Split surface id and simplify connecting to surfaces service

Surface IDs are composed of a namespace component and a local
identifier component. This splits up the two pieces in the mojom to make
it clearer.

This also simplifies the path for connecting to the surfaces service. In
order to perform any operations with surfaces, each client must know
what its id namespace value is. This was done by first connecting to a
SurfacesService interface and then calling CreateSurfaceConnection which
returned a Surface pointer and the namespace. Having to connect to this
extra interface and receive the Surface impl asynchronously complicated
startup code for applications by adding extra states in startup.

Instead of that, this allows connecting to the Surface interface
directly and promises that the ID namespace will be provided as the
first message in the pipe directed at the SurfaceClient. Callers can
then either block on startup or handle setting the ID namespace
asynchronously depending on what other things that thread could be doing
at the time.

In a follow-up, I plan to define the id namespace value 0 as meaning "the
namespace of this connection" which will allow creating surfaces and
submitting frames before learning the connection's namespace. The only
thing the namespace will be passing surface IDs to other clients for them
to embed.

This also removes the Size parameter from CreateSurface, which wasn't
used by anything. The size of submitted frames is part of the embedder /
embedee contract.

R=esprehn@chromium.org, sky@chromium.org

Review URL: https://codereview.chromium.org/807733002
This commit is contained in:
James Robinson
2014-12-16 15:15:22 -08:00
parent 0cbe4deb25
commit 6d53b01c8d
4 changed files with 13 additions and 10 deletions

View File

@@ -16,13 +16,11 @@ SurfaceAllocator::SurfaceAllocator(uint32_t id_namespace)
SurfaceAllocator::~SurfaceAllocator() {
}
uint64_t SurfaceAllocator::CreateSurfaceId() {
// Surface IDs are 64 integers. The high 32 bits are the namespace of the ID,
// which is assigned to us by the surfaces service. The lower 32 bits are ours
// to allocate as we see fit. For simplicity, we just allocate them
// sequentially. In principle, we could run out, but at 60 Hz, it takes
// several years to run out.
return static_cast<uint64_t>(id_namespace_) << 32 | next_id_++;
mojo::SurfaceIdPtr SurfaceAllocator::CreateSurfaceId() {
auto id = mojo::SurfaceId::New();
id->local = next_id_++;
id->id_namespace = id_namespace_;
return id.Pass();
}
} // namespace sky

View File

@@ -6,6 +6,7 @@
#define SKY_COMPOSITOR_SURFACE_ALLOCATOR_H_
#include "base/basictypes.h"
#include "mojo/services/surfaces/public/interfaces/surface_id.mojom.h"
namespace sky {
@@ -14,7 +15,7 @@ class SurfaceAllocator {
SurfaceAllocator(uint32_t id_namespace);
~SurfaceAllocator();
uint64_t CreateSurfaceId();
mojo::SurfaceIdPtr CreateSurfaceId();
private:
uint32_t id_namespace_;

View File

@@ -51,13 +51,16 @@ void SurfaceHolder::SetSize(const gfx::Size& size) {
surface_id_ = mojo::SurfaceId::New();
}
surface_id_->id = surface_allocator_->CreateSurfaceId();
surface_->CreateSurface(surface_id_.Clone(), mojo::Size::From(size));
surface_id_ = surface_allocator_->CreateSurfaceId();
surface_->CreateSurface(surface_id_.Clone());
size_ = size;
client_->OnSurfaceIdAvailable(surface_id_.Clone());
}
void SurfaceHolder::SetIdNamespace(uint32_t id_namespace) {
}
void SurfaceHolder::ReturnResources(
mojo::Array<mojo::ReturnedResourcePtr> resources) {
// TODO(abarth): The surface service shouldn't spam us with empty calls.

View File

@@ -43,6 +43,7 @@ class SurfaceHolder : public mojo::SurfaceClient {
private:
// mojo::SurfaceClient
void SetIdNamespace(uint32_t id_namespace) override;
void ReturnResources(
mojo::Array<mojo::ReturnedResourcePtr> resources) override;