Abstract: A lookup engine of a transactional memory (TM) has multiple hardware lookup structures, each usable to perform a different type of lookup. In response to a lookup command, the lookup engine reads a first block of first information from a memory unit. The first information configures the lookup engine to perform a first type of lookup, thereby identifying a first result value. If the first result value is not a final result value, then the lookup engine uses address information in the first result value to read a second block of second information. The second information configures the lookup engine to perform a second type of lookup, thereby identifying a second result value. This process repeats until a final result value is obtained. The type of lookup performed is determined by the result value of the preceding lookup and/or type information of the block of information for the next lookup.

Abstract: The page tables in existing art are modified to allow virtual address resolution by mapping to multiple overlapping entries, and resolving a physical address from the most specific entry. This enables more efficient use of system resources by allowing smaller frames to shadow larger frames. A page table is selected. When a virtual address in a request corresponds to an entry in the page table, which identifies a next page table associated with the large frame, a determination is made that the virtual address corresponds to an entry in the next page table, the entry in the next page table referencing a small frame overlay for the large frame. The virtual address is mapped to a physical address in the small frame overlay using data of the entry in the next page table. The physical address in a process-specific view of the large frame is returned.

Abstract: An extended translation look-aside buffers (eTLB) for converting virtual addresses into physical addresses is described, which includes both a set associative memory structure (SAM) and a content addressable memory (CAM) structure. An improved approach for operating an eTLB is described in which the same instruction is issued to perform the same task regardless of the exact underlying memory structure within the eTLB being accessed. For flush operations, the same instruction to perform a TLB flush can be provided to the eTLB that operates upon both the CAM and the SAM, which is then handled differently by the underlying implementation system of the eTLB depending upon whether the CAM and/or SAM is being accessed.

Abstract: A method includes implementing, with a memory of a computing device, a memory controller of the memory of the computing device, a storage device coupled to the computing device and/or an external device coupled to the computing device, a scheme for detecting an overlap between a first address range and a second address range. The first address range includes a first starting address and a first ending address, and the second address range includes a second starting address and a second ending address. The method also includes reducing a number of comparators utilized in the address range overlap detection through solely determining whether the first starting address is within the second address range or the second starting address is within the first address range.

Abstract: A request to modify an object in storage that is associated with one or more computing devices may be obtained, the storage organized based on a latch-free B-tree structure. A storage address of the object may be determined, based on accessing a mapping table that includes map indicators mapping logical object identifiers to physical storage addresses. A prepending of a first delta record to a prior object state of the object may be initiated, the first delta record indicating an object modification associated with the obtained request. Installation of a first state change associated with the object modification may be initiated via a first atomic operation on a mapping table entry that indicates the prior object state of the object. For example, the latch-free B-tree structure may include a B-tree like index structure over records as the objects, and logical page identifiers as the logical object identifiers.

Abstract: A computer implemented instruction is executed. One or more translation table entry locations (TLB) are specified by the instruction. Based on a local-clearing (LC) control specified by the instruction being a first value, the processor selectively clears TLBs in a plurality of the CPUs in a configuration of entries corresponding to the determined translation table entry location. Based on the local-clearing (LC) being a second value, the processor selectively clears only the TLBs of the CPU executing the instruction of entries corresponding to the determined translation table entry location. A computer program product, computer system and computer implemented method are provided.

Abstract: Embodiments relate to prefetch address translation in a computer processor. An aspect includes issuing, by prefetch logic, a prefetch request comprising a virtual page address. Another aspect includes, based on the prefetch request missing the TLB and the address translation logic of the processor being busy performing a current translation request, comparing a page of the prefetch request to a page of the current translation request. Yet another aspect includes, based on the page of the prefetch request matching the page of the current translation request, storing the prefetch request in a prefetch buffer.

Abstract: Embodiments relate to managing page table entries in a processing system. A first page table entry (PTE) of a page table for translating virtual addresses to main storage addresses is identified. The page table includes a second page table entry contiguous with the second page table entry. It is determined whether the first PTE may be joined with the second PTE, based on the respective pages of main storage being contiguous. A marker is set in the page table for indicating that the main storage pages identified by the first PTE and second PTEs are contiguous.

Abstract: A first and a second operand are compared. If they are equal, the contents of register R1+1 are stored at the second-operand location, and the specified CPU or CPUs in the configuration are cleared of all TLB table entries of the designated type formed through the use of the replaced entry in storage, and all lower-level TLB table entries formed through the use of the cleared higher-level TLB table entries. A valid DAT table entry is replaced with a new entry, and the Translation Lookaside Buffer (TLB) is purged of any copies of (at least) the single entry on all CPUs in the configuration. If the first and second operands are unequal, the second operand is loaded at the first-operand location. The comparison result is indicated by the condition code. A method, system and a computer program product are provided.

Abstract: Embodiments relate to managing memory page tables in a processing system. A request to access a desired block of memory is received. The request includes an effective address that includes an effective segment identifier (ESID) and a linear address, the linear address including a most significant portion and a byte index. An entry in a buffer that includes the ESID of the effective address is located. Based on the entry including a radix page table pointer (RPTP), performing: using the RPTP to locate a translation table of a hierarchy of translation tables, using the located translation table to translate the most significant portion of the linear address to obtain an address of a block of memory, and based on the obtained address, performing the requested access to the desired block of memory.

Abstract: A computer system includes a translation look-aside (TLB) buffer and a processing unit. The TLB is configured to store an entry that comprises virtual address information, real address information associated with the virtual address information, and additional information corresponding to at least one of the virtual address information and the real address information. The processing unit is configured to control the TLB to modify the additional information while maintaining the entry in a valid state accessible by the processing unit for a translation look-aside operation corresponding to the virtual address information and the real address information.

Abstract: A system for accessing memory locations includes translating, by a processor, a virtual address to locate a first page table entry (PTE) in a page table. The first PTE includes a marker and an address of a page of main storage. It is determined whether a marker is set in the first PTE. The system identifies a large page size of a large page associated with the first PTE based on determining that the marker is set in the first PTE. The large page consists of contiguous pages of main storage. An origin address of the large page is determined based on determining that the marker is set in the first PTE. The virtual address is used to index into the large page at the origin address to access main storage.

Abstract: A system and method for implementation of MMU assisted data breakpoints for any number of data structures within a program application are provided. For each data structure for which a data breakpoint is desired, two distinct MMU entries are created. One MMU entry has access attributes. The other entry has an interrupt triggering sub-entry. According to the preferred embodiment, access to the second MMU entry causes a page fault.

Abstract: Methods and mechanisms for operating a translation lookaside buffer (TLB). A translation lookaside buffer (TLB) includes a plurality of segments, each segment including one or more entries. A control unit is coupled to the TLB. The control unit is configured to determine utilization of segments, and dynamically disable segments in response to determining that segments are under-utilized. The control unit is also configured to dynamically enable segments responsive to determining a given number of segments are over-utilized.

Abstract: An approach is provided in a hypervised computer system where a page table request is at an operating system running in the hypervised computer system. The operating system determines whether the page table request requires the hypervisor to process. If the determination reveals that the page table request requires the hypervisor, then the hypervisor is used to handle the request. However, if the determination reveals that the page table request does not require the hypervisor, then an indicator included in a page table entry corresponding to the request is read to determine if the page table entry is controlled by the operating system or the hypervisor. The operating system is able to update the page table entry if the indicator identifies the page table entry as being operating system controlled.

Abstract: Extended translation look-aside buffers (eTLB) for converting virtual addresses into physical addresses are presented, the eTLB including, a physical memory address storage having a number of physical addresses, a virtual memory address storage configured to store a number of virtual memory addresses corresponding with the physical addresses, the virtual memory address storage including, a set associative memory structure (SAM), and a content addressable memory (CAM) structure; and comparison circuitry for determining whether a requested address is present in the virtual memory address storage, wherein the eTLB is configured to receive an index register for identifying the SAM structure and the CAM structure, and wherein the eTLB is configured to receive an entry register for providing a virtual page number corresponding with the plurality of virtual memory addresses.

Abstract: A system that includes a memory, a tilelet data structure entry, a first tile freelist, and an allocation subsystem. The memory includes a first tilelet on a first tile. The tilelet data structure entry includes a first tilelet preferred pagesize assigned to a first value. The first tile freelist for the first tile includes a first tile in-bounds page freelist, and a first tile out-of-bounds page freelist. The allocation subsystem is configured to detect that a first physical page is freed, store, in the first tile in-bounds page freelist, a first page data structure, detect that a second physical page is freed, store, in the first tile out-of-bounds page freelist, a second page data structure, and coalesce the memory using the second page and at least one of the physical pages associated with the plurality of out-of-bounds page data structures into a third physical page.

Abstract: Embodiments of an invention for using a translation lookaside buffer to manage protected micro-contexts are disclosed. In one embodiment, an apparatus includes an interface and memory management logic. The interface is to perform a transaction to fetch information from a memory. The memory management logic is to translate an untranslated address to a memory address. The memory management logic includes a storage location, a series of translation stages, determination logic, and a translation lookaside buffer. The storage location is to store an address of a data structure for the first translation stage. Each of the translation stages includes translation logic to find an entry in a data structure based on a portion of the untranslated address. Each entry is to store an address of a different data structure for the first translation stage, an address of a data structure for a successive translation stage, or the physical address.

Abstract: According to one embodiment, a memory management method implemented by a computer includes managing each block of a memory region included in the computer based on a buddy allocation algorithm. The method includes managing a correspondence relation between a virtual address and a physical address of one block using one entry of a page table. Each block has a size of a super page. The method includes allocating an empty first block to a process so that the number of empty blocks does not exceed the number of empty entries of a translation look-aside buffer (TLB).

Abstract: One embodiment of the present invention relates to a CPU cache system that stores tag information and cached data in the same SRAM. The system includes an SRAM memory device, a lookup buffer, and a cache controller. The SRAM memory device includes a cache data section and a cache tag section. The cache data section includes data entries and the tag section includes tag entries associated with the data entries. The tag entries include memory addresses that correspond to the data entries. The lookup buffer includes lookup entries associated with at least a portion of the data entries. The number of lookup entries is less than the number of tag entries. The cache controller is configured to perform a speculative read of the cache data section and a cache check of the lookup buffer simultaneously or in a single cycle.

Abstract: Systems and methods for accessing a unified translation lookaside buffer (TLB) are disclosed. A method includes receiving an indicator of a level one translation lookaside buffer (L1TLB) miss corresponding to a request for a virtual address to physical address translation, searching a cache that includes virtual addresses and page sizes that correspond to translation table entries (TTEs) that have been evicted from the L1TLB, where a page size is identified, and searching a second level TLB and identifying a physical address that is contained in the second level TLB. Access is provided to the identified physical address.

Abstract: A shared resource management system and method are described. In one embodiment, a shared resource management system facilitates age based miss replay. In one exemplary implementation, a shared resource management system includes a plurality of engines, and a shared resource a shared resource management unit. The plurality of engines perform processing. The shared resource supports the processing. The shared resource management unit handles multiple outstanding miss requests.

Abstract: Administering registered virtual addresses in a hybrid computing environment that includes a host computer and an accelerator, the accelerator architecture optimized, with respect to the host computer architecture, for speed of execution of a particular class of computing functions, the host computer and the accelerator adapted to one another for data communications by a system level message passing module, where administering registered virtual addresses includes maintaining, by an operating system, a watch list of ranges of currently registered virtual addresses; upon a change in physical to virtual address mappings of a particular range of virtual addresses falling within the ranges included in the watch list, notifying the system level message passing module by the operating system of the change; and updating, by the system level message passing module, a cache of ranges of currently registered virtual addresses to reflect the change in physical to virtual address mappings.

Abstract: Techniques are disclosed relating to integrated circuits that implement a virtual memory. In one embodiment, an integrated circuit is disclosed that includes a translation lookaside buffer configured to store non-prefetched translations and a translation table configured to store prefetched translations. In such an embodiment, the translation lookaside buffer and the translation table share table walk circuitry. In some embodiments, the table walk circuitry is configured to store a translation in the translation table in response to a prefetch request and without updating the translation lookaside buffer. In some embodiments, the translation lookaside buffer, the translation table, and table walk circuitry are included within a memory management unit configured to service memory requests received from a plurality of client circuits via a plurality of direct memory access (DMA) channels.

Abstract: In a computer system having virtual machines, one or more unused bits of a guest virtual address range are allocated for aliasing so that multiple virtually addressed sub-pages can be mapped to a common memory page. When one bit is allocated for aliasing, sub-pages can be virtually addressed at a granularity that is one-half of a memory page. When M bits are allocated for aliasing, sub-pages can be virtually addressed at a granularity that is 1/(2M)-th of a memory page. The granularity of page sizes can be selected according to particular use cases. In the case of COW optimization, page sizes can be set statically between 4 KB and 2 MB or configured dynamically among multiple page sizes.

Abstract: A management information generating method wherein logical and physical block addresses (BAs) of continuous addresses are associated with each other in the BA translation table. When a logical block is constructed, an allowable value is set for the number of defective physical blocks. A logical block having fewer defects than the set number is set usable, and a logical block having more defects than the set number is set unusable. System logical block construction is performed to preferentially select physical blocks from a plane list including a large number of usable blocks to equalize the number of usable blocks in each plane list. It is determined whether the number of free blocks is insufficient on the basis of a first management unit and whether the storage area for the indicated capacity can be reserved on the basis of the management unit different from the first unit.

Abstract: There is provided a method and apparatus for implementing a virtual mirror of a primary storage device (106) on a secondary storage device (108). The method comprises providing a map (124a) for translating primary data storage locations on said primary storage device (106) to secondary data storage locations on said secondary storage device (106) and utilizing said map (124a) to enable data stored on said secondary storage device (108) to mirror data stored on said primary storage device (106). By providing such a method, the requirements of the primary and secondary disks (106, 108) can be decoupled such that a smaller secondary disk (108) could be used with a larger primary (106) which will not be filled to capacity. This reduces the unused capacity on the secondary disk (108) which would otherwise be wasted.

Abstract: Provided are a method, system, and program for utilizing a virtualized data structure table such as an address translation and protection table (TPT), for example, in an I/O device. The virtualized data structure table has virtually contiguous data structures but not necessarily physically contiguous data structures in system memory. The data structure table may be accessed in a virtually contiguous manner. In the illustrated embodiment, the table is subdivided at a first hierarchal level into a plurality of virtually contiguous units or segments. Each unit or segment is in turn subdivided at a second hierarchal level into a plurality of virtually contiguous subunits, subsegments, pages or blocks. Each page or block is in turn subdivided at a third hierarchal level into a plurality of physically contiguous table entries.

Abstract: A method for writing and reading data in memory cells, comprises the steps of: defining a virtual memory, defining write commands and read commands of data (DT) in the virtual memory, providing a first nonvolatile physical memory zone (A1), providing a second nonvolatile physical memory zone (A2), and, in response to a write command of an initial data, searching for a first erased location in the first memory zone, writing the initial data (DT1a) in the first location (PB1(DPP0)), and writing, in the metadata (DSC0) an information (DS(PB1)) allowing the first location to be found and an information (LPA, DS(PB1)) forming a link between the first location and the location of the data in the virtual memory.

Abstract: An illustrative embodiment of a computer-implemented process for managing a staging area creates the staging area for identified candidate cold objects, moves the identified candidate objects into the staging area, tracks application access to memory comprising the staging area and determines whether frequency of use information for a specific object exceeds a predetermined threshold. Responsive to a determination that the frequency of use information for the specific object exceeds a predetermined threshold, move the specific object into a regular area and determine whether a current time exceeds a predetermined threshold. Responsive to a determination that the current time exceeds a predetermined threshold, the computer-implemented process moves remaining objects from the staging area to a cold area.

Abstract: Systems and methods for performing non-allocating memory access instructions with physical address. A system includes a processor, one or more levels of caches, a memory, a translation look-aside buffer (TLB), and a memory access instruction specifying a memory access by the processor and an associated physical address. Execution logic is configured to bypass the TLB for the memory access instruction and perform the memory access with the physical address, while avoiding allocation of one or more intermediate levels of caches where a miss may be encountered.

Abstract: A translation lookaside buffer (TLB) configured for use in a multiple operating system environment includes a plurality of storage locations, each storage location being configured to store a page translation entry configured to relate a virtual address range to a physical address range, each page translation entry having an address space identifier (ASID) associated with an operating system. The TLB also includes flush logic configured to receive a TLB flush request from an operating system having an operating system ASID and flush only TLB page translation entries having a stored ASID that matches the operating system ASID.

Abstract: An application processor includes a system memory unit, peripheral devices, a control unit and a central processing unit (CPU). The system memory unit includes one page table. The peripheral devices share the page table and perform a DMA (Direct Memory Access) operation on the system memory unit using the page table, where each of the peripheral devices includes a memory management unit having a translation lookaside buffer. The control unit divides a total virtual address space corresponding to the page table into sub virtual address spaces, assigns the sub virtual address spaces to the peripheral devices, respectively, allocates and releases a DMA buffer in the system memory unit, and updates the page table, where at least two of the sub virtual address spaces have different sizes from each other. The CPU controls the peripheral devices and the control unit. The application processor reduces memory consumption.

Abstract: Embodiments of the invention include systems and methods for auto-tiering multiple file systems across a common resource pool. Storage resources are allocated as a sub-LUN auto-tiering (SLAT) sub-pool. The sub-pool is managed as a single virtual address space (VAS) with a virtual block address (VBA) for each logical block address of each data block in the sub-pool, and a portion of those VBAs can be allocated to each of a number of file systems. Mappings are maintained between each logical block address in which file system data is physically stored and a VBA in the file system's portion of the virtual address space. As data moves (e.g., is added, auto-tiered, etc.), the mappings can be updated. In this way, multiple SLAT file systems can exploit the full resources of the common SLAT sub-pool and maximize the resource options available to auto-tiering functions.

Abstract: A content addressable memory system, method and computer program product is described. The memory system comprises a location addressable store having data identified by location and multiple levels of content addressable stores each holding ternary content words. The content words are associated with references to data in the location addressable store. The content store levels might be implemented using different technologies that have different performance, capacity, and cost attributes. The memory system includes a content based cache for improved performance and a content addressable memory management unit for managing memory access operations and virtual memory addressing.

Abstract: A computer implemented method optimizes memory page sizes during runtime. A process is identified from a policy file. The policy file contains at least one policy based threshold. A resource usage profiler monitors the process during runtime. The resource usage profiler determines whether the process exceeds the set of stated desired policies from the at least one policy based threshold. If the process exceeds the set of stated desired policies from the set of policy based thresholds, a performance projection for the process is executed to determine whether the process would experience a performance benefit from a different page size. Responsive to determining that the process would experience the performance benefit from the different page size, the page size for the process is changed.

Abstract: A method, apparatus and computer readable media for servicing page fault exceptions in a accelerated processing device (APD). A page fault related to a wavefront is detected. A fault handling request to a translation mechanism is sent when the page fault is detected. A fault handling response corresponding to the detected page fault from the translation mechanism is received. Confirmation that the detected page fault has been handled through performing page mapping based on the fault handling response is received.

Abstract: A system, method and computer program product for providing quiesce filtering for shared memory. The method includes receiving a shared-memory quiesce request at a processor. The request includes a donor zone. The processor includes translation look aside buffer one (TLB1). It is determined that the shared-memory request can be filtered by the processor if there not any shared memory entries in the TLB1 and the donor zone is not equal to a current zone of the processor and the processor is not running in host mode. The shared-memory quiesce request is filtered in response to the determining.

Abstract: Memory management methods and computing apparatus with memory management capabilities are disclosed. One exemplary method includes mapping an address from an address space of a physically-mapped device to a first address of a common address space so as to create a first common mapping instance, and encapsulating an existing processor mapping that maps an address from an address space of a processor to a second address of the common address space to create a second common mapping instance. In addition, a third common mapping instance between an address from an address space of a memory-management-unit (MMU) device and a third address of the common address space is created, wherein the first, second, and third addresses of the common address space may be the same address or different addresses, and the first, second, and third common mapping instances may be manipulated using the same function calls.

Abstract: An apparatus and method for managing a translation look-aside buffer (TLB). The TLB is shared by a plurality of jobs. The method including the steps of: obtaining at least one attribute of each job of the plurality of jobs; assigning a priority level to each job according to at least one attribute of each job; and managing the related TLB entries of each job according to the priority level of each job. The present invention also provides an apparatus for managing TLB corresponding to the above method. The method and apparatus according to the present invention provide an efficient use of the shared TLB.

Abstract: A chip multi-processor (CMP) with virtual domain management. The CMP has a plurality of tiles each including a core and a cache, a mapping storage, a plurality of memory controllers, a communication bus interconnecting the tiles and the memory controllers, and machine-executable instructions. The tiles and memory controllers are responsive to the instructions to group the tiles into a plurality of virtual domains, each virtual domain associated with at least one memory controller, and to store a mapping unique to each virtual domain in the mapping storage.

Abstract: The apparatus includes a virtual address space generation unit generating a virtual address space of a guest operating system, the guest operating system being executed in the virtual address space, and a virtual address space of a virtual machine monitor, the virtual machine monitor being executed in the virtual address space; a gateway page generation unit generating a gateway page allocated to a predetermined region of an actual memory region and mapped to the virtual address space of the guest operating system and the virtual address space of the guest machine monitor; and a memory management unit executing the gateway page to map a kernel region of the guest operating system to the predetermined region of the virtual address space of the virtual machine monitor to perform translation between the virtual address space of the guest operating system and the virtual address space of the virtual machine monitor.

Abstract: An address translation apparatus includes: a first address translation unit to hold, first address information, validity information and correspondence information, the first address information indicating correspondence between a virtual address and a physical address, the validity information indicating a validity of the first address information, and the correspondence information indicating correspondence between first context information corresponding to the first address information and second context information in an access request; an information holding unit to hold context information in the first address information; a comparison unit to compare the first context information with the second context information and update the correspondence information based on a comparison result; and a control unit to search a new first entry having the first address information including the same virtual address as in the access request based on the validity information and the correspondence information and out

Abstract: Example methods, apparatus, and articles of manufacture to access memory are disclosed. A disclosed example method involves receiving at least one runtime characteristic associated with accesses to contents of a memory page and dynamically adjusting a memory fetch width for accessing the memory page based on the at least one runtime characteristic.

Abstract: A microprocessor includes a translation lookaside buffer, a request to load a page table entry into the microprocessor generated in response to a miss of a virtual address in the translation lookaside buffer, and a prefetch unit. The prefetch unit receives a physical address of a first cache line that includes the requested page table entry and responsively generates a request to prefetch into the microprocessor a second cache line that is the next physically sequential cache line to the first cache line.

Abstract: A memory system comprises a storage device and a host. The host classifies pages stored in the storage device into a plurality of data groups according to properties of the pages, and transmits setup information regarding the classified data groups to the storage device.

Abstract: A system and method for efficiently handling translation look-aside buffer (TLB) misses. A memory management unit (MMU) detects when a given virtual address misses in each available translation-lookaside-buffer (TLB). The MMU determines whether a memory access operation associated with the given virtual address is the oldest, uncompleted memory access operation in a scheduler. If this is the case, a demand table walk (TW) request may be stored in an available entry in a TW queue. During this time, the utilization of the memory subsystem resources may be low. While a demand TW request is stored in the TW queue, subsequent speculative TW requests may be stored in the TW queue. When the TW queue does not store a demand TW request, no more entries of the TW queue may be allocated to store TW requests.

Abstract: A system and method to manage a translation lookaside buffer (TLB) is disclosed. In a particular embodiment, a method of managing a first TLB includes in response to starting execution of a memory instruction, setting a first field associated with an entry of the first TLB to indicate use of the entry. The method also includes setting a second field to indicate that the entry in the first TLB matches a corresponding entry in a second TLB.

Abstract: A system and method for accessing memory are provided. The system comprises a lookup buffer for storing one or more page table entries, wherein each of the one or more page table entries comprises at least a virtual page number and a physical page number; a logic circuit for receiving a virtual address from said processor, said logic circuit for matching the virtual address to the virtual page number in one of the page table entries to select the physical page number in the same page table entry, said page table entry having one or more bits set to exclude a memory range from a page.

Abstract: An address translation buffer (TLB) which holds pairs of virtual addresses and physical addresses by respective page sizes and performs an address translation, a storage unit which holds a pair of a virtual address removed from the TLB and page size corresponding thereto when a pair of a new virtual address and physical address read from a page table is registered to the TLB, base registers which hold a base address by each page size are held. The TLB is searched based on a translation object virtual address included in a memory access request, and when a TLB miss occurs, a main storage is searched based on a pointer address generated from information held by the storage unit and the base register, and the translation object virtual address is translated into the physical address.