Abstract: Implementations disclosed describe a system and a method to execute a virtual machine on a processing device, receive a request to access a memory page identified by a guest virtual memory address (GVA) in an address space of the virtual machine, translate the GVA to a guest physical memory address (GPA) using a guest page table (GPT) comprising a GPT entry mapping the GVA to the GPA, translate the GPA to a host physical address (HPA) of the memory page, store, in a translation lookaside buffer (TLB), a TLB entry mapping the GVA to the HPA, modify the GPT entry to designate the memory page as accessed, detect an attempt by an application to modify the GPT entry; generate, in response to the attempt to modify the GPT entry, a page fault; and flush, in response to the page fault, the TLB entry.

Abstract: Systems and methods for container access to graphics processing unit (GPU) resources are disclosed herein. In some embodiments, a computing system may include a physical GPU and kernel-mode driver circuitry, to communicatively couple with the physical GPU to create a plurality of emulated GPUs and a corresponding plurality of device nodes. Each device node may be associated with a single corresponding user-side container to enable communication between the user-side container and the corresponding emulated GPU. Other embodiments may be disclosed and/or claimed.

Abstract: An apparatus and method are described for intelligent cloud based testing of graphics hardware and software. For example, one embodiment of an apparatus comprises: a hardware pool comprising a plurality of test machines to perform cloud-based graphics validation operations; a virtual resource pool comprising data associated a plurality of different graphics hardware resources; a resource manager to coordinate between the hardware pool and the virtual resource pool to cause one or more virtual machines (VMs) to be executed on one or more of the test machines using resources from the virtual resource pool; and a task dispatcher to dispatch graphics validation tasks to the VMs responsive to user input.

Abstract: Implementations disclosed describe a system and a method to execute a virtual machine on a processing device, receive a request to access a memory page identified by a guest virtual memory address (GVA) in an address space of the virtual machine, translate the GVA to a guest physical memory address (GPA) using a guest page table (GPT) comprising a GPT entry mapping the GVA to the GPA, translate the GPA to a host physical address (HPA) of the memory page, store, in a translation lookaside buffer (TLB), a TLB entry mapping the GVA to the HPA, modify the GPT entry to designate the memory page as accessed, detect an attempt by an application to modify the GPT entry; generate, in response to the attempt to modify the GPT entry, a page fault; and flush, in response to the page fault, the TLB entry.

Abstract: Implementations disclosed describe a system and a method to execute a virtual machine on a processing device, receive a request to access a memory page identified by a guest virtual memory address (GVA) in an address space of the virtual machine, translate the GVA to a guest physical memory address (GPA) using a guest page table (GPT) comprising a GPT entry mapping the GVA to the GPA, translate the GPA to a host physical address (HPA) of the memory page, store, in a translation lookaside buffer (TLB), a TLB entry mapping the GVA to the HPA, modify the GPT entry to designate the memory page as accessed, detect an attempt by an application to modify the GPT entry; generate, in response to the attempt to modify the GPT entry, a page fault; and flush, in response to the page fault, the TLB entry.

Abstract: Technologies for scheduling workload submissions for a graphics processing unit (GPU) in a virtualization environment include a GPU scheduler embodied in a computing device. The virtualization environment includes a number of different virtual machines that are configured with a native graphics driver. The GPU scheduler receives GPU commands from the different virtual machines, dynamically selects a scheduling policy, and schedules the GPU commands for processing by the GPU.

Abstract: Implementations disclosed describe a system and a method to execute a virtual machine on a processing device, receive a request to access a memory page identified by a guest virtual memory address (GVA) in an address space of the virtual machine, translate the GVA to a guest physical memory address (GPA) using a guest page table (GPT) comprising a GPT entry mapping the GVA to the GPA, translate the GPA to a host physical address (HPA) of the memory page, store, in a translation lookaside buffer (TLB), a TLB entry mapping the GVA to the HPA, modify the GPT entry to designate the memory page as accessed, detect an attempt by an application to modify the GPT entry; generate, in response to the attempt to modify the GPT entry, a page fault; and flush, in response to the page fault, the TLB entry.

Abstract: Systems and methods for container access to graphics processing unit (GPU) resources are disclosed herein. In some embodiments, a computing system may include a physical GPU and kernel-mode driver circuitry, to communicatively couple with the physical GPU to create a plurality of emulated GPUs and a corresponding plurality of device nodes. Each device node may be associated with a single corresponding user-side container to enable communication between the user-side container and the corresponding emulated GPU. Other embodiments may be disclosed and/or claimed.

Abstract: An apparatus and method are described for intelligent resource provisioning for shadow structures. For example, one embodiment of an apparatus comprises: graphics processing unit (GPU) to process graphics commands and responsively render a plurality of image frames in a graphics memory address space; shadow structure management logic to reserve one or more shadow slots in the graphics memory address space in which to store shadow instances of different GPU contexts; and the shadow structure management logic to implement a partial shadowing policy for shadowing GPU contexts in the shadow slots, the partial shadowing policy based on characteristics of pages of the GPU contexts.

Abstract: Technologies for scheduling workload submissions for a graphics processing unit (GPU) in a virtualization environment include a GPU scheduler embodied in a computing device. The virtualization environment includes a number of different virtual machines that are configured with a native graphics driver. The GPU scheduler receives GPU commands from the different virtual machines, dynamically selects a scheduling policy, and schedules the GPU commands for processing by the GPU.

Abstract: Apparatuses, methods and storage medium associated with live migration of virtual machines (VMs) from/to host computers with graphics virtualization are disclosed herein. In embodiments, an apparatus may include a virtual machine monitor (VMM) having a memory manager to manage accesses of system memory of the apparatus, including tracking of modified memory pages of the system memory. Additionally, the VMM may include a graphics command parser to analyze graphics commands issued to a graphics processor (GPU) of the apparatus to detect writes to the system memory caused by the graphics commands, and augment the tracking of modified memory pages. Further, the VMM may include a live migration function to live migrate a VM to another apparatus, including provision of current memory content of the VM, utilizing modified memory pages tracked by the memory manager, as augmented by the graphics command parser.

Abstract: An apparatus and method are described for intelligent cloud based testing of graphics hardware and software. For example, one embodiment of an apparatus comprises: a hardware pool comprising a plurality of test machines to perform cloud-based graphics validation operations; a virtual resource pool comprising data associated a plurality of different graphics hardware resources; a resource manager to coordinate between the hardware pool and the virtual resource pool to cause one or more virtual machines (VMs) to be executed on one or more of the test machines using resources from the virtual resource pool; and a task dispatcher to dispatch graphics validation tasks to the VMs responsive to user input.

Abstract: An apparatus and method are described for intelligent resource provisioning for shadow structures. For example, one embodiment of an apparatus comprises: graphics processing unit (GPU) to process graphics commands and responsively render a plurality of image frames in a graphics memory address space; shadow structure management logic to reserve one or more shadow slots in the graphics memory address space in which to store shadow instances of different GPU contexts; and the shadow structure management logic to implement a partial shadowing policy for shadowing GPU contexts in the shadow slots, the partial shadowing policy based on characteristics of pages of the GPU contexts.

Abstract: Technologies for scheduling workload submissions for a graphics processing unit (GPU) in a virtualization environment include a GPU scheduler embodied in a computing device. The virtualization environment includes a number of different virtual machines that are configured with a native graphics driver. The GPU scheduler receives GPU commands from the different virtual machines, dynamically selects a scheduling policy, and schedules the GPU commands for processing by the GPU.

Abstract: Systems and methods for container access to graphics processing unit (GPU) resources are disclosed herein. In some embodiments, a computing system may include a physical GPU and kernel-mode driver circuitry, to communicatively couple with the physical GPU to create a plurality of emulated GPUs and a corresponding plurality of device nodes. Each device node may be associated with a single corresponding user-side container to enable communication between the user-side container and the corresponding emulated GPU. Other embodiments may be disclosed and/or claimed.

Abstract: Apparatuses, methods and storage medium associated with live migration of VMs from/to host computers with graphics virtualization are disclosed herein. In embodiments, an apparatus may include a VMM having a memory manager to manage accesses of system memory of the apparatus, including tracking of modified memory pages of the system memory. Additionally, the VMM may include a graphics command parser to analyze graphics commands issued to a GPU of the apparatus to detect writes to the system memory caused by the graphics commands, and augment the tracking of modified memory pages. Further, the VMM may include a live migration function to live migrate a VM to another apparatus, including provision of current memory content of the VM, utilizing modified memory pages tracked by the memory manager, as augmented by the graphics command parser. Other embodiments may be described and/or claimed.

Abstract: Apparatuses, methods and storage medium associated with live migration of virtual machines (VMs) from/to host computers with graphics virtualization are disclosed herein. In embodiments, an apparatus may include a virtual machine monitor (VMM) having a memory manager to manage accesses of system memory of the apparatus, including tracking of modified memory pages of the system memory. Additionally, the VMM may include a graphics command parser to analyze graphics commands issued to a graphics processor (GPU) of the apparatus to detect writes to the system memory caused by the graphics commands, and augment the tracking of modified memory pages. Further, the VMM may include a live migration function to live migrate a VM to another apparatus, including provision of current memory content of the VM, utilizing modified memory pages tracked by the memory manager, as augmented by the graphics command parser.

Abstract: Technologies for scheduling workload submissions for a graphics processing unit (GPU) in a virtualization environment include a GPU scheduler embodied in a computing device. The virtualization environment includes a number of different virtual machines that are configured with a native graphics driver. The GPU scheduler receives GPU commands from the different virtual machines, dynamically selects a scheduling policy, and schedules the GPU commands for processing by the GPU.

Abstract: Apparatuses, methods and storage medium associated with live migration of VMs from/to host computers with graphics virtualization are disclosed herein. In embodiments, an apparatus may include a VMM having a memory manager to manage accesses of system memory of the apparatus, including tracking of modified memory pages of the system memory. Additionally, the VMM may include a graphics command parser to analyze graphics commands issued to a GPU of the apparatus to detect writes to the system memory caused by the graphics commands, and augment the tracking of modified memory pages. Further, the VMM may include a live migration function to live migrate a VM to another apparatus, including provision of current memory content of the VM, utilizing modified memory pages tracked by the memory manager, as augmented by the graphics command parser. Other embodiments may be described and/or claimed.

Abstract: An embodiment of the present invention is a technique to hide latency in program traces. Blocks of instructions between start and end of a critical section are associated with color information. The blocks correspond to a program trace and containing a wait instruction. The wait instruction is sunk down the blocks globally to the end of the critical section using the color information and a dependence constraint on the wait instruction.