WO2009096161A1 - プロセッサ性能解析装置、方法及びシミュレータ - Google Patents
プロセッサ性能解析装置、方法及びシミュレータ Download PDFInfo
- Publication number
- WO2009096161A1 WO2009096161A1 PCT/JP2009/000246 JP2009000246W WO2009096161A1 WO 2009096161 A1 WO2009096161 A1 WO 2009096161A1 JP 2009000246 W JP2009000246 W JP 2009000246W WO 2009096161 A1 WO2009096161 A1 WO 2009096161A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- processor
- information
- access request
- memory
- access
- Prior art date
Links
Images
Classifications
-
- 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
- G06F11/3466—Performance evaluation by tracing or monitoring
- G06F11/3471—Address tracing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2201/00—Indexing scheme relating to error detection, to error correction, and to monitoring
- G06F2201/885—Monitoring specific for caches
Definitions
- the present invention relates to an apparatus for analyzing the performance of a processor in a system LSI (Large Scale Integration), and in particular, analyzes the performance of a multi-thread processor having a plurality of logical processors in the processor and capable of executing a plurality of programs simultaneously. It is related with the apparatus which performs.
- LSI Large Scale Integration
- a system LSI in which a processor and a functional core other than the processor are integrated is widely used in today's digital televisions and digital recorders.
- Functional cores include, for example, general-purpose IF (Interface) circuits such as PCI (Peripheral Component Interconnect) bus and IDE (Integrated Drive Electronics) bus, codec circuits that encode and decode content data such as video and music, and paid content Encryption processing circuit for protecting copyright information.
- general-purpose IF Interface
- PCI Peripheral Component Interconnect
- IDE Integrated Drive Electronics
- FIG. 8 is a block diagram of a conventional processor performance evaluation apparatus described in Patent Document 1.
- FIG. 8 is a block diagram of a conventional processor performance evaluation apparatus described in Patent Document 1.
- 8 includes an instruction unit 401, an arithmetic unit 402, a primary cache unit 403, and a secondary cache unit 404.
- the secondary cache unit 404 includes a secondary cache 405 and an external access unit 406, and outputs each piece of hardware information in the computer.
- the secondary cache 405 outputs information such as the number of accesses, the number of hits, and the request type.
- the external access unit 406 outputs information such as the number of write and read queues clogged in the access buffer in the access between the secondary cache 405 and the memory 40.
- a core ID for identifying the instruction unit 401 and the arithmetic unit 402 is provided, and the secondary cache 405 and Information indicating where the external access unit 406 is used is added and output.
- the output information makes it possible to grasp the operation of the entire computer and facilitate performance bottleneck analysis.
- the conventional configuration described above has a problem that it is not possible to obtain information on performance degradation factors other than the processor, while it can grasp performance degradation factors such as cache misses and TLB misses occurring in the processor. ing.
- the performance degradation factor other than the processor is, for example, an event such as a memory access from the processor being waited for a long time because the DMA (Direct Memory Access) transfer of the functional core occupies the memory IF resource.
- DMA Direct Memory Access
- the present invention has been made in view of the above problems, and provides a processor performance analysis apparatus capable of analyzing system performance deterioration factors including not only the processor but also the operating state of functional cores other than the processor.
- the purpose is to provide.
- a processor performance analysis apparatus includes a plurality of logical processors, performs parallel processing by the plurality of logical processors, and issues a first access request to a memory; A function core that executes processing different from the processing executed by the processor and that issues a second access request to the memory; accepts the first access request and the second access request; A processor performance analysis device for analyzing the performance of the processor in a system LSI including a memory interface for controlling access, wherein the processor monitors the processor and outputs first information indicating an operation state of the processor.
- the memory interface is monitored by monitoring information output means and the memory interface.
- the second information output means for outputting second information indicating the state of the memory access caused by the first and second access requests received by the memory, and the performance of the processor using the first information and the second information.
- the processor performance analysis apparatus further includes third information output means for outputting third information indicating a factor by which the processor issues the first access request by monitoring the processor. May further analyze the performance of the processor using the third information.
- the processor issues the first access request to the memory for each of the logical processors
- the third information output means assigns attribute information specifying the logical processor that has issued the first access request to the first processor. You may output as 3 information.
- the processor issues the first access request when a prefetch or a cache miss occurs, and the third information output means causes the processor to issue the first access request because of the prefetch and the cache. Information indicating which of the mistakes may be output as the third information.
- the cache miss is an instruction cache miss, a data cache miss, and a TLB miss.
- the second information output means may output information indicating whether the access request received by the memory interface is the first access request or the second access request as the second information.
- the second information output means completes the data transfer after receiving the first access request or the information related to the waiting order of the second access request, or the first access request or the second access request.
- the information about the time until may be output as the second information.
- the first information output means may include information indicating whether the logical processor is operating or in a waiting state, a cache hit or miss of the processor, or a prefetch operation hit or miss. You may output as 1st information.
- the system LSI may include a plurality of the processors, and the processor performance analysis device may include the first information output unit corresponding to each of the plurality of processors.
- the processor performance analysis apparatus may further include trigger output means for receiving the analysis result of the processor by the analysis means and outputting a trigger signal when the analysis result satisfies a predetermined condition. .
- the processor performance analysis apparatus further relates to a third access request issued from the processor to the functional core via a bus connecting the processor and the functional core by monitoring the processor.
- Bus access attribute information output means for outputting fourth information may be provided, and the analysis means may further analyze the performance of the processor using the fourth information.
- the present invention includes a plurality of logical processors, performs parallel processing by the plurality of logical processors, and performs processing different from processing executed by the processor that issues a first access request to the memory.
- a system LSI comprising: a functional core that executes and issues a second access request to the memory; and a memory interface that receives the first access request and the second access request and controls access to the memory
- a processor performance analysis simulator for analyzing the performance of the processor by simulating the operation of the processor, the first information output means for outputting the first information indicating the operation state of the processor, and the access accepted by the memory interface Output second information indicating the state of memory access caused by the request
- Second information output means can also be implemented as a processor performance analysis simulator comprising analyzing means for analyzing the performance of the processor by using the second information and the first information.
- the present invention can be realized not only as an apparatus, but also as a method using steps of processing means constituting the apparatus, as a program for causing a computer to execute the steps, or as a computer read recording the program It can also be realized as a possible recording medium such as a CD-ROM, or as information, data or a signal indicating the program.
- These programs, information, data, and signals may be distributed via a communication network such as the Internet.
- processor performance analysis apparatus of the present invention it is possible to evaluate the processor performance including the influence of the memory access operation of the functional core other than the processor included in the system LSI.
- performance bottleneck analysis becomes easy, and it becomes easy to improve performance by correcting software and hardware.
- FIG. 1 is a block diagram of a system LSI including the processor performance analysis apparatus according to the first embodiment.
- FIG. 2 is a flowchart showing the operation of the processor performance analysis apparatus according to the first embodiment.
- FIG. 3 is a block diagram of a system LSI including the processor performance analysis apparatus according to the second embodiment.
- FIG. 4 is a flowchart illustrating the operation of the processor performance analysis apparatus according to the second embodiment.
- FIG. 5 is a block diagram of a system LSI including the processor performance analysis apparatus according to the third embodiment.
- FIG. 6 is a flowchart illustrating the operation of the processor performance analysis apparatus according to the third embodiment.
- FIG. 7 is a block diagram of a system LSI including a plurality of multithread processors.
- FIG. 8 is a block diagram of a conventional processor performance evaluation apparatus.
- FIG. 1 is a block diagram of a system LSI including a processor performance analysis apparatus according to this embodiment.
- the system LSI 10 includes a multithread processor 11, a functional core 12, and a memory interface 13.
- the multi-thread processor 11 has a plurality of logical processors (LPs), and can execute a plurality of programs in parallel by the logical processors. In addition, a memory access request to the memory 20 is issued in order to write an instruction or data to the memory 20 or read from the memory 20 as necessary when executing the program.
- the multithread processor 11 includes a primary cache, a secondary cache, a TLB, and the like (not shown). For example, when a prefetch or a cache miss occurs, the multithread processor 11 issues a memory access request to the memory 20. The memory access request is issued for each logical processor.
- the functional core 12 is a plurality of functional cores that execute processing different from that of the multi-thread processor 11 and issue a memory access request to the memory 20.
- the functional core 12 includes, for example, a DMA controller, an interface circuit to an external device, an AV (Audio Visual) codec circuit that compresses or expands music and video content data, and an encryption / decryption that performs data encryption and decryption Circuit.
- the interface circuit to the external device is, for example, a PCI interface, a USB (Universal Serial Bus) interface, or the like.
- a DMA controller that is one of the functional cores 12 controls access between each functional core 12 and the memory 20. Note that the functional core 12 is not necessarily plural.
- the memory interface 13 receives a memory access request to the memory 20 issued by the multi-thread processor 11 and the functional core 12. Then, the access to the memory 20 is controlled by arbitrating the received memory access request.
- the processor performance analysis apparatus analyzes the operation state of the multithread processor 11 included in the system LSI 10 and the state of memory access from the multithread processor 11 and the functional core 12.
- FIG. 1 also shows the configuration of the processor performance analysis apparatus of this embodiment.
- the processor performance analysis apparatus 100 in FIG. 1 includes an operation information output unit 101, an access attribute information output unit 102, an access information output unit 103, and an analysis information output unit 104.
- the operation information output unit 101 and the access attribute information output unit 102 are provided in the multi-thread processor 11.
- the access information output unit 103 is provided in the memory interface 13.
- the operation information output unit 101 monitors the multi-thread processor 11 to dynamically output operation information indicating the operation state inside the multi-thread processor 11.
- the operation information includes, for example, whether each of the logical processors is operating or is in a data access waiting state, whether the number of operating logical processors exceeds the number of arithmetic units, and whether a waiting state has occurred. Is prefetch access, prefetch operation is hit or miss, instruction cache and data cache are hit or miss, TLB is hit or miss, secondary cache is It is information such as whether it is a hit or a miss.
- the access attribute information output unit 102 monitors the multi-thread processor 11 and outputs memory access attribute information related to a memory access request to the memory 20 issued by the multi-thread processor 11.
- the memory access attribute information is, for example, ID information indicating which logical processor is issuing a memory access request. Also, whether a memory access request is issued due to an instruction or data prefetch, an instruction or data cache miss, a TLB miss, a secondary cache miss, or an access to a non-cacheable area This is information such as access factor information indicating whether it is something.
- the access information output unit 103 monitors the memory interface 13 to output memory access information related to a memory access state generated by a memory access request accepted by the memory interface 13.
- the memory access information is, for example, information indicating whether the accepted memory access request is issued from the multi-thread processor 11 or the functional core 12.
- the access information output unit 103 issues the accepted memory access request to the multi-thread processor 11, the memory access attribute information output from the access attribute information output unit 102, the memory The operation state in the interface 13 is correlated and output as memory access information.
- the memory access information includes information indicating whether the received memory access request is issued by a logical processor having any ID information, prefetch, cache miss, or TLB miss. Is output as As another example of the memory access information, the time from when the access request is received to when the data transfer starts and / or ends, the number of access requests being received when a plurality of access requests overlap, For example, the order of processing queues.
- the analysis information output unit 104 associates the operation information, the memory access attribute information, and the memory access information, and outputs analysis information related to the system performance.
- the analysis information includes, for example, a period in which all the logical processors of the multi-thread processor 11 are not operating and are in a waiting state, cache hit rate for each logical processor, information on the number of memory accesses and memory access wait time, and multi-thread. This is information related to an increase in the memory access completion waiting time due to the memory access of the functional core 12 in memory access from the processor 11.
- FIG. 2 is a flowchart showing the operation of the processor performance analysis apparatus 100 of the present embodiment.
- the operation information output unit 101 outputs operation information indicating the processing status of each logical processor by monitoring a plurality of logical processors included in the multi-thread processor 11 (S101). Specifically, for each logical processor, information indicating whether it is operating or in a data access waiting state, whether the cache is hit or missed, and the like are output as operation information.
- the access attribute information output unit 102 outputs a memory access attribute information related to a memory access request to the memory 20 issued by the multi-thread processor 11 by monitoring a plurality of logical processors (S102). Specifically, information such as ID information that identifies a logical processor that has issued a memory access request and access factor information that indicates a factor that issues a memory access request is output as memory access attribute information.
- the access information output unit 103 monitors the memory interface 13 so that the memory access request being accepted by the memory interface 13 is issued by the multithread processor 11 or issued by the functional core 12. Is determined (S103).
- the access information output unit 103 receives the memory access attribute information output from the access attribute information output unit 102 and the memory interface 13
- the memory access information is output in association with the operation state (S104). Specifically, information specifying the logical processor that issued the accepted memory access request and information indicating whether the memory access request is due to prefetch or cache miss are output as memory access information. .
- the access information output unit 103 indicates that the accepted memory access request is issued by the functional core 12.
- Information indicating the presence or the like is output as memory access information (S105).
- the analysis information output unit 104 uses the operation information (output in S101), the memory access attribute information (output in S102), and the memory access information (output in S104 or S105) to determine the operation status of the system LSI 10
- the analysis information is output by analyzing (S106).
- operation information output (S101) or the memory access attribute information output (S102) may be performed first or may be parallelized.
- the processor performance analysis apparatus can obtain the operation status of the entire system by associating the operation information of the processor with the memory access information from the processor and the functional core. With the above configuration, it is possible to perform an appropriate system bottleneck analysis and system performance improvement study.
- the processor performance analysis apparatus outputs a trigger signal for operating an external apparatus or the like based on the analysis result of the processor performance.
- FIG. 3 is a block diagram of a system LSI provided with the processor performance analysis apparatus of the present embodiment.
- the processor performance analysis apparatus 200 shown in FIG. 3 is different from the processor performance analysis apparatus 100 of FIG. 1 in that a trigger output unit 201 is newly added and an analysis information output unit instead of the analysis information output unit 104 The difference is that 204 is added.
- description of the same components as those in FIG. 1 will be omitted, and different points will be mainly described.
- the trigger output unit 201 When the trigger output unit 201 receives a signal indicating that the system state satisfies a specific condition from the analysis information output unit 104, the trigger output unit 201 outputs the signal to the outside of the system LSI 10 as a trigger signal. For example, a trigger signal is output to a debugger for the multithread processor 11 connected to the outside of the system LSI 10. Further, as an example of the system state detected by the analysis information output unit 104, all logical processors of the multi-thread processor 11 are in a data waiting state, all program executions are stopped, and a specific logical processor A state where a system bottleneck has occurred, such as a state where the memory access waiting time exceeds a predetermined value, can be mentioned.
- the analysis information output unit 204 generates analysis information by associating operation information, memory access attribute information, and memory access information, and outputs the generated analysis information not only to the system LSI 10 but also to the trigger output unit 201.
- a specific example of the analysis information is the same as in the first embodiment.
- FIG. 4 is a flowchart showing the operation of the processor performance analysis apparatus 200 of this embodiment.
- the processing shown in the figure is different from the processing shown in FIG. 2 in that processing for outputting a trigger signal (S207 and S208) is further added.
- FIG. 4 processes denoted by the same reference numerals as those in FIG. 2 are the same processes as those in the first embodiment, and description thereof will be omitted below.
- the analysis information output unit 104 uses operation information (output in S101), memory access attribute information (output in S102), and memory access information (output in S104 or S105).
- the analysis information is output by analyzing the operation status of the system LSI 10 (S106).
- the trigger output unit 201 determines whether or not the state of the system indicated by the analysis information output from the analysis information output unit 104 satisfies the specific condition (S207). When the system state satisfies the specific condition (Yes in S207), the trigger output unit 201 outputs a trigger signal indicating that the system state satisfies the specific condition to the outside of the system LSI 10 (S208).
- the trigger signal is not output and only the analysis information is output to the outside.
- the processor performance analysis apparatus outputs a trigger signal for operating an external apparatus or the like based on the analysis result of the processor performance. This makes it easy to check the software operation when a system bottleneck occurs, and the convenience in system bottleneck analysis is further improved.
- the processor performance analysis apparatus can perform analysis based on information on an access request issued from a processor to a functional core when the processor and the functional core are connected by an IO bus. .
- FIG. 5 is a block diagram of a system LSI provided with the processor performance analysis apparatus of the present embodiment.
- the processor performance analysis device 300 shown in FIG. 5 is different from the processor performance analysis device 100 of FIG. 1 in that an IO bus access attribute information output unit 301 is newly added and instead of the analysis information output unit 104. The difference is that an analysis information output unit 304 is added.
- description of the same components as those in FIG. 1 will be omitted, and different points will be mainly described.
- the IO bus access attribute information output unit 301 outputs the IO bus access attribute information related to the access request transferred via the IO bus connecting the multi-thread processor 11 and the functional core 12 by monitoring the multi-thread processor 11. To do.
- the IO bus access attribute information is attribute information related to access to the functional core 12 via the IO bus used for register access from the multi-thread processor 11 to the functional core 12. Also, ID information indicating which logical processor is issuing an IO bus access request.
- the analysis information output unit 304 generates analysis information by associating operation information, memory access attribute information, memory access information, and IO bus access attribute information, and outputs the generated analysis information to the outside of the system LSI 10.
- FIG. 6 is a flowchart showing the operation of the processor performance analysis apparatus 300 of this embodiment.
- the process shown in the figure is different from the process shown in FIG. 2 in that a process of outputting IO bus access attribute information (S303) is further added.
- S303 IO bus access attribute information
- the IO bus access attribute information output unit 301 outputs the IO bus access attribute information by monitoring the multi-thread processor 11. (S302). If the access request is not transferred via the IO bus, the IO bus access attribute information output unit 301 uses the information indicating that the access request is not transferred via the IO bus as the IO bus access attribute information. It may be output, or IO bus access attribute information may not be output.
- the access information output unit 103 outputs memory access attribute information (S104 or S105).
- the analysis information output unit 304 outputs operation information (output at S101), memory access attribute information (output at S102), IO bus access attribute information (output at S303), and memory access information (output at S104 or S105).
- the analysis information is output by analyzing the operation status of the system LSI 10 (S106).
- the operation information output (S101), the memory access attribute information output (S102), and the IO bus access attribute information output (S303) may be performed first or in parallel. .
- the processor performance analysis apparatus can analyze not only the access from the processor to the memory but also the performance penalty due to the IO bus access from the processor to the functional core. The accuracy in analysis is further improved.
- processor performance analysis apparatus and the processor performance analysis method of the present invention have been described based on the embodiment, the present invention is not limited to this embodiment. Unless it deviates from the meaning of this invention, the form which carried out the various deformation
- the multi-thread processor 11 is provided as the processor of the system LSI 10 of the present embodiment, a multi-processor configuration with a plurality of processors may be employed.
- the system LSI 10 includes a plurality of multithread processors 11.
- the plurality of multi-thread processors 11 includes an operation information output unit 101 and an access attribute information output unit 102 corresponding to each.
- the analysis processing of the performance of the processor included in the system LSI 10 may be simulated.
- the multi-thread processor 11, the functional core 12, the memory interface 13, and the like are realized as software on the computer, and the processor performance analysis method illustrated in FIGS. 2, 4, 6, and the like is executed by the computer.
- the system performance is analyzed by causing the multi-thread processor 11 and the functional core 12 realized on the computer to execute a predetermined program in a pseudo manner.
- the processor performance analysis apparatus of the present invention is useful when analyzing performance bottlenecks of system LSIs and performing performance improvement studies by modifying hardware and software. For example, it can be applied to uses such as debugging of parallel programming processing of a multi-thread processor.
Abstract
Description
11 マルチスレッドプロセッサ
12 機能コア
13 メモリインタフェース
20、40 メモリ
30 コンピュータ
100、200、300 プロセッサ性能解析装置
101 動作情報出力部
102 アクセス属性情報出力部
103 アクセス情報出力部
104、204、304 解析情報出力部
201 トリガ出力部
301 IOバスアクセス属性情報出力部
401 命令ユニット
402 演算ユニット
403 1次キャッシュユニット
404 2次キャッシュユニット
405 2次キャッシュ
406 外部アクセスユニット
まず、本実施の形態のプロセッサ性能解析装置を備えるシステムLSIの構成について説明する。
本実施の形態のプロセッサ性能解析装置は、プロセッサの性能の解析結果に基づいて、外部の装置などを操作するためのトリガ信号を出力する。
本実施の形態のプロセッサ性能解析装置は、プロセッサと機能コアとがIOバスによって接続されている場合に、プロセッサから機能コアに対して発行されるアクセス要求に関する情報を基にして解析することができる。
Claims (13)
- 複数の論理プロセッサを備え、前記複数の論理プロセッサにより並列処理を行い、かつ、メモリへの第1アクセス要求を発行するプロセッサと、
前記プロセッサが実行する処理とは異なる処理を実行し、かつ、前記メモリへの第2アクセス要求を発行する機能コアと、
前記第1アクセス要求と前記第2アクセス要求とを受け付け、前記メモリへのアクセスを制御するメモリインタフェースとを備える
システムLSIにおける前記プロセッサの性能を解析するプロセッサ性能解析装置であって、
前記プロセッサを監視することで、前記プロセッサの動作状態を示す第1情報を出力する第1情報出力手段と、
前記メモリインタフェースを監視することで、前記メモリインタフェースが受け付ける前記第1及び第2アクセス要求によって生じるメモリアクセスの状態を示す第2情報を出力する第2情報出力手段と、
前記第1情報と前記第2情報とを用いて前記プロセッサの性能を解析する解析手段とを備える
ことを特徴とするプロセッサ性能解析装置。 - 前記プロセッサ性能解析装置は、さらに、
前記プロセッサを監視することで、前記プロセッサが前記第1アクセス要求を発行する要因を示す第3情報を出力する第3情報出力手段を備え、
前記解析手段は、さらに、前記第3情報を用いて前記プロセッサの性能を解析する
ことを特徴とする請求項1記載のプロセッサ性能解析装置。 - 前記プロセッサは、前記論理プロセッサ毎に前記メモリへの前記第1アクセス要求を発行し、
前記第3情報出力手段は、前記第1アクセス要求を発行した論理プロセッサを特定する属性情報を前記第3情報として出力する
ことを特徴とする請求項2記載のプロセッサ性能解析装置。 - 前記プロセッサは、プリフェッチ又はキャッシュミスが生じた場合に前記第1アクセス要求を発行し、
前記第3情報出力手段は、前記プロセッサが前記第1アクセス要求を発行した原因が前記プリフェッチ及び前記キャッシュミスのいずれによるものかを示す情報を前記第3情報として出力する
ことを特徴とする請求項2又は3記載のプロセッサ性能解析装置。 - 前記キャッシュミスは、命令キャッシュミス、データキャッシュミス及びTLB(Translation Lookaside Buffer)ミスである
ことを特徴とする請求項4記載のプロセッサ性能解析装置。 - 前記第2情報出力手段は、前記メモリインタフェースが受け付けたアクセス要求が前記第1アクセス要求及び前記第2アクセス要求のいずれであるかを示す情報を前記第2情報として出力する
ことを特徴とする請求項1~5のいずれか1項に記載のプロセッサ性能解析装置。 - 前記第2情報出力手段は、前記第1アクセス要求若しくは前記第2アクセス要求の待ち順序に関する情報、又は、前記第1アクセス要求若しくは前記第2アクセス要求を受け付けてからデータの転送が完了するまでの時間に関する情報を、前記第2情報として出力する
ことを特徴とする請求項1~6のいずれか1項に記載のプロセッサ性能解析装置。 - 前記第1情報出力手段は、前記論理プロセッサ毎に動作しているのか若しくは待ち状態にあるのか、前記プロセッサのキャッシュのヒット若しくはミス、又は、プリフェッチ動作のヒット若しくはミスを示す情報を、前記第1情報として出力する
ことを特徴とする請求項1~7のいずれか1項に記載のプロセッサ性能解析装置。 - 前記システムLSIは、前記プロセッサを複数備え、
前記プロセッサ性能解析装置は、
前記複数のプロセッサのそれぞれに対応する前記第1情報出力手段を備える
ことを特徴とする請求項1~8のいずれか1項に記載のプロセッサ性能解析装置。 - 前記プロセッサ性能解析装置は、さらに、
前記解析手段による前記プロセッサの解析結果を受け取り、前記解析結果が予め定められた条件を満たす場合に、トリガ信号を出力するトリガ出力手段を備える
ことを特徴とする請求項1~9のいずれか1項に記載のプロセッサ性能解析装置。 - 前記プロセッサ性能解析装置は、さらに、
前記プロセッサを監視することで、前記プロセッサと前記機能コアとを接続するバスを介して前記プロセッサから前記機能コアに対して発行される第3アクセス要求に関する第4情報を出力するバスアクセス属性情報出力手段を備え、
前記解析手段は、さらに、前記第4情報を用いて前記プロセッサの性能を解析する
ことを特徴とする請求項1~10のいずれか1項に記載のプロセッサ性能解析装置。 - 複数の論理プロセッサを備え、前記複数の論理プロセッサにより並列処理を行い、かつ、メモリへの第1アクセス要求を発行するプロセッサと、
前記プロセッサが実行する処理とは異なる処理を実行し、かつ、前記メモリへの第2アクセス要求を発行する機能コアと、
前記第1アクセス要求と前記第2アクセス要求とを受け付け、前記メモリへのアクセスを制御するメモリインタフェースとを備える
システムLSIにおける前記プロセッサの性能を解析するプロセッサ性能解析方法であって、
前記プロセッサを監視することで、前記プロセッサの動作状態を示す第1情報を出力する第1情報出力ステップと、
前記メモリインタフェースを監視することで、前記メモリインタフェースが受け付ける前記第1及び第2アクセス要求によって生じるメモリアクセスの状態を示す第2情報を出力する第2情報出力ステップと、
前記第1情報と前記第2情報とを用いて前記プロセッサの性能を解析する解析ステップとを含む
ことを特徴とするプロセッサ性能解析方法。 - 複数の論理プロセッサを備え、前記複数の論理プロセッサにより並列処理を行い、かつ、メモリへの第1アクセス要求を発行するプロセッサと、
前記プロセッサが実行する処理とは異なる処理を実行し、かつ、前記メモリへの第2アクセス要求を発行する機能コアと、
前記第1アクセス要求と前記第2アクセス要求とを受け付け、前記メモリへのアクセスを制御するメモリインタフェースとを備える
システムLSIの動作をシミュレートすることで、前記プロセッサの性能を解析するプロセッサ性能解析シミュレータであって、
前記プロセッサを監視することで、前記プロセッサの動作状態を示す第1情報を出力する第1情報出力手段と、
前記メモリインタフェースを監視することで、前記メモリインタフェースが受け付ける前記第1及び第2アクセス要求によって生じるメモリアクセスの状態を示す第2情報を出力する第2情報出力手段と、
前記第1情報と前記第2情報とを用いて前記プロセッサの性能を解析する解析手段とを備える
ことを特徴とするプロセッサ性能解析シミュレータ。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US12/864,935 US20100332690A1 (en) | 2008-01-29 | 2009-01-23 | Processor performance analysis device, method, and simulator |
JP2009551426A JPWO2009096161A1 (ja) | 2008-01-29 | 2009-01-23 | プロセッサ性能解析装置、方法及びシミュレータ |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2008017714 | 2008-01-29 | ||
JP2008-017714 | 2008-01-29 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2009096161A1 true WO2009096161A1 (ja) | 2009-08-06 |
Family
ID=40912518
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2009/000246 WO2009096161A1 (ja) | 2008-01-29 | 2009-01-23 | プロセッサ性能解析装置、方法及びシミュレータ |
Country Status (3)
Country | Link |
---|---|
US (1) | US20100332690A1 (ja) |
JP (1) | JPWO2009096161A1 (ja) |
WO (1) | WO2009096161A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015020092A1 (ja) * | 2013-08-07 | 2015-02-12 | 株式会社 東芝 | 情報処理方法、情報処理装置及びプログラム |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR102285749B1 (ko) | 2014-11-10 | 2021-08-05 | 삼성전자주식회사 | 세마포어 기능을 갖는 시스템 온 칩 및 그것의 세마포어 구현 방법 |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08263310A (ja) * | 1994-12-16 | 1996-10-11 | Internatl Business Mach Corp <Ibm> | 多重プロセッサ・システムにおけるcpuのモニタリング・システム及び方法 |
JP2001142869A (ja) * | 1999-11-17 | 2001-05-25 | Matsushita Electric Ind Co Ltd | システム集積回路 |
JP2001256079A (ja) * | 2000-03-10 | 2001-09-21 | Matsushita Electric Ind Co Ltd | ソフトウェアのデバッグ方法および論理回路のデバッグ方法 |
JP2002024201A (ja) * | 2000-07-10 | 2002-01-25 | Toshiba Corp | 半導体集積回路 |
WO2004055672A1 (ja) * | 2002-12-18 | 2004-07-01 | Fujitsu Limited | 性能シミュレーション装置、性能シミュレーションプログラムおよび性能シミュレーション方法 |
JP2006040090A (ja) * | 2004-07-29 | 2006-02-09 | Fujitsu Ltd | キャッシュメモリ制御装置、キャッシュメモリ制御方法、中央処理装置、情報処理装置、中央処理方法。 |
JP2006113716A (ja) * | 2004-10-13 | 2006-04-27 | Matsushita Electric Ind Co Ltd | 半導体集積回路 |
WO2006049090A1 (ja) * | 2004-11-04 | 2006-05-11 | Matsushita Electric Industrial Co., Ltd. | 集積回路、及び集積回路パッケージ |
JP2006172234A (ja) * | 2004-12-17 | 2006-06-29 | Matsushita Electric Ind Co Ltd | システム性能評価方法およびシステム性能評価装置 |
JP2006293694A (ja) * | 2005-04-11 | 2006-10-26 | Sony Computer Entertainment Inc | 情報処理装置、コンピュータの制御方法及びプログラム |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5426741A (en) * | 1991-02-20 | 1995-06-20 | Digital Equipment Corporation | Bus event monitor |
US20050182611A1 (en) * | 2002-12-18 | 2005-08-18 | Fujitsu Limited | Apparatus and method for simulating performance, and computer product |
US7478198B2 (en) * | 2004-05-24 | 2009-01-13 | Intel Corporation | Multithreaded clustered microarchitecture with dynamic back-end assignment |
JP4839164B2 (ja) * | 2006-09-15 | 2011-12-21 | 株式会社日立製作所 | ハードウェアモニタを用いた性能評価システム及び再構築可能な計算機システム |
JP2008123045A (ja) * | 2006-11-08 | 2008-05-29 | Matsushita Electric Ind Co Ltd | プロセッサ |
US7886162B2 (en) * | 2007-05-29 | 2011-02-08 | International Business Machines Corporation | Cryptographic secure program overlays |
-
2009
- 2009-01-23 WO PCT/JP2009/000246 patent/WO2009096161A1/ja active Application Filing
- 2009-01-23 JP JP2009551426A patent/JPWO2009096161A1/ja not_active Withdrawn
- 2009-01-23 US US12/864,935 patent/US20100332690A1/en not_active Abandoned
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH08263310A (ja) * | 1994-12-16 | 1996-10-11 | Internatl Business Mach Corp <Ibm> | 多重プロセッサ・システムにおけるcpuのモニタリング・システム及び方法 |
JP2001142869A (ja) * | 1999-11-17 | 2001-05-25 | Matsushita Electric Ind Co Ltd | システム集積回路 |
JP2001256079A (ja) * | 2000-03-10 | 2001-09-21 | Matsushita Electric Ind Co Ltd | ソフトウェアのデバッグ方法および論理回路のデバッグ方法 |
JP2002024201A (ja) * | 2000-07-10 | 2002-01-25 | Toshiba Corp | 半導体集積回路 |
WO2004055672A1 (ja) * | 2002-12-18 | 2004-07-01 | Fujitsu Limited | 性能シミュレーション装置、性能シミュレーションプログラムおよび性能シミュレーション方法 |
JP2006040090A (ja) * | 2004-07-29 | 2006-02-09 | Fujitsu Ltd | キャッシュメモリ制御装置、キャッシュメモリ制御方法、中央処理装置、情報処理装置、中央処理方法。 |
JP2006113716A (ja) * | 2004-10-13 | 2006-04-27 | Matsushita Electric Ind Co Ltd | 半導体集積回路 |
WO2006049090A1 (ja) * | 2004-11-04 | 2006-05-11 | Matsushita Electric Industrial Co., Ltd. | 集積回路、及び集積回路パッケージ |
JP2006172234A (ja) * | 2004-12-17 | 2006-06-29 | Matsushita Electric Ind Co Ltd | システム性能評価方法およびシステム性能評価装置 |
JP2006293694A (ja) * | 2005-04-11 | 2006-10-26 | Sony Computer Entertainment Inc | 情報処理装置、コンピュータの制御方法及びプログラム |
Non-Patent Citations (2)
Title |
---|
HIROYUKI MORISHITA ET AL.: "HDTV Taio Media Processor MCP2 ni Okeru Stream Shoribu no Kaihatsu", INFORMATION PROCESSING SOCIETY OF JAPAN KENKYU HOKOKU, vol. 2001, no. 22, 8 March 2001 (2001-03-08), pages 32 * |
TOSHIKAZU MUNEZANE ET AL.: "Digital STB/TV-yo Tanmatsu Architecture", MATSUSHITA TECHNICAL JOURNAL, vol. 44, no. 1, 18 February 1998 (1998-02-18), pages 77 - 78 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015020092A1 (ja) * | 2013-08-07 | 2015-02-12 | 株式会社 東芝 | 情報処理方法、情報処理装置及びプログラム |
JP2015035028A (ja) * | 2013-08-07 | 2015-02-19 | 株式会社東芝 | 情報処理方法、情報処理装置及びプログラム |
US10025504B2 (en) | 2013-08-07 | 2018-07-17 | Kabushiki Kaisha Toshiba | Information processing method, information processing apparatus and non-transitory computer readable medium |
Also Published As
Publication number | Publication date |
---|---|
JPWO2009096161A1 (ja) | 2011-05-26 |
US20100332690A1 (en) | 2010-12-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11748240B2 (en) | Scheduling of scenario models for execution within different computer threads and scheduling of memory regions for use with the scenario models | |
US7689867B2 (en) | Multiprocessor breakpoint | |
US9892024B2 (en) | Backward compatibility testing of software in a mode that disrupts timing | |
US8527812B2 (en) | Information processing device | |
US20060242389A1 (en) | Job level control of simultaneous multi-threading functionality in a processor | |
US20040030870A1 (en) | Software breakpoints with tailoring for multiple processor shared memory or multiple thread systems | |
US8255673B2 (en) | Monitoring transactions in a data processing apparatus | |
KR20090117700A (ko) | 프로세서 성능 모니터링 | |
WO2009096161A1 (ja) | プロセッサ性能解析装置、方法及びシミュレータ | |
US10481913B2 (en) | Token-based data dependency protection for memory access | |
JPH1049373A (ja) | パイプライン・デジタル・プロセッサにおいて多重で高精度の事象を操作する方法と装置 | |
US11907724B2 (en) | In-memory trace with overlapping processing and logout | |
EP1235139B1 (en) | System and method for supporting precise exceptions in a data processor having a clustered architecture | |
JP2002163126A (ja) | デバッグ用cpuに内蔵のイベント検出回路、イベント検出方法および外部周辺回路 | |
US20050268074A1 (en) | Method and apparatus for determining the criticality of a micro-operation | |
JP5387231B2 (ja) | マイクロプロセッサ及びマイクロプロセッサのデバッグ方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 09707046 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2009551426 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 12864935 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: 09707046 Country of ref document: EP Kind code of ref document: A1 |