WO2011045931A1 - 情報処理装置 - Google Patents
情報処理装置 Download PDFInfo
- Publication number
- WO2011045931A1 WO2011045931A1 PCT/JP2010/006098 JP2010006098W WO2011045931A1 WO 2011045931 A1 WO2011045931 A1 WO 2011045931A1 JP 2010006098 W JP2010006098 W JP 2010006098W WO 2011045931 A1 WO2011045931 A1 WO 2011045931A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- memory
- information processing
- processing apparatus
- measuring
- cpu
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F12/00—Accessing, addressing or allocating within memory systems or architectures
- G06F12/02—Addressing or allocation; Relocation
- G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3037—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a memory, e.g. virtual memory, cache
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
- G06F2212/25—Using a specific main memory architecture
- G06F2212/251—Local memory within processor subsystem
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
- G06F2212/25—Using a specific main memory architecture
- G06F2212/253—Centralized memory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
- G06F2212/60—Details of cache memory
- G06F2212/601—Reconfiguration of cache memory
- G06F2212/6012—Reconfiguration of cache memory of operating mode, e.g. cache mode or local memory mode
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Definitions
- the present invention relates to a memory and bus control technology used for improving performance in an information processing apparatus.
- a cache memory of an information processing apparatus has means for allocating an arbitrary area as a local memory, and a built-in memory that can have a memory configuration optimum for processing contents is known.
- the built-in memory is faster and consumes less power than an externally connected memory, the performance of the information processing apparatus is improved by changing the memory configuration so that the built-in memory can be used effectively.
- Patent Document 2 As a means to maintain this optimal memory configuration, the processing contents are analyzed during compilation, the processing contents are divided into several phases, the optimal memory configuration is derived in each phase, and the code for switching the memory configuration is generated The technique of doing is known.
- Patent Document 2 since the method described in Patent Document 2 is an analysis of static memory access at the time of compilation, it can be applied to a system in which a generated program occupies a CPU and operates alone. There has been a problem that an optimum memory configuration cannot be obtained by detecting a change in memory utilization efficiency due to a change in device operation, power saving state, operation of other software, and the like.
- the processing speed can be improved by using most of the built-in memory as cache memory, but other bus master devices are frequently used as memory.
- the process to access is performed and the process is a process with higher priority than the process performed by the CPU. It is more efficient to release the cache memory, allocate it as local memory, and let the bus master device use it.
- the present invention has been made in view of such conventional problems, and dynamically detects a decrease in memory utilization efficiency that cannot be predicted in advance, and reconfigures the allocation size of the cache memory and the local memory.
- an information processing apparatus capable of improving performance is provided.
- the information processing apparatus includes a system-on-chip built-in memory (first storage unit), an external memory (second storage unit) connected to the system-on-chip, the first storage unit, and the first storage unit.
- An area of the first storage means or the second storage means used by the CPU and other bus master devices (information processing means), or moving means for moving the data stored in the means or the second storage means; Second changing means for changing is provided.
- This configuration dynamically changes the internal memory configuration based on changes in memory usage efficiency, and allows the CPU and other bus master devices to use the internal memory with priority over the external memory.
- the tightness of the bus, power consumption can be improved.
- the measuring unit with this configuration may be a thing that measures the hit rate of the cache memory and the number of accesses. With this configuration, it is possible to improve a decrease in CPU processing speed caused by a shortage of cache memory. Further, when the cache memory is excessive, the cache memory can be released to other bus master devices as a local memory.
- the measuring unit with this configuration may be a device that measures the load factor and operating frequency of the CPU.
- the measuring unit having this configuration may be a VRAM size and an object for measuring the frequency of screen update.
- the VRAM usage status of the graphic controller is detected, and when the display performance decreases due to an increase in VRAM access, the size of the local memory is increased and used as a VRAM. it can. Further, when the VRAM size is reduced while the local memory is used as the VRAM, the local memory can be released to another bus master device or released to the CPU as a cache.
- the measuring unit with this configuration may be a type that measures the type of process being executed and the state change in the process.
- this configuration it is possible to improve the CPU processing speed by predicting the memory utilization efficiency from the change of the software executed by the CPU and having an optimal memory configuration.
- the cache memory can be opened to other bus master devices as local memory.
- the measuring unit with this configuration may be an object that measures the occupancy rate of the bus band.
- the bus master device including the CPU
- the bus master device whose processing speed is low due to insufficient bus bandwidth can use the built-in memory to improve the processing speed.
- the measuring unit having this configuration may be a device that measures the working set of the CPU.
- the processing speed of the CPU can be improved by allocating more cache memory to the CPU as the working set increases. Further, when the working set decreases, the cache memory can be released to other bus master devices as a local memory.
- the measuring unit with this configuration may be an object that measures interrupt events.
- an optimal memory configuration can be maintained in response to changes in memory usage conditions triggered by events.
- a memory use efficiency change that cannot be predicted by static analysis is detected, the configuration of the first storage means is dynamically changed, and the memory area used by the information processing means is moved. Can rearrange and maintain optimal memory configuration. As a result, it is possible to improve performance such as an improvement in processing speed of the information processing means, a reduction in bus bandwidth load, and a reduction in power consumption.
- FIG. 1 is a schematic configuration diagram of an information processing apparatus according to Embodiment 1 of the present invention.
- Schematic configuration diagram of an information processing apparatus in Embodiment 2 of the present invention The flowchart figure explaining operation
- Schematic configuration diagram of an information processing apparatus in Embodiment 3 of the present invention The flowchart figure explaining operation
- Schematic configuration diagram of an information processing apparatus in Embodiment 4 of the present invention The flowchart figure explaining operation
- Schematic configuration diagram of an information processing apparatus in Embodiment 5 of the present invention The flowchart figure explaining operation
- Schematic configuration diagram of an information processing apparatus in Embodiment 6 of the present invention The flowchart figure explaining operation
- FIG. 1 is a schematic configuration diagram of an information processing apparatus according to Embodiment 1 of the present invention
- FIG. 2 is a flowchart for explaining the operation of the information processing apparatus.
- the information processing apparatus includes a CPU 101 (information processing means), a built-in memory 102 (first storage means), and an external memory 103 (second storage means). And another bus master device 104 (information processing means) that is not a CPU, a cache measurement unit 106 that measures the state of the built-in memory 102, and a control unit 105 (first changing means, moving means, second changing means) And an internal bus 108 that connects the CPU 101 and the internal memory 102, and an external bus 107 that connects the internal memory 102, the bus master device 104, and the external memory 103.
- the control unit 105 monitors the measurement value of the cache measurement unit 106, and when the memory use efficiency can be expected to be improved, the control unit 105 changes the memory configuration of the built-in memory 102, and the CPU 101 changes the memory configuration. And a moving means for moving data in the internal memory 102 and the external memory 103 used by the bus master device 104, and a second changing means for causing the CPU 101 and the bus master device 104 to use the destination area. .
- control unit 105 performs initial allocation between the cache memory and the local memory of the internal memory 102 (S101), and then measures the cache hit rate at regular intervals (S102). (S103).
- the cache memory Process determines that there is a possibility that the size of the cache memory is excessive when the cache hit rate is a numerical value greater than a certain value (for example, 90% or more) (S104).
- the cache memory Process to reduce.
- the current cache memory allocation size is determined (S105). If the cache memory size can be reduced (for example, the cache allocation size of the internal memory 102 is 5% or less), the configuration of the internal memory 102 is changed. Then, the cache memory is decreased (for example, 5%) and the local memory is increased (S106).
- the data in the area of the external memory 103 currently used by the bus master device 104 is moved to the local memory, the setting of the bus master device 104 is changed, and the local Use the area to which the data of the memory is moved.
- the control unit 105 determines that the cache memory size may be insufficient (S104).
- the current cache memory allocation size is determined (S108), and when the cache memory size can be increased (for example, the cache allocation size of the built-in memory 102 is 95% or less), the local memory is allocated to the cache memory.
- the data in the local memory area used by the bus master device 104 is moved to the external memory 103, the setting of the bus master device 104 is changed, and the data transfer destination area in the external memory 103 is used (S109).
- the configuration of the internal memory 102 is changed to increase the cache size (for example, 5%) and decrease the local memory (S110).
- the configuration of the built-in memory 102 can be changed from the cache utilization efficiency. Therefore, when the cache memory size is insufficient, the cache memory is increased and the local memory is decreased. The processing speed of the CPU 101 can be improved. Further, when the cache memory is excessively allocated, the cache memory size can be decreased, the local memory can be increased, and the bus master device 104 can be released.
- FIG. 3 is a schematic configuration diagram of the information processing apparatus according to Embodiment 2 of the present invention
- FIG. 4 is a flowchart for explaining the operation of the information processing apparatus
- FIG. 5 is a table showing configurations of the cache memory and the local memory.
- the constituent elements 201 to 205, 207, and 208 are equivalent to the constituent elements 101 to 105, 107, and 108 of the first embodiment.
- a CPU measurement unit 206 that measures the state of the CPU 201 is provided.
- control unit 205 performs initial allocation of the cache memory and the local memory of the internal memory 202 (S201), and thereafter the CPU frequency and the load factor every certain time (S202). Is measured (S203).
- the control unit 205 determines the allocation size of the cache memory and the local memory using the table shown in FIG. 5 based on the frequency and load factor of the CPU 201 (S204).
- the frequency and load factor of the CPU 201 are lower and the operation rate is lower, the frequency of memory access of the CPU 201 is reduced and the cache memory is unnecessary, and conversely, the frequency and load factor are higher and the operation rate is higher.
- the table is configured such that the cache memory size is large when the frequency and load factor of the CPU 201 are high, and the cache memory size is small when the frequency and load factor of the CPU 201 are low.
- control unit 205 compares the current allocation size of the cache memory with the currently determined size (S205). If the cache memory size decreases, the configuration of the internal memory 202 is changed to reduce the cache memory to reduce the local memory. Is increased (S206). In order to use the newly allocated local memory for another bus master device, the data in the area of the external memory 203 currently used by the bus master device 204 is moved to the local memory, the setting of the bus master device 204 is changed, and the data in the local memory is changed. The destination area is used (S207). When the size of the cache memory increases, the local memory is allocated to the cache.
- the data in the local memory area currently used by the bus master device 204 is moved to the external memory 203, the setting of the bus master device 204 is changed, and the external memory
- the data transfer destination area 203 is used (S208).
- the configuration of the internal memory 202 is changed to increase the cache size and decrease the local memory (S209).
- the configuration of the internal memory 202 can be changed based on the operation status of the CPU 201. Therefore, the operating rate of the CPU 201 that is in a sleep state in a low power consumption state is low, and the cache memory is When it is not necessary, the cache memory can be reduced, the local memory can be increased, and the bus master device 204 can be released. Further, when the operating rate of the CPU 201 increases and the CPU 201 requires a cache memory, the local memory can be reduced, the cache memory can be increased, and the processing speed of the CPU 201 can be improved.
- FIG. 3 is a schematic configuration diagram of the information processing apparatus according to the third embodiment of the present invention
- FIG. 7 is a flowchart for explaining the operation of the information processing apparatus
- FIG. 8 is a table showing configurations of the cache memory and the local memory.
- the components 301 to 303, 305, 307, and 308 are equivalent to the components 101 to 103, 105, 107, and 108 of the first embodiment.
- a graphic controller 304 information processing means
- a VRAM measuring unit 306 that measures the state of the VRAM in the built-in memory 302 and the external memory 303.
- control unit 305 performs initial allocation between the cache memory and the local memory in the internal memory 302 (S301), and then updates and updates the size of the VRAM at regular intervals (S302). The number of times is measured (S303).
- the control unit 305 determines the allocation size of the cache memory and the local memory using the table shown in FIG. 5 based on the size of the VRAM and the number of updates (S304).
- the memory access frequency of the graphic controller 304 decreases as the VRAM size is small and the number of updates is small, and conversely increases as the VRAM size is large and the number of updates is large.
- the local memory allocation of the VRAM may be increased. Therefore, the table is configured such that the size of the VRAM is large, the size of the local memory is large when the number of updates is large, the size of the VRAM is large, and the size of the local memory is small when the number of updates is large.
- the subsequent control is equivalent to the control (S205 to S209) of the second embodiment, except that the bus master device 204 of the second embodiment is replaced with the graphic controller 304.
- the configuration of the internal memory 302 can be changed from the state of the VRAM, when the screen size is large and the number of drawing updates is large, the cache memory is reduced and the local memory is reduced.
- the access performance of the VRAM can be improved, and the drawing performance of the graphic controller 304 can be kept above a certain level.
- the drawing performance of the graphic controller 304 may be low, the local memory allocated as VRAM is reduced, the cache memory is increased, and the processing speed of the CPU 301 is improved. be able to.
- FIG. 9 is a schematic configuration diagram of the information processing apparatus according to Embodiment 4 of the present invention
- FIG. 10 is a flowchart for explaining the operation of the information processing apparatus
- FIG. 11 is a table showing the configurations of the cache memory and the local memory.
- the components 401 to 405, 407, and 408 are equivalent to the components 101 to 105, 107, and 108 of the first embodiment.
- a process measurement unit 406 that measures the process state from the CPU 401, the built-in memory 402, and the external memory 403 is provided.
- the process measurement unit 406 is configured to measure the state of the process by measuring the CPU 401, the built-in memory 402, and the external memory 403 by the OS executed by the CPU 401.
- control unit 405 performs initial allocation of the cache memory and the local memory of the built-in memory 402 (S401), and monitors changes in the process state (S402). When a change in the process state is detected, the execution process and the state of the process are measured (S403).
- the control unit 405 determines the allocation size of the cache memory and the local memory using the table shown in the figure based on the execution process and the state of the process (404).
- the memory access of the CPU 401 varies depending on the process being executed, and also varies depending on the state of the same process.
- the process that performs the polling process occupies the CPU 401 for a long time, but does not use the memory, so the cache memory may be small. Therefore, the table is configured so that the cache memory size is large in the execution and process state of a process with a large memory access, and the cache memory size is small in the execution and process state of a process with a small memory access. Yes.
- the subsequent control (S405 to S409) is equivalent to the control (S205 to S209) of the second embodiment.
- the configuration of the built-in memory 402 can be changed from the state of the process. Therefore, the cache memory is increased and the local memory is decreased during the execution of a process with many memory accesses. The memory access performance can be improved. In addition, when the process is executed with less memory access, the cache memory size can be reduced, the local memory can be increased, and the bus master device 404 can be released.
- FIG. 12 is a schematic configuration diagram of an information processing apparatus according to Embodiment 5 of the present invention
- FIG. 13 is a flowchart for explaining the operation of the information processing apparatus.
- the components 501 to 505, 507, and 508 are equivalent to the components 101 to 105, 107, and 108 of the first embodiment.
- a bus measurement unit 506 that measures the external bus 507 and the internal bus 508 is provided.
- the bus measurement unit 506 is configured to monitor the signal lines of the external bus 507 and the internal bus 508, and count the bus masters (including the CPU 501) that occupy the bus at regular intervals to measure the bus occupancy rate. Is done.
- control unit 505 performs initial allocation between the cache memory and the local memory in the built-in memory 502 (S501), and then the bus occupancy of each bus master at regular time intervals (S502). The rate is measured (S503).
- the control unit 505 distributes the memory area used by each bus master to the internal memory 502 and the external memory 503 based on the bus occupancy rate of each bus master. At this time, since the built-in memory 502 has higher performance, it is preferentially assigned to the built-in memory 502 to a bus master having a high bus occupancy rate and frequently accessing the memory.
- the bus master using the internal memory 502 is the CPU 501
- the use area is allocated to the cache, and when it is another bus master device 504, it is allocated as a local memory. If the size or bandwidth of the internal memory 502 is insufficient, an area of the external memory 503 is allocated (S504).
- the subsequent control (S505 to S509) is equivalent to the control (S205 to S209) of the second embodiment.
- the configuration of the internal memory 502 is changed based on the bus usage status, the bus occupation is distributed to the internal bus 508 and the external bus 507, and the bus master does not have sufficient bus bandwidth. It is possible to improve the reduction in processing speed when performance cannot be achieved and to preferentially use the high-performance built-in memory 502.
- bus measuring unit 506. there may be a plurality of bus master devices measured by the bus measuring unit 506. In this way, it is possible to improve the processing speed reduction when the buses used by a plurality of bus master devices such as DSP and DMA compete and performance cannot be achieved.
- FIG. 14 is a schematic configuration diagram of the information processing apparatus according to Embodiment 6 of the present invention
- FIG. 15 is a flowchart for explaining the operation of the information processing apparatus.
- the components 601 to 605, 607, and 608 are equivalent to the components 101 to 105, 107, and 108 of the first embodiment.
- a working set measuring unit 606 that measures the working set of the CPU 601 from the internal bus 608 is provided.
- the working set measuring unit 606 is configured to count the address signals of the internal bus 608 every predetermined time and measure the working set during that time.
- control unit 605 performs initial allocation of the cache memory and the local memory of the built-in memory 602 (S601), and then measures the working set at regular time intervals (S602). (S603).
- the control unit 605 compares the working set of the CPU 601 with the size of the built-in memory 602 (S604). If the working set is equal to or smaller than the size of the built-in memory 602, the cache memory size is the same as the working set, and the rest is stored in the local memory. If the working set exceeds the size of the built-in memory 602, the size of the cache memory is determined to be 0 for all the built-in memories 602 and the local memory.
- the subsequent control (S607 to S611) is equivalent to the control (S205 to S209) of the second embodiment.
- the configuration of the built-in memory 602 can be changed from the working set, so that only the necessary cache memory can be allocated to the CPU 601 and the processing speed of the CPU 601 can be improved. . Further, by assigning an unnecessary cache to the local memory, it can be released to another bus master device 604.
- FIG. 16 is a schematic configuration diagram of an information processing apparatus according to Embodiment 7 of the present invention
- FIG. 17 is a flowchart for explaining the operation of the information processing apparatus
- FIG. 18 is a table showing the configuration of state transition, cache memory, and local memory. is there.
- the components 701 to 705, 707, and 708 are equivalent to the components 101 to 105, 107, and 108 of the first embodiment.
- Various peripherals 710, an interrupt controller 709 that generates an interrupt from the signals of the various peripherals 710 to the CPU 701, and an interrupt measurement unit 706 that measures an interrupt generated by the interrupt controller 710 are provided.
- control unit 705 performs initial allocation between the cache memory and the local memory in the built-in memory 702 (S701), and then the state of various peripherals 710 changes, and the interrupt controller 709 is changed.
- the CPU 701 generates an interrupt (S702)
- the interrupt type is measured (S703).
- the control unit 705 performs state transition using a table from the current state and the interrupt type (S704), and determines the allocation size of the cache memory and the local memory (S705).
- the table is configured to have an optimal memory configuration from the current state and the interrupt that has occurred. Examples are given below.
- a key interrupt occurs when the information processing apparatus is in a power saving state, a large number of caches are allocated to the CPU 701 for the recovery process.
- a key interrupt occurs while the information processing apparatus is in a moving image playback state, a certain size of the local memory used by the bus master device 704 such as a graphic controller or DSP is secured to ensure the memory access performance.
- the subsequent control (S706 to S710) is equivalent to the control (S205 to S209) of the second embodiment.
- the state of the information processing apparatus can be determined from the interrupt type, and the internal memory configuration suitable for the state can be maintained. Access performance can be secured.
- the information processing apparatus includes a first storage unit that can be used by switching an arbitrary area as a local memory or a cache memory, a second storage unit that is different from the first storage unit, and the first storage unit. And a measuring unit for detecting a change in usage status of the second storage unit, a first changing unit for changing the configuration of the first storage unit based on the measurement result of the measuring unit, A moving means for moving data stored in the first storage means or the second storage means, and a second for changing the area of the first storage means or the second storage means used by the information processing means The changing means is provided.
- control unit monitors the measurement result of the measurement unit and has the first changing unit, the moving unit, and the second changing unit, it can be changed from the state of the information processing apparatus to the optimum memory configuration.
- processing performance can be ensured, such as information on mobile phones and PCs that perform real-time decoding of video and audio data and processing such as hibernation Useful for processing equipment.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Quality & Reliability (AREA)
- Mathematical Physics (AREA)
- Computer Hardware Design (AREA)
- Memory System Of A Hierarchy Structure (AREA)
Abstract
Description
図1は本発明の実施の形態1における情報処理装置の概略構成図、図2は同情報処理装置の動作を説明するフローチャートである。
図3は本発明の実施の形態2における情報処理装置の概略構成図、図4は同情報処理装置の動作を説明するフローチャート、図5はキャッシュメモリとローカルメモリの構成を示すテーブルである。
図6は本発明の実施の形態3における情報処理装置の概略構成図、図7は同情報処理装置の動作を説明するフローチャート、図8はキャッシュメモリとローカルメモリの構成を示すテーブルである。
図9は本発明の実施の形態4における情報処理装置の概略構成図、図10は同情報処理装置の動作を説明するフローチャート、図11はキャッシュメモリとローカルメモリの構成を示すテーブルである。
図12は本発明の実施の形態5における情報処理装置の概略構成図、図13は同情報処理装置の動作を説明するフローチャートである。
図14は本発明の実施の形態6における情報処理装置の概略構成図、図15は同情報処理装置の動作を説明するフローチャートである。
図16は本発明の実施の形態7における情報処理装置の概略構成図、図17は同情報処理装置の動作を説明するフローチャート、図18は状態遷移とキャッシュメモリとローカルメモリの構成を示すテーブルである。
102、202、302、402、502、602、702 内蔵メモリ
103、203、303、403、503、603、703 外部メモリ
104、204、404、504、604、704 バスマスタデバイス
304 グラフィックコントローラ
105、205、305、405、505、605、705 制御部
106 キャッシュ計測部
107、207、307、407、507、607、707 外部バス
108、208、308、408、508、608,708 内部バス
206 CPU計測部
306 VRAM計測部
406 プロセス計測部
506 バス計測部
606 ワーキングセット計測部
706 割り込み計測部
709 割り込みコントローラ
710 各種ペリフェラル
Claims (16)
- 任意の領域をローカルメモリ又はキャッシュメモリとして切り替えて使用可能な第一の記憶手段と、
前記第一の記憶手段と異なる第二の記憶手段と、
前記第一の記憶手段と前記第二の記憶手段の利用状況の変化を検出するための計測部と、
前記計測部の計測結果を元に前記第一の記憶手段の構成を変更する第一の変更手段と、
前記第一の記憶手段又は前記第二の記憶手段に格納されたデータを移動する移動手段と、
情報処理手段が使用する前記第一の記憶手段又は前記第二の記憶手段の領域を変更する第二の変更手段と、を有する情報処理装置。 - 前記移動手段は、前記第一の変更手段により前記第一の記憶手段の構成を変更した場合に、前記情報処理手段が使用している領域のデータを移動させ、
前記第二の変更手段は、前記情報処理手段に移動先のデータ領域を使用させることを特徴とする請求項1記載の情報処理装置。 - 前記計測部が、CPUのメモリ利用効率を計測する手段を有する請求項1記載の情報処理装置。
- 前記CPUのメモリ利用効率を計測する手段が、キャッシュの利用効率を計測する手段であることを特徴とする請求項3記載の情報処理装置。
- 前記キャッシュの利用効率を計測する手段は、キャッシュヒット率、またはキャッシュアクセス数を計測することを特徴とする請求項4記載の情報処理装置。
- 前記CPUのメモリ利用効率を計測する手段が、CPUの状態を計測する手段であることを特徴とする請求項3記載の情報処理装置。
- 前記CPUの状態を計測する手段は、CPUの負荷率、または周波数を計測することを特徴とする請求項6記載の情報処理装置。
- 前記CPUのメモリ利用効率を計測する手段が、ソフトウェアの状態を計測する手段であることを特徴とする請求項3記載の情報処理装置。
- 前記ソフトウェアの状態を計測する手段は、実行されているプロセス、または実行されているプロセス内部の状態を計測することを特徴とする請求項8記載の情報処理装置。
- 前記CPUのメモリ利用効率を計測する手段が、ワーキングセットサイズを計測する手段であることを特徴とする請求項3記載の情報処理装置。
- 前記計測部が、バスの利用効率を計測する手段を有する請求項1記載の情報処理装置。
- 前記バスの利用効率を計測する手段が、表示状態を計測する手段であることを特徴とする請求項11記載の情報処理装置。
- 前記表示状態を計測する手段は、VRAMサイズ、または表示更新頻度を計測することを特徴とする請求項12記載の情報処理装置。
- 前記バスの利用効率を計測する手段が、バス占有率を計測する手段であることを特徴とする請求項11記載の情報処理装置。
- 前記計測部が、各種イベントを計測する手段を有する請求項1記載の情報処理装置。
- 前記各種イベントを計測する手段が、割込みを計測する手段であることを特徴とする請求項15記載の情報処理装置。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2011536042A JPWO2011045931A1 (ja) | 2009-10-14 | 2010-10-13 | 情報処理装置 |
CN2010800465615A CN102713867A (zh) | 2009-10-14 | 2010-10-13 | 信息处理设备 |
US13/500,494 US20120198159A1 (en) | 2009-10-14 | 2010-10-13 | Information processing device |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2009-236941 | 2009-10-14 | ||
JP2009236941 | 2009-10-14 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2011045931A1 true WO2011045931A1 (ja) | 2011-04-21 |
Family
ID=43875986
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2010/006098 WO2011045931A1 (ja) | 2009-10-14 | 2010-10-13 | 情報処理装置 |
Country Status (4)
Country | Link |
---|---|
US (1) | US20120198159A1 (ja) |
JP (1) | JPWO2011045931A1 (ja) |
CN (1) | CN102713867A (ja) |
WO (1) | WO2011045931A1 (ja) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2013058049A (ja) * | 2011-09-07 | 2013-03-28 | Nec Corp | I/oデバイス共有システム、i/oデバイス共有方法、及びプログラム |
JP2015064863A (ja) * | 2013-08-26 | 2015-04-09 | 富士ゼロックス株式会社 | 情報処理装置、演算処理装置及びプログラム |
WO2015132877A1 (ja) * | 2014-03-04 | 2015-09-11 | 株式会社日立製作所 | 計算機およびメモリ制御方法 |
JP2016181030A (ja) * | 2015-03-23 | 2016-10-13 | 富士通株式会社 | 情報処理装置、記憶装置制御方法、記憶装置制御プログラム及び情報処理システム |
JP2017527884A (ja) * | 2014-07-17 | 2017-09-21 | クアルコム,インコーポレイテッド | セットおよびウェイによるコンポーネントキャッシュへの柔軟なキャッシュパーティショニングのための方法および装置 |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102819313B (zh) * | 2012-07-17 | 2015-05-06 | 腾讯科技(深圳)有限公司 | 终端设备的运行方法和终端设备 |
US9455913B2 (en) * | 2013-02-15 | 2016-09-27 | Broadcom Corporation | Management of traffic buffering in internal and external memories in a passive optical network |
US20150026406A1 (en) * | 2013-07-19 | 2015-01-22 | Advanced Micro Devices, Inc. | Size adjusting caches by way |
US9436608B1 (en) | 2015-02-12 | 2016-09-06 | International Business Machines Corporation | Memory nest efficiency with cache demand generation |
US10552327B2 (en) | 2016-08-23 | 2020-02-04 | Apple Inc. | Automatic cache partitioning |
CN110582003B (zh) * | 2019-01-12 | 2020-10-27 | 安徽省安泰科技股份有限公司 | 兼容性大数据采集终端 |
US20230094030A1 (en) * | 2021-09-30 | 2023-03-30 | Advanced Micro Devices, Inc. | Cache resizing based on processor workload |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0934776A (ja) * | 1995-07-21 | 1997-02-07 | Sony Computer Entertainment:Kk | 情報処理装置および方法 |
JPH0950401A (ja) * | 1995-08-09 | 1997-02-18 | Toshiba Corp | キャッシュメモリ及びそれを備えた情報処理装置 |
JP3071752B2 (ja) * | 1998-03-24 | 2000-07-31 | 三菱電機株式会社 | ブリッジ方法、バスブリッジ及びマルチプロセッサシステム |
JP3516326B2 (ja) * | 1997-03-13 | 2004-04-05 | 株式会社日立製作所 | 共有キャッシュメモリを有するメモリコントローラ及びこれを備えたコンピュータシステム |
JP3822885B2 (ja) * | 1994-09-09 | 2006-09-20 | 株式会社ルネサステクノロジ | 1チップデータプロセッサ |
JP2008077255A (ja) * | 2006-09-20 | 2008-04-03 | Nec Corp | 携帯端末装置及びそれに用いるメモリ管理方法並びにそのプログラム |
JP2008234074A (ja) * | 2007-03-16 | 2008-10-02 | Fujitsu Ltd | キャッシュ装置 |
WO2008155822A1 (ja) * | 2007-06-19 | 2008-12-24 | Fujitsu Limited | キャッシュ制御装置及び制御方法 |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7136110B2 (en) * | 2000-06-14 | 2006-11-14 | Canon Kabushiki Kaisha | Image signal processing apparatus |
US6678814B2 (en) * | 2001-06-29 | 2004-01-13 | International Business Machines Corporation | Method and apparatus for allocating data usages within an embedded dynamic random access memory device |
-
2010
- 2010-10-13 WO PCT/JP2010/006098 patent/WO2011045931A1/ja active Application Filing
- 2010-10-13 CN CN2010800465615A patent/CN102713867A/zh active Pending
- 2010-10-13 JP JP2011536042A patent/JPWO2011045931A1/ja not_active Withdrawn
- 2010-10-13 US US13/500,494 patent/US20120198159A1/en not_active Abandoned
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3822885B2 (ja) * | 1994-09-09 | 2006-09-20 | 株式会社ルネサステクノロジ | 1チップデータプロセッサ |
JPH0934776A (ja) * | 1995-07-21 | 1997-02-07 | Sony Computer Entertainment:Kk | 情報処理装置および方法 |
JPH0950401A (ja) * | 1995-08-09 | 1997-02-18 | Toshiba Corp | キャッシュメモリ及びそれを備えた情報処理装置 |
JP3516326B2 (ja) * | 1997-03-13 | 2004-04-05 | 株式会社日立製作所 | 共有キャッシュメモリを有するメモリコントローラ及びこれを備えたコンピュータシステム |
JP3071752B2 (ja) * | 1998-03-24 | 2000-07-31 | 三菱電機株式会社 | ブリッジ方法、バスブリッジ及びマルチプロセッサシステム |
JP2008077255A (ja) * | 2006-09-20 | 2008-04-03 | Nec Corp | 携帯端末装置及びそれに用いるメモリ管理方法並びにそのプログラム |
JP2008234074A (ja) * | 2007-03-16 | 2008-10-02 | Fujitsu Ltd | キャッシュ装置 |
WO2008155822A1 (ja) * | 2007-06-19 | 2008-12-24 | Fujitsu Limited | キャッシュ制御装置及び制御方法 |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2013058049A (ja) * | 2011-09-07 | 2013-03-28 | Nec Corp | I/oデバイス共有システム、i/oデバイス共有方法、及びプログラム |
JP2015064863A (ja) * | 2013-08-26 | 2015-04-09 | 富士ゼロックス株式会社 | 情報処理装置、演算処理装置及びプログラム |
WO2015132877A1 (ja) * | 2014-03-04 | 2015-09-11 | 株式会社日立製作所 | 計算機およびメモリ制御方法 |
US9588896B2 (en) | 2014-03-04 | 2017-03-07 | Hitachi, Ltd. | Computer and memory control method |
JPWO2015132877A1 (ja) * | 2014-03-04 | 2017-03-30 | 株式会社日立製作所 | 計算機およびメモリ制御方法 |
JP2017527884A (ja) * | 2014-07-17 | 2017-09-21 | クアルコム,インコーポレイテッド | セットおよびウェイによるコンポーネントキャッシュへの柔軟なキャッシュパーティショニングのための方法および装置 |
JP2016181030A (ja) * | 2015-03-23 | 2016-10-13 | 富士通株式会社 | 情報処理装置、記憶装置制御方法、記憶装置制御プログラム及び情報処理システム |
Also Published As
Publication number | Publication date |
---|---|
CN102713867A (zh) | 2012-10-03 |
JPWO2011045931A1 (ja) | 2013-03-04 |
US20120198159A1 (en) | 2012-08-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2011045931A1 (ja) | 情報処理装置 | |
EP1691288B1 (en) | Memory management for a mobile multimedia processor | |
KR102114388B1 (ko) | 전자 장치의 메모리 압축 방법 및 장치 | |
EP2513748B1 (en) | Power efficient memory | |
EP3872604B1 (en) | Hardware automatic performance state transitions in system on processor sleep and wake events | |
JP6038699B2 (ja) | 電子機器 | |
EP3602310B1 (en) | Power-conserving cache memory usage | |
US9601180B2 (en) | Automatic partial array self-refresh | |
WO2012014285A1 (ja) | 割込制御方法、マルチコアプロセッサシステム、および割込制御プログラム | |
JP2012022616A (ja) | 共有メモリシステム及びその制御方法 | |
JP2008234320A (ja) | キャッシュ制御回路 | |
US7944770B2 (en) | Static random access memory system and control method for static random access memory system | |
KR102117511B1 (ko) | 프로세서 및 메모리 제어 방법 | |
JP2009251713A (ja) | キャッシュメモリ制御装置 | |
JP2008262451A (ja) | メモリ電源管理装置及びメモリ電源管理方法 | |
US20140351546A1 (en) | Method and apparatus for mapping a physical memory having a plurality of memory regions | |
JP5783348B2 (ja) | 制御装置、制御プログラム、画像形成装置 | |
CN110888749B (zh) | 电子设备中执行任务级高速缓存管理的方法和装置 | |
US20170322614A1 (en) | Power management techniques | |
JP5505195B2 (ja) | メモリの制御装置、及び制御方法 | |
JP2008165318A (ja) | 計算機システム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201080046561.5 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 10823199 Country of ref document: EP Kind code of ref document: A1 |
|
DPE1 | Request for preliminary examination filed after expiration of 19th month from priority date (pct application filed from 20040101) | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2011536042 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13500494 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 10823199 Country of ref document: EP Kind code of ref document: A1 |