JPH03210643A - Task executing time measuring method - Google Patents

Task executing time measuring method

Info

Publication number
JPH03210643A
JPH03210643A JP2005123A JP512390A JPH03210643A JP H03210643 A JPH03210643 A JP H03210643A JP 2005123 A JP2005123 A JP 2005123A JP 512390 A JP512390 A JP 512390A JP H03210643 A JPH03210643 A JP H03210643A
Authority
JP
Japan
Prior art keywords
task
time
processing
execution time
processing period
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP2005123A
Other languages
Japanese (ja)
Inventor
Shukichi Shimada
周吉 嶋田
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fuji Electric Co Ltd
Fuji Facom Corp
Original Assignee
Fuji Electric Co Ltd
Fuji Facom Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Fuji Electric Co Ltd, Fuji Facom Corp filed Critical Fuji Electric Co Ltd
Priority to JP2005123A priority Critical patent/JPH03210643A/en
Publication of JPH03210643A publication Critical patent/JPH03210643A/en
Pending legal-status Critical Current

Links

Landscapes

  • Debugging And Monitoring (AREA)

Abstract

PURPOSE:To obtain the executing time for each task that is started by an OS by performing the additional processing operations to calculate the executing time before and after the processing of an original task. CONSTITUTION:The executing time integrating areas a10 and a20 are provided for each task, and a 1st additional processing P1 is prepared before each task to read the task start time out of a clock and to store it. At the same time, a 2nd additional processing P2 is prepared after each task. In this processing P2, the task end time is read out of the clock, the task processing period is obtained from the difference between the end time and the start time, and this obtained processing period is added to the area a10 of the task. Furthermore the presence or absence is checked for a lower rank task where the relevant task is interrupted. When the presence of the lower rank task is confirmed, the processing period is subtracted from the area a20 of the lower rank task. Then the values stored in both areas a10 and a20 are defined as the task executing time. Thus it is possible to calculate the executing time of each task that is started by the OS.

Description

【発明の詳細な説明】[Detailed description of the invention]

r産業上の利用分野] 本発明は上位タスクを優先しながら複数のタスクを並行
して実行し得る計算機システムにおいて、各タスクの実
行時間を測定する方法に関する。 なお以下各図において同一の符号は同一もしくは相当部
分を示す。
r Industrial Application Field] The present invention relates to a method for measuring the execution time of each task in a computer system that can execute a plurality of tasks in parallel while giving priority to higher-level tasks. Note that in the following figures, the same reference numerals indicate the same or corresponding parts.

【従来の技術】[Conventional technology]

従来、この種の実行時間を求める方法としては、事前に
処理時間のわかっているプログラム(空き時間測定用プ
ログラム)をCPUの空き時間(つまり本来、O3によ
るタスクの起動がなされていない時間)中に実行させ、
カウンタによりそのプログラムの実行回数を計測し、空
き時間測定用プログラムの処理時間と実行回数とからC
PUの空き時間を算出し、この空き時間を用いてCPU
の実行時間(つまりO3により本来のタスクが起動され
ている時間)を算出する方法が知られている。
Conventionally, the method of determining this type of execution time is to run a program whose processing time is known in advance (a program for measuring free time) during the CPU's free time (that is, the time when tasks are not being started by O3). run it,
The number of executions of the program is measured using a counter, and C
Calculate the free time of the PU, and use this free time to
There is a known method for calculating the execution time (that is, the time during which the original task is activated by O3).

【発明が解決しようとする課題】[Problem to be solved by the invention]

しかしながらこのような従来の方法では、CPUの実行
時間の算出はできるが、タスク毎の実行時間の測定は不
可能であった。 そこでこの発明の課題は、事前に処理時間のわかってい
るプログラムを新たに設けることなく、O8により起動
されるタスクごとに実行時間を求めることにある。
However, with such conventional methods, although it is possible to calculate the execution time of the CPU, it is impossible to measure the execution time of each task. Therefore, an object of the present invention is to obtain the execution time for each task started by the O8 without creating a new program whose processing time is known in advance.

【課題を解決するための手段】[Means to solve the problem]

前記の課題を解決するために本発明の方法は、r時計(
システムクロックなど)を持ち、上位のタスクを優先し
ながら複数のタスクを並行して実行し得る計算機システ
ムにおける前記タスク毎の実行時間を測定する方法にお
いて、 前記タスク毎に実行時間積算領域(alO,a20など
)を設け、 各タスクの前に、当該のタスクの起動時刻を前記時計か
ら読取って記憶する第1の追加処理(PIなど)を付加
すると共に、 同じく各タスクの後に、当該タスクの終了時刻を前記時
計から読取り、この終了時刻と当該タスクの前記起動時
刻との差から当該タスクの処理期間を求めて、この処理
期間を当該タスクの前記実行時間積算領域へ加算し、さ
らに当該タスクが割込んだ下位タスクの有無を調べ、該
下位タスクが有るときは、この下位タスクについての前
記実行時間積算領域から前記処理期間を減算する第2の
追加処理(P2など)を付加し、 前記実行時間積算領域の格納値を当該タスクの実行時間
とするようにするものとする。
In order to solve the above problems, the method of the present invention provides an r clock (
In a method for measuring the execution time of each task in a computer system that has a system clock (such as a system clock) and can execute multiple tasks in parallel while giving priority to higher-level tasks, an execution time accumulation area (alO, a20, etc.), and before each task, add a first additional process (PI, etc.) that reads and stores the start time of the task from the clock, and also after each task, add the first additional process (PI, etc.) that reads and stores the start time of the task. The time is read from the clock, the processing period of the task is determined from the difference between the end time and the start time of the task, this processing period is added to the execution time accumulation area of the task, and the task is Check whether there is an interrupted lower task, and if there is such a lower task, add a second additional process (such as P2) that subtracts the processing period from the execution time accumulation area for this lower task, and execute the execution. The value stored in the time integration area is assumed to be the execution time of the task.

【作 用】[For use]

各タスクの前後に付加された追加処理によって、先ず当
該タスクの実行開始時刻から終了時刻までの処理期間を
求める。 この処理期間は、当該タスクがその実行中に上位タスク
によって割込まれた場合は、当該タスクの実行時間とそ
の上位タスクの処理期間との和である。従って当該タス
クの実行時間を求めるには、当該タスクの処理期間から
当該タスクに割込んだ上位タスクの処理期間を減算しな
ければならない。 同様に当該タスクが、下位タスクに割込んで実行された
場合、下位タスクの実行時間を求めるには、下位タスク
の処理期間から当該タスクの処理期間を減算しなければ
ならない。 そこで各タスク毎にその実行時間を記憶させる実行時間
積算領域を設け、タスクの終了のつど前記追加処理によ
って、当該タスクの処理期間を対応する実行時間積算領
域へ加算すると共に、当該タスクが割込んだ下位タスク
の有無を調べ、その下位タスクが有ればその下位タスク
の実行時間積算領域から当該タスクの処理期間を減算す
るようにする。
First, the processing period from the execution start time to the end time of the task is determined by additional processing added before and after each task. If the task is interrupted by a higher-level task during execution, this processing period is the sum of the execution time of the task and the processing period of the higher-level task. Therefore, in order to obtain the execution time of the task, the processing period of the higher-level task that interrupted the task must be subtracted from the processing period of the task. Similarly, if the task is executed by interrupting a subordinate task, the processing period of the task must be subtracted from the processing period of the subordinate task to obtain the execution time of the subordinate task. Therefore, an execution time accumulation area is provided for each task to store its execution time, and each time a task ends, the processing period of the task is added to the corresponding execution time accumulation area by the additional process, and when the task is interrupted, The presence or absence of a subordinate task is checked, and if there is a subordinate task, the processing period of the task is subtracted from the execution time accumulation area of the subordinate task.

【実施例】【Example】

以下第1図ないし第4図に基づいて本発明の詳細な説明
する。 第1図は本発明の実施例としてのタスクの実行時間を求
める処理のフローチャートを示す。即ち本発明では本来
のタスクの処理POの前後にそれぞれ実行時間算出のた
めの追加処理PI、P2を行っている。 第2図はO8によるタスク起動の例を示す、この例では
タスク実行の優先レジスタの高さは、タスク1〉タスク
2とする。即ち時刻T1にタスクlの処理が開始され、
時刻T2にこのタスクlの処理が終了する。次に時刻T
3においてタスク2の処理が開始されるが、この処理の
途中の時刻T4にタスク1の処理が割込によって開始さ
れ、時刻T5にこのタスク1の処理が終り、ここでタス
ク2の処理が再開され、時刻T6でタスク2の処理が終
了するものとする。 第3図は第2図に示すタスク起動例の処理の流れを示し
、第4図は第2図のタスク起動例における、追加処理に
よるデータのメモリM上への書込の様子を示す。 次に第3図の具体的な処理内容を第1図、第2図、第4
図を参照しつつ説明する。 今、第2図のようにタスクがO3により起動されたとす
る。なお以下の文章中の■から@までの数字は第3図中
のステップ番号に対応している。 同様に以下の81〜S4の符号は第1図の追加処理P2
中のステップを示す。 まず、■タスク1がO3により起動されると、タスク1
は、■第1図の追加処理P1を実行する。 即ち図外のシステムクロックを読取り、第4図のように
タスク起動時刻TIをメモリM上の領域alに一時記憶
する。その後、■タスクlによる本来の処理(第1図中
PO)を行い、その処理が終了した後、■第1図の追加
処理P2を実行する。 即ちシステムクロックを読取り(Sl)、このタスク終
了時刻T2と先にメモリM上の領域alに一時記tαさ
れたタスク起動時刻T1とから、タスクlの実行時間(
T2−TI)を算出し、メモリM内のタスク1実行時間
積算領域aloに加算する(S2)。さらに、O8から
の情報をもとにタスク1によって割込まれた相手がない
ことを判断しくS31分岐N)、追加処理P2を終了す
る。 次に、■タスク2がO8により起動されると、前述のタ
スク1の時と同様に、■まず追加処理Plを実行する。 即ちシステムクロックを読取り、第4図のようにタスク
起動時間T3をメモリM上の領域a2に一時記憶した後
、■タスク2の本来の処理POを開始する。■タスク2
の起動中にタスクlが割込むと、■まず、この割込処理
(タスク1の処理)についての追加処理P1を実行する
。 即ちシステムクロックを読取り、第4図のようにこのと
きのタスク起動時間T4をメモリM上の領域a1に一時
記憶する。続いて[相]タスク1の本来の処理POを実
行する。この処理が終了した後、■タスクlについての
追加処理P2を実行する。 即チタスク1はシステムクロックを読取り(31)、こ
のタスク終了時間T5と先にメモリM上の領域a1に一
時記憶したタスク起動時間T4とからタスクlの実行時
間(T5−74)を算出して、第4図のようにメモリM
内のタスクl実行時間積算領域aloに加算する(S2
)。さらに、O8からの情報をもとに、自分によって割
込まれた相手がタスク2であることを判断しくS39分
岐Y)、タスク2実行時間積算領域a20から自分自身
の実行時間(T5−T4)を減算する(S4)。以上、
■から■までの割込処理の後に、@タスクlに割込まれ
中断していたタスク2が処理を再開し、タスク2本来の
処理POを終了後、■タスク2の追加処理P2を実行す
る。即ちタスク2はシステムクロックを読取り(Sl)
、このタスク終了時刻T6とメモリ上の領域a2に一時
記憶したタスク起動時刻T3とから、タスクlの実行時
間を含んだ実行時間(つまり処理時間)(T6−T3)
を算出する。さらにその実行時間(T6−T3)をメモ
リ内のタスク2実行時間積算領域a20に加算する(S
2)。この実行時間(T6−T3)の加算処理を行うこ
とによりメモリMのタスク2実行時間積算領域a20に
は、割込んだタスク1の実行時間を含まないタスク2の
正味の実行時間((T6−T3)−(T5−74))が
記憶されることになる。 以上の処理PL、P2を、CPUによって起動される各
タスクに対して追加することにより、各タスクの実行時
間を求める。 この方式では、本来のタスクの処理POの前後に追加処
理PI、P2が行われるが、追加処理P1゜P2の実行
時間はタスクの本来の処理時間に対して充分小さ(、タ
スクの実行時間測定においては誤差範囲として扱える。 【発明の効果] 本発明によれば、システムクロックを持ち、上位のタス
クを優先しながら複数のタスクを並行して実行し得る計
算機システムにおける、前記タスク毎の実行時間を測定
する方法において、前記タスク毎に実行時間積算領域(
alO,a20など)を設け、 各タスクの前に、当該タスクの起動時刻を前記システム
クロックから読取って記憶する第1の追加処理P1を追
加すると共に、 同じく各タスクの後に、当該タスクの終了時刻を前記時
計から読取り、この終了時刻と当該タスクの前記起動時
刻との差から当該タスクの処理期間を求めて、この処理
期間を当該タスクの前記実行時間積算領域へ加算し、さ
らに当該タスクが割込んだ下位タスクの有無を調べ、該
下位タスクが有るきとは、この下位タスクについての前
記実行時間積算領域から前記処理時間を減算する第2の
追加処理P2を付加し、 前記実行時間積算領域の格納値を当該タスクの実行時間
とするようにしたので、 従来の技術では不可能であったO3により起動されるタ
スクごとの実行時間の算出を行うことが可能となる。さ
らに、このタスクごとの実行時間の算出のために、従来
の技術のように、空き時間測定用プログラムのような、
事前に処理時間を測定しておかなくてはならないプログ
ラムを新たに設けるという必要はない。
The present invention will be explained in detail below based on FIGS. 1 to 4. FIG. 1 shows a flowchart of a process for determining the execution time of a task as an embodiment of the present invention. That is, in the present invention, additional processing PI and P2 for calculating the execution time are performed before and after the original task processing PO. FIG. 2 shows an example of task activation by O8. In this example, the heights of the task execution priority registers are task 1>task 2. That is, the processing of task l is started at time T1,
The processing of this task 1 ends at time T2. Next time T
3, the processing of task 2 is started, but the processing of task 1 is started by an interrupt at time T4 in the middle of this processing, the processing of task 1 is finished at time T5, and the processing of task 2 is restarted here. It is assumed that the processing of task 2 is completed at time T6. FIG. 3 shows the flow of processing in the task activation example shown in FIG. 2, and FIG. 4 shows how data is written into the memory M by additional processing in the task activation example shown in FIG. Next, the specific processing contents in Figure 3 are explained in Figures 1, 2, and 4.
This will be explained with reference to the figures. Now, assume that the task is started by O3 as shown in FIG. Note that the numbers from ■ to @ in the text below correspond to the step numbers in FIG. 3. Similarly, the symbols 81 to S4 below are the additional processing P2 in FIG.
Showing the steps inside. First, ■ When task 1 is started by O3, task 1
(1) Executes additional processing P1 in FIG. That is, the system clock (not shown) is read and the task start time TI is temporarily stored in the area al on the memory M as shown in FIG. Thereafter, (1) the original processing by task l (PO in FIG. 1) is performed, and after that processing is completed, (2) additional processing P2 in FIG. 1 is executed. That is, the system clock is read (Sl), and the execution time of task l (
T2-TI) is calculated and added to the task 1 execution time accumulation area alo in the memory M (S2). Furthermore, based on the information from O8, it is determined that there is no other party interrupted by task 1 (S31 branch N), and the additional process P2 is ended. Next, (1) When task 2 is activated by O8, (1) first executes additional processing Pl, as in the case of task 1 described above. That is, after reading the system clock and temporarily storing the task start time T3 in the area a2 of the memory M as shown in FIG. 4, the original processing PO of task 2 is started. ■Task 2
When task l interrupts during startup, (1) First, additional processing P1 regarding this interrupt processing (processing of task 1) is executed. That is, the system clock is read and the task activation time T4 at this time is temporarily stored in the area a1 on the memory M as shown in FIG. Next, [phase] the original processing PO of task 1 is executed. After this processing is completed, ① additional processing P2 for task l is executed. In other words, task 1 reads the system clock (31) and calculates the execution time (T5-74) of task l from this task end time T5 and the task start time T4 temporarily stored in area a1 on memory M. , as shown in Figure 4, the memory M
(S2
). Furthermore, based on the information from O8, it is determined that the other party interrupted by the task is task 2 (S39 branch Y), and the own execution time (T5-T4) is calculated from the task 2 execution time accumulation area a20. (S4). that's all,
After the interrupt processing from ■ to ■, task 2, which was interrupted by @task l, resumes processing, and after completing task 2's original processing PO, ■ executes additional processing P2 of task 2. . That is, task 2 reads the system clock (Sl)
, from this task end time T6 and the task start time T3 temporarily stored in area a2 on the memory, the execution time (that is, processing time) including the execution time of task l (T6-T3)
Calculate. Furthermore, the execution time (T6-T3) is added to the task 2 execution time accumulation area a20 in the memory (S
2). By performing the addition process of this execution time (T6-T3), the task 2 execution time accumulation area a20 of memory M stores the net execution time of task 2 ((T6- T3)-(T5-74)) will be stored. By adding the above processes PL and P2 to each task activated by the CPU, the execution time of each task is determined. In this method, additional processing PI and P2 are performed before and after the original task processing PO, but the execution time of additional processing P1 and P2 is sufficiently small compared to the original processing time of the task (the task execution time measurement [Effects of the Invention] According to the present invention, the execution time of each task in a computer system that has a system clock and can execute multiple tasks in parallel while giving priority to higher-ranking tasks. In the method of measuring the execution time accumulation area (
alO, a20, etc.), and before each task, add a first additional process P1 that reads and stores the start time of the task from the system clock, and also after each task, add the end time of the task. is read from the clock, the processing period of the task is determined from the difference between the end time and the start time of the task, this processing period is added to the execution time accumulation area of the task, and the task is The presence or absence of a sub-task that has entered the sub-task is checked, and if the sub-task is present, a second additional process P2 is added to subtract the processing time from the execution time accumulation area for this sub-task; Since the stored value of is set as the execution time of the task, it becomes possible to calculate the execution time of each task activated by O3, which was impossible with the conventional technology. Furthermore, in order to calculate the execution time for each task, as in the conventional technology, a free time measurement program, etc.
There is no need to create a new program whose processing time must be measured in advance.

【図面の簡単な説明】[Brief explanation of drawings]

第1図は本発明の実施例としてのタスク実行時間を求め
る処理のフローチャート、 第2図はタスクの実行例を示すタイムチャート、第3図
は第2図に対応する本発明の処理の流れを示すフローチ
ャート、 第4図は第3図の処理におけるメモリへのデータ書込の
実施例を示す図である。 PO:タスク本来の処理、Pl、P2 :追加処理、M
:メモリ、al :タスクl起動時刻記憶領域、a2 
:タスク2起動時刻記憶領域、alO:タスク1実行時
間積算領域、a20:タスク2実行時間積牙 図 タスク1の実行時間 :  (T2− TI) + (
T5−74)タスク2の実行時間 =(Tε−73) 
−(T5−74)第 図
FIG. 1 is a flowchart of a process for determining task execution time as an embodiment of the present invention, FIG. 2 is a time chart showing an example of task execution, and FIG. 3 is a flowchart of the process of the present invention corresponding to FIG. FIG. 4 is a diagram showing an example of writing data to memory in the process of FIG. 3. PO: original processing of the task, Pl, P2: additional processing, M
:Memory, al :Task l startup time storage area, a2
: Task 2 startup time storage area, alO: Task 1 execution time accumulation area, a20: Task 2 execution time accumulation diagram Task 1 execution time: (T2-TI) + (
T5-74) Task 2 execution time = (Tε-73)
-(T5-74)Figure

Claims (1)

【特許請求の範囲】 1)時計を持ち、上位のタスクを優先しながら複数のタ
スクを並行して実行し得る計算機システムにおける、前
記タスク毎の実行時間を測定する方法において、 前記タスク毎に実行時間積算領域を設け、 各タスクの前に、当該タスクの起動時刻を前記時計から
読取って記憶する第1の追加処理を付加すると共に、 同じく各タスクの後に、当該タスクの終了時刻を前記時
計から読取り、この終了時刻と当該タスクの前記起動時
刻との差から当該タスクの処理期間を求めて、この処理
期間を当該タスクの前記実行時間積算領域へ加算し、さ
らに当該タスクが割込んだ下位タスクの有無を調べ、該
下位タスクが有るときは、この下位タスクについての前
記実行時間積算領域から前記処理期間を減算する第2の
追加処理をを付加し、 前記実行時間積算領域の格納値を当該タスクの実行時間
とするようにしたことを特徴とするタスク実行時間測定
方法。
[Claims] 1) A method for measuring the execution time of each task in a computer system that has a clock and can execute multiple tasks in parallel while giving priority to higher-ranking tasks, comprising: A time accumulation area is provided, and before each task, a first additional process is added that reads and stores the start time of the task from the clock, and also after each task, the end time of the task is read from the clock. Read, calculate the processing period of the task from the difference between this end time and the start time of the task, add this processing period to the execution time accumulation area of the task, and further calculate the processing period of the task that was interrupted by the lower task. If the lower task exists, add a second additional process to subtract the processing period from the execution time accumulation area for this lower task, and set the stored value in the execution time accumulation area to the corresponding A task execution time measuring method characterized in that the task execution time is taken as the task execution time.
JP2005123A 1990-01-13 1990-01-13 Task executing time measuring method Pending JPH03210643A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2005123A JPH03210643A (en) 1990-01-13 1990-01-13 Task executing time measuring method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2005123A JPH03210643A (en) 1990-01-13 1990-01-13 Task executing time measuring method

Publications (1)

Publication Number Publication Date
JPH03210643A true JPH03210643A (en) 1991-09-13

Family

ID=11602540

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2005123A Pending JPH03210643A (en) 1990-01-13 1990-01-13 Task executing time measuring method

Country Status (1)

Country Link
JP (1) JPH03210643A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002006949A3 (en) * 2000-07-15 2002-12-27 Bosch Gmbh Robert Method and device for measuring the execution time of a task in a real time system
JP2006260542A (en) * 2005-03-17 2006-09-28 Microsoft Corp Determination of actual amount of time a processor consumes in executing code portion
WO2008072674A1 (en) 2006-12-12 2008-06-19 International Business Machines Corporation Detection device, program, and detection method
US8140915B2 (en) 2007-03-12 2012-03-20 International Business Machines Corporation Detecting apparatus, system, program, and detecting method
JP2012141864A (en) * 2011-01-05 2012-07-26 Hitachi Ltd Method for calculating processor activity ratio in smt processor

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2002006949A3 (en) * 2000-07-15 2002-12-27 Bosch Gmbh Robert Method and device for measuring the execution time of a task in a real time system
JP2006260542A (en) * 2005-03-17 2006-09-28 Microsoft Corp Determination of actual amount of time a processor consumes in executing code portion
WO2008072674A1 (en) 2006-12-12 2008-06-19 International Business Machines Corporation Detection device, program, and detection method
US8271981B2 (en) 2006-12-12 2012-09-18 International Business Machines Corporation Detecting an extraordinary behavior
US8140915B2 (en) 2007-03-12 2012-03-20 International Business Machines Corporation Detecting apparatus, system, program, and detecting method
JP2012141864A (en) * 2011-01-05 2012-07-26 Hitachi Ltd Method for calculating processor activity ratio in smt processor

Similar Documents

Publication Publication Date Title
US9218209B2 (en) Apparatus and method for parallel processing of a query
KR0138468B1 (en) Microcomputer
US7844971B2 (en) Method and apparatus for detecting cross-thread stack access in multithreaded programs
EP2176763B1 (en) Memory transaction grouping
US20060004882A1 (en) Custom Atomic Transactions in Programming Environments
US7779393B1 (en) System and method for efficient verification of memory consistency model compliance
JP2010510590A (en) Managing software transaction commit order and conflicts
JP2011501323A (en) How to switch threads in a routine
TW200903339A (en) Parallelizing sequential frameworks using transactions
US8856753B2 (en) Granular measurement of processor performance in executing sections of software code
EP2171589B1 (en) Transactional debugger for a transactional memory system
JP2008310748A (en) Task execution time recording device, task execution time recording method, and task execution recording program
US8359572B2 (en) Self-describing re-usable software components
CN113360270A (en) Data cleaning task processing method and device
JPH03210643A (en) Task executing time measuring method
CN115905040B (en) Counter processing method, graphics processor, device and storage medium
US9135058B2 (en) Method for managing tasks in a microprocessor or in a microprocessor assembly
US8196123B2 (en) Object model for transactional memory
US20060064671A1 (en) Creating and using a building block
JPH0973407A (en) Data management device
CN116185940B (en) Atomic counter operation method, device, equipment and storage medium
JPS62245443A (en) Measuring method for execution time of program
JPS59142652A (en) Managing system of elapsed time
CN117033497B (en) Method for updating and reading data on data lake and related equipment
US10185569B2 (en) Precise-restartable parallel execution of programs