total_damage;
enum gl_border_status border_status = BORDER_STATUS_CLEAN;
libweston: Support zwp_surface_synchronization_v1.get_release Implement the get_release request of the zwp_surface_synchronization_v1 interface. This commit implements the zwp_buffer_release_v1 interface. It supports the zwp_buffer_release_v1.fenced_release event for surfaces rendered by the GL renderer, and the zwp_buffer_release_v1.immediate_release event for other cases. Note that the immediate_release event is safe to be used for surface buffers used as planes in the DRM backend, since the backend releases them only after the next page flip that doesn't use the buffers has finished. Changes in v7: - Remove "partial" from commit title and description. - Fix inverted check when clearing used_in_output_repaint flag. Changes in v5: - Use the new, generic explicit sync server error reporting function. - Introduce and use weston_buffer_release_move. - Introduce internally and use weston_buffer_release_destroy. Changes in v4: - Support the zwp_buffer_release_v1.fenced_release event. - Support release fences in the GL renderer. - Assert that pending state buffer_release is always NULL after a commit. - Simplify weston_buffer_release_reference. - Move removal of destroy listener before resource destruction to avoid concerns about use-after-free in weston_buffer_release_reference - Rename weston_buffer_release_reference.busy_count to ref_count. - Add documentation for weston_buffer_release and ..._reference. Changes in v3: - Raise NO_BUFFER for get_release if no buffer has been committed, don't raise UNSUPPORTED_BUFFER for non-dmabuf buffers, so get_release works for all valid buffers. - Destroy the buffer_release object after sending an event. - Track lifetime of buffer_release objects per commit, independently of any buffers. - Use updated protocol interface names. - Use correct format specifier for resource ids. Changes in v2: - Raise UNSUPPORTED_BUFFER at commit if client has requested a buffer_release, but the committed buffer is not a valid linux_dmabuf. - Remove tests that are not viable anymore due to our inability to create dmabuf buffers and fences in a unit-test environment. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
struct weston_view *view;
if (use_output(output) < 0)
return;
libweston: Support zwp_surface_synchronization_v1.get_release Implement the get_release request of the zwp_surface_synchronization_v1 interface. This commit implements the zwp_buffer_release_v1 interface. It supports the zwp_buffer_release_v1.fenced_release event for surfaces rendered by the GL renderer, and the zwp_buffer_release_v1.immediate_release event for other cases. Note that the immediate_release event is safe to be used for surface buffers used as planes in the DRM backend, since the backend releases them only after the next page flip that doesn't use the buffers has finished. Changes in v7: - Remove "partial" from commit title and description. - Fix inverted check when clearing used_in_output_repaint flag. Changes in v5: - Use the new, generic explicit sync server error reporting function. - Introduce and use weston_buffer_release_move. - Introduce internally and use weston_buffer_release_destroy. Changes in v4: - Support the zwp_buffer_release_v1.fenced_release event. - Support release fences in the GL renderer. - Assert that pending state buffer_release is always NULL after a commit. - Simplify weston_buffer_release_reference. - Move removal of destroy listener before resource destruction to avoid concerns about use-after-free in weston_buffer_release_reference - Rename weston_buffer_release_reference.busy_count to ref_count. - Add documentation for weston_buffer_release and ..._reference. Changes in v3: - Raise NO_BUFFER for get_release if no buffer has been committed, don't raise UNSUPPORTED_BUFFER for non-dmabuf buffers, so get_release works for all valid buffers. - Destroy the buffer_release object after sending an event. - Track lifetime of buffer_release objects per commit, independently of any buffers. - Use updated protocol interface names. - Use correct format specifier for resource ids. Changes in v2: - Raise UNSUPPORTED_BUFFER at commit if client has requested a buffer_release, but the committed buffer is not a valid linux_dmabuf. - Remove tests that are not viable anymore due to our inability to create dmabuf buffers and fences in a unit-test environment. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
/* Clear the used_in_output_repaint flag, so that we can properly track
* which surfaces were used in this output repaint. */
wl_list_for_each_reverse(view, &compositor->view_list, link) {
if (view->plane == &compositor->primary_plane) {
struct gl_surface_state *gs =
get_surface_state(view->surface);
gs->used_in_output_repaint = false;
}
}
if (go->begin_render_sync != EGL_NO_SYNC_KHR)
gr->destroy_sync(gr->egl_display, go->begin_render_sync);
if (go->end_render_sync != EGL_NO_SYNC_KHR)
gr->destroy_sync(gr->egl_display, go->end_render_sync);
go->begin_render_sync = create_render_sync(gr);
/* Calculate the viewport */
glViewport(go->borders[GL_RENDERER_BORDER_LEFT].width,
go->borders[GL_RENDERER_BORDER_BOTTOM].height,
output->current_mode->width,
output->current_mode->height);
/* Calculate the global GL matrix */
go->output_matrix = output->matrix;
weston_matrix_translate(&go->output_matrix,
-(output->current_mode->width / 2.0),
-(output->current_mode->height / 2.0), 0);
weston_matrix_scale(&go->output_matrix,
2.0 / output->current_mode->width,
-2.0 / output->current_mode->height, 1);
/* In fan debug mode, redraw everything to make sure that we clear any
* fans left over from previous draws on this buffer.
* This precludes the use of EGL_EXT_swap_buffers_with_damage and
* EGL_KHR_partial_update, since we damage the whole area. */
if (gr->fan_debug) {
pixman_region32_t undamaged;
pixman_region32_init(&undamaged);
pixman_region32_subtract(&undamaged, &output->region,
output_damage);
gr->fan_debug = false;
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
repaint_views(output, &undamaged);
gr->fan_debug = true;
pixman_region32_fini(&undamaged);
}
/* previous_damage covers regions damaged in previous paints since we
* last used this buffer */
pixman_region32_init(&previous_damage);
pixman_region32_init(&total_damage); /* total area to redraw */
/* Update previous_damage using buffer_age (if available), and store
* current damaged region for future use. */
output_get_damage(output, &previous_damage, &border_status);
output_rotate_damage(output, output_damage, go->border_status);
/* Redraw both areas which have changed since we last used this buffer,
* as well as the areas we now want to repaint, to make sure the
* buffer is up to date. */
pixman_region32_union(&total_damage, &previous_damage, output_damage);
border_status |= go->border_status;
if (gr->has_egl_partial_update && !gr->fan_debug) {
int n_egl_rects;
EGLint *egl_rects;
/* For partial_update, we need to pass the region which has
* changed since we last rendered into this specific buffer;
* this is total_damage. */
pixman_region_to_egl_y_invert(output, &total_damage,
&egl_rects, &n_egl_rects);
gr->set_damage_region(gr->egl_display, go->egl_surface,
egl_rects, n_egl_rects);
free(egl_rects);
}
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
repaint_views(output, &total_damage);
pixman_region32_fini(&total_damage);
pixman_region32_fini(&previous_damage);
draw_output_borders(output, border_status);
wl_signal_emit(&output->frame_signal, output_damage);
go->end_render_sync = create_render_sync(gr);
if (gr->swap_buffers_with_damage && !gr->fan_debug) {
int n_egl_rects;
EGLint *egl_rects;
/* For swap_buffers_with_damage, we need to pass the region
* which has changed since the previous SwapBuffers on this
* surface - this is output_damage. */
pixman_region_to_egl_y_invert(output, output_damage,
&egl_rects, &n_egl_rects);
ret = gr->swap_buffers_with_damage(gr->egl_display,
go->egl_surface,
egl_rects, n_egl_rects);
free(egl_rects);
} else {
ret = eglSwapBuffers(gr->egl_display, go->egl_surface);
}
if (ret == EGL_FALSE && !errored) {
errored = 1;
weston_log("Failed in eglSwapBuffers.\n");
gl_renderer_print_egl_error_state();
}
go->border_status = BORDER_STATUS_CLEAN;
/* We have to submit the render sync objects after swap buffers, since
* the objects get assigned a valid sync file fd only after a gl flush.
*/
timeline_submit_render_sync(gr, compositor, output,
go->begin_render_sync,
TIMELINE_RENDER_POINT_TYPE_BEGIN);
timeline_submit_render_sync(gr, compositor, output, go->end_render_sync,
TIMELINE_RENDER_POINT_TYPE_END);
libweston: Support zwp_surface_synchronization_v1.get_release Implement the get_release request of the zwp_surface_synchronization_v1 interface. This commit implements the zwp_buffer_release_v1 interface. It supports the zwp_buffer_release_v1.fenced_release event for surfaces rendered by the GL renderer, and the zwp_buffer_release_v1.immediate_release event for other cases. Note that the immediate_release event is safe to be used for surface buffers used as planes in the DRM backend, since the backend releases them only after the next page flip that doesn't use the buffers has finished. Changes in v7: - Remove "partial" from commit title and description. - Fix inverted check when clearing used_in_output_repaint flag. Changes in v5: - Use the new, generic explicit sync server error reporting function. - Introduce and use weston_buffer_release_move. - Introduce internally and use weston_buffer_release_destroy. Changes in v4: - Support the zwp_buffer_release_v1.fenced_release event. - Support release fences in the GL renderer. - Assert that pending state buffer_release is always NULL after a commit. - Simplify weston_buffer_release_reference. - Move removal of destroy listener before resource destruction to avoid concerns about use-after-free in weston_buffer_release_reference - Rename weston_buffer_release_reference.busy_count to ref_count. - Add documentation for weston_buffer_release and ..._reference. Changes in v3: - Raise NO_BUFFER for get_release if no buffer has been committed, don't raise UNSUPPORTED_BUFFER for non-dmabuf buffers, so get_release works for all valid buffers. - Destroy the buffer_release object after sending an event. - Track lifetime of buffer_release objects per commit, independently of any buffers. - Use updated protocol interface names. - Use correct format specifier for resource ids. Changes in v2: - Raise UNSUPPORTED_BUFFER at commit if client has requested a buffer_release, but the committed buffer is not a valid linux_dmabuf. - Remove tests that are not viable anymore due to our inability to create dmabuf buffers and fences in a unit-test environment. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
update_buffer_release_fences(compositor, output);
}
static int
gl_renderer_read_pixels(struct weston_output *output,
pixman_format_code_t format, void *pixels,
uint32_t x, uint32_t y,
uint32_t width, uint32_t height)
{
GLenum gl_format;
struct gl_output_state *go = get_output_state(output);
x += go->borders[GL_RENDERER_BORDER_LEFT].width;
y += go->borders[GL_RENDERER_BORDER_BOTTOM].height;
switch (format) {
case PIXMAN_a8r8g8b8:
gl_format = GL_BGRA_EXT;
break;
case PIXMAN_a8b8g8r8:
gl_format = GL_RGBA;
break;
default:
return -1;
}
if (use_output(output) < 0)
return -1;
glPixelStorei(GL_PACK_ALIGNMENT, 1);
glReadPixels(x, y, width, height, gl_format,
GL_UNSIGNED_BYTE, pixels);
return 0;
}
static GLenum gl_format_from_internal(GLenum internal_format)
{
switch (internal_format) {
case GL_R8_EXT:
return GL_RED_EXT;
case GL_RG8_EXT:
return GL_RG_EXT;
default:
return internal_format;
}
}
static void
gl_renderer_flush_damage(struct weston_surface *surface)
{
struct gl_renderer *gr = get_renderer(surface->compositor);
struct gl_surface_state *gs = get_surface_state(surface);
struct weston_buffer *buffer = gs->buffer_ref.buffer;
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
struct weston_view *view;
bool texture_used;
pixman_box32_t *rectangles;
uint8_t *data;
int i, j, n;
pixman_region32_union(&gs->texture_damage,
&gs->texture_damage, &surface->damage);
if (!buffer)
return;
/* Avoid upload, if the texture won't be used this time.
* We still accumulate the damage in texture_damage, and
* hold the reference to the buffer, in case the surface
* migrates back to the primary plane.
*/
texture_used = false;
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
wl_list_for_each(view, &surface->views, surface_link) {
if (view->plane == &surface->compositor->primary_plane) {
texture_used = true;
Split the geometry information from weston_surface out into weston_view The weston_surface structure is split into two structures: * The weston_surface structure storres everything required for a client-side or server-side surface. This includes buffers; callbacks; backend private data; input, damage, and opaque regions; and a few other bookkeeping bits. * The weston_view structure represents an entity in the scenegraph and storres all of the geometry information. This includes clip region, alpha, position, and the transformation list as well as all of the temporary information derived from the geometry state. Because a view, and not a surface, is a scenegraph element, the view is what is placed in layers and planes. There are a few things worth noting about the surface/view split: 1. This is *not* a modification to the protocol. It is, instead, a modification to Weston's internal scenegraph to allow a single surface to exist in multiple places at a time. Clients are completely unaware of how many views to a particular surface exist. 2. A view is considered a direct child of a surface and is destroyed when the surface is destroyed. Because of this, the view.surface pointer is always valid and non-null. 3. The compositor's surface_list is replaced with a view_list. Due to subsurfaces, building the view list is a little more complicated than it used to be and involves building a tree of views on the fly whenever subsurfaces are used. However, this means that backends can remain completely subsurface-agnostic. 4. Surfaces and views both keep track of which outputs they are on. 5. The weston_surface structure now has width and height fields. These are populated when a new buffer is attached before surface.configure is called. This is because there are many surface-based operations that really require the width and height and digging through the views didn't work well. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
11 years ago
break;
}
}
if (!texture_used)
return;
if (!pixman_region32_not_empty(&gs->texture_damage) &&
!gs->needs_full_upload)
goto done;
data = wl_shm_buffer_get_data(buffer->shm_buffer);
if (!gr->has_unpack_subimage) {
wl_shm_buffer_begin_access(buffer->shm_buffer);
for (j = 0; j < gs->num_textures; j++) {
glBindTexture(GL_TEXTURE_2D, gs->textures[j]);
glTexImage2D(GL_TEXTURE_2D, 0,
gs->gl_format[j],
gs->pitch / gs->hsub[j],
buffer->height / gs->vsub[j],
0,
gl_format_from_internal(gs->gl_format[j]),
gs->gl_pixel_type,
data + gs->offset[j]);
}
wl_shm_buffer_end_access(buffer->shm_buffer);
goto done;
}
if (gs->needs_full_upload) {
glPixelStorei(GL_UNPACK_SKIP_PIXELS_EXT, 0);
glPixelStorei(GL_UNPACK_SKIP_ROWS_EXT, 0);
wl_shm_buffer_begin_access(buffer->shm_buffer);
for (j = 0; j < gs->num_textures; j++) {
glBindTexture(GL_TEXTURE_2D, gs->textures[j]);
7 years ago
glPixelStorei(GL_UNPACK_ROW_LENGTH_EXT,
gs->pitch / gs->hsub[j]);
glTexImage2D(GL_TEXTURE_2D, 0,
gs->gl_format[j],
gs->pitch / gs->hsub[j],
buffer->height / gs->vsub[j],
0,
gl_format_from_internal(gs->gl_format[j]),
gs->gl_pixel_type,
data + gs->offset[j]);
}
wl_shm_buffer_end_access(buffer->shm_buffer);
goto done;
}
rectangles = pixman_region32_rectangles(&gs->texture_damage, &n);
wl_shm_buffer_begin_access(buffer->shm_buffer);
for (i = 0; i < n; i++) {
pixman_box32_t r;
r = weston_surface_to_buffer_rect(surface, rectangles[i]);
for (j = 0; j < gs->num_textures; j++) {
glBindTexture(GL_TEXTURE_2D, gs->textures[j]);
7 years ago
glPixelStorei(GL_UNPACK_ROW_LENGTH_EXT,
gs->pitch / gs->hsub[j]);
glPixelStorei(GL_UNPACK_SKIP_PIXELS_EXT,
r.x1 / gs->hsub[j]);
glPixelStorei(GL_UNPACK_SKIP_ROWS_EXT,
r.y1 / gs->hsub[j]);
glTexSubImage2D(GL_TEXTURE_2D, 0,
r.x1 / gs->hsub[j],
r.y1 / gs->vsub[j],
(r.x2 - r.x1) / gs->hsub[j],
(r.y2 - r.y1) / gs->vsub[j],
gl_format_from_internal(gs->gl_format[j]),
gs->gl_pixel_type,
data + gs->offset[j]);
}
}
wl_shm_buffer_end_access(buffer->shm_buffer);
done:
pixman_region32_fini(&gs->texture_damage);
pixman_region32_init(&gs->texture_damage);
gs->needs_full_upload = false;
weston_buffer_reference(&gs->buffer_ref, NULL);
libweston: Support zwp_surface_synchronization_v1.get_release Implement the get_release request of the zwp_surface_synchronization_v1 interface. This commit implements the zwp_buffer_release_v1 interface. It supports the zwp_buffer_release_v1.fenced_release event for surfaces rendered by the GL renderer, and the zwp_buffer_release_v1.immediate_release event for other cases. Note that the immediate_release event is safe to be used for surface buffers used as planes in the DRM backend, since the backend releases them only after the next page flip that doesn't use the buffers has finished. Changes in v7: - Remove "partial" from commit title and description. - Fix inverted check when clearing used_in_output_repaint flag. Changes in v5: - Use the new, generic explicit sync server error reporting function. - Introduce and use weston_buffer_release_move. - Introduce internally and use weston_buffer_release_destroy. Changes in v4: - Support the zwp_buffer_release_v1.fenced_release event. - Support release fences in the GL renderer. - Assert that pending state buffer_release is always NULL after a commit. - Simplify weston_buffer_release_reference. - Move removal of destroy listener before resource destruction to avoid concerns about use-after-free in weston_buffer_release_reference - Rename weston_buffer_release_reference.busy_count to ref_count. - Add documentation for weston_buffer_release and ..._reference. Changes in v3: - Raise NO_BUFFER for get_release if no buffer has been committed, don't raise UNSUPPORTED_BUFFER for non-dmabuf buffers, so get_release works for all valid buffers. - Destroy the buffer_release object after sending an event. - Track lifetime of buffer_release objects per commit, independently of any buffers. - Use updated protocol interface names. - Use correct format specifier for resource ids. Changes in v2: - Raise UNSUPPORTED_BUFFER at commit if client has requested a buffer_release, but the committed buffer is not a valid linux_dmabuf. - Remove tests that are not viable anymore due to our inability to create dmabuf buffers and fences in a unit-test environment. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
weston_buffer_release_reference(&gs->buffer_release_ref, NULL);
}
static void
ensure_textures(struct gl_surface_state *gs, int num_textures)
{
int i;
if (num_textures <= gs->num_textures)
return;
for (i = gs->num_textures; i < num_textures; i++) {
glGenTextures(1, &gs->textures[i]);
glBindTexture(gs->target, gs->textures[i]);
glTexParameteri(gs->target,
GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(gs->target,
GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
}
gs->num_textures = num_textures;
glBindTexture(gs->target, 0);
}
static void
gl_renderer_attach_shm(struct weston_surface *es, struct weston_buffer *buffer,
struct wl_shm_buffer *shm_buffer)
{
struct weston_compositor *ec = es->compositor;
struct gl_renderer *gr = get_renderer(ec);
struct gl_surface_state *gs = get_surface_state(es);
GLenum gl_format[3] = {0, 0, 0};
GLenum gl_pixel_type;
int pitch;
int num_planes;
buffer->shm_buffer = shm_buffer;
buffer->width = wl_shm_buffer_get_width(shm_buffer);
buffer->height = wl_shm_buffer_get_height(shm_buffer);
num_planes = 1;
gs->offset[0] = 0;
gs->hsub[0] = 1;
gs->vsub[0] = 1;
switch (wl_shm_buffer_get_format(shm_buffer)) {
case WL_SHM_FORMAT_XRGB8888:
gs->shader = &gr->texture_shader_rgbx;
pitch = wl_shm_buffer_get_stride(shm_buffer) / 4;
gl_format[0] = GL_BGRA_EXT;
gl_pixel_type = GL_UNSIGNED_BYTE;
es->is_opaque = true;
break;
case WL_SHM_FORMAT_ARGB8888:
gs->shader = &gr->texture_shader_rgba;
pitch = wl_shm_buffer_get_stride(shm_buffer) / 4;
gl_format[0] = GL_BGRA_EXT;
gl_pixel_type = GL_UNSIGNED_BYTE;
es->is_opaque = false;
break;
case WL_SHM_FORMAT_RGB565:
gs->shader = &gr->texture_shader_rgbx;
pitch = wl_shm_buffer_get_stride(shm_buffer) / 2;
gl_format[0] = GL_RGB;
gl_pixel_type = GL_UNSIGNED_SHORT_5_6_5;
es->is_opaque = true;
break;
case WL_SHM_FORMAT_YUV420:
gs->shader = &gr->texture_shader_y_u_v;
pitch = wl_shm_buffer_get_stride(shm_buffer);
gl_pixel_type = GL_UNSIGNED_BYTE;
num_planes = 3;
gs->offset[1] = gs->offset[0] + (pitch / gs->hsub[0]) *
(buffer->height / gs->vsub[0]);
gs->hsub[1] = 2;
gs->vsub[1] = 2;
gs->offset[2] = gs->offset[1] + (pitch / gs->hsub[1]) *
(buffer->height / gs->vsub[1]);
gs->hsub[2] = 2;
gs->vsub[2] = 2;
if (gr->has_gl_texture_rg) {
gl_format[0] = GL_R8_EXT;
gl_format[1] = GL_R8_EXT;
gl_format[2] = GL_R8_EXT;
} else {
gl_format[0] = GL_LUMINANCE;
gl_format[1] = GL_LUMINANCE;
gl_format[2] = GL_LUMINANCE;
}
es->is_opaque = true;
break;
case WL_SHM_FORMAT_NV12:
pitch = wl_shm_buffer_get_stride(shm_buffer);
gl_pixel_type = GL_UNSIGNED_BYTE;
num_planes = 2;
gs->offset[1] = gs->offset[0] + (pitch / gs->hsub[0]) *
(buffer->height / gs->vsub[0]);
gs->hsub[1] = 2;
gs->vsub[1] = 2;
if (gr->has_gl_texture_rg) {
gs->shader = &gr->texture_shader_y_uv;
gl_format[0] = GL_R8_EXT;
gl_format[1] = GL_RG8_EXT;
} else {
gs->shader = &gr->texture_shader_y_xuxv;
gl_format[0] = GL_LUMINANCE;
gl_format[1] = GL_LUMINANCE_ALPHA;
}
es->is_opaque = true;
break;
case WL_SHM_FORMAT_YUYV:
gs->shader = &gr->texture_shader_y_xuxv;
pitch = wl_shm_buffer_get_stride(shm_buffer) / 2;
gl_pixel_type = GL_UNSIGNED_BYTE;
num_planes = 2;
gs->offset[1] = 0;
gs->hsub[1] = 2;
gs->vsub[1] = 1;
if (gr->has_gl_texture_rg)
gl_format[0] = GL_RG8_EXT;
else
gl_format[0] = GL_LUMINANCE_ALPHA;
gl_format[1] = GL_BGRA_EXT;
es->is_opaque = true;
break;
default:
weston_log("warning: unknown shm buffer format: %08x\n",
wl_shm_buffer_get_format(shm_buffer));
return;
}
/* Only allocate a texture if it doesn't match existing one.
* If a switch from DRM allocated buffer to a SHM buffer is
* happening, we need to allocate a new texture buffer. */
if (pitch != gs->pitch ||
buffer->height != gs->height ||
gl_format[0] != gs->gl_format[0] ||
gl_format[1] != gs->gl_format[1] ||
gl_format[2] != gs->gl_format[2] ||
gl_pixel_type != gs->gl_pixel_type ||
gs->buffer_type != BUFFER_TYPE_SHM) {
gs->pitch = pitch;
gs->height = buffer->height;
gs->target = GL_TEXTURE_2D;
gs->gl_format[0] = gl_format[0];
gs->gl_format[1] = gl_format[1];
gs->gl_format[2] = gl_format[2];
gs->gl_pixel_type = gl_pixel_type;
gs->buffer_type = BUFFER_TYPE_SHM;
gs->needs_full_upload = true;
gs->y_inverted = true;
gs->direct_display = false;
gs->surface = es;
ensure_textures(gs, num_planes);
}
}
static void
gl_renderer_attach_egl(struct weston_surface *es, struct weston_buffer *buffer,
uint32_t format)
{
struct weston_compositor *ec = es->compositor;
struct gl_renderer *gr = get_renderer(ec);
struct gl_surface_state *gs = get_surface_state(es);
EGLint attribs[3];
int i, num_planes;
buffer->legacy_buffer = (struct wl_buffer *)buffer->resource;
gr->query_buffer(gr->egl_display, buffer->legacy_buffer,
EGL_WIDTH, &buffer->width);
gr->query_buffer(gr->egl_display, buffer->legacy_buffer,
EGL_HEIGHT, &buffer->height);
gr->query_buffer(gr->egl_display, buffer->legacy_buffer,
EGL_WAYLAND_Y_INVERTED_WL, &buffer->y_inverted);
for (i = 0; i < gs->num_images; i++) {
egl_image_unref(gs->images[i]);
gs->images[i] = NULL;
}
gs->num_images = 0;
gs->target = GL_TEXTURE_2D;
es->is_opaque = false;
switch (format) {
case EGL_TEXTURE_RGB:
es->is_opaque = true;
/* fallthrough */
case EGL_TEXTURE_RGBA:
default:
num_planes = 1;
gs->shader = &gr->texture_shader_rgba;
break;
case EGL_TEXTURE_EXTERNAL_WL:
num_planes = 1;
gs->target = GL_TEXTURE_EXTERNAL_OES;
gs->shader = &gr->texture_shader_egl_external;
break;
case EGL_TEXTURE_Y_UV_WL:
num_planes = 2;
gs->shader = &gr->texture_shader_y_uv;
es->is_opaque = true;
break;
case EGL_TEXTURE_Y_U_V_WL:
num_planes = 3;
gs->shader = &gr->texture_shader_y_u_v;
es->is_opaque = true;
break;
case EGL_TEXTURE_Y_XUXV_WL:
num_planes = 2;
gs->shader = &gr->texture_shader_y_xuxv;
es->is_opaque = true;
break;
}
ensure_textures(gs, num_planes);
for (i = 0; i < num_planes; i++) {
attribs[0] = EGL_WAYLAND_PLANE_WL;
attribs[1] = i;
attribs[2] = EGL_NONE;
gs->images[i] = egl_image_create(gr,
EGL_WAYLAND_BUFFER_WL,
buffer->legacy_buffer,
attribs);
if (!gs->images[i]) {
weston_log("failed to create img for plane %d\n", i);
continue;
}
gs->num_images++;
glActiveTexture(GL_TEXTURE0 + i);
glBindTexture(gs->target, gs->textures[i]);
gr->image_target_texture_2d(gs->target,
gs->images[i]->image);
}
gs->pitch = buffer->width;
gs->height = buffer->height;
gs->buffer_type = BUFFER_TYPE_EGL;
gs->y_inverted = buffer->y_inverted;
}
static void
gl_renderer_destroy_dmabuf(struct linux_dmabuf_buffer *dmabuf)
{
struct dmabuf_image *image = linux_dmabuf_buffer_get_user_data(dmabuf);
dmabuf_image_destroy(image);
}
static struct egl_image *
import_simple_dmabuf(struct gl_renderer *gr,
struct dmabuf_attributes *attributes)
{
struct egl_image *image;
EGLint attribs[50];
int atti = 0;
bool has_modifier;
/* This requires the Mesa commit in
* Mesa 10.3 (08264e5dad4df448e7718e782ad9077902089a07) or
* Mesa 10.2.7 (55d28925e6109a4afd61f109e845a8a51bd17652).
* Otherwise Mesa closes the fd behind our back and re-importing
* will fail.
* https://bugs.freedesktop.org/show_bug.cgi?id=76188
*/
attribs[atti++] = EGL_WIDTH;
attribs[atti++] = attributes->width;
attribs[atti++] = EGL_HEIGHT;
attribs[atti++] = attributes->height;
attribs[atti++] = EGL_LINUX_DRM_FOURCC_EXT;
attribs[atti++] = attributes->format;
if (attributes->modifier[0] != DRM_FORMAT_MOD_INVALID) {
if (!gr->has_dmabuf_import_modifiers)
return NULL;
has_modifier = true;
} else {
has_modifier = false;
}
if (attributes->n_planes > 0) {
attribs[atti++] = EGL_DMA_BUF_PLANE0_FD_EXT;
attribs[atti++] = attributes->fd[0];
attribs[atti++] = EGL_DMA_BUF_PLANE0_OFFSET_EXT;
attribs[atti++] = attributes->offset[0];
attribs[atti++] = EGL_DMA_BUF_PLANE0_PITCH_EXT;
attribs[atti++] = attributes->stride[0];
if (has_modifier) {
attribs[atti++] = EGL_DMA_BUF_PLANE0_MODIFIER_LO_EXT;
attribs[atti++] = attributes->modifier[0] & 0xFFFFFFFF;
attribs[atti++] = EGL_DMA_BUF_PLANE0_MODIFIER_HI_EXT;
attribs[atti++] = attributes->modifier[0] >> 32;
}
}
if (attributes->n_planes > 1) {
attribs[atti++] = EGL_DMA_BUF_PLANE1_FD_EXT;
attribs[atti++] = attributes->fd[1];
attribs[atti++] = EGL_DMA_BUF_PLANE1_OFFSET_EXT;
attribs[atti++] = attributes->offset[1];
attribs[atti++] = EGL_DMA_BUF_PLANE1_PITCH_EXT;
attribs[atti++] = attributes->stride[1];
if (has_modifier) {
attribs[atti++] = EGL_DMA_BUF_PLANE1_MODIFIER_LO_EXT;
attribs[atti++] = attributes->modifier[1] & 0xFFFFFFFF;
attribs[atti++] = EGL_DMA_BUF_PLANE1_MODIFIER_HI_EXT;
attribs[atti++] = attributes->modifier[1] >> 32;
}
}
if (attributes->n_planes > 2) {
attribs[atti++] = EGL_DMA_BUF_PLANE2_FD_EXT;
attribs[atti++] = attributes->fd[2];
attribs[atti++] = EGL_DMA_BUF_PLANE2_OFFSET_EXT;
attribs[atti++] = attributes->offset[2];
attribs[atti++] = EGL_DMA_BUF_PLANE2_PITCH_EXT;
attribs[atti++] = attributes->stride[2];
if (has_modifier) {
attribs[atti++] = EGL_DMA_BUF_PLANE2_MODIFIER_LO_EXT;
attribs[atti++] = attributes->modifier[2] & 0xFFFFFFFF;
attribs[atti++] = EGL_DMA_BUF_PLANE2_MODIFIER_HI_EXT;
attribs[atti++] = attributes->modifier[2] >> 32;
}
}
if (gr->has_dmabuf_import_modifiers) {
if (attributes->n_planes > 3) {
attribs[atti++] = EGL_DMA_BUF_PLANE3_FD_EXT;
attribs[atti++] = attributes->fd[3];
attribs[atti++] = EGL_DMA_BUF_PLANE3_OFFSET_EXT;
attribs[atti++] = attributes->offset[3];
attribs[atti++] = EGL_DMA_BUF_PLANE3_PITCH_EXT;
attribs[atti++] = attributes->stride[3];
attribs[atti++] = EGL_DMA_BUF_PLANE3_MODIFIER_LO_EXT;
attribs[atti++] = attributes->modifier[3] & 0xFFFFFFFF;
attribs[atti++] = EGL_DMA_BUF_PLANE3_MODIFIER_HI_EXT;
attribs[atti++] = attributes->modifier[3] >> 32;
}
}
attribs[atti++] = EGL_NONE;
image = egl_image_create(gr, EGL_LINUX_DMA_BUF_EXT, NULL,
attribs);
return image;
}
/* The kernel header drm_fourcc.h defines the DRM formats below. We duplicate
* some of the definitions here so that building Weston won't require
* bleeding-edge kernel headers.
*/
#ifndef DRM_FORMAT_R8
#define DRM_FORMAT_R8 fourcc_code('R', '8', ' ', ' ') /* [7:0] R */
#endif
#ifndef DRM_FORMAT_GR88
#define DRM_FORMAT_GR88 fourcc_code('G', 'R', '8', '8') /* [15:0] G:R 8:8 little endian */
#endif
#ifndef DRM_FORMAT_XYUV8888
#define DRM_FORMAT_XYUV8888 fourcc_code('X', 'Y', 'U', 'V') /* [31:0] X:Y:Cb:Cr 8:8:8:8 little endian */
#endif
struct yuv_format_descriptor yuv_formats[] = {
{
.format = DRM_FORMAT_YUYV,
.input_planes = 1,
.output_planes = 2,
.texture_type = TEXTURE_Y_XUXV_WL,
{{
.width_divisor = 1,
.height_divisor = 1,
.format = DRM_FORMAT_GR88,
.plane_index = 0
}, {
.width_divisor = 2,
.height_divisor = 1,
.format = DRM_FORMAT_ARGB8888,
.plane_index = 0
}}
}, {
.format = DRM_FORMAT_NV12,
.input_planes = 2,
.output_planes = 2,
.texture_type = TEXTURE_Y_UV_WL,
{{
.width_divisor = 1,
.height_divisor = 1,
.format = DRM_FORMAT_R8,
.plane_index = 0
}, {
.width_divisor = 2,
.height_divisor = 2,
.format = DRM_FORMAT_GR88,
.plane_index = 1
}}
}, {
.format = DRM_FORMAT_YUV420,
.input_planes = 3,
.output_planes = 3,
.texture_type = TEXTURE_Y_U_V_WL,
{{
.width_divisor = 1,
.height_divisor = 1,
.format = DRM_FORMAT_R8,
.plane_index = 0
}, {
.width_divisor = 2,
.height_divisor = 2,
.format = DRM_FORMAT_R8,
.plane_index = 1
}, {
.width_divisor = 2,
.height_divisor = 2,
.format = DRM_FORMAT_R8,
.plane_index = 2
}}
}, {
.format = DRM_FORMAT_YUV444,
.input_planes = 3,
.output_planes = 3,
.texture_type = TEXTURE_Y_U_V_WL,
{{
.width_divisor = 1,
.height_divisor = 1,
.format = DRM_FORMAT_R8,
.plane_index = 0
}, {
.width_divisor = 1,
.height_divisor = 1,
.format = DRM_FORMAT_R8,
.plane_index = 1
}, {
.width_divisor = 1,
.height_divisor = 1,
.format = DRM_FORMAT_R8,
.plane_index = 2
}}
}, {
.format = DRM_FORMAT_XYUV8888,
.input_planes = 1,
.output_planes = 1,
.texture_type = TEXTURE_XYUV_WL,
{{
.width_divisor = 1,
.height_divisor = 1,
.format = DRM_FORMAT_XBGR8888,
.plane_index = 0
}}
}
};
static struct egl_image *
import_dmabuf_single_plane(struct gl_renderer *gr,
const struct dmabuf_attributes *attributes,
struct yuv_plane_descriptor *descriptor)
{
struct dmabuf_attributes plane;
struct egl_image *image;
char fmt[4];
plane.width = attributes->width / descriptor->width_divisor;
plane.height = attributes->height / descriptor->height_divisor;
plane.format = descriptor->format;
plane.n_planes = 1;
plane.fd[0] = attributes->fd[descriptor->plane_index];
plane.offset[0] = attributes->offset[descriptor->plane_index];
plane.stride[0] = attributes->stride[descriptor->plane_index];
plane.modifier[0] = attributes->modifier[descriptor->plane_index];
image = import_simple_dmabuf(gr, &plane);
if (!image) {
weston_log("Failed to import plane %d as %.4s\n",
descriptor->plane_index,
dump_format(descriptor->format, fmt));
return NULL;
}
return image;
}
static bool
import_yuv_dmabuf(struct gl_renderer *gr,
struct dmabuf_image *image)
{
unsigned i;
int j;
int ret;
struct yuv_format_descriptor *format = NULL;
struct dmabuf_attributes *attributes = &image->dmabuf->attributes;
char fmt[4];
for (i = 0; i < ARRAY_LENGTH(yuv_formats); ++i) {
if (yuv_formats[i].format == attributes->format) {
format = &yuv_formats[i];
break;
}
}
if (!format) {
weston_log("Error during import, and no known conversion for format "
"%.4s in the renderer\n",
dump_format(attributes->format, fmt));
return false;
}
if (attributes->n_planes != format->input_planes) {
weston_log("%.4s dmabuf must contain %d plane%s (%d provided)\n",
dump_format(format->format, fmt),
format->input_planes,
(format->input_planes > 1) ? "s" : "",
attributes->n_planes);
return false;
}
for (j = 0; j < format->output_planes; ++j) {
image->images[j] = import_dmabuf_single_plane(gr, attributes,
&format->plane[j]);
if (!image->images[j]) {
while (j) {
ret = egl_image_unref(image->images[--j]);
assert(ret == 0);
}
return false;
}
}
image->num_images = format->output_planes;
switch (format->texture_type) {
case TEXTURE_Y_XUXV_WL:
image->shader = &gr->texture_shader_y_xuxv;
break;
case TEXTURE_Y_UV_WL:
image->shader = &gr->texture_shader_y_uv;
break;
case TEXTURE_Y_U_V_WL:
image->shader = &gr->texture_shader_y_u_v;
break;
case TEXTURE_XYUV_WL:
image->shader = &gr->texture_shader_xyuv;
break;
default:
assert(false);
}
return true;
}
static GLenum
choose_texture_target(struct dmabuf_attributes *attributes)
{
if (attributes->n_planes > 1)
return GL_TEXTURE_EXTERNAL_OES;
switch (attributes->format & ~DRM_FORMAT_BIG_ENDIAN) {
case DRM_FORMAT_YUYV:
case DRM_FORMAT_YVYU:
case DRM_FORMAT_UYVY:
case DRM_FORMAT_VYUY:
case DRM_FORMAT_AYUV:
case DRM_FORMAT_XYUV8888:
return GL_TEXTURE_EXTERNAL_OES;
default:
return GL_TEXTURE_2D;
}
}
static struct dmabuf_image *
import_dmabuf(struct gl_renderer *gr,
struct linux_dmabuf_buffer *dmabuf)
{
struct egl_image *egl_image;
struct dmabuf_image *image;
image = dmabuf_image_create();
image->dmabuf = dmabuf;
egl_image = import_simple_dmabuf(gr, &dmabuf->attributes);
if (egl_image) {
image->num_images = 1;
image->images[0] = egl_image;
image->import_type = IMPORT_TYPE_DIRECT;
image->target = choose_texture_target(&dmabuf->attributes);
switch (image->target) {
case GL_TEXTURE_2D:
image->shader = &gr->texture_shader_rgba;
break;
default:
image->shader = &gr->texture_shader_egl_external;
}
} else {
if (!import_yuv_dmabuf(gr, image)) {
dmabuf_image_destroy(image);
return NULL;
}
image->import_type = IMPORT_TYPE_GL_CONVERSION;
image->target = GL_TEXTURE_2D;
}
return image;
}
static void
gl_renderer_query_dmabuf_formats(struct weston_compositor *wc,
int **formats, int *num_formats)
{
struct gl_renderer *gr = get_renderer(wc);
static const int fallback_formats[] = {
DRM_FORMAT_ARGB8888,
DRM_FORMAT_XRGB8888,
DRM_FORMAT_YUYV,
DRM_FORMAT_NV12,
DRM_FORMAT_YUV420,
DRM_FORMAT_YUV444,
DRM_FORMAT_XYUV8888,
};
bool fallback = false;
EGLint num;
assert(gr->has_dmabuf_import);
if (!gr->has_dmabuf_import_modifiers ||
!gr->query_dmabuf_formats(gr->egl_display, 0, NULL, &num)) {
num = gr->has_gl_texture_rg ? ARRAY_LENGTH(fallback_formats) : 2;
fallback = true;
}
*formats = calloc(num, sizeof(int));
if (*formats == NULL) {
*num_formats = 0;
return;
}
if (fallback) {
memcpy(*formats, fallback_formats, num * sizeof(int));
*num_formats = num;
return;
}
if (!gr->query_dmabuf_formats(gr->egl_display, num, *formats, &num)) {
*num_formats = 0;
free(*formats);
return;
}
*num_formats = num;
}
static void
gl_renderer_query_dmabuf_modifiers(struct weston_compositor *wc, int format,
uint64_t **modifiers,
int *num_modifiers)
{
struct gl_renderer *gr = get_renderer(wc);
int num;
assert(gr->has_dmabuf_import);
if (!gr->has_dmabuf_import_modifiers ||
!gr->query_dmabuf_modifiers(gr->egl_display, format, 0, NULL,
NULL, &num) ||
num == 0) {
*num_modifiers = 0;
return;
}
*modifiers = calloc(num, sizeof(uint64_t));
if (*modifiers == NULL) {
*num_modifiers = 0;
return;
}
if (!gr->query_dmabuf_modifiers(gr->egl_display, format,
num, *modifiers, NULL, &num)) {
*num_modifiers = 0;
free(*modifiers);
return;
}
*num_modifiers = num;
}
static bool
gl_renderer_import_dmabuf(struct weston_compositor *ec,
struct linux_dmabuf_buffer *dmabuf)
{
struct gl_renderer *gr = get_renderer(ec);
struct dmabuf_image *image;
int i;
assert(gr->has_dmabuf_import);
for (i = 0; i < dmabuf->attributes.n_planes; i++) {
/* return if EGL doesn't support import modifiers */
if (dmabuf->attributes.modifier[i] != DRM_FORMAT_MOD_INVALID)
if (!gr->has_dmabuf_import_modifiers)
return false;
/* return if modifiers passed are unequal */
if (dmabuf->attributes.modifier[i] !=
dmabuf->attributes.modifier[0])
return false;
}
/* reject all flags we do not recognize or handle */
if (dmabuf->attributes.flags & ~ZWP_LINUX_BUFFER_PARAMS_V1_FLAGS_Y_INVERT)
return false;
image = import_dmabuf(gr, dmabuf);
if (!image)
return false;
wl_list_insert(&gr->dmabuf_images, &image->link);
linux_dmabuf_buffer_set_user_data(dmabuf, image,
gl_renderer_destroy_dmabuf);
return true;
}
static bool
import_known_dmabuf(struct gl_renderer *gr,
struct dmabuf_image *image)
{
switch (image->import_type) {
case IMPORT_TYPE_DIRECT:
image->images[0] = import_simple_dmabuf(gr, &image->dmabuf->attributes);
if (!image->images[0])
return false;
image->num_images = 1;
break;
case IMPORT_TYPE_GL_CONVERSION:
if (!import_yuv_dmabuf(gr, image))
return false;
break;
default:
weston_log("Invalid import type for dmabuf\n");
return false;
}
return true;
}
static bool
dmabuf_is_opaque(struct linux_dmabuf_buffer *dmabuf)
{
const struct pixel_format_info *info;
info = pixel_format_get_info(dmabuf->attributes.format &
~DRM_FORMAT_BIG_ENDIAN);
if (!info)
return false;
return pixel_format_is_opaque(info);
}
static void
gl_renderer_attach_dmabuf(struct weston_surface *surface,
struct weston_buffer *buffer,
struct linux_dmabuf_buffer *dmabuf)
{
struct gl_renderer *gr = get_renderer(surface->compositor);
struct gl_surface_state *gs = get_surface_state(surface);
struct dmabuf_image *image;
int i;
int ret;
if (!gr->has_dmabuf_import) {
linux_dmabuf_buffer_send_server_error(dmabuf,
"EGL dmabuf import not supported");
return;
}
buffer->width = dmabuf->attributes.width;
buffer->height = dmabuf->attributes.height;
gl-renderer, simple-dmabuf-v4l: fix dmabuf y-invert Invert the Y_INVERT flag for the EGL import fo dmabufs. This fixes weston-simple-dmabuf-intel to show the same image on both GL-composited and with direct scanout on a hardware plane. Before, the image would y-flip when switching between these two cases. Now the orientation also matches the color values written in simple-dmabuf-intel.c. The GL-renderer uses the OpenGL convention of texture coordinates, where the origin is at the bottom-left of an image. This can be observed in texture_region() where the texcoords are inverted if y_invert is false, since the surface coordinates have origin at top-left. Both wl_shm and dmabuf buffers have origin at the top-left. When wl_shm buffer is imported with glTexImage2D, it gets inverted because glTexImage2D is defined to read in the bottom row first. The shm data is top row first. This incidentally also means, that buffer pixel 0,0 ends up at texture coordinates 0,0. This is now inverted compared to the GL coordinate convention, and therefore gl_renderer_attach_shm() sets y_inverted to true. This causes texture_region() to NOT invert the texcoords. Wayland surface coordinates have origin at top-left, hence the double-inversion. Dmabuf buffers also have the origin at top-left. However, they are imported via EGL to GL, where they should get the GL oriented coordinates but they do not. It is as if pixel 0,0 ends up at texcoords 0,0 - the same thing as with wl_shm buffers. Therefore we need to invert the invert flag. Too bad EGL_EXT_image_dma_buf_import does not seem to specify the image orientation. The GL spec implied result seems to conflict with the reality in Mesa 11.2.2. I asked about this in the Mesa developer mailing list. The question with no answers: https://lists.freedesktop.org/archives/mesa-dev/2016-June/120249.html and the thread I hijacked to get some answers: https://lists.freedesktop.org/archives/mesa-dev/2016-June/120733.html which culminated to the conclusion: https://lists.freedesktop.org/archives/mesa-dev/2016-June/120955.html that supports this patch. simple-dmabuf-v4l is equally fixed to not add Y_INVERT. There is no rational reason to have it, and removing is necessary together with the GL-renderer change to keep the image the right way up. This has been tested with VIVID. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk> Reviewed-by: Quentin Glidic <sardemff7+git@sardemff7.net>
8 years ago
/*
* GL-renderer uses the OpenGL convention of texture coordinates, where
* the origin is at bottom-left. Because dmabuf buffers have the origin
* at top-left, we must invert the Y_INVERT flag to get the image right.
*/
buffer->y_inverted =
gl-renderer, simple-dmabuf-v4l: fix dmabuf y-invert Invert the Y_INVERT flag for the EGL import fo dmabufs. This fixes weston-simple-dmabuf-intel to show the same image on both GL-composited and with direct scanout on a hardware plane. Before, the image would y-flip when switching between these two cases. Now the orientation also matches the color values written in simple-dmabuf-intel.c. The GL-renderer uses the OpenGL convention of texture coordinates, where the origin is at the bottom-left of an image. This can be observed in texture_region() where the texcoords are inverted if y_invert is false, since the surface coordinates have origin at top-left. Both wl_shm and dmabuf buffers have origin at the top-left. When wl_shm buffer is imported with glTexImage2D, it gets inverted because glTexImage2D is defined to read in the bottom row first. The shm data is top row first. This incidentally also means, that buffer pixel 0,0 ends up at texture coordinates 0,0. This is now inverted compared to the GL coordinate convention, and therefore gl_renderer_attach_shm() sets y_inverted to true. This causes texture_region() to NOT invert the texcoords. Wayland surface coordinates have origin at top-left, hence the double-inversion. Dmabuf buffers also have the origin at top-left. However, they are imported via EGL to GL, where they should get the GL oriented coordinates but they do not. It is as if pixel 0,0 ends up at texcoords 0,0 - the same thing as with wl_shm buffers. Therefore we need to invert the invert flag. Too bad EGL_EXT_image_dma_buf_import does not seem to specify the image orientation. The GL spec implied result seems to conflict with the reality in Mesa 11.2.2. I asked about this in the Mesa developer mailing list. The question with no answers: https://lists.freedesktop.org/archives/mesa-dev/2016-June/120249.html and the thread I hijacked to get some answers: https://lists.freedesktop.org/archives/mesa-dev/2016-June/120733.html which culminated to the conclusion: https://lists.freedesktop.org/archives/mesa-dev/2016-June/120955.html that supports this patch. simple-dmabuf-v4l is equally fixed to not add Y_INVERT. There is no rational reason to have it, and removing is necessary together with the GL-renderer change to keep the image the right way up. This has been tested with VIVID. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk> Reviewed-by: Quentin Glidic <sardemff7+git@sardemff7.net>
8 years ago
!(dmabuf->attributes.flags & ZWP_LINUX_BUFFER_PARAMS_V1_FLAGS_Y_INVERT);
for (i = 0; i < gs->num_images; i++)
egl_image_unref(gs->images[i]);
gs->num_images = 0;
gs->pitch = buffer->width;
gs->height = buffer->height;
gs->buffer_type = BUFFER_TYPE_EGL;
gs->y_inverted = buffer->y_inverted;
gs->direct_display = dmabuf->direct_display;
surface->is_opaque = dmabuf_is_opaque(dmabuf);
/*
* We try to always hold an imported EGLImage from the dmabuf
* to prevent the client from preventing re-imports. But, we also
* need to re-import every time the contents may change because
* GL driver's caching may need flushing.
*
* Here we release the cache reference which has to be final.
*/
if (dmabuf->direct_display)
return;
image = linux_dmabuf_buffer_get_user_data(dmabuf);
/* The dmabuf_image should have been created during the import */
assert(image != NULL);
for (i = 0; i < image->num_images; ++i) {
ret = egl_image_unref(image->images[i]);
assert(ret == 0);
}
if (!import_known_dmabuf(gr, image)) {
linux_dmabuf_buffer_send_server_error(dmabuf, "EGL dmabuf import failed");
return;
}
gs->num_images = image->num_images;
for (i = 0; i < gs->num_images; ++i)
gs->images[i] = egl_image_ref(image->images[i]);
gs->target = image->target;
ensure_textures(gs, gs->num_images);
for (i = 0; i < gs->num_images; ++i) {
glActiveTexture(GL_TEXTURE0 + i);
glBindTexture(gs->target, gs->textures[i]);
gr->image_target_texture_2d(gs->target, gs->images[i]->image);
}
gs->shader = image->shader;
}
static void
gl_renderer_attach(struct weston_surface *es, struct weston_buffer *buffer)
{
struct weston_compositor *ec = es->compositor;
struct gl_renderer *gr = get_renderer(ec);
struct gl_surface_state *gs = get_surface_state(es);
struct wl_shm_buffer *shm_buffer;
struct linux_dmabuf_buffer *dmabuf;
EGLint format;
int i;
weston_buffer_reference(&gs->buffer_ref, buffer);
libweston: Support zwp_surface_synchronization_v1.get_release Implement the get_release request of the zwp_surface_synchronization_v1 interface. This commit implements the zwp_buffer_release_v1 interface. It supports the zwp_buffer_release_v1.fenced_release event for surfaces rendered by the GL renderer, and the zwp_buffer_release_v1.immediate_release event for other cases. Note that the immediate_release event is safe to be used for surface buffers used as planes in the DRM backend, since the backend releases them only after the next page flip that doesn't use the buffers has finished. Changes in v7: - Remove "partial" from commit title and description. - Fix inverted check when clearing used_in_output_repaint flag. Changes in v5: - Use the new, generic explicit sync server error reporting function. - Introduce and use weston_buffer_release_move. - Introduce internally and use weston_buffer_release_destroy. Changes in v4: - Support the zwp_buffer_release_v1.fenced_release event. - Support release fences in the GL renderer. - Assert that pending state buffer_release is always NULL after a commit. - Simplify weston_buffer_release_reference. - Move removal of destroy listener before resource destruction to avoid concerns about use-after-free in weston_buffer_release_reference - Rename weston_buffer_release_reference.busy_count to ref_count. - Add documentation for weston_buffer_release and ..._reference. Changes in v3: - Raise NO_BUFFER for get_release if no buffer has been committed, don't raise UNSUPPORTED_BUFFER for non-dmabuf buffers, so get_release works for all valid buffers. - Destroy the buffer_release object after sending an event. - Track lifetime of buffer_release objects per commit, independently of any buffers. - Use updated protocol interface names. - Use correct format specifier for resource ids. Changes in v2: - Raise UNSUPPORTED_BUFFER at commit if client has requested a buffer_release, but the committed buffer is not a valid linux_dmabuf. - Remove tests that are not viable anymore due to our inability to create dmabuf buffers and fences in a unit-test environment. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
weston_buffer_release_reference(&gs->buffer_release_ref,
es->buffer_release_ref.buffer_release);
if (!buffer) {
for (i = 0; i < gs->num_images; i++) {
egl_image_unref(gs->images[i]);
gs->images[i] = NULL;
}
gs->num_images = 0;
glDeleteTextures(gs->num_textures, gs->textures);
gs->num_textures = 0;
gs->buffer_type = BUFFER_TYPE_NULL;
gs->y_inverted = true;
gs->direct_display = false;
es->is_opaque = false;
return;
}
shm_buffer = wl_shm_buffer_get(buffer->resource);
if (shm_buffer)
gl_renderer_attach_shm(es, buffer, shm_buffer);
else if (gr->has_bind_display &&
gr->query_buffer(gr->egl_display, (void *)buffer->resource,
EGL_TEXTURE_FORMAT, &format))
gl_renderer_attach_egl(es, buffer, format);
else if ((dmabuf = linux_dmabuf_buffer_get(buffer->resource)))
gl_renderer_attach_dmabuf(es, buffer, dmabuf);
else {
weston_log("unhandled buffer type!\n");
if (gr->has_bind_display) {
weston_log("eglQueryWaylandBufferWL failed\n");
gl_renderer_print_egl_error_state();
}
weston_buffer_reference(&gs->buffer_ref, NULL);
libweston: Support zwp_surface_synchronization_v1.get_release Implement the get_release request of the zwp_surface_synchronization_v1 interface. This commit implements the zwp_buffer_release_v1 interface. It supports the zwp_buffer_release_v1.fenced_release event for surfaces rendered by the GL renderer, and the zwp_buffer_release_v1.immediate_release event for other cases. Note that the immediate_release event is safe to be used for surface buffers used as planes in the DRM backend, since the backend releases them only after the next page flip that doesn't use the buffers has finished. Changes in v7: - Remove "partial" from commit title and description. - Fix inverted check when clearing used_in_output_repaint flag. Changes in v5: - Use the new, generic explicit sync server error reporting function. - Introduce and use weston_buffer_release_move. - Introduce internally and use weston_buffer_release_destroy. Changes in v4: - Support the zwp_buffer_release_v1.fenced_release event. - Support release fences in the GL renderer. - Assert that pending state buffer_release is always NULL after a commit. - Simplify weston_buffer_release_reference. - Move removal of destroy listener before resource destruction to avoid concerns about use-after-free in weston_buffer_release_reference - Rename weston_buffer_release_reference.busy_count to ref_count. - Add documentation for weston_buffer_release and ..._reference. Changes in v3: - Raise NO_BUFFER for get_release if no buffer has been committed, don't raise UNSUPPORTED_BUFFER for non-dmabuf buffers, so get_release works for all valid buffers. - Destroy the buffer_release object after sending an event. - Track lifetime of buffer_release objects per commit, independently of any buffers. - Use updated protocol interface names. - Use correct format specifier for resource ids. Changes in v2: - Raise UNSUPPORTED_BUFFER at commit if client has requested a buffer_release, but the committed buffer is not a valid linux_dmabuf. - Remove tests that are not viable anymore due to our inability to create dmabuf buffers and fences in a unit-test environment. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
weston_buffer_release_reference(&gs->buffer_release_ref, NULL);
gs->buffer_type = BUFFER_TYPE_NULL;
gs->y_inverted = true;
es->is_opaque = false;
weston_buffer_send_server_error(buffer,
"disconnecting due to unhandled buffer type");
}
}
static void
gl_renderer_surface_set_color(struct weston_surface *surface,
float red, float green, float blue, float alpha)
{
struct gl_surface_state *gs = get_surface_state(surface);
struct gl_renderer *gr = get_renderer(surface->compositor);
gs->color[0] = red;
gs->color[1] = green;
gs->color[2] = blue;
gs->color[3] = alpha;
gs->buffer_type = BUFFER_TYPE_SOLID;
gs->pitch = 1;
gs->height = 1;
gs->shader = &gr->solid_shader;
}
static void
gl_renderer_surface_get_content_size(struct weston_surface *surface,
int *width, int *height)
{
struct gl_surface_state *gs = get_surface_state(surface);
if (gs->buffer_type == BUFFER_TYPE_NULL) {
*width = 0;
*height = 0;
} else {
*width = gs->pitch;
*height = gs->height;
}
}
static uint32_t
pack_color(pixman_format_code_t format, float *c)
{
uint8_t r = round(c[0] * 255.0f);
uint8_t g = round(c[1] * 255.0f);
uint8_t b = round(c[2] * 255.0f);
uint8_t a = round(c[3] * 255.0f);
switch (format) {
case PIXMAN_a8b8g8r8:
return (a << 24) | (b << 16) | (g << 8) | r;
default:
assert(0);
return 0;
}
}
static int
gl_renderer_surface_copy_content(struct weston_surface *surface,
void *target, size_t size,
int src_x, int src_y,
int width, int height)
{
static const GLfloat verts[4 * 2] = {
0.0f, 0.0f,
1.0f, 0.0f,
1.0f, 1.0f,
0.0f, 1.0f
};
static const GLfloat projmat_normal[16] = { /* transpose */
2.0f, 0.0f, 0.0f, 0.0f,
0.0f, 2.0f, 0.0f, 0.0f,
0.0f, 0.0f, 1.0f, 0.0f,
-1.0f, -1.0f, 0.0f, 1.0f
};
static const GLfloat projmat_yinvert[16] = { /* transpose */
2.0f, 0.0f, 0.0f, 0.0f,
0.0f, -2.0f, 0.0f, 0.0f,
0.0f, 0.0f, 1.0f, 0.0f,
-1.0f, 1.0f, 0.0f, 1.0f
};
const pixman_format_code_t format = PIXMAN_a8b8g8r8;
const size_t bytespp = 4; /* PIXMAN_a8b8g8r8 */
const GLenum gl_format = GL_RGBA; /* PIXMAN_a8b8g8r8 little-endian */
struct gl_renderer *gr = get_renderer(surface->compositor);
struct gl_surface_state *gs = get_surface_state(surface);
int cw, ch;
GLuint fbo;
GLuint tex;
GLenum status;
const GLfloat *proj;
int i;
gl_renderer_surface_get_content_size(surface, &cw, &ch);
switch (gs->buffer_type) {
case BUFFER_TYPE_NULL:
return -1;
case BUFFER_TYPE_SOLID:
*(uint32_t *)target = pack_color(format, gs->color);
return 0;
case BUFFER_TYPE_SHM:
gl_renderer_flush_damage(surface);
/* fall through */
case BUFFER_TYPE_EGL:
break;
}
glGenTextures(1, &tex);
glBindTexture(GL_TEXTURE_2D, tex);
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, cw, ch,
0, GL_RGBA, GL_UNSIGNED_BYTE, NULL);
glBindTexture(GL_TEXTURE_2D, 0);
glGenFramebuffers(1, &fbo);
glBindFramebuffer(GL_FRAMEBUFFER, fbo);
glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0,
GL_TEXTURE_2D, tex, 0);
status = glCheckFramebufferStatus(GL_FRAMEBUFFER);
if (status != GL_FRAMEBUFFER_COMPLETE) {
weston_log("%s: fbo error: %#x\n", __func__, status);
glDeleteFramebuffers(1, &fbo);
glDeleteTextures(1, &tex);
return -1;
}
glViewport(0, 0, cw, ch);
glDisable(GL_BLEND);
use_shader(gr, gs->shader);
if (gs->y_inverted)
proj = projmat_normal;
else
proj = projmat_yinvert;
glUniformMatrix4fv(gs->shader->proj_uniform, 1, GL_FALSE, proj);
glUniform1f(gs->shader->alpha_uniform, 1.0f);
for (i = 0; i < gs->num_textures; i++) {
glUniform1i(gs->shader->tex_uniforms[i], i);
glActiveTexture(GL_TEXTURE0 + i);
glBindTexture(gs->target, gs->textures[i]);
glTexParameteri(gs->target, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glTexParameteri(gs->target, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
}
/* position: */
glVertexAttribPointer(0, 2, GL_FLOAT, GL_FALSE, 0, verts);
glEnableVertexAttribArray(0);
/* texcoord: */
glVertexAttribPointer(1, 2, GL_FLOAT, GL_FALSE, 0, verts);
glEnableVertexAttribArray(1);
glDrawArrays(GL_TRIANGLE_FAN, 0, 4);
glDisableVertexAttribArray(1);
glDisableVertexAttribArray(0);
glPixelStorei(GL_PACK_ALIGNMENT, bytespp);
glReadPixels(src_x, src_y, width, height, gl_format,
GL_UNSIGNED_BYTE, target);
glDeleteFramebuffers(1, &fbo);
glDeleteTextures(1, &tex);
return 0;
}
static void
surface_state_destroy(struct gl_surface_state *gs, struct gl_renderer *gr)
{
int i;
wl_list_remove(&gs->surface_destroy_listener.link);
wl_list_remove(&gs->renderer_destroy_listener.link);
gs->surface->renderer_state = NULL;
glDeleteTextures(gs->num_textures, gs->textures);
for (i = 0; i < gs->num_images; i++)
egl_image_unref(gs->images[i]);
weston_buffer_reference(&gs->buffer_ref, NULL);
libweston: Support zwp_surface_synchronization_v1.get_release Implement the get_release request of the zwp_surface_synchronization_v1 interface. This commit implements the zwp_buffer_release_v1 interface. It supports the zwp_buffer_release_v1.fenced_release event for surfaces rendered by the GL renderer, and the zwp_buffer_release_v1.immediate_release event for other cases. Note that the immediate_release event is safe to be used for surface buffers used as planes in the DRM backend, since the backend releases them only after the next page flip that doesn't use the buffers has finished. Changes in v7: - Remove "partial" from commit title and description. - Fix inverted check when clearing used_in_output_repaint flag. Changes in v5: - Use the new, generic explicit sync server error reporting function. - Introduce and use weston_buffer_release_move. - Introduce internally and use weston_buffer_release_destroy. Changes in v4: - Support the zwp_buffer_release_v1.fenced_release event. - Support release fences in the GL renderer. - Assert that pending state buffer_release is always NULL after a commit. - Simplify weston_buffer_release_reference. - Move removal of destroy listener before resource destruction to avoid concerns about use-after-free in weston_buffer_release_reference - Rename weston_buffer_release_reference.busy_count to ref_count. - Add documentation for weston_buffer_release and ..._reference. Changes in v3: - Raise NO_BUFFER for get_release if no buffer has been committed, don't raise UNSUPPORTED_BUFFER for non-dmabuf buffers, so get_release works for all valid buffers. - Destroy the buffer_release object after sending an event. - Track lifetime of buffer_release objects per commit, independently of any buffers. - Use updated protocol interface names. - Use correct format specifier for resource ids. Changes in v2: - Raise UNSUPPORTED_BUFFER at commit if client has requested a buffer_release, but the committed buffer is not a valid linux_dmabuf. - Remove tests that are not viable anymore due to our inability to create dmabuf buffers and fences in a unit-test environment. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
weston_buffer_release_reference(&gs->buffer_release_ref, NULL);
pixman_region32_fini(&gs->texture_damage);
free(gs);
}
static void
surface_state_handle_surface_destroy(struct wl_listener *listener, void *data)
{
struct gl_surface_state *gs;
struct gl_renderer *gr;
gs = container_of(listener, struct gl_surface_state,
surface_destroy_listener);
gr = get_renderer(gs->surface->compositor);
surface_state_destroy(gs, gr);
}
static void
surface_state_handle_renderer_destroy(struct wl_listener *listener, void *data)
{
struct gl_surface_state *gs;
struct gl_renderer *gr;
gr = data;
gs = container_of(listener, struct gl_surface_state,
renderer_destroy_listener);
surface_state_destroy(gs, gr);
}
static int
gl_renderer_create_surface(struct weston_surface *surface)
{
struct gl_surface_state *gs;
struct gl_renderer *gr = get_renderer(surface->compositor);
gs = zalloc(sizeof *gs);
if (gs == NULL)
return -1;
/* A buffer is never attached to solid color surfaces, yet
* they still go through texcoord computations. Do not divide
* by zero there.
*/
gs->pitch = 1;
gs->y_inverted = true;
gs->direct_display = false;
gs->surface = surface;
pixman_region32_init(&gs->texture_damage);
surface->renderer_state = gs;
gs->surface_destroy_listener.notify =
surface_state_handle_surface_destroy;
wl_signal_add(&surface->destroy_signal,
&gs->surface_destroy_listener);
gs->renderer_destroy_listener.notify =
surface_state_handle_renderer_destroy;
wl_signal_add(&gr->destroy_signal,
&gs->renderer_destroy_listener);
if (surface->buffer_ref.buffer) {
gl_renderer_attach(surface, surface->buffer_ref.buffer);
gl_renderer_flush_damage(surface);
}
return 0;
}
static const char vertex_shader[] =
"uniform mat4 proj;\n"
"attribute vec2 position;\n"
"attribute vec2 texcoord;\n"
"varying vec2 v_texcoord;\n"
"void main()\n"
"{\n"
" gl_Position = proj * vec4(position, 0.0, 1.0);\n"
" v_texcoord = texcoord;\n"
"}\n";
/* Declare common fragment shader uniforms */
#define FRAGMENT_CONVERT_YUV \
" y *= alpha;\n" \
" u *= alpha;\n" \
" v *= alpha;\n" \
" gl_FragColor.r = y + 1.59602678 * v;\n" \
" gl_FragColor.g = y - 0.39176229 * u - 0.81296764 * v;\n" \
" gl_FragColor.b = y + 2.01723214 * u;\n" \
" gl_FragColor.a = alpha;\n"
static const char fragment_debug[] =
" gl_FragColor = vec4(0.0, 0.3, 0.0, 0.2) + gl_FragColor * 0.8;\n";
static const char fragment_brace[] =
"}\n";
static const char texture_fragment_shader_rgba[] =
"precision mediump float;\n"
"varying vec2 v_texcoord;\n"
"uniform sampler2D tex;\n"
"uniform float alpha;\n"
"void main()\n"
"{\n"
" gl_FragColor = alpha * texture2D(tex, v_texcoord)\n;"
;
static const char texture_fragment_shader_rgbx[] =
"precision mediump float;\n"
"varying vec2 v_texcoord;\n"
"uniform sampler2D tex;\n"
"uniform float alpha;\n"
"void main()\n"
"{\n"
" gl_FragColor.rgb = alpha * texture2D(tex, v_texcoord).rgb\n;"
" gl_FragColor.a = alpha;\n"
;
static const char texture_fragment_shader_egl_external[] =
"#extension GL_OES_EGL_image_external : require\n"
"precision mediump float;\n"
"varying vec2 v_texcoord;\n"
"uniform samplerExternalOES tex;\n"
"uniform float alpha;\n"
"void main()\n"
"{\n"
" gl_FragColor = alpha * texture2D(tex, v_texcoord)\n;"
;
static const char texture_fragment_shader_y_uv[] =
"precision mediump float;\n"
"uniform sampler2D tex;\n"
"uniform sampler2D tex1;\n"
"varying vec2 v_texcoord;\n"
"uniform float alpha;\n"
"void main() {\n"
" float y = 1.16438356 * (texture2D(tex, v_texcoord).x - 0.0625);\n"
" float u = texture2D(tex1, v_texcoord).r - 0.5;\n"
" float v = texture2D(tex1, v_texcoord).g - 0.5;\n"
FRAGMENT_CONVERT_YUV
;
static const char texture_fragment_shader_y_u_v[] =
"precision mediump float;\n"
"uniform sampler2D tex;\n"
"uniform sampler2D tex1;\n"
"uniform sampler2D tex2;\n"
"varying vec2 v_texcoord;\n"
"uniform float alpha;\n"
"void main() {\n"
" float y = 1.16438356 * (texture2D(tex, v_texcoord).x - 0.0625);\n"
" float u = texture2D(tex1, v_texcoord).x - 0.5;\n"
" float v = texture2D(tex2, v_texcoord).x - 0.5;\n"
FRAGMENT_CONVERT_YUV
;
static const char texture_fragment_shader_y_xuxv[] =
"precision mediump float;\n"
"uniform sampler2D tex;\n"
"uniform sampler2D tex1;\n"
"varying vec2 v_texcoord;\n"
"uniform float alpha;\n"
"void main() {\n"
" float y = 1.16438356 * (texture2D(tex, v_texcoord).x - 0.0625);\n"
" float u = texture2D(tex1, v_texcoord).g - 0.5;\n"
" float v = texture2D(tex1, v_texcoord).a - 0.5;\n"
FRAGMENT_CONVERT_YUV
;
static const char texture_fragment_shader_xyuv[] =
"precision mediump float;\n"
"uniform sampler2D tex;\n"
"varying vec2 v_texcoord;\n"
"uniform float alpha;\n"
"void main() {\n"
" float y = 1.16438356 * (texture2D(tex, v_texcoord).b - 0.0625);\n"
" float u = texture2D(tex, v_texcoord).g - 0.5;\n"
" float v = texture2D(tex, v_texcoord).r - 0.5;\n"
FRAGMENT_CONVERT_YUV
;
static const char solid_fragment_shader[] =
"precision mediump float;\n"
"uniform vec4 color;\n"
"uniform float alpha;\n"
"void main()\n"
"{\n"
" gl_FragColor = alpha * color\n;"
;
static int
compile_shader(GLenum type, int count, const char **sources)
{
GLuint s;
char msg[512];
GLint status;
s = glCreateShader(type);
glShaderSource(s, count, sources, NULL);
glCompileShader(s);
glGetShaderiv(s, GL_COMPILE_STATUS, &status);
if (!status) {
glGetShaderInfoLog(s, sizeof msg, NULL, msg);
weston_log("shader info: %s\n", msg);
return GL_NONE;
}
return s;
}
static int
shader_init(struct gl_shader *shader, struct gl_renderer *renderer,
const char *vertex_source, const char *fragment_source)
{
char msg[512];
GLint status;
int count;
const char *sources[3];
shader->vertex_shader =
compile_shader(GL_VERTEX_SHADER, 1, &vertex_source);
if (shader->vertex_shader == GL_NONE)
return -1;
if (renderer->fragment_shader_debug) {
sources[0] = fragment_source;
sources[1] = fragment_debug;
sources[2] = fragment_brace;
count = 3;
} else {
sources[0] = fragment_source;
sources[1] = fragment_brace;
count = 2;
}
shader->fragment_shader =
compile_shader(GL_FRAGMENT_SHADER, count, sources);
if (shader->fragment_shader == GL_NONE)
return -1;
shader->program = glCreateProgram();
glAttachShader(shader->program, shader->vertex_shader);
glAttachShader(shader->program, shader->fragment_shader);
glBindAttribLocation(shader->program, 0, "position");
glBindAttribLocation(shader->program, 1, "texcoord");
glLinkProgram(shader->program);
glGetProgramiv(shader->program, GL_LINK_STATUS, &status);
if (!status) {
glGetProgramInfoLog(shader->program, sizeof msg, NULL, msg);
weston_log("link info: %s\n", msg);
return -1;
}
shader->proj_uniform = glGetUniformLocation(shader->program, "proj");
shader->tex_uniforms[0] = glGetUniformLocation(shader->program, "tex");
shader->tex_uniforms[1] = glGetUniformLocation(shader->program, "tex1");
shader->tex_uniforms[2] = glGetUniformLocation(shader->program, "tex2");
shader->alpha_uniform = glGetUniformLocation(shader->program, "alpha");
shader->color_uniform = glGetUniformLocation(shader->program, "color");
return 0;
}
static void
shader_release(struct gl_shader *shader)
{
glDeleteShader(shader->vertex_shader);
glDeleteShader(shader->fragment_shader);
glDeleteProgram(shader->program);
shader->vertex_shader = 0;
shader->fragment_shader = 0;
shader->program = 0;
}
static void
log_extensions(const char *name, const char *extensions)
{
const char *p, *end;
int l;
int len;
l = weston_log("%s:", name);
p = extensions;
while (*p) {
end = strchrnul(p, ' ');
len = end - p;
if (l + len > 78)
l = weston_log_continue("\n" STAMP_SPACE "%.*s",
len, p);
else
l += weston_log_continue(" %.*s", len, p);
for (p = end; isspace(*p); p++)
;
}
weston_log_continue("\n");
}
static void
log_egl_info(EGLDisplay egldpy)
{
const char *str;
str = eglQueryString(egldpy, EGL_VERSION);
weston_log("EGL version: %s\n", str ? str : "(null)");
str = eglQueryString(egldpy, EGL_VENDOR);
weston_log("EGL vendor: %s\n", str ? str : "(null)");
str = eglQueryString(egldpy, EGL_CLIENT_APIS);
weston_log("EGL client APIs: %s\n", str ? str : "(null)");
str = eglQueryString(egldpy, EGL_EXTENSIONS);
log_extensions("EGL extensions", str ? str : "(null)");
}
static void
log_gl_info(void)
{
const char *str;
str = (char *)glGetString(GL_VERSION);
weston_log("GL version: %s\n", str ? str : "(null)");
str = (char *)glGetString(GL_SHADING_LANGUAGE_VERSION);
weston_log("GLSL version: %s\n", str ? str : "(null)");
str = (char *)glGetString(GL_VENDOR);
weston_log("GL vendor: %s\n", str ? str : "(null)");
str = (char *)glGetString(GL_RENDERER);
weston_log("GL renderer: %s\n", str ? str : "(null)");
str = (char *)glGetString(GL_EXTENSIONS);
log_extensions("GL extensions", str ? str : "(null)");
}
static void
gl_renderer_output_set_border(struct weston_output *output,
enum gl_renderer_border_side side,
int32_t width, int32_t height,
int32_t tex_width, unsigned char *data)
{
struct gl_output_state *go = get_output_state(output);
if (go->borders[side].width != width ||
go->borders[side].height != height)
/* In this case, we have to blow everything and do a full
* repaint. */
go->border_status |= BORDER_SIZE_CHANGED | BORDER_ALL_DIRTY;
if (data == NULL) {
width = 0;
height = 0;
}
go->borders[side].width = width;
go->borders[side].height = height;
go->borders[side].tex_width = tex_width;
go->borders[side].data = data;
go->border_status |= 1 << side;
}
static int
gl_renderer_setup(struct weston_compositor *ec, EGLSurface egl_surface);
static EGLSurface
gl_renderer_create_window_surface(struct gl_renderer *gr,
EGLNativeWindowType window_for_legacy,
void *window_for_platform,
const uint32_t *drm_formats,
unsigned drm_formats_count)
{
EGLSurface egl_surface = EGL_NO_SURFACE;
EGLConfig egl_config;
egl_config = gl_renderer_get_egl_config(gr, EGL_WINDOW_BIT,
drm_formats, drm_formats_count);
if (egl_config == EGL_NO_CONFIG_KHR)
return EGL_NO_SURFACE;
log_egl_config_info(gr->egl_display, egl_config);
if (gr->create_platform_window)
egl_surface = gr->create_platform_window(gr->egl_display,
egl_config,
window_for_platform,
NULL);
else
egl_surface = eglCreateWindowSurface(gr->egl_display,
egl_config,
window_for_legacy, NULL);
return egl_surface;
}
static int
gl_renderer_output_create(struct weston_output *output,
EGLSurface surface)
{
struct gl_output_state *go;
int i;
go = zalloc(sizeof *go);
if (go == NULL)
return -1;
go->egl_surface = surface;
for (i = 0; i < BUFFER_DAMAGE_COUNT; i++)
pixman_region32_init(&go->buffer_damage[i]);
wl_list_init(&go->timeline_render_point_list);
go->begin_render_sync = EGL_NO_SYNC_KHR;
go->end_render_sync = EGL_NO_SYNC_KHR;
output->renderer_state = go;
return 0;
}
static int
gl_renderer_output_window_create(struct weston_output *output,
EGLNativeWindowType window_for_legacy,
void *window_for_platform,
const uint32_t *drm_formats,
unsigned drm_formats_count)
{
struct weston_compositor *ec = output->compositor;
struct gl_renderer *gr = get_renderer(ec);
EGLSurface egl_surface = EGL_NO_SURFACE;
int ret = 0;
egl_surface = gl_renderer_create_window_surface(gr,
window_for_legacy,
window_for_platform,
drm_formats,
drm_formats_count);
if (egl_surface == EGL_NO_SURFACE) {
weston_log("failed to create egl surface\n");
return -1;
}
ret = gl_renderer_output_create(output, egl_surface);
if (ret < 0)
weston_platform_destroy_egl_surface(gr->egl_display, egl_surface);
return ret;
}
static int
gl_renderer_output_pbuffer_create(struct weston_output *output,
int width,
int height,
const uint32_t *drm_formats,
unsigned drm_formats_count)
{
struct gl_renderer *gr = get_renderer(output->compositor);
EGLConfig pbuffer_config;
EGLSurface egl_surface;
int ret;
EGLint pbuffer_attribs[] = {
EGL_WIDTH, width,
EGL_HEIGHT, height,
EGL_NONE
};
pbuffer_config = gl_renderer_get_egl_config(gr, EGL_PBUFFER_BIT,
drm_formats,
drm_formats_count);
if (pbuffer_config == EGL_NO_CONFIG_KHR) {
weston_log("failed to choose EGL config for PbufferSurface\n");
return -1;
}
log_egl_config_info(gr->egl_display, pbuffer_config);
egl_surface = eglCreatePbufferSurface(gr->egl_display, pbuffer_config,
pbuffer_attribs);
if (egl_surface == EGL_NO_SURFACE) {
weston_log("failed to create egl surface\n");
gl_renderer_print_egl_error_state();
return -1;
}
ret = gl_renderer_output_create(output, egl_surface);
if (ret < 0)
eglDestroySurface(gr->egl_display, egl_surface);
return ret;
}
static void
gl_renderer_output_destroy(struct weston_output *output)
{
struct gl_renderer *gr = get_renderer(output->compositor);
struct gl_output_state *go = get_output_state(output);
struct timeline_render_point *trp, *tmp;
int i;
for (i = 0; i < 2; i++)
pixman_region32_fini(&go->buffer_damage[i]);
eglMakeCurrent(gr->egl_display,
EGL_NO_SURFACE, EGL_NO_SURFACE,
EGL_NO_CONTEXT);
weston_platform_destroy_egl_surface(gr->egl_display, go->egl_surface);
if (!wl_list_empty(&go->timeline_render_point_list))
weston_log("warning: discarding pending timeline render"
"objects at output destruction");
wl_list_for_each_safe(trp, tmp, &go->timeline_render_point_list, link)
timeline_render_point_destroy(trp);
if (go->begin_render_sync != EGL_NO_SYNC_KHR)
gr->destroy_sync(gr->egl_display, go->begin_render_sync);
if (go->end_render_sync != EGL_NO_SYNC_KHR)
gr->destroy_sync(gr->egl_display, go->end_render_sync);
free(go);
}
static int
gl_renderer_create_fence_fd(struct weston_output *output)
{
struct gl_output_state *go = get_output_state(output);
struct gl_renderer *gr = get_renderer(output->compositor);
int fd;
if (go->end_render_sync == EGL_NO_SYNC_KHR)
return -1;
fd = gr->dup_native_fence_fd(gr->egl_display, go->end_render_sync);
if (fd == EGL_NO_NATIVE_FENCE_FD_ANDROID)
return -1;
return fd;
}
static void
gl_renderer_destroy(struct weston_compositor *ec)
{
struct gl_renderer *gr = get_renderer(ec);
struct dmabuf_image *image, *next;
wl_signal_emit(&gr->destroy_signal, gr);
if (gr->has_bind_display)
gr->unbind_display(gr->egl_display, ec->wl_display);
/* Work around crash in egl_dri2.c's dri2_make_current() - when does this apply? */
eglMakeCurrent(gr->egl_display,
EGL_NO_SURFACE, EGL_NO_SURFACE,
EGL_NO_CONTEXT);
wl_list_for_each_safe(image, next, &gr->dmabuf_images, link)
dmabuf_image_destroy(image);
if (gr->dummy_surface != EGL_NO_SURFACE)
weston_platform_destroy_egl_surface(gr->egl_display,
gr->dummy_surface);
eglTerminate(gr->egl_display);
eglReleaseThread();
wl_list_remove(&gr->output_destroy_listener.link);
wl_array_release(&gr->vertices);
wl_array_release(&gr->vtxcnt);
if (gr->fragment_binding)
weston_binding_destroy(gr->fragment_binding);
if (gr->fan_binding)
weston_binding_destroy(gr->fan_binding);
free(gr);
}
/** Checks whether a platform EGL client extension is supported
*
* \param ec The weston compositor
* \param extension_suffix The EGL client extension suffix
* \return 1 if supported, 0 if using fallbacks, -1 unsupported
*
* This function checks whether a specific platform_* extension is supported
* by EGL.
*
* The extension suffix should be the suffix of the platform extension (that
* specifies a platform argument as defined in EGL_EXT_platform_base). For
* example, passing "foo" will check whether either "EGL_KHR_platform_foo",
* "EGL_EXT_platform_foo", or "EGL_MESA_platform_foo" is supported.
*
* The return value is 1:
* - if the supplied EGL client extension is supported.
* The return value is 0:
* - if the platform_base client extension isn't supported so will
* fallback to eglGetDisplay and friends.
* The return value is -1:
* - if the supplied EGL client extension is not supported.
*/
static int
gl_renderer_supports(struct weston_compositor *ec,
const char *extension_suffix)
{
static const char *extensions = NULL;
char s[64];
if (!extensions) {
extensions = (const char *) eglQueryString(
EGL_NO_DISPLAY, EGL_EXTENSIONS);
if (!extensions)
return 0;
log_extensions("EGL client extensions",
extensions);
}
if (!weston_check_egl_extension(extensions, "EGL_EXT_platform_base"))
return 0;
snprintf(s, sizeof s, "EGL_KHR_platform_%s", extension_suffix);
if (weston_check_egl_extension(extensions, s))
return 1;
snprintf(s, sizeof s, "EGL_EXT_platform_%s", extension_suffix);
if (weston_check_egl_extension(extensions, s))
return 1;
snprintf(s, sizeof s, "EGL_MESA_platform_%s", extension_suffix);
if (weston_check_egl_extension(extensions, s))
return 1;
/* at this point we definitely have some platform extensions but
* haven't found the supplied platform, so chances are it's
* not supported. */
return -1;
}
static const char *
platform_to_extension(EGLenum platform)
{
switch (platform) {
case EGL_PLATFORM_GBM_KHR:
return "gbm";
case EGL_PLATFORM_WAYLAND_KHR:
return "wayland";
case EGL_PLATFORM_X11_KHR:
return "x11";
case EGL_PLATFORM_SURFACELESS_MESA:
return "surfaceless";
default:
assert(0 && "bad EGL platform enum");
}
}
static void
output_handle_destroy(struct wl_listener *listener, void *data)
{
struct gl_renderer *gr;
struct weston_output *output = data;
gr = container_of(listener, struct gl_renderer,
output_destroy_listener);
if (wl_list_empty(&output->compositor->output_list))
eglMakeCurrent(gr->egl_display, gr->dummy_surface,
gr->dummy_surface, gr->egl_context);
}
static int
gl_renderer_create_pbuffer_surface(struct gl_renderer *gr) {
EGLConfig pbuffer_config;
static const EGLint pbuffer_attribs[] = {
EGL_WIDTH, 10,
EGL_HEIGHT, 10,
EGL_NONE
};
pbuffer_config = gl_renderer_get_egl_config(gr, EGL_PBUFFER_BIT,
NULL, 0);
if (pbuffer_config == EGL_NO_CONFIG_KHR) {
weston_log("failed to choose EGL config for PbufferSurface\n");
return -1;
}
gr->dummy_surface = eglCreatePbufferSurface(gr->egl_display,
pbuffer_config,
pbuffer_attribs);
if (gr->dummy_surface == EGL_NO_SURFACE) {
weston_log("failed to create PbufferSurface\n");
return -1;
}
return 0;
}
static int
gl_renderer_display_create(struct weston_compositor *ec,
EGLenum platform,
void *native_display,
EGLint egl_surface_type,
const uint32_t *drm_formats,
unsigned drm_formats_count)
{
struct gl_renderer *gr;
EGLint major, minor;
int supports = 0;
if (platform) {
supports = gl_renderer_supports(
ec, platform_to_extension(platform));
if (supports < 0)
return -1;
}
/* Surfaceless is unusable without platform_base extension */
if (supports == 0 && platform == EGL_PLATFORM_SURFACELESS_MESA)
return -1;
gr = zalloc(sizeof *gr);
if (gr == NULL)
return -1;
gr->base.read_pixels = gl_renderer_read_pixels;
gr->base.repaint_output = gl_renderer_repaint_output;
gr->base.flush_damage = gl_renderer_flush_damage;
gr->base.attach = gl_renderer_attach;
gr->base.surface_set_color = gl_renderer_surface_set_color;
gr->base.destroy = gl_renderer_destroy;
gr->base.surface_get_content_size =
gl_renderer_surface_get_content_size;
gr->base.surface_copy_content = gl_renderer_surface_copy_content;
gr->platform = platform;
gr->egl_display = NULL;
/* extension_suffix is supported */
if (supports) {
if (!get_platform_display) {
get_platform_display = (void *) eglGetProcAddress(
"eglGetPlatformDisplayEXT");
}
/* also wrap this in the supports check because
* eglGetProcAddress can return non-NULL and still not
* support the feature at runtime, so ensure the
* appropriate extension checks have been done. */
if (get_platform_display && platform) {
gr->egl_display = get_platform_display(platform,
native_display,
NULL);
}
}
if (!gr->egl_display) {
weston_log("warning: either no EGL_EXT_platform_base "
"support or specific platform support; "
"falling back to eglGetDisplay.\n");
gr->egl_display = eglGetDisplay(native_display);
}
if (gr->egl_display == EGL_NO_DISPLAY) {
weston_log("failed to create display\n");
goto fail;
}
if (!eglInitialize(gr->egl_display, &major, &minor)) {
weston_log("failed to initialize display\n");
goto fail_with_error;
}
log_egl_info(gr->egl_display);
ec->renderer = &gr->base;
if (gl_renderer_setup_egl_extensions(ec) < 0)
goto fail_with_error;
if (!gr->has_configless_context) {
if (!gr->has_surfaceless_context)
egl_surface_type |= EGL_PBUFFER_BIT;
gr->egl_config = gl_renderer_get_egl_config(gr,
egl_surface_type,
drm_formats,
drm_formats_count);
if (gr->egl_config == EGL_NO_CONFIG_KHR) {
weston_log("failed to choose EGL config\n");
goto fail_terminate;
}
}
libweston: Support zwp_surface_synchronization_v1.set_acquire_fence Implement the set_acquire_fence request of the zwp_surface_synchronization_v1 interface. The implementation uses the acquire fence in two ways: 1. If the associated buffer is used as GL render source, an EGLSyncKHR is created from the fence and used to synchronize access. 2. If the associated buffer is used as a plane framebuffer, the acquire fence is treated as an in-fence for the atomic commit operation. If in-fences are not supported and the buffer has an acquire fence, we don't consider it for plane placement. If the used compositor/renderer doesn't support explicit synchronization, we don't advertise the protocol at all. Currently only the DRM and X11 backends when using the GL renderer advertise the protocol for production use. Issues for discussion --------------------- a. Currently, a server-side wait of EGLSyncKHR is performed before using the EGLImage/texture during rendering. Unfortunately, it's not clear from the specs whether this is generally safe to do, or we need to sync before glEGLImageTargetTexture2DOES. The exception is TEXTURE_EXTERNAL_OES where the spec mentions it's enough to sync and then glBindTexture for any changes to take effect. Changes in v5: - Meson support. - Make explicit sync server error reporting more generic, supporting all explicit sync related interfaces not just wp_linux_surface_synchronization. - Fix typo in warning for missing EGL_KHR_wait_sync extension. - Support minor version 2 of the explicit sync protocol (i.e., support fences for opaque EGL buffers). Changes in v4: - Introduce and use fd_clear and and fd_move helpers. - Don't check for a valid buffer when updating surface acquire fence fd from state. - Assert that pending state acquire fence fd is always clear after a commit. - Clarify that WESTON_CAP_EXPLICIT_SYNC applies to just the renderer. - Check for EGL_KHR_wait_sync before using eglWaitSyncKHR. - Dup the acquire fence before passing to EGL. Changes in v3: - Keep acquire_fence_fd in surface instead of buffer. - Clarify that WESTON_CAP_EXPLICIT_SYNC applies to both backend and renderer. - Move comment about non-ownership of in_fence_fd to struct drm_plane_state definition. - Assert that we don't try to use planes with in-fences when using the legacy KMS API. - Remove unnecessary info from wayland error messages. - Handle acquire fence for subsurface commits. - Guard against self-update in fd_update. - Disconnect the client if acquire fence EGLSyncKHR creation or wait fails. - Use updated protocol interface names. - User correct format specifier for resource ids. - Advertise protocol for X11 backend with GL renderer. Changes in v2: - Remove sync file wait fallbacks. - Raise UNSUPPORTED_BUFFER error at commit if we have an acquire fence, but the committed buffer is not a valid linux_dmabuf. - Don't put buffers with in-fences on planes that don't support in-fences. - Don't advertise explicit sync protocol if backend does not support explicit sync. Signed-off-by: Alexandros Frantzis <alexandros.frantzis@collabora.com>
6 years ago
ec->capabilities |= WESTON_CAP_ROTATION_ANY;
ec->capabilities |= WESTON_CAP_CAPTURE_YFLIP;
ec->capabilities |= WESTON_CAP_VIEW_CLIP_MASK;
if (gr->has_native_fence_sync && gr->has_wait_sync)
ec->capabilities |= WESTON_CAP_EXPLICIT_SYNC;
wl_list_init(&gr->dmabuf_images);
if (gr->has_dmabuf_import) {
gr->base.import_dmabuf = gl_renderer_import_dmabuf;
gr->base.query_dmabuf_formats =
gl_renderer_query_dmabuf_formats;
gr->base.query_dmabuf_modifiers =
gl_renderer_query_dmabuf_modifiers;
}
if (gr->has_surfaceless_context) {
weston_log("EGL_KHR_surfaceless_context available\n");
gr->dummy_surface = EGL_NO_SURFACE;
} else {
weston_log("EGL_KHR_surfaceless_context unavailable. "
"Trying PbufferSurface\n");
if (gl_renderer_create_pbuffer_surface(gr) < 0)
goto fail_with_error;
}
wl_display_add_shm_format(ec->wl_display, WL_SHM_FORMAT_RGB565);
wl_display_add_shm_format(ec->wl_display, WL_SHM_FORMAT_YUV420);
wl_display_add_shm_format(ec->wl_display, WL_SHM_FORMAT_NV12);
wl_display_add_shm_format(ec->wl_display, WL_SHM_FORMAT_YUYV);
wl_signal_init(&gr->destroy_signal);
if (gl_renderer_setup(ec, gr->dummy_surface) < 0) {
if (gr->dummy_surface != EGL_NO_SURFACE)
weston_platform_destroy_egl_surface(gr->egl_display,
gr->dummy_surface);
goto fail_with_error;
}
return 0;
fail_with_error:
gl_renderer_print_egl_error_state();
fail_terminate:
eglTerminate(gr->egl_display);
fail:
free(gr);
ec->renderer = NULL;
return -1;
}
static int
compile_shaders(struct weston_compositor *ec)
{
struct gl_renderer *gr = get_renderer(ec);
gr->texture_shader_rgba.vertex_source = vertex_shader;
gr->texture_shader_rgba.fragment_source = texture_fragment_shader_rgba;
gr->texture_shader_rgbx.vertex_source = vertex_shader;
gr->texture_shader_rgbx.fragment_source = texture_fragment_shader_rgbx;
gr->texture_shader_egl_external.vertex_source = vertex_shader;
gr->texture_shader_egl_external.fragment_source =
texture_fragment_shader_egl_external;
gr->texture_shader_y_uv.vertex_source = vertex_shader;
gr->texture_shader_y_uv.fragment_source = texture_fragment_shader_y_uv;
gr->texture_shader_y_u_v.vertex_source = vertex_shader;
gr->texture_shader_y_u_v.fragment_source =
texture_fragment_shader_y_u_v;
gr->texture_shader_y_xuxv.vertex_source = vertex_shader;
gr->texture_shader_y_xuxv.fragment_source =
texture_fragment_shader_y_xuxv;
gr->texture_shader_xyuv.vertex_source = vertex_shader;
gr->texture_shader_xyuv.fragment_source = texture_fragment_shader_xyuv;
gr->solid_shader.vertex_source = vertex_shader;
gr->solid_shader.fragment_source = solid_fragment_shader;
return 0;
}
static void
fragment_debug_binding(struct weston_keyboard *keyboard,
const struct timespec *time,
uint32_t key, void *data)
{
struct weston_compositor *ec = data;
struct gl_renderer *gr = get_renderer(ec);
struct weston_output *output;
gr->fragment_shader_debug = !gr->fragment_shader_debug;
shader_release(&gr->texture_shader_rgba);
shader_release(&gr->texture_shader_rgbx);
shader_release(&gr->texture_shader_egl_external);
shader_release(&gr->texture_shader_y_uv);
shader_release(&gr->texture_shader_y_u_v);
shader_release(&gr->texture_shader_y_xuxv);
shader_release(&gr->texture_shader_xyuv);
shader_release(&gr->solid_shader);
/* Force use_shader() to call glUseProgram(), since we need to use
* the recompiled version of the shader. */
gr->current_shader = NULL;
wl_list_for_each(output, &ec->output_list, link)
weston_output_damage(output);
}
static void
fan_debug_repaint_binding(struct weston_keyboard *keyboard,
const struct timespec *time,
uint32_t key, void *data)
{
struct weston_compositor *compositor = data;
struct gl_renderer *gr = get_renderer(compositor);
gr->fan_debug = !gr->fan_debug;
weston_compositor_damage_all(compositor);
}
static uint32_t
get_gl_version(void)
{
const char *version;
int major, minor;
version = (const char *) glGetString(GL_VERSION);
if (version &&
(sscanf(version, "%d.%d", &major, &minor) == 2 ||
sscanf(version, "OpenGL ES %d.%d", &major, &minor) == 2)) {
return GR_GL_VERSION(major, minor);
}
return GR_GL_VERSION_INVALID;
}
static int
gl_renderer_setup(struct weston_compositor *ec, EGLSurface egl_surface)
{
struct gl_renderer *gr = get_renderer(ec);
const char *extensions;
EGLBoolean ret;
EGLint context_attribs[16] = {
EGL_CONTEXT_CLIENT_VERSION, 0,
};
unsigned int nattr = 2;
if (!eglBindAPI(EGL_OPENGL_ES_API)) {
weston_log("failed to bind EGL_OPENGL_ES_API\n");
gl_renderer_print_egl_error_state();
return -1;
}
/*
* Being the compositor we require minimum output latency,
* so request a high priority context for ourselves - that should
* reschedule all of our rendering and its dependencies to be completed
* first. If the driver doesn't permit us to create a high priority
* context, it will fallback to the default priority (MEDIUM).
*/
if (gr->has_context_priority) {
context_attribs[nattr++] = EGL_CONTEXT_PRIORITY_LEVEL_IMG;
context_attribs[nattr++] = EGL_CONTEXT_PRIORITY_HIGH_IMG;
}
assert(nattr < ARRAY_LENGTH(context_attribs));
context_attribs[nattr] = EGL_NONE;
/* try to create an OpenGLES 3 context first */
context_attribs[1] = 3;
gr->egl_context = eglCreateContext(gr->egl_display, gr->egl_config,
EGL_NO_CONTEXT, context_attribs);
if (gr->egl_context == NULL) {
/* and then fallback to OpenGLES 2 */
context_attribs[1] = 2;
gr->egl_context = eglCreateContext(gr->egl_display,
gr->egl_config,
EGL_NO_CONTEXT,
context_attribs);
if (gr->egl_context == NULL) {
weston_log("failed to create context\n");
gl_renderer_print_egl_error_state();
return -1;
}
}
if (gr->has_context_priority) {
EGLint value = EGL_CONTEXT_PRIORITY_MEDIUM_IMG;
eglQueryContext(gr->egl_display, gr->egl_context,
EGL_CONTEXT_PRIORITY_LEVEL_IMG, &value);
if (value != EGL_CONTEXT_PRIORITY_HIGH_IMG) {
weston_log("Failed to obtain a high priority context.\n");
/* Not an error, continue on as normal */
}
}
ret = eglMakeCurrent(gr->egl_display, egl_surface,
egl_surface, gr->egl_context);
if (ret == EGL_FALSE) {
weston_log("Failed to make EGL context current.\n");
gl_renderer_print_egl_error_state();
return -1;
}
gr->gl_version = get_gl_version();
if (gr->gl_version == GR_GL_VERSION_INVALID) {
weston_log("warning: failed to detect GLES version, "
"defaulting to 2.0.\n");
gr->gl_version = GR_GL_VERSION(2, 0);
}
log_gl_info();
gr->image_target_texture_2d =
(void *) eglGetProcAddress("glEGLImageTargetTexture2DOES");
extensions = (const char *) glGetString(GL_EXTENSIONS);
if (!extensions) {
weston_log("Retrieving GL extension string failed.\n");
return -1;
}
if (!weston_check_egl_extension(extensions, "GL_EXT_texture_format_BGRA8888")) {
weston_log("GL_EXT_texture_format_BGRA8888 not available\n");
return -1;
}
if (weston_check_egl_extension(extensions, "GL_EXT_read_format_bgra"))
ec->read_format = PIXMAN_a8r8g8b8;
else
ec->read_format = PIXMAN_a8b8g8r8;
if (gr->gl_version >= GR_GL_VERSION(3, 0) ||
weston_check_egl_extension(extensions, "GL_EXT_unpack_subimage"))
gr->has_unpack_subimage = true;
if (gr->gl_version >= GR_GL_VERSION(3, 0) ||
weston_check_egl_extension(extensions, "GL_EXT_texture_rg"))
gr->has_gl_texture_rg = true;
if (weston_check_egl_extension(extensions, "GL_OES_EGL_image_external"))
gr->has_egl_image_external = true;
glActiveTexture(GL_TEXTURE0);
if (compile_shaders(ec))
return -1;
gr->fragment_binding =
weston_compositor_add_debug_binding(ec, KEY_S,
fragment_debug_binding,
ec);
gr->fan_binding =
weston_compositor_add_debug_binding(ec, KEY_F,
fan_debug_repaint_binding,
ec);
gr->output_destroy_listener.notify = output_handle_destroy;
wl_signal_add(&ec->output_destroyed_signal,
&gr->output_destroy_listener);
weston_log("GL ES 2 renderer features:\n");
weston_log_continue(STAMP_SPACE "read-back format: %s\n",
ec->read_format == PIXMAN_a8r8g8b8 ? "BGRA" : "RGBA");
weston_log_continue(STAMP_SPACE "wl_shm sub-image to texture: %s\n",
gr->has_unpack_subimage ? "yes" : "no");
weston_log_continue(STAMP_SPACE "EGL Wayland extension: %s\n",
gr->has_bind_display ? "yes" : "no");
return 0;
}
WL_EXPORT struct gl_renderer_interface gl_renderer_interface = {
.display_create = gl_renderer_display_create,
.output_window_create = gl_renderer_output_window_create,
.output_pbuffer_create = gl_renderer_output_pbuffer_create,
.output_destroy = gl_renderer_output_destroy,
.output_set_border = gl_renderer_output_set_border,
.create_fence_fd = gl_renderer_create_fence_fd,
};