Invention Grant
US08390631B2 Synchronizing queued data access between multiple GPU rendering contexts 有权
在多个GPU渲染上下文之间同步排队的数据访问

Synchronizing queued data access between multiple GPU rendering contexts
Abstract:
Synchronized access to a shared surface from multiple rendering contexts is provided. Only one rendering context is allowed to access a shared surface at a given time to read from and write to the surface. Other non-owning rendering contexts are prevented from accessing and rendering to the shared surface while the surface is currently owned by another rendering context. A non-owning rendering context makes an acquire call and waits for the surface to be released. When the currently owning rendering context finishes rendering to the shared surface, it release the surface. The rendering context that made the acquire call then acquires access and renders to the shared surface.
Information query
Patent Agency Ranking
0/0