Abstract: Examples are disclosed for adjusting a performance state of a graphics subsystem and/or a processor based on a comparison of an average frame rate to a target frame rate and also based on whether the graphics subsystem is in a burst mode or sustained mode of operation.

Abstract: Embodiments of an invention for dynamic error correction using parity and redundant rows are disclosed. In one embodiment, an apparatus includes a storage structure, parity logic, an error storage space, and an error event generator. The storage structure is to store a plurality of data values. The parity logic is to detect a parity error in a data value stored in the storage structure. The error storage space is to store an indication of a detection of the parity error. The error event generator is to generate an event in response to the indication of the parity error being stored in the error storage space.

Abstract: A graphics processing unit's workload duration is monitored across a number of frames. A threshold “k” may be used to determine if the workload is Burst or Sustained for a number of frames and another time constant “t” may be used to monitor the burst behavior. If the device continues to be in burst mode over time “t” and the performance state is not an energy efficient state, then the system may lower the performance state to “Pe” and monitor if the same workload remains as Burst. If not, the performance state may be raised to the next higher performance state.

Abstract: In accordance with some embodiments, a system may detect whether or not a workload currently being worked on by two processors is serialized or concurrent. A workload is serialized or a producer consumer workload when the workload is such that one processor must receive the output from the other processor before it can begin. A workload is concurrent if both processors can work on the workload at the same time. In one embodiment, the nature of memory accesses can be used to determine the workload type. For example, when both processors use a shared virtual memory, the memory accesses can be tracked to detect whether serialized or concurrent workloads are involved.

Abstract: The graphics pipeline produces real time utilization data for each of a plurality of functional units making up an overall graphics processor or graphics system on a chip. This information may be used for fine grain management of power consumption and performance at the functional unit level as opposed the overall device level. As a result, the graphics functional units may be managed dynamically based on real time hardware metrics to improve performance and reduce power consumption. The technique may be implemented in a software module in one embodiment.

Abstract: A method and apparatus for supporting programmable software context state execution during hardware context restore flow is described. In one example, a context ID is assigned to graphics applications including a unique context memory buffer, a unique indirect context pointer and a corresponding size to each context ID, an indirect context offset, and an indirect context buffer address range. When execution of the first context workload is indirected, the state of the first context workload is saved to the assigned context memory buffer. The indirect context pointer, the indirect context offset and a size of the indirect context buffer address range are saved to registers that are independent of the saved context state. The context is restored by accessing the saved indirect context pointer, the indirect context offset and the buffer size.

Abstract: Mid-command buffer preemption is described for graphics workloads in a graphics processing environment. In one example, instructions of a first context are executed at a graphics processor, the first context has a sequence of instructions in an addressable buffer and at least one of the instructions is a preemption instruction. Upon executing the preemption instruction, execution of the first context is stopped before the sequence of instructions is completed. An address is stored for an instruction with which the first context will be resumed. The second context is executed, and upon completion of the execution of the second context, the execution of the first context is resumed at the stored address.

Abstract: An electronic device is described herein. The electronic device may include a page walker module to receive a page request of a graphics processing unit (GPU). The page walker module may detect a page fault associated with the page request. The electronic device may include a controller, at least partially comprising hardware logic. The controller is to monitor execution of the page request having the page fault. The controller determines whether to suspend execution of a work item at the GPU associated with the page request having the page fault, or to continue execution of the work item based on factors associated with the page request.

Abstract: Examples are disclosed for adjusting a performance state of a graphics subsystem and/or a processor based on a comparison of an average frame rate to a target frame rate and also based on whether the graphics subsystem is in a burst mode or sustained mode of operation.

Abstract: A method, device, and system to distribute code and data stores between volatile and non-volatile memory are described. In one embodiment, the method includes storing one or more static code segments of a software application in a phase change memory with switch (PCMS) device, storing one or more static data segments of the software application in the PCMS device, and storing one or more volatile data segments of the software application in a volatile memory device. The method then allocates an address mapping table with at least a first address pointer to point to each of the one or more static code segments, at least a second address pointer to point to each of the one or more static data segments, and at least a third address pointer to point to each of the one or more volatile data segments.

Abstract: Methods and devices to augment volatile memory in a graphics subsystem with certain types of non-volatile memory are described. In one embodiment, includes storing one or more static or near-static graphics resources in a non-volatile random access memory (NVRAM). The NVRAM is directly accessible by a graphics processor using at least memory store and load commands. The method also includes a graphics processor executing a graphics application. The graphics processor sends a request using a memory load command for an address corresponding to at least one static or near-static graphics resources stored in the NVRAM. The method also includes directly loading the requested graphics resource from the NVRAM into a cache for the graphics processor in response to the memory load command.

Abstract: Techniques are disclosed for processing rendering engine workload of a graphics system in a secure fashion, wherein at least some security processing of the workload is offloaded from software-based security parsing to hardware-based security parsing. In some embodiments, commands from a given application are received by a user mode driver (UMD), which is configured to generate a command buffer delineated into privileged and/or non-privileged command sections. The delineated command buffer can then be passed by the UMD to a kernel-mode driver (KMD), which is configured to parse and validate only privileged buffer sections, but to issue all other batch buffers with a privilege indicator set to non-privileged.

Abstract: Transitions to ring 0, each time an application wants to use an adjunct processor, are avoided, saving central processor operating cycles and improving efficiency. Instead, initially each application is registered and setup to use adjunct processor resources in ring 3.

Abstract: A computing device for performing scheduling operations for graphics hardware is described herein. The computing device includes a central processing unit (CPU) that is configured to execute an application. The computing device also includes a graphics scheduler configured to operate independently of the CPU. The graphics scheduler is configured to receive work queues relating to workloads from the application that are to execute on the CPU and perform scheduling operations for any of a number of graphics engines based on the work queues.

Abstract: In accordance with some embodiments, a graphics process frame generation frame rate may be monitored in combination with a utilization or work load metric for the graphics process in order to allocate performance resources to the graphics process and in some cases, between the graphics process and a central processing unit.

Abstract: Embodiments of an invention for dynamic error correction using parity and redundant rows are disclosed. In one embodiment, an apparatus includes a storage structure, parity logic, an error storage space, and an error event generator. The storage structure is to store a plurality of data values. The parity logic is to detect a parity error in a data value stored in the storage structure. The error storage space is to store an indication of a detection of the parity error. The error event generator is to generate an event in response to the indication of the parity error being stored in the error storage space.

Abstract: In accordance with some embodiments, a graphics process frame generation frame rate may be monitored in combination with a utilization or work load metric for the graphics process in order to allocate performance resources to the graphics process and in some cases, between the graphics process and a central processing unit.

Abstract: According to some embodiments, a power budget allocation engine of a multi-component computer system may receive a power budget allocation adjustment request signal from a first component. Based on the received budget allocation adjustment request signal (and, in some embodiments, a component preference), the power budget allocation engine may determine whether to adjust a power budget allocation signal provided to the first component.

Abstract: In accordance with some embodiments, a graphics process frame generation frame rate may be monitored in combination with a utilization or work load metric for the graphics process in order to allocate performance resources to the graphics process and in some cases, between the graphics process and a central processing unit.

Abstract: In accordance with some embodiments, a graphics process frame generation frame rate may be monitored in combination with a utilization or work load metric for the graphics process in order to allocate performance resources to the graphics process and in some cases, between the graphics process and a central processing unit.