专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US08341628B2 Controlling depth and latency of exit of a virtual processor's idle state in a power management environment 失效
标题翻译：在电源管理环境中控制虚拟处理器空闲状态退出的深度和延迟
公开(公告)号：US08341628B2
公开(公告)日：2012-12-25
申请号：US12645597
申请日：2009-12-23
申请人： Richard Louis Arndt , Christopher Francois , Naresh Nayar , Karthick Rajamani , Freeman Leigh Rawson, III , Randal Craig Swanberg
发明人： Richard Louis Arndt , Christopher Francois , Naresh Nayar , Karthick Rajamani , Freeman Leigh Rawson, III , Randal Craig Swanberg
IPC分类号： G06F9/455 , G06F1/00
CPC分类号： G06F9/45533 , G06F9/45541 , G06F2009/45575
摘要： A mechanism is provided in a logically partitioned data processing system for controlling depth and latency of exit of a virtual processor's idle state. A virtualization layer generates a cede latency setting information (CLSI) data. Responsive to booting a logical partition, the virtualization layer communicates the CLSI data to an operating system (OS) of the logical partition. The OS determines, based on the CLSI data, a particular idle state of a virtual processor under a control of the OS. Responsive to the OS calling the virtualization layer, the OS communicates the particular idle state of the virtual processor to the virtualization layer for assigning the particular idle state and wake-up characteristics to the virtual processor.
摘要翻译：在逻辑分区的数据处理系统中提供一种机制，用于控制虚拟处理器的空闲状态的退出的深度和等待时间。一个虚拟化层产生一个雪松延迟设置信息（CLSI）数据。响应于引导逻辑分区，虚拟化层将CLSI数据传送到逻辑分区的操作系统（OS）。 OS根据CLSI数据确定在OS控制下的虚拟处理器的特定空闲状态。响应于调用虚拟化层的OS，OS将虚拟处理器的特定空闲状态传送到虚拟化层，以将特定的空闲状态和唤醒特性分配给虚拟处理器。

2. 发明授权

US07653799B2 Method and apparatus for managing memory for dynamic promotion of virtual memory page sizes 有权
标题翻译：用于管理虚拟存储器页面大小的动态提升的存储器的方法和装置
公开(公告)号：US07653799B2
公开(公告)日：2010-01-26
申请号：US11751004
申请日：2007-05-19
申请人： David Alan Hepkin , Randal Craig Swanberg
发明人： David Alan Hepkin , Randal Craig Swanberg
IPC分类号： G06F12/12
CPC分类号： G06F12/023
摘要： A computer implemented method, apparatus, and computer usable program code for managing real memory. In response to a request for a page to be moved into real memory, a contiguous range of real memory is reserved for the page corresponding to a contiguous virtual memory range to form a reservation within a plurality of reservations for the real memory. This reservation enables efficient promotion of pages to a larger page size. The page only occupies a portion of the contiguous range of real memory for the reservation. In response to a need for real memory, a selected reservation is released within the plurality of reservations based on an age of the selected reservation within the plurality of reservations.
摘要翻译：用于管理实际存储器的计算机实现的方法，装置和计算机可用程序代码。响应于要移动到实际存储器的页面的请求，为与相邻虚拟存储器范围相对应的页面保留连续范围的实际存储器，以在真实存储器的多个预留内形成预留。此预订可以有效地将页面升级到更大的页面大小。该页面仅占用预留的实际存储器的连续范围的一部分。响应于对真实存储器的需要，基于多个保留期间所选保留的年龄，在多个保留内释放所选择的保留。

3. 发明授权

US07448037B2 Method and data processing system having dynamic profile-directed feedback at runtime 失效
标题翻译：方法和数据处理系统在运行时具有动态配置文件导向的反馈
公开(公告)号：US07448037B2
公开(公告)日：2008-11-04
申请号：US10755878
申请日：2004-01-13
申请人： Ravi Kumar Arimilli , Randal Craig Swanberg
发明人： Ravi Kumar Arimilli , Randal Craig Swanberg
IPC分类号： G06F9/46 , G06F9/30
CPC分类号： G06F9/5011 , G06F2209/507
摘要： Software communicates to a processing unit a classification each of at least one schedulable software entity that the processing unit executes. A resource manager within the processing unit dynamically allocates hardware resources within the processing unit to the schedulable software entity during execution in accordance with the classification. The classification may be retrieved by the software from in data storage, and operating system software may schedule the schedulable software entity for execution by reference to the classification. The processing unit may also monitor, in hardware, execution of each of a plurality of schedulable software entities within the processing unit in accordance with a monitoring parameter set. The processing unit may then report to software the utilization of hardware resources by each of the plurality of schedulable software entities so that the software may develop or refine a classification for the schedulable software entity.
摘要翻译：软件向处理单元传送处理单元执行的至少一个可调度软件实体的分类。处理单元内的资源管理器根据分类在执行期间将处理单元内的硬件资源动态地分配给可调度软件实体。分类可以由软件从数据存储中检索，并且操作系统软件可以通过参考分类来调度可调度的软件实体以供执行。处理单元还可以根据监视参数集在硬件中监视处理单元内的多个可调度软件实体中的每一个的执行。处理单元然后可以通过多个可调度软件实体中的每一个向软件报告硬件资源的利用，使得软件可以开发或改进可调度软件实体的分类。

4. 发明授权

US06918023B2 Method, system, and computer program product for invalidating pretranslations for dynamic memory removal 失效
标题翻译：方法，系统和计算机程序产品，用于无效动态内存删除的预翻译
公开(公告)号：US06918023B2
公开(公告)日：2005-07-12
申请号：US10262176
申请日：2002-09-30
申请人： Luke Matthew Browning , Bruce G. Mealey , Randal Craig Swanberg
发明人： Luke Matthew Browning , Bruce G. Mealey , Randal Craig Swanberg
IPC分类号： G06F12/00 , G06F12/10
CPC分类号： G06F12/1081
摘要： A system, method, and computer program product are disclosed for invalidating specified pretranslations maintained in a data processing system which maintains decentralized copies of pretranslations. A centralized mapping of virtual addresses to their associated physical addresses is established. The centralized mapping includes a listing of pretranslations of the virtual addresses to their associated physical addresses. Multiple lists of pretranslations are generated. Control of the lists may be passed from one entity to another, such that the lists are not owned by any particular entity. Each one of the lists includes a copy of pretranslations for virtual addresses. A particular one of the physical addresses is specified. Each list that includes a pretranslation of a virtual address to the specified physical addresses is located. The pretranslation of the virtual address to the specified physical address is then invalidated within each one of the lists.
摘要翻译：公开了一种系统，方法和计算机程序产品，用于使数据处理系统中维护的指定的预翻译失效，该数据处理系统维护预翻译的分散副本。建立虚拟地址与其相关物理地址的集中映射。集中式映射包括虚拟地址与其相关联的物理地址的预翻译列表。生成多个预翻译列表。列表的控制可以从一个实体传递到另一个实体，使得列表不是由任何特定实体拥有的。每个列表包括虚拟地址的预翻译副本。指定了一个特定的物理地址。每个包含虚拟地址到指定物理地址的转换的列表都位于。然后虚拟地址到指定物理地址的预翻译在每个列表中无效。

5. 发明授权

US06895508B1 Stack memory protection 失效
标题翻译：堆栈内存保护
公开(公告)号：US06895508B1
公开(公告)日：2005-05-17
申请号：US09657121
申请日：2000-09-07
申请人： Randal Craig Swanberg , Michael Stephen Williams
发明人： Randal Craig Swanberg , Michael Stephen Williams
IPC分类号： G06F9/30 , G06F9/312 , G06F12/14
CPC分类号： G06F9/30134 , G06F9/30043 , G06F12/145
摘要： A method and system for memory page protection wherein new stack memory load/store instructions are defined for memory management. A corresponding operating system and compiler utilize these new stack memory load/store instructions. Whenever it is desired to have a block of memory used as a stack memory, the stack memory load/store instructions are used. A stack memory attribute is stored in a page table associated with the block of memory. Memory blocks having a stack memory attribute may be read and written into using only stack memory load/store instructions. If a normal load/store is attempted to a memory block having a stack memory attribute a error condition is indicated. Likewise a stack memory load/store to a block of memory not have a stack memory attribute will cause a error condition. Stack memory load/stores meant for one type of stack memory (e.g., program stack attribute) will also cause a fault if the stack load/store is attempted to another type of stack memory (e.g., processor stack). Stack memory (processor stacks), transparent to a programmer writing code for a processor employing stack memory attributes, would have a processor stack attributes assigned by the processor or CPU. Using this method and system, stack memory may be assigned anywhere in memory without creating wasted protected pages or having data corrupted by stack memory overruns or under runs. The operating system no longer needs to allocate specific space in memory as stack memory and likewise does not have to estimate how much memory will be needed for program stacks and processor stacks (e.g.IA64 register stacks).
摘要翻译：一种用于存储器页面保护的方法和系统，其中为存储器管理定义新的堆栈存储器加载/存储指令。相应的操作系统和编译器利用这些新的堆栈存储器加载/存储指令。每当需要将一块存储器用作堆栈存储器时，使用堆栈存储器加载/存储指令。堆栈存储器属性存储在与存储器块相关联的页表中。具有堆栈存储器属性的存储器块可以仅使用堆栈存储器加载/存储指令来读取和写入。如果对具有堆栈存储器属性的存储器块尝试正常加载/存储，则指示错误条件。同样，堆栈内存加载/存储到一个内存块不具有堆栈内存属性将导致错误条件。对于一种类型的堆栈存储器（例如，程序堆栈属性）的堆栈存储器加载/存储（如果堆栈加载/存储被尝试到另一种类型的堆栈存储器（例如，处理器堆栈））也将引起故障。对于使用堆栈存储器属性的处理器编写代码的编程器，堆栈存储器（处理器堆栈）将具有由处理器或CPU分配的处理器堆栈属性。使用此方法和系统，堆栈内存可能会分配到内存中的任何位置，而不会创建浪费的受保护页面，或者由堆栈内存超限或运行中的数据损坏。操作系统不再需要在内存中分配特定空间作为堆栈存储器，同样也不需要估计程序堆栈和处理器堆栈（例如，ISA64寄存器堆栈）将需要多少内存。

6. 发明授权

US06654878B1 Register bit scanning 有权
标题翻译：注册位扫描
公开(公告)号：US06654878B1
公开(公告)日：2003-11-25
申请号：US09657117
申请日：2000-09-07
申请人： Randal Craig Swanberg , Michael Stephen Williams
发明人： Randal Craig Swanberg , Michael Stephen Williams
IPC分类号： G06F9308
CPC分类号： G06F9/30018 , G06F9/30072
摘要： Testing register bits and in particular bitmask registers is a method employed in many computer architectures (e.g., IBM PowerPC, IA32, VAX, etc.) to manage instruction flow within a processor. Since the testing or scanning of bitmask registers for the first occurrence of a logic state (e.g., logic one) is done quite often, register scanning is implemented in hardware in these processors. Other computer architectures (e.g., Intel IA64) manage instruction flow with alternate methods and therefore do register scanning as a software construct. When software written for the first computer architecture (e.g., IBM PowerPC) is ported to a system with IA64 architecture, the program would execute with reduced speed. The IA64 architecture uses the EPIC instruction protocol and as such executes predicate instructions that employ a predicate register where each bit of the predicate register can be associated as the true or false result of a comparison. To scan a register in the IA64 architecture the register contents are loaded into the predicate register and a sequence of predicate instructions are executed in the order that the bits are to be scanned for the desired condition. The sequence of predicate instruction sequence returns the register bit that passes the predicate condition. In this manner the speed of register scanning necessary for the ported software can be increased over the software scanning of the IA64 architecture.
摘要翻译：测试寄存器位和特定位掩码寄存器是许多计算机体系结构（例如，IBM PowerPC，IA32，VAX等）中用于管理处理器内的指令流的方法。由于对逻辑状态（例如，逻辑1）的第一次发生的位掩码寄存器的测试或扫描经常进行，所以在这些处理器中的硬件中实现寄存器扫描。其他计算机体系结构（例如，Intel IA64）利用替代方法管理指令流，因此进行注册扫描作为软件构造。当为第一台计算机体系结构（例如IBM PowerPC）编写的软件移植到具有IA64架构的系统时，该程序将以较低的速度执行。 IA64架构使用EPIC指令协议，因此执行使用谓词寄存器的谓词指令，其中谓词寄存器的每个位可以与比较的真或否结果相关联。要扫描IA64体系结构中的寄存器，将寄存器内容加载到谓词寄存器中，并按照要扫描所需条件的顺序执行一系列谓词指令。谓词指令序列的序列返回通过谓词条件的寄存器位。以这种方式，可以通过IA64架构的软件扫描来增加移植软件所需的寄存器扫描速度。

7. 发明授权

US5832280A Method and system in a data processing system for interfacing an operating system with a power management controller. 失效
标题翻译：用于将操作系统与电力管理控制器接口的数据处理系统中的方法和系统。
公开(公告)号：US5832280A
公开(公告)日：1998-11-03
申请号：US539658
申请日：1995-10-05
申请人： Randal Craig Swanberg
发明人： Randal Craig Swanberg
IPC分类号： G06F1/26 , G06F1/32
CPC分类号： G06F1/26
摘要： In a data processing system having an operating system and a power management controller coupled to one or more power-managed devices, each of the power-managed devices is assigned a device identifier. An architected power-managed device select register and an architected power mode select register are provided within the data processing system. To modify a power mode of a power-managed device, the operating system writes a selected one of the device identifiers to the architected power-managed device select register for selecting an identified one of the power-managed devices. Thereafter, the operating system writes a power mode identifier to the architect power mode select register for selecting one of a plurality of power modes within the selected power-managed device. The device identifier and the power mode identifier are translated into control signals for the power management controller within the data processing system. Such control signals are then transmitted to the power management controller and the identified power-managed device is operated in the selected power mode, wherein the operating system controls the power modes of power-managed devices without programming a particular implementation of power management control in the operating system.
摘要翻译：在具有耦合到一个或多个功率管理设备的操作系统和功率管理控制器的数据处理系统中，每个功率管理设备被分配设备标识符。在数据处理系统中提供了一种架构化的功率管理设备选择寄存器和架构化的功率模式选择寄存器。为了修改功率管理设备的功率模式，操作系统将所选择的一个设备标识符写入到架构化的功率管理设备选择寄存器中，以选择被识别的功率管理设备之一。此后，操作系统将功率模式标识符写入架构师功率模式选择寄存器，用于选择所选择的功率管理设备内的多个功率模式中的一个。设备标识符和功率模式标识符被转换为数据处理系统内的功率管理控制器的控制信号。然后将这样的控制信号发送到电力管理控制器，并且所识别的功率管理设备以所选择的功率模式运行，其中操作系统控制功率管理设备的功率模式，而无需编程功率管理设备中的功率管理控制的特定实现操作系统。

8. 发明授权

US08762680B2 Scaling energy use in a virtualized environment 有权
标题翻译：扩展虚拟化环境中的能源使用
公开(公告)号：US08762680B2
公开(公告)日：2014-06-24
申请号：US13358636
申请日：2012-01-26
申请人： Richard Loise Arndt , Randal Craig Swanberg
发明人： Richard Loise Arndt , Randal Craig Swanberg
IPC分类号： G06F12/00 , G06F9/50
CPC分类号： G06F9/5077 , G06F1/3203 , G06F1/3287 , G06F3/0625 , G06F3/0634 , G06F3/0653 , G06F3/0665 , G06F3/067 , G06F9/5094 , Y02D10/154 , Y02D10/171 , Y02D10/22 , Y02D10/36 , Y02D50/20
摘要： A set of physical input/output adapters (PIOAs) is configured such that each PIOAs in the set of PIOAs is a functional equivalent of another PIOAs in the set of PIOAs. A utilization of each PIOA in the set of PIOAs is measured. A number of PIOAs needed to service a workload is determined. A first subset of PIOAs from the set of PIOAs is powered down if the number of PIOAs needed to service the workload is smaller than a number of operational PIOAs. The I/O operations associated with the first subset of PIOAs are transferred to a second subset of PIOAs remaining operational in the set of PIOAs.
摘要翻译：一组物理输入/输出适配器（PIO）被配置为使得该组PIO中的每个PIO是与该组PIO中的另一个PIO的功能等价物。测量该组PIO中每个PIOA的利用率。确定了服务工作负载所需的一些PIO。如果需要维护工作负载的PIO数量小于多个可操作PIO，PIO的PIOA的第一个子集将被关闭。与PIO的第一子集相关联的I / O操作被传送到在该组PIO中保持操作的PIO的第二子集。

9. 发明授权

US08745307B2 Multiple page size segment encoding 有权
标题翻译：多页大小段编码
公开(公告)号：US08745307B2
公开(公告)日：2014-06-03
申请号：US12779563
申请日：2010-05-13
申请人： Sundeep Chadha , Cathy May , Naresh Nayar , Randal Craig Swanberg
发明人： Sundeep Chadha , Cathy May , Naresh Nayar , Randal Craig Swanberg
IPC分类号： G06F12/00 , G06F12/10
CPC分类号： G06F12/1018 , G06F12/1009 , G06F12/1027 , G06F12/1036 , G06F2212/652
摘要： An approach identifies an amount of high order bits used to store a memory address in a memory address field that is included in a memory. This approach calculates at least one minimum number of low order bits not used to store the address with the calculation being based on the identified amount of high order bits. The approach retrieves a data element from one of the identified minimum number of low order bits of the address field and also retrieves a second data element from one of the one of the identified minimum number of low order bits of the address field.
摘要翻译：一种方法识别用于在存储器中包括的存储器地址字段中存储存储器地址的高位数量。该方法计算不用于存储地址的至少一个最低数量的低阶位，其中计算基于所识别的高位位数。该方法从所识别的地址字段的最低位数的最低位数中的一个中检索数据元素，并且还从所识别的地址字段的最低位数中的一个中检索第二数据元素。

10. 发明申请

US20080288718A1 METHOD AND APPARATUS FOR MANAGING MEMORY FOR DYNAMIC PROMOTION OF VIRTUAL MEMORY PAGE SIZES 有权
标题翻译：管理虚拟存储器页面尺寸动态记忆的方法和装置
公开(公告)号：US20080288718A1
公开(公告)日：2008-11-20
申请号：US11751004
申请日：2007-05-19
申请人： DAVID ALAN HEPKIN , RANDAL CRAIG SWANBERG
发明人： DAVID ALAN HEPKIN , RANDAL CRAIG SWANBERG
IPC分类号： G06F12/12
CPC分类号： G06F12/023
摘要： A computer implemented method, apparatus, and computer usable program code for managing real memory. In response to a request for a page to be moved into real memory, a contiguous range of real memory is reserved for the page corresponding to a contiguous virtual memory range to form a reservation within a plurality of reservations for the real memory. This reservation enables efficient promotion of pages to a larger page size. The page only occupies a portion of the contiguous range of real memory for the reservation. In response to a need for real memory, a selected reservation is released within the plurality of reservations based on an age of the selected reservation within the plurality of reservations.
摘要翻译：用于管理实际存储器的计算机实现的方法，装置和计算机可用程序代码。响应于要移动到实际存储器的页面的请求，为与相邻虚拟存储器范围相对应的页面保留连续范围的实际存储器，以在真实存储器的多个预留内形成预留。此预订可以有效地将页面升级到更大的页面大小。该页面仅占用预留的实际存储器的连续范围的一部分。响应于对真实存储器的需要，基于多个保留期间所选保留的年龄，在多个保留内释放所选择的保留。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式