JPH0635738A - Data processor - Google Patents

Abstract

PURPOSE:To improve the overhead of a diagnostic device by easily monitoring the retry of a processor by the diagnostic device by using a hardware. CONSTITUTION:A diagnostic device 1 is equipped with a diagnostic bus control part 6 which sets and starts a retry monitoring timer 3 of a processor 2 through diagnostic bus control parts 6 and 5, and an interruption processing part 8 which operates an interruption processing from the retry monitoring timer 3. And also, a microcommand execution control part 4 of the processor 2 is equipped with a function which stops the retry monitoring timer 3, and a clock control part 7 stops the clock of the processor 2 when the retry monitoring timer 3 is time-out.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention comprises a data processing device, and more particularly, a diagnostic device and a processor. The diagnostic device has means for activating a retry of the processor when a fault occurs in the processor. The present invention relates to a data processing device that retries execution instructions.

[0002]

2. Description of the Related Art In a conventional data processing apparatus of this type, the monitoring of the processor when it is retried is realized by the firmware of the diagnostic apparatus.

[0003]

In this conventional data processing device, since the processor retry is monitored by the firmware of the diagnostic device, the diagnostic device stores the processor that initiated the retry and monitors the time. Had to be prepared for a plurality of processors, resulting in a complicated program, and the overhead of the monitoring of the retry sometimes caused a problem in the performance of the diagnostic device.

[0004]

The apparatus of the present invention comprises a diagnostic device and a processor, and the fault is detected by the diagnostic device having means for activating the retry of the processor when the fault of the processor occurs. In a data processing device that retries an execution instruction in a phase, the processor has a retry monitoring timer, means for activating the retry monitoring timer when the diagnostic device restarts the retry, and when the retry is successful, A means for stopping the retry monitoring timer by a microinstruction executed in a processor, and means for stopping the processor when the retry monitoring timer times out, such as when the retry fails, are provided. It is characterized by monitoring.

[0005]

Embodiments of the present invention will now be described with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of the first embodiment of the present invention. In FIG. 1, the diagnostic device 1 is connected to the processor 2 by a diagnostic bus control unit 6, and an interrupt processing unit 8 for processing an interrupt from the processor 2 is prepared.

Of the processor 2, the diagnostic bus control unit 5 connected to the diagnostic bus control unit 6 of the diagnostic device 1 issues a command to each control unit in the processor 2 according to a command from the diagnostic device 1. The microinstruction execution control unit 4 outputs the decode signal of the retry monitoring end instruction to the retry monitoring timer 3.

The retry monitoring timer 3 stops the time count by the retry monitoring end instruction decode signal from the micro instruction execution control unit 4, and outputs a time-out signal to the clock control unit 7 when a time-out occurs.
The clock controller 7 stops the clock of the processor 2 by the time-out signal from the retry monitoring timer 3 and outputs a clock stop interrupt signal to the diagnostic device 1.

Next, the overall operation will be described. Diagnostic device 1
Issues a retry activation command to the processor 2 by the diagnostic bus control unit 6. At that time, the diagnostic device 1 stores that the processor 2 has activated the retry.

When the diagnostic bus control unit 5 of the processor 2 receives the retry activation command, it sends the activation of the retry to the microinstruction execution control unit 4 and the retry monitoring timer 3. The microinstruction execution control unit 4 retries in response to a retry activation command from the diagnostic bus control unit 5, and the retry monitoring timer 3 resets the count value and starts counting.

Upon successful retry, the processor 3 stops the retry watch timer 3 by executing a microinstruction retry watch end. Processor 2 if retry fails
The microinstruction retry monitoring is not executed, and the retry monitoring timer 3 sends a time-out signal to the clock control unit 7, whereby the clock control unit 7 stops the clock of the processor 2 and issues a clock stop interrupt to the diagnostic device 1. Occur.

The diagnostic device 1 recognizes that a retry has been performed in the past in the processing of the clock stop interrupt from the processor 2, and disconnects the processor 2 from the system.

FIG. 2 shows a detailed view of the retry monitoring timer 3 in FIG. The retry monitoring timer 3 is composed of a timer control unit 9, a timer unit 10 and a time value holding unit 11.

The timer control unit 9 is connected to the diagnostic bus control unit 5 and sets a time value in the time value holding unit 11 or starts the count of the timer unit 10 according to an instruction from the diagnostic bus control unit 5 to execute a microinstruction. The control unit 4 causes the timer unit 10 to stop counting.

The timer unit 10 fetches the count value from the time value holding unit 11 when the timer control unit 9 starts counting and starts counting, and outputs a timeout signal to the clock control unit 7 when counting out. The diagnostic device 1 activates the retry after setting the monitoring time in the retry monitoring timer 3 of the processor 2 in the processor 2 via the diagnostic bus control unit 6. As described above, the monitoring time can be made variable.

FIG. 3 is a block diagram showing a second embodiment of the present invention. The same components as those in FIG. 1 are designated by the same reference numerals. In this embodiment, the retry monitoring timer 3
From the clock control unit 7 to the diagnostic bus control unit 5 so that the diagnostic device 1 can detect the time.
Also, the time-out signal of the retry monitoring timer 3 can be read out via the diagnostic bus control unit 5.

In this embodiment, the diagnostic device 1 does not store the restart of the processor 2 in the first embodiment when issuing the retry start instruction to the processor 2. The operation after the retry activation until the clock control unit 7 generates the clock stop interrupt to the diagnostic device 1 is the same as that described in the first embodiment. The diagnostic device 1 reads the retry monitoring timer 3 and the timeout signal via the diagnostic bus control unit and the diagnostic bus control unit 5 in the processing of the clock stop interrupt by the clock control unit 7 of the processor 2, and the timeout has occurred. Sometimes processor 2 is disconnected from the system.

FIG. 4 is a block diagram showing a third embodiment of the present invention. Also in the present embodiment, common reference numerals are given to constituent elements common to FIG. In this embodiment, the timeout signal from the retry monitoring timer 3 to the clock controller 7 is used as the interrupt signal to the diagnostic device 1.

Also in this embodiment, the diagnostic device 1
Does not store the restart in the processor 2 as in the first embodiment when issuing the restart start instruction to the processor 2. After the retry activation, the retry monitoring timer 3 times out and sends a time-out signal to the clock control unit 7, so that the operation until the clock control unit 7 stops the clock of the processor 2 will be described in the first embodiment. Is the same as. The diagnostic apparatus 1 receives the timeout signal of the retry monitoring timer 3 in the interrupt processing unit 8, recognizes the retry failure due to the interrupt factor, and disconnects the processor 2 from the system.

[0020]

As a result of the configuration as described above, the present invention can simplify the monitoring of retries by the diagnostic device, eliminate the overhead even with a plurality of processors, improve the performance, and improve the performance of each processor. Detailed monitoring is possible without burdening the diagnostic device by changing the time value.

[Brief description of drawings]

FIG. 1 is a block diagram of a first embodiment of the present invention.

FIG. 2 is a detailed diagram of a retry monitoring timer in FIG.

FIG. 3 is a block diagram of a second embodiment of the present invention.

FIG. 4 is a block diagram of a third embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 diagnostic device 2 processor 3 retry monitoring timer 4 micro instruction execution control unit 5 diagnostic bus control unit 6 diagnostic bus control unit 7 clock control unit 8 interrupt processing unit 9 timer control unit 10 timer unit 11 time value holding unit

Claims (4)

[Claims] 1. A diagnostic device and a processor, wherein the diagnostic device has means for activating a retry of the processor when a fault occurs in the processor, thereby retrying an execution instruction in a phase in which a fault is detected. In the data processing device, the processor includes a retry monitoring timer, a means for starting the retry monitoring timer when the diagnostic device restarts the retry, and a micro instruction executed in the processor when the retry is successful. A data processing device comprising means for stopping a retry monitoring timer and means for stopping the processor when the retry monitoring timer times out, such as when the retry fails, and for monitoring the retry of the processor. . 2. The data processing device according to claim 1, wherein the diagnostic device has means for setting an arbitrary value in the retry monitoring timer. 3. The data processing apparatus according to claim 1, further comprising means for the diagnostic apparatus to read out the timeout of the retry monitoring timer. 4. The data processing apparatus according to claim 1, further comprising means for interrupting the timeout of the retry monitoring timer to the diagnostic apparatus.