KR101476078B1 - Data returned responsive to executing a start subchannel instruction - Google Patents

Abstract

There is provided a method of executing a Start Subchannel command in a computing environment that includes main storage and storage class memory, the method comprising: determining, responsive to determining that the subchannel identified by the Start Subchannel command is an asynchronous data mover (ADM) Performing the following steps: obtaining an operation request block from main storage, the operation request block including an address of an operation block; Obtaining an operation block from the main storage based on an address of the operation block, the operation block comprising a request block, a response block, and one or more first movement specification blocks (MSBs) Wherein the request block comprises an MSB count field, the MSB count field having a value indicating the number of one or more MSBs referenced by it included in the operation block, the response block being configured to hold exception conditions Each move specification block is configured to include an operation code field, a block count field, a main storage address field for main storage addressing, a storage class memory address field for storage class memory addressing, a block size field and flags field, ; Obtaining a move specification block of the one or more move specification blocks; Determining an operation configured to perform the obtained move specification block on blocks of storage class memory based on an operation code in the obtained move specification block; Initiating the operation on blocks of storage class memory associated with a storage class memory address of the storage class memory address field based on the move specification block; And storing information in the response block in response to initiating at least the operation and meeting an exception condition, the response block including an exception flag field, a field validity field, a failed MSB address field, a failed AIDAW ) Address field, a failure main storage address field, and a failure storage class memory address field, said information being associated with one or more exceptions resulting from said operation, said one or more exceptions being: A storage class memory associated with a storage class memory address of the storage class memory address field, or the obtained move specification block, the information being associated with an indication of an exception and an exception And related Includes performing the - and include additional information, said additional information is also express one or more address exception.

Description

Data returned in response to executing the command {DATA RETURNED RESPONSIVE TO EXECUTING A START SUBCHANNEL INSTRUCTION}

One or more embodiments of the present invention generally relate to secondary storage of a computing environment, and more particularly to managing secondary storage.

The computing environment may include main storage (also known as main memory) and secondary storage. Main storage is a storage accessible by the processor, for example, addressable randomly by an absolute address. Main storage is considered faster access storage than secondary storage such as direct access storage devices (DASD) or storage class memory. Also, the addressing of main storage is considered to be simpler than the addressing of DASD or storage class memory.

Storage class memory is an external storage space outside the classical main storage, providing faster access than direct access storage devices. Unlike DASD, storage class memories are not typically implemented as mechanical-arm spinning disks, but rather are implemented as non-mechanical solid-state parts. Typically, the storage class memory is a solid state that is coupled to the computing system via various input / output (I / O) adapters, which are used to map the description of the I / O device to the memory bus Lt; / RTI > devices.

Disadvantages of the prior art are solved and advantages are provided through the provision of a computer program product that executes a Start Subchannel command in a computing environment including main storage and storage class memory. The computer program product includes a computer readable storage medium having instructions readable by the processing circuitry and storing instructions to be executed by the processing circuitry to perform any method. The method may comprise performing the following steps in response to, for example, determining that the subchannel identified by the Start Subchannel command is an asynchronous data mover (ADM) subchannel: Obtaining an operation request block from the main storage, the operation request block including an address of an operation block; Obtaining an operation block from the main storage based on an address of the operation block, the operation block comprising a request block, a response block, and one or more first movement specification blocks (MSBs) Wherein the request block comprises an MSB count field, the MSB count field having a value indicating the number of one or more MSBs referenced by it included in the operation block, the response block being configured to hold exception conditions Each move specification block is configured to include an operation code field, a block count field, a main storage address field for main storage addressing, a storage class memory address field for storage class memory addressing, a block size field and flags field, ; Obtaining a move specification block of the one or more move specification blocks; Determining an operation configured to perform the obtained move specification block on blocks of storage class memory based on an operation code in the obtained move specification block; Initiating the operation on blocks of storage class memory associated with a storage class memory address of the storage class memory address field based on the move specification block; And storing information in the response block in response to initiating at least the operation and meeting an exception condition, the response block including an exception flag field, a field validity field, a failed MSB address field, a failed AIDAW ) Address field, a failure main storage address field, and a failure storage class memory address field, said information being associated with one or more exceptions resulting from said operation, said one or more exceptions being: A storage class memory associated with a storage class memory address of the storage class memory address field, or the obtained move specification block, the information being associated with an indication of an exception and an exception And related Includes performing the - and include additional information, said additional information is also express one or more address exception.

Methods and systems related to one or more embodiments of the invention are also described and claimed herein. Also, services related to one or more embodiments of the present invention are also described and claimed herein.

Additional features and advantages are implemented through the techniques of the present invention. Other embodiments and embodiments of the invention are described in detail herein and are considered a part of the claimed invention.

Preferred embodiments of the present invention will now be described, by way of example only, with reference to the following drawings.
FIG. 1A illustrates one embodiment of a computing environment for implementing and using one or more embodiments of the present invention.
Figure IB illustrates another embodiment of a computing environment for implementing and using one or more embodiments of the present invention.
FIG. 2A illustrates an operation request block of an Extended Asynchronous Data Mover used in accordance with an embodiment of the present invention.
Figure 2B illustrates an embodiment of an operation block of an extended asynchronous data mover used in accordance with an embodiment of the present invention.
2C illustrates an embodiment of a request block of an extended asynchronous data mover used in accordance with an embodiment of the present invention.
Figure 2D illustrates one embodiment of a response block of an extended asynchronous data mover used in accordance with an embodiment of the present invention.
Figure 2E illustrates one embodiment of a move specification block of an extended asynchronous data mover used in accordance with one embodiment of the present invention.
2F illustrates an embodiment of an indirect data address word of an extended asynchronous data mover used in accordance with an embodiment of the present invention.
3A shows an embodiment of a Start Subchannel command used in accordance with an embodiment of the present invention.
Figure 3B illustrates one embodiment of logic associated with a Start Subchannel instruction in accordance with an embodiment of the present invention.
3C illustrates an embodiment of a Clear Subchannel command used in accordance with an embodiment of the present invention.
FIG. 3D illustrates one embodiment of a Test Subchannel command used in accordance with one embodiment of the present invention.
Figure 3E illustrates one embodiment of a Modify Subchannel command used in accordance with an embodiment of the present invention.
Figure 3F illustrates one embodiment of a Store Subchannel command used in accordance with one embodiment of the present invention.
4A illustrates an embodiment of a subchannel information block of an extended asynchronous data mover used in accordance with an embodiment of the present invention.
4B illustrates an embodiment of a path management control word of an extended asynchronous data mover used in accordance with an embodiment of the present invention.
4C illustrates an embodiment of a subchannel status word used in accordance with an embodiment of the present invention.
4D illustrates one embodiment of an extended status word of an extended asynchronous data mover used in accordance with an embodiment of the present invention.
FIG. 5A illustrates an embodiment of a request block of Store Storage Class Memory Information (store storage class memory information) used in accordance with an embodiment of the present invention.
FIG. 5B illustrates an embodiment of a response block of Store Storage Class Memory Information used in accordance with an embodiment of the present invention.
5C illustrates one embodiment of a storage class memory address list entry used in accordance with an example embodiment of the present invention.
5D illustrates one embodiment of the logic associated with the Store Storage Class Memory Information command, in accordance with one embodiment of the present invention.
6A illustrates an embodiment of a state diagram illustrating various states of a storage class memory, in accordance with an embodiment of the present invention.
6B illustrates an embodiment of a state diagram illustrating operation states and data states, in accordance with an embodiment of the invention.
Figure 7A illustrates one embodiment of a request block of Configure Storage Class Memory (configure storage class memory) used in accordance with an embodiment of the present invention.
Figure 7B illustrates one embodiment of a response block of Configure Storage Class Memory used in accordance with an embodiment of the present invention.
Figures 7C-7D illustrate one embodiment of logic associated with the Configure Storage Class Memory command used in accordance with an embodiment of the present invention.
8A shows an example of a notification response block used in accordance with an embodiment of the present invention.
FIG. 8B shows an embodiment of a request block of Store Event Information (store event information) used in accordance with an embodiment of the present invention.
8C illustrates an embodiment of a response block of Store Event Information used in accordance with an embodiment of the present invention.
FIG. 9A illustrates an embodiment of a request block of a Deconfigure Storage Class Memory (unconfigure storage class memory) used in accordance with an embodiment of the present invention.
Figure 9B illustrates one embodiment of an increment request list entry of a storage class memory used in accordance with an embodiment of the invention.
FIG. 9C illustrates an embodiment of a response block of a Deconfigure Storage Class Memory used in accordance with an embodiment of the present invention.
Figures 9D-9E illustrate one embodiment of logic associated with a Deconfigure Storage Class Memory command used in accordance with an embodiment of the present invention.
Figure 10 illustrates an embodiment of a computer program product embodying one or more embodiments of the present invention.
Figure 11 illustrates one embodiment of a host computer system for implementing and using one or more embodiments of the present invention.
Figure 12 illustrates a further example of a computer system for implementing and using one or more embodiments of the present invention.
Figure 13 illustrates another example of a computer system including a computer network for implementing and using one or more embodiments of the present invention.
Figure 14 illustrates one embodiment of various elements of a computer system for implementing and using one or more embodiments of the present invention.
Figure 15A illustrates an embodiment of an execution unit of the computer system of Figure 14 for implementing and using one or more embodiments of the present invention.
Figure 15B illustrates an embodiment of a branch unit of the computer system of Figure 14 for implementing and using one or more embodiments of the present invention.
Figure 15C illustrates one embodiment of a load / store unit of the computer system of Figure 14 for implementing and using one or more embodiments of the present invention.
Figure 16 illustrates one embodiment of an emulated host computer system for implementing and using one or more embodiments of the present invention.

According to one or more embodiments, an abstraction for a storage class memory is provided, which hides the details of the implementation of the storage class memory from a program (e.g., the operating system) and moves the data between the main storage and the storage class memory And provides a standard channel programming interface for performing specific actions such as control.

In one example, a facility called "Extended Asynchronous Data Mover (EADM) Facility" is provided, which allows programs to request the transfer of blocks of data between main storage and storage class memory , And also requests other operations. Embodiments of this facility are described in detail below.

In addition, an optional facility called EADM Release Facility is associated with the EADM Facility. Once installed, the EADM Release Facility provides a means for the program, which specifies that the program no longer requires retention of data in one or more blocks of storage class memory. Whether the release operation is supported for all blocks of storage class memory or only for a subset of storage class memory is model dependent.

Once a block of storage class memory is freed, the program can transfer new data to that block, which would otherwise not succeed in subsequent requests to transfer data from the block to main storage.

One embodiment of a computing environment for implementing and / or using one or more embodiments of the invention is described with reference to FIG. Computing environment 100 includes, for example, the New York State Amon greater material International Business Machines Corporation (International Business Machines Corporation) be based on the z / Architecture ^® is supplied by the (IBM ^®). Examples of the z / Architecture ^® are described in "z / Architecture Principles of Operation ( z / Architecture Operating Principle)" IBM ^® publication entitled IBM Publication No. SA22-7832-08 (August 2010), which The entirety of which is incorporated herein by reference. In one example, a computing environment based on the z / Architecture ^® include the zEnterprise 196 (zl96) from International Business Machines Corporation System for feeding. IBM ^® and z / Architecture ^® are registered trademarks, zEnterprise 196 and zl96 is a trademark of International Business Machines NY Amon larger material Corporation. Other names used herein may be registered trademarks, trademarks or product names of International Business Machines Corporation or other companies.

As one example, computing environment 100 includes one or more central processing units 102, which are coupled to main memory 104 via one or more buses 106. One or more of the central processing units may execute an operating system 108, such as the z / OS ( ^R) operating system supplied by International Business Machines Corporation. In other instances, one or more of the central processing units may execute other operating systems or may not be running an operating system. z / OS ^® is a registered trademark of International Business Machines NY Amon larger material Corporation.

Central processing units 102 and main memory 104 may be further coupled to I / O hub 120 via one or more connections 122 (e.g., buses or other connections). The I / O hub provides connections to one or more I / O adapters 130, which are further coupled to one or more solid state devices 140. These adapters and solid state devices are implementations of storage class memory (eg flash memory). The I / O hub is part of an I / O system 145 that facilitates implementation of one or more embodiments of the embodiment.

In a further embodiment, the computing environment 150, as shown in Figure 1b includes a central processing includes complex (CPC, 152), this be based on the z / Architecture ^® supplied by International Business Machines Corporation. The central processing complex 152 may include one or more partitions 154, a hypervisor 156, one or more central processors 158, and one of the input / output subsystems 160, for example. Or more components. In this example, partitions 154 are logical partitions (e. G., LPARs), each of which includes a set of system hardware resources virtualized into a separate system.

Each logical partition 154 is capable of functioning as a separate system. That is, each logical partition may be independently reset, loaded with the operating system at first, or loaded with other control codes, if desired, and may operate with other programs. An operating system or application running on a logical partition appears to be able to access a full and complete system, but only a subset of it is actually available. The combination of hardware and licensed internal code (LIC), called firmware, prevents a program in one logical partition from interfering with a program in another logical partition. This allows several different logical partitions to operate on a single or multiple physical processors in a time-slice manner. When used herein, the firmware includes, for example, microcode, millicode, and / or macrocode of a processor (or an entity that performs processing). The firmware includes, for example, hardware-level instructions and / or data structures used in the implementation of higher level machine code. In one embodiment, the firmware includes, for example, proprietary code that is typically transmitted in microcode, which includes trusted software or microcode specific to the underlying hardware And controls the operating system's access to system hardware.

In this example, some logical partitions have a resident operating system (OS) 170, which may be different in one or more logical partitions. In one embodiment, the at least one logical partition runs the z / OS ( ^R) operating system supplied by International Business Machines Corporation of Amonk, New York, USA.

The logical partitions 154 are managed by the hypervisor 156, which is implemented by firmware running on the central processors 158. Logical partitions 154 and hypervisor 156 each include one or more programs residing in respective portions of main memory 159 associated with central processors. An example of a hypervisor 156 is the Processor Resource / Systems Manager (PR / SM), which is supplied by International Business Machines Corporation of Amonk, NY.

Central processors 158 are physical processor resources that are allocated to logical partitions. For example, logical partition 154 includes one or more logical processors, each of which represents all or a share of physical processor resources 158 that are allocated to that partition . The logical processors of a particular partition 154 may be dedicated to that partition such that underlying processor resources are reserved for that partition or may be shared with other partitions such that the underlying processor resources are potentially available to other partitions. have.

The input / output subsystem 160 (only a portion of which is depicted) provides connectivity to the storage class memory 180. In this example, an address space is provided for the storage class memory, the storage class memory provides the memory flat, and obscures the details of the physical implementation from the program. As an example, there is one system-wide address space for storage class memory, but in terms of configuration (e.g., LPAR or, in another embodiment, virtualized guest) For each configuration, there is one address space occupied by storage class memory increments. The storage class memory address space is separate and separate from the main storage address space.

In one particular example of the z / Architecture ^®, I / O sub-system includes a channel subsystem, which, as well as the flow of information between the main memory, the I / O peripheral control units (and devices), storage class memory and It also controls the flow of information between main memory. However, the I / O subsystem may be different from the channel subsystem.

In the case of a channel subsystem, the subchannels are used to perform EADM operations. These subchannels are called asynchronous data mover (ADM) -type subchannels and are associated with EADM operations, and are not I / O devices such as other I / O-type subchannels. The ADM-type subchannel does not include the device number nor the channel path information. The number of ADM-type sub-channels provided for configuration is model-specific. The ADM-type subchannels are addressed by a subsystem identification word (SID).

The ADM-type subchannels are used by the Extended Asynchronous Data Mover Facility, which is an extension to the channel subsystem. As described herein, the EADM facility allows a program to request the transfer of data blocks between main memory and storage class memory, and also allows other operations such as clearing a block of storage class memory or releasing a block of storage class memory It also allows you to perform. In one embodiment, when the EADM facilities are installed, the following occurs:

하나 또는 그 이상의 ADM-타입 서브채널들이 제공되고 이들은 EADM 오퍼레이션들을 위해 사용된다.

One or more ADM-type subchannels are provided and they are used for EADM operations.

EADM 오퍼레이션들은 명시된(specified) EADM-오퍼레이션 블록(AOB)에 의해 지정된다(designated). AOB는 EADM-요청 블록(ARQB)와 EADM-응답 블록(ARSB)를 포함하며, EADM 이동-명세 블록들(MSB들)의 리스트를 지정한다. 이동 오퍼레이션을 위해, MSB들은 블록들의 사이즈, 블록들의 소스 위치(source location)와 목적지 위치(destination location), 데이터 이동 방향 등의 이동될 데이터의 블록들에 관한 정보를 포함한다.

EADM operations are designated by the specified EADM-operation block (AOB). The AOB includes an EADM-Request Block (ARQB) and an EADM-Response Block (ARSB) and specifies a list of EADM Move-Specification Blocks (MSBs). For a move operation, the MSBs contain information about the blocks of data to be moved, such as the size of the blocks, the source location and destination location of the blocks, and the direction of data movement.

The maximum number of MSBs that can be specified by the AOB depends on the model. The maximum count of the blocks to be moved or operated that the MSB can specify is also model dependent.

프로그램이 ADM-타입 서브채널과 EADM 오퍼레이션 요청 블록(ORB)를 지정하는 Start Subchannel 명령을 발행함으로써 EADM 오퍼레이션들을 개시한다. 이어서, EADM ORB가 AOB를 지정한다. 상기 명령은 EADM ORB의 콘텐츠들을 지정된 서브채널에 전달한다.

The program initiates EADM operations by issuing a Start Subchannel command specifying an ADM-type subchannel and an EADM operation request block (ORB). The EADM ORB then designates the AOB. The command conveys the contents of the EADM ORB to the designated subchannel.

Start Subchannel이 발행되어 EADM 오퍼레이션들을 개시하면, 채널 서브시스템은 명시된 오퍼레이션을 비동기적으로 수행한다.

When a Start Subchannel is issued and initiates EADM operations, the channel subsystem performs the specified operation asynchronously.

프로그램에 의해 관측될 때, 스토리지 클래스 메모리는 모델 종속적인 최소 블록 사이즈에서 동시처리 블록(block concurrent)으로 보여진다. 이 모델 종속 값은 SCM 블록 동시처리 사이즈(SCM block concurrency size)라 불린다.

When observed by the program, the storage class memory is viewed as a block concurrent at a model-dependent minimum block size. This model-dependent value is called the SCM block concurrency size.

EADM 오퍼레이션들이 완료되면, 상기 오퍼레이션들이 개시된 ADM-타입 서브채널을 위한 프로그램에 대해서 I/O 인터럽션이 펜딩된다.

Upon completion of the EADM operations, the I / O interruption is pending for the program for the ADM-type subchannel on which the operations are initiated.

The EADM ORB contains the specification of a subchannel key and the address of the AOB to be used. The AOB may be specified, for example, on a 4 K-byte boundary and up to 4 K bytes in size. If more MSBs are needed than fit for a 4K byte AOB, then the AOB can specify a list of MSBs that extend into additional storage areas, which instead of specifying the storage area to be used for data transfer, next) MSBs that specify the address of the MSB.

To fetch the MSB, the first EADM operation is initiated by the channel subsystem, using the information in the specified EADM ORB and the ARQB in the specified AOB. The MSB contains information that specifies and controls the EADM operations to be processed.

Each EADM operation is represented by one MSB. The MSB specifies, for example, the transfer of data blocks from main storage to storage class memory, the transfer of data blocks from storage class to main storage, the clearing of blocks of storage class memory, and the release of blocks of storage class memory. can do.

If the blocks of storage to be transferred are not contiguous in main storage, a new MSB is used, or the MSB specifies indirect addressing by specifying non-contiguous blocks by specifying a list of EADM indirect data address words (AIDAWs) Can be used.

Since the MSB specifies transmission of data in only one direction, the new MSB is used when there is a change in the direction of transmission.

A conclusion of the EADM operation is usually indicated by the combined condition conditions of the channel end and the device end. This combination of states indicates a combination of primary and secondary states indicating that the subchannel is available for another start function after the state is cleared.

The EADM operation may be terminated early by the Clear Subchannel command. Execution of the Clear Subchannel command terminates the execution of AOB in the subchannel, clears the AOB indications in the subchannel, and performs the clear function asynchronously.

Further details regarding the EADM ORB and related control structures are described below with reference to Figures 2a-2f. Initially, referring to FIG. 2A, one embodiment of an EADM ORB is described.

As an example, the EADM ORB 200 includes:

Interruption parameter 202: This field is kept unmodified in the subchannel until it is replaced by a subsequent Modify Subchannel or Start Subchannel command. These bits are used in I / O interruption codes when an I / O interruption occurs for a subchannel and when an interruption request is cleared, for example, by the execution of Test Pending Interruption .

Subchannel key 204: This field forms a subchannel key for the EADM operations specified by the ARQB and accesses the main storage for fetching ARQBs, fetching MSBs, storing ARSBs, and transferring data. . The value of this field is the defined value; Otherwise, the program check condition is recognized by the channel subsystem or an operand exception is recognized.

ORB Extension Control (X) (205): This field specifies whether the ORB is extended. This field is the value specified when the ADM-type subchannel is specified; Otherwise, an operand exception or program check condition is recognized.

EADM Operation Block (AOB) Address 206: This field specifies the address of the EADM Operation Block (AOB). If some of the bits in this field do not contain a defined value, an operand exception or program check condition is recognized.

If this field specifies a protected location for fetching or a location outside the configuration, the start function is not started. In this case, the subchannel is status pending with primary, secondary, and alert status.

Channel Subsystem (CSS) Primality 208: This field is assigned to a designated subchannel when the start function for one or more subchannels that are starting pending is to be initiated, And a channel subsystem priority number used to command the selection of the channel subsystem.

Storage Class Memory (SCM) Priority 210: This field contains a Storage Class Memory (SCM) priority number that specifies the priority level applied to all EADM operations associated with the start function.

Whether the contents of the SCM priority field are recognized by EADM facilities depends on the model. In models that do not recognize this field, the field contents are ignored and all EADM operations associated with the start function are assigned an implicit priority number.

Format (FMT) (212): This field specifies the layout of the ORB. This field shall contain the value specified when the ADM-type subchannel is specified; Otherwise, an operand exception is recognized or a specific condition code is set.

The EADM operation block (AOB) specified by the EADM AOB address 206 of the EADM ORB contains the information used to invoke the EADM operations. The AOB, in one example, is allocated within a 4K byte range and the length is variable.

2B, EADM AOB 220 includes three parts: an EADM request block (ARQB) 222, an EADM response block (ARSB) 224, and a defined number (e.g., And an MSB area 226 that includes at most the MSBs of the MSs 124-124. ARQB can specify more usage than defined MSBs when using an MSB branch (i.e., using a branch to the next MSB flag in the MSB).

One embodiment of an EADM request block (ARQB) 222 is described with reference to FIG. 2C. In one example, the ARQB 222 includes the following:

Format (FMT) 230: This field specifies the layout of the ARQB. The value of this field is the defined value; Otherwise, a program check condition is recognized by the channel subsystem or a command code error is indicated in the exception qualifier code field of the ARSB.

Command code 232: This field shall specify the command of the EADM move blocks; Otherwise, the program check condition is recognized by the channel subsystem and the command code error is displayed in the exception modifier code field of the ARSB.

MSB Count 234: This field specifies the count of the MSBs making up the EADM request. The maximum count of MSBs that can be specified depends on the model. The value of this field shall be greater than 0 and less than or equal to the model-dependent maximum MSB count value; Otherwise, the program check condition is recognized by the channel subsystem and the MSB count error is indicated in the exception modifier code field of the ARSB.

In addition to the EADM request block, the EADM ORB also specifies an EADM response block (ARSB). The EADM response block is meaningful only in this embodiment when an exception condition is recognized. Specifically, the ARSB determines when an alarm condition exists in the EADM subchannel status word (SCSW), when the EADM extended status word (ESW) is meaningful, and when the EADM response block store (R) Only in 1 is meaningful, each of which is described below. When the ARSB is meaningless, the contents of the ARSB in the AOB are unpredictable.

If the program is stored in the ARSB while the associated subchannel is subchannel active, unpredictable results may occur.

When ARSB is stored, the amount of data transmitted, even if it is, can not be predicted.

One embodiment of an EADM response block is described with reference to Figure 2D. In one example, the ARSB 224 includes:

Format (FMT) 240: This field specifies the layout of the ARSB. When ARSB is stored, the value of this field is stored as the defined value.

Exception Flags (EF) 242: When the ARSB is stored, this field specifies the exception reason for the ARSB to be stored when set. Exemplary exception reasons include:

Program check: A programming error is detected.

Protection check: Storage access is prohibited by the protection mechanism. The protection is applied to the fetching of ARQB, MSB, AIDAWs and data to be transferred to the storage class memory, and also applies to the storage of the information in the ARSB and the data transmitted from the storage class memory.

Channel data check: An uncorrected storage error is detected in relation to the data contained in the main storage and currently used to perform the EADM operation. If detected, the condition may be displayed, although the data is not used when prefetched. Channel data checking is indicated when the data is referenced by the channel subsystem or when the associated key has an invalid check block code (CBC) in main storage.

Channel control check: The channel control check is caused by a machine malfunction that affects the channel subsystem control. The condition includes an invalid CBC on ARQB, ARSB, MSB, AIDAW, or each associated key. Although no ARQB, MSB, or AIDAW is used, the condition is indicated when an invalid CBC is detected on the pre-fetched ARQB, MSB, AIDAW, or each associated key.

Extended asynchronous data movement facility check: An uncorrected error is detected with respect to the data contained in the storage class memory and currently being used to perform the EADM operation.

Exception Control Block Identifiers (ECBI) 244: When the ARSB is stored, this field is a multiple bit mask, and the bits of this mask, when set, Specifies the following single or combination of components associated with:

컨트롤 블록 또는 블록들.

Control block or blocks.

메인 스토리지 데이터 영역.

Main storage data area.

스토리지 클래스 메모리.

Storage class memory.

The bits that can be set indicate, for example, an EADM move specification block, an EADM indirect data address, data in main storage, and / or data in storage class memory.

The bits in the ECBI field describe the components associated with a single exception condition. If the components can not be identified for the exception condition, this field contains, for example, zeros.

Field validity flag (FVF) 246: When the ARSB is stored, this field contains a multiple bit mask, the bits of which indicate the validity of certain fields within the ARSB. When the validity bit is set, its corresponding field is stored and available for recovery purposes. The bits that may be set indicate, for example, a failing MSB address field, a failed AIDAW field, a failed main storage address field, and / or a failed storage class memory address field.

Exception Qualifier Code (EQC) (248): When the ARSB is stored, this field contains a code value that further describes the exception specified by the exception flags field. The code values may be displayed as an example as follows:

No further explanation is provided. In this case, the fields validated by the Exception Control Block Identifiers (ECBI) field and the Field Validity Flags field may identify the control blocks for which an exception is recognized.

Format error: The format specified by the format field is reserved. In this case, the fields validated by the Exception Control Block Identifiers (ECBI) field and the field validity flags field may identify control blocks for which an exception is recognized.

Command code error: The value specified in the command code field of ARQB is not recognized.

MSB Count Error: The value specified in the MSB Count field of ARQB is 0 or exceeds the model dependent maximum of MSBs that can be specified.

The flag bits specified by the flags error: flags field are reserved. In this case, the fields validated by the Exception Control Block Identifiers (ECBI) field and the Field Validity Flags field may identify the control blocks for which an exception is recognized.

Operation code error: The reserved operation code value is specified. If the field validity flags field indicates that the MSB address field is valid, the field contains the address of the MSB where the exception is recognized.

Block size error: The reserved block size value is specified. If the field validity flags indicate that the MSB address field is valid, then the field contains the address of the MSB where the exception is recognized.

Block count error: The value specified in the block count field of the MSB is either 0 or exceeds the model-dependent maximum count of blocks that can be specified by the MSB. If the field validity flags field indicates that the MSB address field is valid, the field contains the address of the MSB where the exception is recognized.

Main storage address specification error: The main storage address is specified as an inaccurate range. These addresses may have been specified by MSB or AIDAW. If the field validity flags field indicates that the MSB address field is valid, the field contains the address of the MSB where the exception is recognized. If the field validity flags field indicates that the AIDAW address field is valid, then the field contains the address of the AIDAW where the exception is acknowledged. If the field validity flags field indicates that the main storage address field is valid, then the field contains the main storage address for which an exception is acknowledged.

Storage class memory address specification error: The storage class memory address is specified as an incorrect range. These addresses are specified by the MSB. If the field validity flags field indicates that the MSB address field is valid, the field contains the address of the MSB where the exception is recognized. If the field validity flags field indicates that the AIDAW address field is valid, then the field contains the address of the AIDAW where the exception is acknowledged. If the field validity flags field indicates that the storage class memory address field is valid, then this field contains the storage class memory address where the exception is recognized.

Main storage address Exception: EADM Facility has attempted to use an address that is not available in the configuration or has wrapped the maximum storage address. These addresses may have been assigned by the MSB or may have resulted from incrementing main storage addresses during data transfer. If the field validity flags field indicates that the failing MSB address field is valid, the field contains the address of the MSB and an exception is noted for this address. If the field validity flags field indicates that the AIDAW address field is valid, the field contains the address of AIDAW, and an exception is recognized for this address. If the field validity flags field indicates that the main storage address field is valid, then the field contains the main storage address and an exception is noted for this address.

Storage class memory address Exception: EADM Facility has attempted to use a storage class memory address that is not available in the configuration. These addresses may have been assigned by the MSB or may have resulted from incrementing storage class memory addresses during data transfer. If the field validity flags field indicates that the failing MSB address field is valid, the field contains the address of the MSB for which the exception is known. If the field validity flags field indicates that the failing AIDAW address field is valid, then the field contains the address of the AIDAW where the exception is acknowledged. If the field validity flags field indicates that the failing storage class memory address field is valid, the field contains the storage class memory address for which an exception is recognized.

Main storage error: An uncorrected main storage error is detected. If the field validity flags field indicates that the MSB address field is valid, the field contains the address of the MSB where the exception is recognized. If the field validity flags field indicates that the AIDAW address field is valid, then the field contains the address of the AIDAW where the exception is acknowledged. If the field validity flags field indicates that the failing main storage address field is valid, then the field contains the main storage address for which the exception is acknowledged.

MSB List Error: AOB specifies a list of MSBs crossing a 4K byte range without specifying a branch-to-next MSB (BNM) to cross the 4K byte range, Is the last MSB in the specified MSB list and the BNM is specified by the MSB. If the field validity flags field indicates that the MSB address field is valid, the field contains the address of the MSB where the exception is recognized.

MSB Branch Error: The AOB specifies a list of MSBs in which the MSB (branch-source MSB) specifies the branch-to-next MSB (BNM) and the MSB which is the branch target also specifies the BNM . If the field validity flags field indicates that the MSB address field is valid, then the field contains the address of the branch source MSB where the exception is acknowledged.

AIDAW List Error: MSB specifies a list of EADM indirect data address words (AIDAW) crossing the 4K byte range without specifying branch-to-next AIDAW (BNA) to cross the 4K byte range. If the field validity flags field indicates that the MSB address field is valid, the field contains the address of the MSB where the exception is recognized. If the field validity flags field indicates that the AIDAW address field is valid, then the field contains the address of the AIDAW where the exception is acknowledged.

AIDAW branch error: The MSB specifies a list of EADM indirect data address words (AIDAW), where AIDAW (branch source AIDAW) specifies a branch-to-source AIDAW (BNA) The target AIDAW also specifies the BNA. If the field validity flags field indicates that the MSB address field is valid, the field contains the address of the MSB where the exception is recognized. If the field validity flags field indicates that the AIDAW address field is valid, then the field contains the address of the branch-source AIDAW where the exception is acknowledged.

Storage class memory temporary error: A recoverable storage class memory error is detected. If the field validity flags field indicates that the MSB address field is valid, the field contains the address of the MSB where the exception is recognized. If the field validity flags field indicates that the AIDAW address field is valid, then the field contains the address of the AIDAW where the exception is acknowledged. If the field validity flags field indicates that the storage class memory address field is valid, then this field contains the storage class memory address where the exception is recognized.

Release operation not supported error: A release operation has been specified for a storage class memory for which the release operation is not supported. If the field validity flags field indicates that the MSB address field is valid, the field contains the address of the MSB where the exception is recognized. If the field validity flags field indicates that the AIDAW address field is valid, then the field contains the address of the AIDAW where the exception is acknowledged. If the field validity flags field indicates that the storage class memory address field is valid, then this field contains the storage class memory address where the exception is recognized.

Released data read error: The read operation has been specified for the storage class memory for which the last successful operation was the release operation. If the field validity flags field indicates that the MSB address field is valid, the field contains the address of the MSB where the exception is recognized. If the field validity flags field indicates that the AIDAW address field is valid, then the field contains the address of the AIDAW where the exception is acknowledged. If the field validity flags field indicates that the storage class memory address field is valid, then this field contains the storage class memory address where the exception is recognized.

Failed MSB address 250: When the specified bit of the field validity flags field is set, this field indicates the address of the MSB for which an exception is recognized.

Failed AIDAW Address 252: Field Validation Flags When the specified bit of the field is set, this field indicates the address of the AIDAW where the exception is acknowledged.

Failed Main Storage Address 254: Field Validation Flags When the specified bit of the field is set, this field indicates the address of the block of main storage for which an exception is acknowledged.

Failed Storage Class Memory Address 256: When the specified bit of the Field Validity Flags field is set, this field contains the SCM address (e.g., 64 bits) of the block of storage class memory in which the exception is acknowledged.

In addition to the EADM request block and the EADM reply block, the EADM operation block (AOB) also specifies one or more EADM move specification blocks (MSBs). The AOB may include, for example, a maximum of 124 MSBs. However, the program branches to a continuation of MSB list by specifying a larger number of MSBs in the count field of the ARQB and using the branch-to-next-MSB (BNM) Can be specified. There may be multiple sequences of MSB lists, but in one embodiment, none of these sequences can cross the 4K byte range without using the BNM to cross the 4K byte range. Sequences of MSB lists need not be contiguous. Each continuation of the MSB list may contain up to 128 MSBs, if specified to start with a 4K byte range. The total number of MSBs in the MSB list is specified by the MSB count field in the ARQB.

The EADM move specification block describes, for example, operations to be performed on storage blocks or blocks of storage to be moved between main storage and storage class memory. One embodiment of MSB 226 is described with reference to Figure 2E, which includes, for example, the following fields:

Format (FMT) 260: This field specifies the layout of the MSB. The value of this field is the defined value; Otherwise, an MSB format error is displayed in the exception modifier code field of the ARSB.

Operation Code (OC) 262: This field specifies the operation to be performed. In one example, when the specified bit of the flags field described below is a defined value, this field specifies the operation to be performed. The example operations include the following code:

Clear storage class memory: When this code is specified, the storage class memory specified by the storage class memory address, block size, and block count fields is cleared (for example, set to zero do). The main storage address field has no meaning for this operation.

Read: When this code is specified, data is specified to be transferred from storage class memory to main memory. The block count, block size, main storage address, and storage class memory address fields are meaningful for this operation.

Write: When this code is specified, the data is specified to be transferred from main storage to storage class memory. The block count, block size, main storage address, and storage class memory address fields are meaningful for this operation.

Release: When this code is specified and the EADM Release facility is installed, the storage class memory specified by the storage class memory address, block size, and block count fields is released. The main storage address field has no meaning for this operation. When EADM Release Facility is not installed, this code is reserved.

If the reserved value is specified, the program check condition is recognized by the channel subsystem and the operation code error is displayed in the exception modifier code field of the ARSB.

When the flag bit specified in the flag field is not a defined value, this field is reserved and not checked.

Flags 264: This field identifies one or more flags that may be set. The reserved flag bits are set to a defined value; Otherwise, the flags error is displayed in the exception modifier code field of the ARSB. Example flags include the following:

브랜치 투 넥스트 MSB(BNM): 세트될 때, 이 플래그는 MSB가 EADM 오퍼레이션을 명시하지 않고 데이터를 전송하기 위해 사용되지 않는다고 표시한다. 그 대신에, 메인 스토리지 주소 필드가 EADM 오퍼레이션을 명시하기 위해 사용될 다음 MSB의 주소를 명시한다.

Branch NeXT MSB (BNM): When set, this flag indicates that the MSB is not used to transmit data without specifying the EADM operation. Instead, the main storage address field specifies the address of the next MSB to be used to specify the EADM operation.

When the BNM flag is set (e.g., 0) and the main storage address field specifies the MSB - the BNM flag is also set in that MSB - the program check condition is recognized by the channel subsystem and the branch source MSB The address is stored in the failed MSB address field of the ARSB, and the MSB branch error is displayed in the exception modifier field of the ARSB.

If the MSB ends in a 4K byte range, the AOB specifies additional MSBs, and the BNM flag is not set, a program check condition is recognized by the channel subsystem and the address of the MSB is stored in the failed MSB address field of the ARSB , And an MSB list error is displayed in the exception modifier code field of the ARSB.

If the MSB is the last MSB in the specified MSB list and the BNM flag is set, the program check condition is recognized by the channel subsystem, the address of the MSB is stored in the failed MSB address field of the ARSB, and the MSB list error is an exception to the ARSB It is displayed in the modifier code field.

When the BNM flag is set, the remaining flags, operation code field, block size field, block set count field, and storage class memory address are not significant.

간접 주소지정(Indirect addressing): 세트될 때, 이 플래그는 메인 스토리지 주소 필드가 데이터 전송을 위해 사용될 메인 스토리지 위치 또는 위치들을 각각 지정하는 EADM 간접 데이터 주소 워드(AIDAW)의 주소 또는 AIDAW들의 리스트의 제1 AIDAW의 주소를 지정한다고 표시한다. 세트되지 않을 때, 이 플래그는 메인 스토리지 주소 필드가 데이터 전송을 위해 사용될 메인 스토리지 위치의 주소를 지정한다고 표시한다.

Indirect addressing: When set, this flag indicates the address of the EADM indirect data address word (AIDAW) or the list of AIDAWs, each of which specifies the main storage location or locations to be used for data transfer. 1 Indicates that the address of AIDAW is specified. When not set, this flag indicates that the main storage address field specifies the address of the main storage location to be used for data transfer.

Block Size (BS) 266: This field contains the size and range of data blocks to be transferred for both main storage and storage class memory, or the size of the block of storage class memory to be operated (e.g., to be cleared or released) And range. Examples include:

4K: When the code value is one defined value, the data blocks to be transmitted are in the range of 4K bytes and the size is 4K bytes.

1M: When the code value is another defined value, the data blocks to be transmitted are in the 1M byte range and the size is 1M byte.

If the reserved value is specified, the program check condition is recognized by the channel subsystem and the block size error is displayed in the exception modifier code field of the ARSB.

When the BNM flag is set, this field is ignored.

Block count 268: This field contains a count of blocks of data to be transmitted or operated (e.g., cleared or released). Based on the block size field, this is a count of one of 4K byte or 1M byte blocks.

The value of this field shall be less than or equal to the model-dependent maximum block count, which is greater than 0 and can be specified by the MSB; Otherwise, the program check condition is recognized by the channel subsystem and a block count error is indicated in the exception modifier code field of the ARSB.

When the BNM flag is set, this field is reserved and not checked.

Main storage address 270: specifies the read or write operation in which the operation code field is meaningful (e.g., the BNM flag is not set), and an indirect addressing indicator in the flags field When not set, this field contains the main storage address to be used for data transfer and the following applies:

블록 사이즈 필드가 4K 바이트 블록들을 명시할 때, 이 필드의 명시된 비트들은 정의된 값(예를 들어, 0)을 포함한다; 그렇지 않으면, 프로그램 검사 조건이 채널 서브시스템에 의해 인지되고 메인 스토리지 명세 에러가 ARSB의 예외 한정자 코드 필드에 표시된다.

When the block size field specifies 4K byte blocks, the specified bits of this field contain a defined value (e.g., 0); Otherwise, the program check condition is recognized by the channel subsystem and the main storage specification error is displayed in the exception modifier code field of the ARSB.

블록 사이즈 필드가 1M 바이트 블록들을 명시할 때, 이 필드의 명시된 비트들은 정의된 값(예를 들어, 0들)을 포함한다; 그렇지 않으면, 프로그램 검사 조건이 채널 서브시스템에 의해 인지되고 메인 스토리지 명세 에러가 ARSB의 예외 한정자 코드 필드에 표시된다.

When the block size field specifies 1 Mbyte blocks, the specified bits of this field contain a defined value (e.g., 0's); Otherwise, the program check condition is recognized by the channel subsystem and the main storage specification error is displayed in the exception modifier code field of the ARSB.

When the operation code field is meaningful and one of the read or write operations is specified and the indirect addressing indicator in the flags field is set, this field indicates the first main storage address of the list of AIDAWs or AIDAWs to be used for data transfer . In this case, certain bits of this field contain the defined value; Otherwise, the program check condition is recognized by the channel subsystem and the main storage specification error is displayed in the exception modifier code field of the ARSB.

When the BNM flag is set, the operation code has no meaning and the MSB does not specify the EADM operation. Instead, this field contains the main storage address of the next MSB that specifies the EADM operation. In this case, the specified bits of this field contain the defined value; Otherwise, the program check condition is recognized by the channel subsystem and the main storage specification error is displayed in the exception modifier code field of the ARSB.

Storage Class Memory Address 272: When the Operation Code field has a meaning, this field contains the storage class memory address to be used for data transfer or to be operated (e.g., cleared, released) And the following applies:

블록 사이즈 필드가 4K 바이트 블록들을 명시할 때, 이 필드의 명시된 비트들은 정의된 값(예를 들어, 0들)을 포함한다; 그렇지 않으면, 프로그램 검사 조건이 채널 서브시스템에 의해 인지되고 스토리지 클래스 메모리 명세 에러가 ARSB의 예외 한정자 코드 필드에 표시된다.

When the block size field specifies 4K byte blocks, the specified bits of this field contain a defined value (e.g., 0s); Otherwise, the program check condition is recognized by the channel subsystem and the storage class memory specification error is displayed in the exception modifier code field of the ARSB.

블록 사이즈 필드가 1M 바이트 블록들을 명시할 때, 이 필드의 명시된 비트들은 정의된 값 포함한다; 그렇지 않으면, 프로그램 검사 조건이 채널 서브시스템에 의해 인지되고 스토리지 클래스 메모리 명세 에러가 ARSB의 예외 한정자 코드 필드에 표시된다.

When the block size field specifies 1-Mbyte blocks, the specified bits of this field contain the defined value; Otherwise, the program check condition is recognized by the channel subsystem and the storage class memory specification error is displayed in the exception modifier code field of the ARSB.

When the BNM flag is set, this field is reserved and not checked.

As indicated above, the EADM indirect data address word may be specified. The EADM Indirect Data Address Word (AIDAW) allows a program to specify the transfer of blocks of data between non-contiguous blocks of storage class memory and main storage. The list of AIDAWs or AIDAWs is specified by the MSB when the indirect addressing flag in the MSB is set.

The amount of data transmitted by a single AIDAW is specified by the Block Size in MSB field. The number of AIDAWs in the AIDAW list is the sum of the number specified by the block count field in the MSB and the number of AIDAWs specifying the branch-to-next-AIDAW. Data transmissions may be processed in a different order than specified in the AIDAW list. Also, data transmissions specified by multiple AIDAWs in the AIDAW list may be processed concurrently.

AIDAW is assigned, for example, to a quadword range. The list of AIDAWs may be of arbitrary length, but in one example, the branch punctuate AIDAW (BNA) does not cross the 4K byte range unless it is specified to cross the 4K byte range. It is not mandatory that AIDAW, the target of the branch, be adjacent to AIDAW, which specifies the BNA. However, the program must generate the AIDAW list in as few 4K byte blocks as possible; Otherwise, performance degradation may occur.

Referring to FIG. 2F, in one example, the EADM indirect data address word (AIDAW) 280 includes:

Flags 282: An example of a flag is:

Branch Auxiliary AIDAW (BNA): When set, this flag indicates that the main storage address field does not specify the main storage address to be used for data transfer. Instead, the main storage address field specifies the address of the next AIDAW to be used for data transfer.

When the BNA flag is set and the main storage address field specifies AIDAW where the BNA flag is also set, the address of the MSB is stored in the failed MSB address field of the ARSB and the address of the branch source AIDAW is stored in the failed main storage address field of the ARSB And the AIDAW branch error is displayed in the exception modifier code field in the ARSB.

If the AIDAW ends in the 4K byte range, the MSB specifies additional AIDAWs, and the BNA flag is not set, the MSB's address is stored in the failed MSB address field of the ARSB, and the address of the AIDAW is stored in the failed main storage address of the ARSB And an AIDAW list error is displayed in the Exception Qualifier Code field in the ARSB.

Main storage address (284): When the BNA field is not set, this field contains the address in the main storage to be used for data transfer and the following applies:

MSB 내 블록 사이즈 필드가 4K 바이트 블록들을 명시할 때, 이 필드의 명시된 비트들은 정의된 값(예를 들어, 0)을 포함한다; 그렇지 않으면, 프로그램 검사 조건이 채널 서브시스템에 의해 인지되고 메인 스토리지 명세 에러가 ARSB의 예외 한정자 코드 필드에 표시된다.

When the in-MSB block size field specifies 4K byte blocks, the specified bits of this field contain a defined value (e.g., 0); Otherwise, the program check condition is recognized by the channel subsystem and the main storage specification error is displayed in the exception modifier code field of the ARSB.

MSB 내 블록 사이즈 필드가 1M 바이트 블록들을 명시할 때, 이 필드의 명시된 비트들은 정의된 값(예를 들어, 0들)을 포함한다; 그렇지 않으면, 프로그램 검사 조건이 채널 서브시스템에 의해 인지되고 메인 스토리지 명세 에러가 ARSB의 예외 한정자 코드 필드에 표시된다.

When the in-MSB block size field specifies 1 Mbyte blocks, the specified bits of this field contain a defined value (e.g., 0's); Otherwise, the program check condition is recognized by the channel subsystem and the main storage specification error is displayed in the exception modifier code field of the ARSB.

BNA 플래그가 세트될 때, 이 필드는 데이터 전송을 위해 사용될 다음 AIDAW의 메인 스토리지 주소를 포함한다.

When the BNA flag is set, this field contains the main storage address of the next AIDAW to be used for data transfer.

As described above, EADM operations are specified by the Start Subchannel command. That is, the program initiates EADM operations by issuing a Start Subchannel command specifying an ADM-type subchannel and an EADM ORB. The execution of the instruction transfers the contents of the EADM ORB to the designated subchannel. The EADM ORB contains the specification of the subchannel key (used for protection) and the address of the AOB to be used.

In one example, as shown in FIG. 3A, a Start Subchannel command 300 includes an operation code 302 that specifies a Start Subchannel function; A first operand 304, which is, for example, an implied operand located in general register 1, including a subsystem identifier that specifies an ADM-type subchannel to be started; And a second operand 306, which is the logical address of the EADM ORB. The EADM ORB specifies the parameters used to control the start function. The contents of the EADM ORB are placed in the subchannel specified during execution of the Start Subchannel before setting the condition code. If the result is a set of condition codes other than the code indicating that the execution of the Start Subchannel succeeded, the contents of the EADM ORB are not placed in the specified subchannel.

Following the execution of the Start Subchannel for the ADM-type subchannel, the channel subsystem performs the start function asynchronously to initiate EADM operations with EADM facilities. The start function includes, for example, the following:

1. Fetch ARQB from AOB.

2. Execute EADM operations as specified by the ARQB and its MSBs specified by ARQB.

3. Conditionally store completion information in ARSB in AOB.

4. The ADM-type subchannel is status pending, indicating the completion of the start function.

In one embodiment, referring to FIG. 3B, when the Start Subchannel command is executed and the SID specifies an ADM-type subchannel and the second operand specifies an EADM ORB, an EADM operation is specified (step 320). The parameters in the EADM ORB are passed to the designated subchannel (step 322) and the channel subsystem is asked to perform the start function with EADM facility (step 324). The channel subsystem initiates EADM operations (step 326) by performing the start function asynchronously by using the information of the subchannel including information conveyed during execution of the Start Subchannel command.

Execution of an EADM operation (e.g., the first operation) includes acquiring an AOB using information from the EADM ORB (step 328), acquiring an EADM request block (ARQB) using the information in the AOB Obtain the designation of one or more EADM movement specification blocks (MSBs) (step 330). One or more designated MSBs are then fetched from main storage (step 332) and the information specified in the MSBs is used to control the requested EADM operation. The first operation is considered to be started when the channel subsystem attempts to initiate a data transfer or attempts a clear or a release operation.

The channel subsystem performs the requested operation (s) in the MSB (s) (step 334). This processing is asynchronous with the execution of the start command.

When the EADM operations initiated by the Start Subchannel are terminated (step 336), the channel subsystem generates status conditions (step 338). The generation of these conditions invokes the attention of the program by I / O interruption (step 339). The program can also query these conditions by executing the Test Pending Interrupt command.

The generated condition conditions are provided to the program in the form of an EADM subchannel status word (SCSW). The EADM SCSW is stored as part of the EADM interruption response block (IRB) by execution of the Test Subchannel command.

When the EADM operations specified in the EADM operation block (AOB) are terminated, the channel subsystem generates the primary and secondary interruption states. EADM operations may be terminated by a Clear Subchannel or an abnormal condition recognized during the performance of the start function.

For a move operation, the processing of the MSB by the channel subsystem controls the flow of a unit of information into or out of the main storage. To change the direction of data movement during MSB processing, a new MSB is needed. The ARQB specifies a count of the MSBs containing the request.

Each EADM operation is represented by one MSB. The MSB specifies one of the following:

메인 스토리지로부터 스토리지 클래스 메모리로 데이터 블록들의 전송.

Transfer of data blocks from main storage to storage class memory.

스토리지 클래스 메모리로부터 메인 스토리지로 데이터 블록들의 전송.

Transfer of data blocks from storage class memory to main storage.

스토리지 클래스 메모리의 블록들을 클리어함.

Clears blocks in storage class memory.

EADM 해제 퍼실리티가 설치되어 있을 때, 스토리지 클래스 메모리의 블록들을 해제함.

Releases blocks of storage class memory when EADM Release Facility is installed.

If the blocks of storage to be transferred for the move operation are not contiguous in main storage, a new MSB is used or the MSB specifies a list of EADM indirect data address words (AIDAWs) to specify non-contiguous blocks Indirect addressing can be used. Since the MSB can only specify the transmission of data in one direction, a new MSB is used when there is a change in the direction of the data transmission.

The following are the characteristics of EADM data transmissions:

데이터 전송들은 명시된 MSB 리스트의 순서에 관해서 순서와 다르게 처리될 수 있다.

Data transmissions may be processed differently in order for the order of the explicit MSB list.

MSB 리스트에서 다수 MSB들에 의해 명시된 데이터 전송들은 동시에 처리될 수 있다.

Data transmissions specified by multiple MSBs in the MSB list can be processed simultaneously.

데이터 전송들은 명시된 AIDAW 리스트의 순서와 관련하여 순서와 다르게 처리될 수 있다.

Data transmissions may be handled differently in the order with respect to the order of the explicit AIDAW list.

AIDAW 리스트에서 다수 AIDAW들에 의해 명시된 데이터 전송들은 동시에 처리될 수 있다.

Data transmissions specified by multiple AIDAWs in the AIDAW list can be processed simultaneously.

메인 스토리지와 스토리지 클래스 메모리에 대한 액세스는, 프로그램과 기타 CPU들에서 관찰되는 바와 같이, 필수적으로 단일-액세스 참조들일 필요는 없으며 필수적으로 좌-에서-우(left-to-right) 방향으로 수행될 필요도 없다.

Access to main storage and storage class memory does not necessarily have to be single-access references, as is observed in programs and other CPUs, and is performed essentially in a left-to-right direction There is no need.

둘 또는 그 이상의 EADM 오퍼레이션들이 현재 활성이고 동일한 SCM 위치들, 메인 스토리지 위치, 또는 둘 모두를 주소지정 한다면, 상기 오퍼레이션들은 동시에 수행될 수 있고 다른 오퍼레이션들로부터 콘텐츠가 인터리브될 (interleaved) 수 있다; 그러나:

If two or more EADM operations are currently active and address the same SCM locations, main storage location, or both, the operations may be performed concurrently and content from other operations may be interleaved; But:

For input operations, the data stored in each block of main storage, which is the same size as the SCM block concurrency size by the EADM facility, is the data transferred from the storage class memory only by one of the concurrent processing EADM operations. .

- For output operations, each block in the storage class memory, which is in range and is equal in size to the concurrent processing size of the SCM block, contains data specified only by one of the concurrent processing EADM operations.

The above is true regardless of whether EADM operations are specified by a single AOB and processed by the same instance of the start function, or whether EADM operations are specified by other AOBs and processed by other instances of the start function It is true.

EADM 오퍼레이션들이 서브채널에 대하여 활성일 때, 프로그램에 의해 ARQB, MSB들, AIDAW들, 및 그 활성인 오퍼레이션과 연관된 전송 데이터에 행해진 변경들이 EADM 퍼실리티에 의해 관찰되는지는 예측할 수 없다.

When EADM operations are active for a subchannel, it is unpredictable that changes made to the transmission data associated with the ARQB, MSBs, AIDAWs, and their active operations by the program are observed by EADM facilities.

When all the blocks specified by all the MSBs specified by the AOB are completed to be transmitted, cleared or released, the subchannel generates a state stored in that subchannel and requests I / O interruption for the ADM-type subchannel do.

The termination of the EADM operation is usually indicated by the combined state conditions of channel termination and device termination. This combination of states indicates a combination of primary and secondary states, indicating that the subchannel is available for another start function after the state is cleared.

As described above, the MSB may specify a data transfer operation, a clear operation, or a release operation, each of which is described below.

When a move operation is requested, one or more blocks of data move between main storage and storage class memory. For example, in a lead operation, SCM data is obtained from SSDs that provide the content specified by the specified SCM address, and then the content is stored in main memory. This process is reversed in the write operation. The adapter (s) controlling the SSDs perform the storage. More specifically, to perform the move operation, the system firmware first translates a given SCM address into an adapter address (e.g., a logical volume address where the logical volume contains one or more SSDs). For example, a translation table is used to correlate the SCM address with the adapter address. The system firmware then submits one or more corresponding adapter migration commands (e.g., a lead or a write) to one or more I / O adapters. The adapter move command includes the main storage address, the adapter address, and the transfer size. The adapter then uses another translation table to find one or more physical SSD addresses corresponding to the adapter address. The I / O adapter executes the move operation by fetching data from main storage and storing it in SSDs, or by fetching data from SSDs and storing it in main memory. Additional details are further provided in the patented POU 920110090US1 entitled " Firmware Management of Storage Class Memory ", which is incorporated herein by reference in its entirety.

When a clear operation is performed, the specified increments of the storage class memory are cleared by setting the contents to zeros.

Also, when the release operation is performed, one or more standard trim (TRIM) commands may be submitted to the SSDs containing the corresponding SCM addresses. The Trim command allows the program to give hints about the use of blocks, allowing for better garbage collection by SSDs. The Trim command causes the operating system to notify the SSD which blocks in the storage are no longer considered for use and can be erased.

The EADM operation may be terminated early by the Clear Subchannel command. Execution of the Clear Subchannel command terminates the execution of the AOB in the subchannel, clears the AOB indications in execution on the subchannel, and performs the clear function asynchronously. When the clear function is performed, before the subchannel is status pending, the data transmission is terminated and the amount of data transmitted is unpredictable. The execution of the clear function does not create a state, but causes the channel subsystem to make I / O interruption pending.

In one embodiment, referring to FIG. 3C, the Clear Subchannel command 350 includes an operation code 352 that specifies a Clear Subchannel function. The subchannel to be cleared is designated, for example, by a subsystem identification word in general register 1.

The clear function for an ADM-type subchannel includes:

1. Ensure that the current data transfer is terminated.

2. Change the fields in the subchannel and conditionally change the ARSB. For example, the subchannel status word is changed to display the clear function in the Function Control field and the Activity Control field. The ARSB may be changed to reflect all detected errors.

3. The subchannel is pending to indicate the completion of the clear function.

Other instructions that can specify an ADM-type subchannel, including Test Subchannel, Modify Subchannel, and Store Subchannel, may also be issued, each of which is described below.

Referring to FIG. 3D, in one example, the Test Subchannel command 360 includes, for example, an operation code 362 that specifies a Test Subchannel function; A first operand 364, which is, for example, an implied operand located in general-purpose register 1, including a subsystem identification word that specifies a subchannel to be tested; And a second operand 366, which is the logical address of an information response block (IRB) where information is stored.

When the test subchannel is executed to specify an ADM-type subchannel, the subchannel is state pending, and information is stored in the designated EADM IRB (Interruption Response Block), the specified condition code is set. When the subchannel is not state pending and the state information is stored in the specified EADM IRB, the defined condition code is set. When the subchannel is not provided or enabled, no action occurs.

In one example, the EADM IRB contains additional model dependency information that includes the Subchannel Status Word (SSW) and the Extended Status Word (ESW) and describes conditions that may be present in the Facility (Extended Control Word) that can be used. Each of these words will be further described below after discussion of various instructions that can specify an ADM-type subchannel.

Referring to FIG. 3E, in one embodiment, the Modify Subchannel command 370 includes an operation code 372 that specifies a Modify Subchannel function; A first operand 374, which is, for example, an implied operand located in general register 1, including a subsystem identification word that specifies a subchannel to be modified; And a second operand 376, which is the logical address of a subchannel information block (SCHIB) associated with the subchannel.

A Modify Subchannel is executed to specify an ADM-type subchannel, and when information from a specified subchannel information block (SCHIB) is placed in a subchannel, a specific condition code is set. When the subchannel is in state pending, no action is taken and the defined condition code is set. If the subchannel is busy for the start or clear function, no action is taken.

Referring to FIG. 3F, an example of a Store Subchannel command is described. In one example, the Store Subchannel command 380 includes an operation code 382 that identifies the Store Subchannel function; A first operand 384, which is an implied operand located, for example, in general register 1, containing a subsystem identification word that specifies a subchannel - information is stored for this subchannel; And a second operand 386, which is the logical address of the SCHIB.

A Store Subchannel is issued to specify an ADM-type subchannel, and when SCHIB is stored, the specified condition code is set. If the specified subchannel is not provided in the channel subsystem, no action is taken.

An example of a subchannel information block (EADM SCHIB) for an ADM-type subchannel is described with reference to FIG. In one example, (EADM SCHIB) 400 includes a model dependent area 401 containing model dependent information. SCHIB 400 also includes a path management control word (PMCW) 402 and a subchannel status word (SCSW) 404, each of which is described below.

In one example, the EADM PMCW 402 includes the following fields, for example, as shown in Figure 4B:

Interruption parameter 410: This field contains the interruption parameter stored in the I / O interruption code. The interruption parameter can be set to any value by the Start Subchannel and the Modify Subchannel. The initial value of the interruption parameter field in the subchannel is zero.

Interrupt subclass (ISC) 412: This field is used to indicate the bit position of the I / O interruption subclass mask bit in the specified control register of each CPU in the configuration, , And an unsigned binary integer. The set of mask bits in the control register of the CPU controls the acknowledgment of interrupt requests related to the subchannel by that CPU. The ISC can be set to any value by the Modify Subchannel. The initial value of the ISC field in the subchannel is, for example, zero.

Enabled 414: This field, when set, indicates that the subchannel is enabled for all EADM functions.

Subchannel Type (ST) 416: This field specifies the subchannel type. Depending on the model and configuration, one or more I / O subchannel types or ADM subchannel types may be provided.

The value of this field is determined when the subchannel is configured and can not be modified by the Modify Subchannel.

When the Modify Subchannel command is executed and an ADM-type subchannel is specified, the ST displays the ADM subchannel; Otherwise, an operand exception is recognized.

4A, the subchannel information block also includes a subchannel status word 404. The EADM subchannel status word (SCSW) provides the program with indications describing the status of the ADM-type subchannel and its associated EADM operations. In one example, as shown in FIG. 4C, the subchannel status word 404 includes the following:

Sub Channel Key (Key) 420: When the EADM start function indicator in the function control field (described below) is set, this field contains the storage access key used by the channel subsystem. These bits are identical to the key specified in the EADM ORB when the Start Subchannel is executed.

Extended Status Word Format (L) 422: When the status pending indicator of the Status Controls field (described below) is set, this field, when set, . Format-0 ESW is stored when an interruption condition, including any of the following indications, is cleared by the Test Subchannel:

프로그램 검사

Program check

보호 검사

Protection check

채널 데이터 검사

Check channel data

채널 컨트롤 검사

Channel control checks

확장 비동기 데이터 이동 퍼실리티(EADMF) 검사

Extended Asynchronous Data Movement Facility (EADMF) checks

Deferred Condition Code (CC) (424): When the EADM start function indicator is set and the state pending indicator is also set, this field indicates the general reason why the subchannel was state pending when the Test Subchannel or Store Subchannel was executed Display. The deferred condition code is only meaningful when the subchannel is in any combination of states and state pending and when the start function indicator of the function control field in the SCSW is set.

The delay condition code is used to indicate if it has met conditions that exclude subchannels from being subchannel active while the subchannel is pending start.

Example delay condition codes include the following:

정상 I/O 인터럽션이 제출되었다.

A normal I / O interruption has been submitted.

EADM 시작 펑션이 성공적으로 개시되는 것을 배제하는 조건들을 위해 채널 서브시스템에 의해서 생성된 상태가 EADM SCSW에 존재한다. 즉, 서브채널은 서브채널 활성 상태(subchannel active state)로 전환되지 않았다.

A condition created by the channel subsystem exists in the EADM SCSW for conditions that preclude the successful start of the EADM start function. That is, the subchannels are not switched to the subchannel active state.

CCW Format (F) (426): When the EADM start function indicator is set, this field is stored with the defined value.

Prefetch (P) (428): When the EADM start function indicator is set, this field is stored with the defined value.

Extension Control (E) 430: This field indicates that model-dependent information is stored in the EADM Extended Control Word (ECW) when set.

Function Control (FC) (432): The Function Control field displays the EADM functions displayed on the subchannel. The example functions include:

EADM 시작 펑션: 세트될 때, EADM 시작 펑션은 요청 완료되었으며 ADM-타입 서브채널에서 펜딩 또는 진행중이라는 것을 표시한다. EADM 시작 펑션은 성공적인 조건 코드가 Start Subchannel에 대해서 세트될 때 서브채널에 표시된다. EADM 시작 펑션은 Test Subchannel이 실행되고 서브채널이 상태 펜딩일 때 서브채널에서 클리어된다. EADM 시작 펑션은 또한 Clear Subchannel의 실행 동안에 서브채널에서 클리어된다.

EADM start function: When set, the EADM start function indicates that the request has been completed and is pending or in progress on the ADM-type subchannel. The EADM start function is indicated on the subchannel when a successful condition code is set for the Start Subchannel. The EADM start function is cleared on the subchannel when the Test Subchannel is executed and the subchannel is a status pen. The EADM start function is also cleared on the subchannel during the execution of the Clear Subchannel.

EADM 클리어 펑션: 세트될 때, EADM 클리어 펑션은 요청 완료되었고 ADM-타입 서브채널에서 펜딩 또는 진행중이라는 것을 표시한다. EADM 클리어 펑션은 성공적인 조건 코드가 Clear Subchannel에 대해서 세트될 때 서브채널에 표시된다. EADM 클리어 펑션 표시는 Test Subchannel이 실행되고 서브채널이 상태 펜딩일 때 서브채널에서 클리어된다.

EADM clear function: When set, the EADM clear function indicates that the request is complete and is pending or in progress on the ADM-type subchannel. The EADM clear function is indicated in the subchannel when a successful condition code is set for the Clear Subchannel. The EADM clear function display is cleared on the subchannel when the test subchannel is executed and the subchannel is in the status pendant.

ACTION CONTROL (AC) (434): The ACTIVITY CONTROL field indicates the current progress of the previously received EADM function in the subchannel.

All the conditions indicated by the bits in the activity control field are reset in the ADM-type subchannel when the Test Subchannel is executed and the subchannel is in status pending.

Exemplary activities include:

시작 펜딩(Start pending): 세트될 때, 서브채널이 시작 펜딩임을 표시한다. 채널 서브시스템은 EADM 시작 펑션의 수행을 처리하는 중일 수도 있고 그렇지 않을 수도 있다. 서브채널은 Start Subchannel에 대해서 성공적인 조건 코드가 세트될 때 시작 펜딩이 된다. 서브채널은 EADM 시작 펑션을 수행중일 때 시작 펜딩으로 남아있고 채널 서브시스템은 ARQB의 페치를 막는 조건들이 존재함을 결정한다.

Start pending: When set, indicates that the subchannel is starting pending. The channel subsystem may or may not be processing the performance of the EADM start function. The subchannel begins pending when a successful condition code is set for the Start Subchannel. The subchannel remains on pending when performing the EADM start function and the channel subsystem determines that there are conditions to prevent ARQB from fetching.

The subchannel is no longer pending when either of the following occurs:

채널 서브시스템이 AOB에 의해 명시된 제1 데이터 전송의 개시를 시도한다.

The channel subsystem attempts to initiate the first data transmission specified by the AOB.

Clear Subchannel이 실행된다.

Clear Subchannel is executed.

Test Subchannel이 서브채널에서 상태 조건을 클리어한다.

Test Subchannel clears the status condition on this subchannel.

Clear pending: When set, the subchannel is clear pending. The channel subsystem may or may not be processing the performance of the EADM clear function. The subchannel can be pending when the specified condition code is set for the Clear Subchannel.

If any of the following occurs, the subchannel is no longer clear pending:

EADM 클리어 펑션이 수행된다.

EADM clear function is executed.

Test Subchannel이 상태 펜딩 조건만을 클리어한다.

Test Subchannel clears only the condition pending condition.

Subchannel active: When set, indicates that the ADM-type subchannel is subchannel active. An ADM-type subchannel is referred to as a subchannel active when the channel subsystem attempts to initiate the first data transmission or perform the first operation specified by the AOB (whichever occurs first).

The subchannel is no longer subchannel active when one of the following occurs:

서브채널이 상태 펜딩이 된다.

The subchannel becomes state pending.

Clear Subchannel이 실행된다.

Clear Subchannel is executed.

Status Control (SC) 436: The Status Control field provides the program with a summary level indication of the interruption condition described by information in the Subchannel Status field and the Device Status field. More than one state control indicator may be set as a result of the conditions in the subchannel.

Example state controls include:

Alert status: When set, indicates that an alarm interruption condition exists. The alarm interruption condition is recognized when an alarm condition exists in the subchannel. An alarm condition is generated by the channel subsystem under one of the following conditions:

ADM-타입 서브채널이 시작 펜딩이고 상태 조건이 제1 데이터 전송의 개시를 금지한다.

The ADM-type subchannel is a pending start and the condition condition prohibits the initiation of the first data transmission.

서브채널은 서브채널 활성이고 서브채널 상태로 보고되는 비정상 조건이 EADM 오퍼레이션들을 종료 완료했다.

The subchannel is subchannel active and the abnormal condition reported in the subchannel state has completed the EADM operations.

When Test Subchannel or Clear Subchannel is executed, the alarm condition is cleared on the subchannel.

Primary status: When set, indicates that the primary interruption condition exists. The primary interruption condition is recognized when the primary state exists in the subchannel. The primary interruption condition is a solicited interruption condition that indicates the completion of the EADM start function in the subchannel when accompanied by a secondary interruption condition.

When a test subchannel or a clear subchannel is executed, the primary interruption condition is cleared in the subchannel.

Secondary status: When set, it indicates that a secondary interruption condition exists. The secondary interruption condition is recognized when the secondary state exists in the subchannel. The secondary interruption condition is a solicited interruption condition that indicates the completion of the EADM start function in the subchannel when accompanied by the primary interruption condition.

When a test subchannel or a clear subchannel is executed, the secondary interruption condition is cleared in the subchannel.

Status pending: When set, indicates that the subchannel is state pending and information describing the cause of the interruption condition is available. When the Test Subchannel is executed and stores the EADM SCSW with the status pending bit set, all EADM SCSW indications are cleared on the subchannel and the subchannel is idle. The state pending condition is also cleared on the subchannel during the execution of the Clear Subchannel.

When state pending is set, all accesses to main storage and storage class memory for the subchannel ARSB are terminated.

Subchannel state 438: ADM-type subchannel state conditions are detected and stored in the subchannel state field by the channel subsystem. The subchannel status field is meaningful when the subchannel is a status pendant. Except for the conditions caused by equipment malfunctions, the subchannel state can only occur when the channel subsystem is involved in processing EADM functions.

Examples of condition conditions include the following:

프로그램 검사: 프로그램 검사는 프로그래밍 에러들이 채널 서브시스템에 의해 검출될 때 발생한다.

Program Check: Program Check occurs when programming errors are detected by the channel subsystem.

보호 검사: 보호 검사는 채널 서브시스템이 보호 메커니즘에 의해 금지된 스토리지 액세스를 시도할 때 발생한다. 보호는 스토리지 클래스 메모리로 전송될 ARQB, MSB들, AIDAW들, 및 데이터의 페치와, ARSB 내 정보와 스토리지 클래스 메모리로부터 전송된 데이터의 저장에 적용된다.

Protection Audit: A protection audit occurs when a channel subsystem attempts to access a storage that is prohibited by a protection mechanism. Protection is applied to the fetching of ARQB, MSBs, AIDAWs, and data to be transferred to the storage class memory, and the storage of the information in the ARSB and the data sent from the storage class memory.

채널 데이터 검사: 채널 데이터 검사는 메인 스토리지로부터 데이터를 페치하는 것 또는 데이터를 메인 스토리지로 저장하는 것과 관련하여 교정되지 않은 스토리지 에러가 검출되었다는 것을 표시한다.

Channel Data Inspection: Channel data inspection indicates that an uncorrected storage error has been detected in relation to fetching data from main storage or storing data to main storage.

채널 컨트롤 검사: 채널 컨트롤 검사는 AOB, MSB들, 또는 AIDAW들의 페치 또는 저장과 관련하여 교정되지 않은 스토리지 에러가 검출되었다는 것, 또는 채널 서브시스템에 기계 오동작이 일어나서 그 오동작이 EADM 오퍼레이션들에 영향을 주었다는 것을 표시한다.

Channel Control Check: The channel control check indicates that an uncorrected storage error has been detected with respect to fetching or storing AOB, MSBs, or AIDAWs, or that a malfunction has occurred in the channel subsystem causing the erroneous operation to affect EADM operations .

확장 비동기 데이터 이동(EADM) 퍼실리티 검사: EADM 퍼실리티 검사는 스토리지 클래스 메모리 내부 또는 외부로 데이터의 전송과 관련하여 또는 스토리지 클래스 메모리에서 오퍼레이션을 수행하는 것과 관련하여 EADM 퍼실리티에 의해 에러가 검출되었다는 것을 표시한다.

Extended Asynchronous Data Movement (EADM) Facility Check: The EADM Facility Check indicates that an error has been detected by the EADM Facility in connection with the transfer of data into or out of the storage class memory or with respect to performing operations in the storage class memory .

When a program check, a protection check, a channel data check, a channel check check, or an EADM facilities check condition is recognized by the channel subsystem, the EADM operations are terminated and the channel is pending state in the primary, secondary, .

EADM Operation Block Address (440): This field contains the EADM operation block address.

Device state 442: This includes device termination or channel termination.

When the ESW format 422 of the subchannel status word is set and the subchannel is state pending, an EADM subchannel extended status word (EADM ESW) is provided to specify additional information about the ADM-type subchannel.

In one example, with reference to FIG. 4D, the EADM subchannel extension status word 450 includes the following:

EADM subchannel logout 452: EADM subchannel logout includes, in one example, the following:

확장 상태 플래그들(ESF): 비트들이, 1일 때, 채널 서브시스템에 의해 에러가 검출되었다고 명시하는 필드.

Extended Status Flags (ESF): A field that specifies that an error was detected by the channel subsystem when the bits are 1.

Examples of extended status flags include:

Key check: When set, when the channel subsystem references data in the EADM operation block (AOB), in the EADM move specification block (MSB), or in the EADM indirect data address word (AIDAW) Indicates that an invalid check block code (CBC) has been detected in the key.

AOB 주소 유효성: 세트될 때, SCSW의 AOB 주소 필드에 저장된 주소는 복구 목적으로 사용 가능하다는 것을 표시한다.

AOB Address Validity: When set, the address stored in the AOB address field of the SCSW indicates that it is available for recovery purposes.

EADM Extended Report Word (454): Includes, for example:

EADM Operations Block Error (B) Indicator: When set, indicates that the exception state stored in the EADM SCSW is associated with the specified EADM operation block (AOB); EADM Response Block Save (R) Indicator: When set, indicates that the EADM response block (ARSB) is stored.

When the SCSW's extended control indicator and the SCSW's extended status word format indicator are set, the EADM extended control word provides additional information about the model dependent nature that describes the conditions that may exist in EADM facilities.

The following channel reporting words (CRWs) may then be reported for ADM-type subchannels: subchannel installed parameters are initialized; Sub-channel installed parameters are restored; Sub-channels available; Pending channel event information.

The unsolicited events and malfunctions that occur in the EADM facility may be reported by the channel event information pending CRW.

Described in detail above is Extended Asynchronous Data Mover Facility used to move data blocks between main storage and storage class memory and perform other operations in the storage class memory. In one embodiment, information about EADM facilities and storage class memory is obtained using EADM commands. Specifically, since the storage class memory is not directly accessible, capability is provided to determine whether the storage class memory is allocated and, if assigned, to obtain information about its configuration. Specifically, in one example, the ability to communicate to a control program (e.g., the operating system) is provided to determine if and how much storage class memory has been allocated and if so, if so. The ability to determine available storage class memory is referred to herein as discovery and an example of a discovery function is provided by the Store Storage Class Memory (SCM) Information (SSI) command.

The Store Storage Class Memory Information (SSI) command is used to obtain information about storage class memory and extended asynchronous data mover facilities. The SSI command returns, for example, the next information obtained from the channel subsystem. This information is described in more detail below:

1. Characteristics of EADM Facilities (including:

AOB 당(per) 이동 명세 블록들 (MSB들)의 최대 카운트.

Maximum count of AOB per move specification blocks (MSBs).

MSB 당 최대 블록 카운트.

Maximum block count per MSB.

2. Characteristics of storage class memory, including:

SCM 증분 사이즈.

SCM incremental size.

SCM 주소 공간 내의 SCM 주소 증분들의 리스트.

List of SCM address increments within the SCM address space.

모델 종속 최대 SCM 주소.

Model dependent maximum SCM address.

The execution of the Store SCM Information command is synchronous and does not change any information contained in the channel subsystem.

More details regarding the SSI command are described with reference to Figures 5a-5d. Referring first to Figure 5A, in one embodiment, the command request block 500 for the Store SCM Information command includes, for example:

Length 502: A value that specifies the length of the command request block.

Command code 504: Value specifying the command code for Store SCM Information command.

Format (FMT) 506: A value specifying the format of the command request block.

Continuation Token 508: A value that can request a continuation point to resume from a previous response that was not completed. If the value of the continuation token is 0, a fresh start is made. If the value of the continuation token is not zero and is not recognized, a new start is made.

One embodiment of the response block 520 of the SSI command is described with reference to FIG. 5B, including, for example, the following:

Length 522: A value indicating the length of the command response block in bytes.

Response code 524: A value that describes the results of an attempt to execute an SSI command. The response code value indicates the length of the response block. For example, if the selected response code is stored, its length specifies 96 + N x 16 bytes, where N is the number of storage class memory address list entries and is described below. In one example, N is in the range 1? N? 248.

Format (FMT) 526: Value indicating the format of the command response block. The value of this field is, for example, zero.

RQ 528: Response qualifier value, as defined below:

There is no response qualification.

The specified sequence of tokens is not recognized and processed as though 0 were specified.

Maximum Block Count (MBC) 530 per MSB: A value that indicates the maximum value that can be used in the Block Count field of the Move Specification Block (MSB).

Maximum SCM Address (MSA) (532): A value that indicates the model dependent maximum SCM address. This value is the SCM address of the last byte in the highest addressable SCM increment.

SCM incremental size (IS) 534: A value indicating the size of each SCM increment in the SCM address list, for example, power-of-two.

Maximum MSB Count (MMC) 536: A value that indicates the maximum count of Movement Specification Blocks (MSBs) that can be specified in the EADM Operation Block (AOB).

Maximum Configurable SCM increments (MCI) (538): A value that is the maximum number of SCM increments that can be configured in the requesting configuration.

In one embodiment, the MCI does not exceed 2 ^{(64- IS )} . For example, for a 16-Gbyte SCM incremental size, MCI ≤ 2 ^(64-34) because all 16-Gbyte SCM increments must be addressable within the addressing limits of the 64-bit address. Also, ((MCI + 1) x IS) -1 does not exceed the model dependent maximum SCM address.

The number of configured SCM increments (NCI) reduces the total size (TS) that can be specified when a successful Configure Storage Class Memory command is executed, described below, to be TS ≤ (MCI-NCI). However, if the number of SCM increments in the initialized state does not effectively constitute the maximum MCI limit of the requesting configuration, based on the total capacity of the system and allocations already assigned to other configurations, It may or may not be fulfilled.

Total initialized SCM increments (540) of the CPC: A value indicating the number of SCM increments in the initialized state (e.g., for the central processing complex (CPC)) for the system. If the system is logically partitioned, this is the total number of available increments to be assigned to or by partitions.

Total Uninitialized SCM increments of CPC (542): A value indicating the number of SCM increments that have not been initialized for the system (e.g., for CPC).

SCM Measurement Block Size (544): Value that is the block size (BS) in bytes of the SCM measurement block. In one example, this value is a power of 2 and the maximum SCM measurement block size is, for example, 4096 bytes.

Maximum number of SCM resource parts (546): A value that is the maximum number of SCM resource parts (RP) (for example, I / O adapters) in the CPC. Each SCM increment is associated with an SCM resource part. Each SCM resource (e.g., one or more I / O adapters and one or more SSDs) includes one or more parts. The maximum number of SCM resource parts is, for example, 509.

In one example, the term "resource part" is defined for use in acquiring measurement information about the storage class memory. Each SCM increment can be distributed across multiple adapters and each adapter provides some utilization / measurement data. Thus, each returned measurement block is identified by a tuple consisting of the sum of the increment identifier and the resource identifier.

SCM data unit size 548: A model dependent value that indicates the number of bytes contained in the SCM data unit (SCM data unit). In one example, the data unit is defined for use in acquiring measurement information regarding the storage class memory. The reported count is a count of data units rather than bytes.

Continuation token 550: a model dependent value by which the subsequent issuance of the SSI command may continue at the succession point indicated by this token. The content of the continuation token depends on the model.

Storage class memory address list 552: Origin of SCM address list. When the stored response code is a predefined value, a plurality of SCM address list entries are stored (e.g., (length - 96) / 16 SALEs are stored).

In one example, the number of stored SALES depends on the number of SCM increments in the requestor's configuration, the respective state when the command is executed, and the channel subsystem model. The actual number of zero or more SALES to be stored and stored is determined, in one example, by subtracting 96 from the size (length) of the response block and dividing the result by 16.

Each SCM address list entry (SALE) represents one SCM increment occupying a range of SCM addresses. The starting SCM address of the SCM increment represented by SALE is the SCM address of the first byte of the corresponding SCM increment that is contained in SALE. The ending address is calculated, in one example, by adding the SCM increment size, in bytes, to the starting SCM address and then subtracting one. This is the SCM address of the last byte of the SCM increment. The storage class memory represented by SALE is a set of contiguous bytes of SCM position, the set, and in one embodiment, start from the natural 2 ^IS byte range ^{(on a natural 2 IS byte boundary} ).

SALE is stored when the corresponding SCM increment is in a configured state and space is available in the response block to SALE. If the space in the SCM address list of the response block is exhausted, the value is stored in the continuation token and execution is completed with the specific response code.

Two or more SALES are stored in ascending order of their SCM addresses.

One embodiment of SALE is described with reference to Figure 5C. In one example, SALE 552 includes, for example, the following:

SCM Address (SA) (560): A value that is the starting SCM address of byte 0 of the corresponding SCM increment in the SCM address space, aligned to the natural boundary as determined by the SCM increment size (2 ^IS bytes).

Persistence Attribute (P) (562): A value indicating the current persistence rule applied to the SCM increment. All positions within the SCM increment inherit the persistence rule. The possible persistence rules are:

Rule 1 - Preserve data when power is off.

Rule 2 - Preserve data until power on reset or IML.

Operation (op) state 564: A value indicating the operation state of the storage class memory increment indicated by SALE. The operation state is valid only when the associated SCM increment is in the configured state.

Examples of operation states are:

Operable (Op): The storage class memory indicated by SALE is available for all I / O operations. When successfully configured, it can enter an operational state and re-enter when leaving a temporary or permanent error state.

Temporary error (TE): Storage class memory indicated by SALE is not possible for all I / O operations. The data state is invalid, but the data content is preserved when switching from operable to temporary error. A temporary error state enters from an operable state when there is no access to an SCM increment.

Permanent Error (PE): The storage class memory indicated by SALE is not available for all I / O operations. The data state is invalid and the data is lost. A permanent error state enters from a temporary error state or an operable state when an uncorrectable error condition is recognized.

If the operation completes with an indication of a permanent error set in the exception qualifier code of the EADM response block, at least the corresponding SCM increment has entered the permanent error state. However, more than one SCM increment may have entered the permanent error state.

If the SCM increment is not in the operable state, then the I / O operation that refers to the incremental location is in a state in which either the temporary or permanent error state is set in the exception qualifier code of the EADM response block I know.

Data State 566: A value indicating the data state of the contents of the storage class memory increment indicated by SALE. The data state is valid when the associated SCM increment is in a configured state and in an operable state.

Exemplary data states include:

Zeroed - The contents of the SCM increment are all zeros.

Valid - The SCM incremental content is an accumulation of all successful write type operations. Incremental positions that have not yet been written remain zero or remain unpredictable.

Unpredictable - the content of the SCM increment before the light type I / O operation is unknown. After one or more light type operations are performed, the data contents of the other unlit locations will remain in an unpredictable state, even though the successfully written locations will make the data state of the SCM increment valid.

A transition from a zeroed state or an unpredictable state to a valid state occurs with a first successful write. Due to any difference between the size of the written data and the size of the target SCM increments, the change in validity does not describe the actual condition of all data that has not yet been written. This position, which has never been accessed for the write, is still described as valid or unpredictable.

Rank 568: A value indicating the conceptual quality of the storage class memory represented by SALE. The rank is only valid when the associated SCM increment is in the configured state and the operable state. A value of 0 (zero) means that there is no rank. A nonzero value in the stated range means that a rank exists. In this example, a rank value of 1 is the highest or the best rank. A rank of 15 is the lowest or worst rank. If all else is the same, the SCM with the higher rank is preferred to the SCM with the lower rank.

R (570): This field indicates that the SCM increment recognizes the release operation. It is related to the following actions:

1. The released block must be written before it is read, otherwise an error is recognized in the read operation if the read precedes the write. Because of this error, the SCM increment remains in the operable state.

2. Upon initial configuration, the data state is zero.

3. The program can perform a special operation, called release, which places the specified block into a released condition.

Resource ID 572: The resource identifier (RID) of the resource providing the SCM increment represented by SALE is a non-zero value. If RID is zero, the resource ID is not displayed. In one specific example, the RID represents a number of adapters and SSDs that provide storage for SCM increments. When specific RAID algorithms are applied or striped to improve performance by allowing concurrent processing I / O operations across multiple adapters / SSDs, the RID may represent a compound entity .

Further details regarding the configuration states of the storage class memory and the operation state and data state of the storage class memory address list are described below.

First, referring to FIG. 6A, configuration states and events / actions that are switched within these states are described. As shown, the SCM states are configured, standby, and reserved. The SCM may be deployed from Reserved to Standby, then in the Atmosphere. From the configured state, the SCM can be deconfigured and enter the Reserved state.

Referring to FIG. 6B, operation states and events that are switched within these states are shown. The SCM increment should be in the idle state to be in the configured state and in the operational state when the configure action is successfully completed. In the zeroed state, the first write for the SCM increment moves the SCM increment to the valid state. Intervening power off and subsequent power on of the SCM increment not represented with rule 1 persistence moves the SCM increment to the unpredictable state.

The error E may cause a transition to a temporary error (TE) state or a permanent error (PE) state, depending on the model's specifics of the error. Acquisition (A) of connectivity may cause a transition from a temporary error state to an operable (Op) state. The deconfiguration of the SCM increment may occur regardless of the operation state of the SCM increment.

FIG. 6B also illustrates data states when in an operable state, depending on how they entered the operable state. The data state is valid when in the configured and operational state and is applied to the corresponding SCM increment. Valid data states are zeroed, unpredictable, and valid. The following are possible data states from various entries into the operable state:

대기중으로부터 - 영이됨(z)

From the atmosphere - becoming spirit (z)

일시 에러로부터 - 유효(v)

From the temporary error - valid (v)

영구 에러로부터 - 예측 불가능(u) 또는 영이됨(z)

From permanent error - unpredictable (u) or zero (z)

오퍼레이션 가능으로부터 - 유효(v) - 제1 라이트

From operability enabled - Valid (v) - First light

오퍼레이션 가능으로부터 - 예측 불가능(u) - 파워가 순환되고(power cycled) 지속성(persistence)은 규칙 1이 아니다.
[00346]

From possible operation - unpredictable (u) - power cycled and persistence is not rule 1.

When not in the operable state, the data state is invalid.

When it is configured for the first time and before the first write, the data of the SCM increment is in the null state, which means that the contents are all zeros.

The incremental data content of the SCM increment is not changed when moving to the temporary error state or in the temporary error state, whereas the increment is inaccessible. Therefore, saying that the data is valid is descriptive, but it does not make much sense because of the lack of program accessibility. Therefore, the data state is invalid in this situation. Also, if data integrity is affected, based on an error that causes a transition from operational availability to a temporary error, a permanent error state is entered, the data state is left in an invalid state, and the data is lost. Even if concurrent repair is able to move the SCM increments in the persistent error state to the operable state without doing both disassembly and then reconfiguring, the original data is still lost State, and whether the data state is unpredictable or zero depends on the model.

The persistence of the SCM increment and its RAS (Reliability, Accessibility and Service Availability) characteristics can also determine changes from valid to unpredictable data states. If persistence is exceeded, the data state is expected to transition from valid to unpredictable.

The transition from zero or unpredictable data states to valid data states occurs with the first successful write. Due to the difference between the size of the written data and the size of the target SCM increments, the change in validity does not describe the actual condition of all data locations that have not yet been written. This position before the first write access is still effectively null or described as unpredictable.

After the SCM increment is configured, the unsolicited notification is pending when one or more events that can be observed in the response of the Store SCM Information command occur. Examples include:

1. The operation state changes from operable to temporary or permanent, but is not reported by the failed operation.

2. The operation status changes from temporary error to operable.

3. Operation status changes from temporary error to permanent error.

4. The rank changes.

Examples in which voluntary notifications are not pending are:

1. The data state changes from zero to zero or from unpredictable to valid.

2. The Configure Storage Class Memory command is completed.

3. The Deconfigure Storage Class Memory command is completed.

When the notification is pending, the program can issue a Store SCM Information command to observe the notification and obtain information. The Store SCM Information command may also be issued at other times when the program wishes to acquire information about the SCM and / or SALE.

In one example, the notification includes a machine check interruption issued to the program with a corresponding CRW indicating an event report. The program issues a CHSC Store Event Information command and obtains a response block with the content code signaling a storage class memory change notification.

In one embodiment, the Store SCM Information (SSI) command is a channel subsystem command issued by a program (e.g., operating system) to obtain information regarding storage class memory and / or SCM address list entries. In one example, when the program issues a Channel Subsystem Call command, the SSI command is displayed in the command block of the command, which is sent to the channel subsystem. If this command is performed in the channel subsystem, the response is returned to the response block, which is the remainder of the 4K byte control block (i.e., the requested information is stored in the main storage area designated for the response block). More details regarding the operation of this command are described with reference to Figure 5d.

Initially, the program generates the request block indicated above and requests a Store SCM Information command (step 580). The request block is obtained (step 582) by the channel subsystem and one or more validation checks are performed on the validity of the request block (e.g., valid length field, valid command request block format, installed command, etc.). If the request is not valid (query 584), a response code indicating the problem is placed in the response block (step 586) and the response block is returned (step 592).

However, if the request is valid (query 584), the channel subsystem obtains information from the machine (e.g., processor, etc.) (step 588) and populates the response block (step 590). The response block is then returned (step 592). For example, information is loaded by the firmware into the main storage, which is stored in the nonvolatile storage of the machine and accessible only by the firmware during system initialization. The channel subsystem (i. E., Firmware in this case) acquires that information by reading information from main storage accessible only by the firmware.

In response to receiving information regarding the storage class memory or the like, a determination can be made regarding changing the configuration of the storage class memory. This determination can be made manually or automatically by a program or other entity. The configuration can be changed by adding the increments or deleting the increments, as described below.

In one example, a Configure Storage Class Memory command is used to configure the storage class memory. This command requests that a certain amount of storage class memory be configured from an available pool of the system. The amount is specified in size and is encoded with a count of SCM increments.

Unless otherwise stated, the number of SCM increments used to satisfy the request is in the initialized state. If the number of requested SCM increments exceeds the limit of the maximum configurable SCM increments, a specific response code is provided.

The contents of each increment are zero due to valid CBC. The corresponding persistence rules associated with each configured SCM increment are set, for example, by manual controls.

One embodiment of a command request block for the Configure Storage Class Memory command is shown in FIG. 7A. In one example, the request block 700 of Configure Storage Class Memory includes:

Length 702: A value that specifies the length of the command request block.

Command code 704: Value specifying the command code for the Configure Storage Class Memory command.

Format (FMT) 706: A value specifying the format of the command request block.

Total Size (TS) 708: A value that specifies the size of the requested storage class memory, encoded as a count of SCM increments. Adding TS to the count of already configured SCM increments should not exceed the Maximum Configurable SCM Increments (MCI) threshold. If the number of SCM increments in the initialized state is less than the specified total size, a specific response code is provided.

Asynchronous Completed Correlator (ACC) 710: The value returned in the asynchronous Completion Notification field of the notification response described below. The correlator serves to resume the original thread that initiated the request.

One embodiment of a command response block for the Configure Storage Class Memory command is shown in Figure 7B. In one embodiment, the command response block 730 includes:

Length 732: A value indicating the length of the command response block.

Response Code 734: A value describing the results of an attempt to execute the Configure Storage Class Memory command.

If the defined response code is stored in the response code field, the asynchronous process is initiated to complete processing of the command. If a response code other than the defined code is stored in the response code field, the SCM increment is not configured, the asynchronous process is not started, and no subsequent notification is made. Completion of the asynchronous process is indicated in the notification response.

Format (FMT) 736: A value that indicates the format of the command response block.

The Configure Storage Class Memory command is issued by the program to request a certain amount of storage class memory to be configured into the SCM address space. One embodiment of the logic used to construct the SCM is described with reference to Figure 7C.

Initially, the program issues a Channel Subsystem Call command that includes a Configure SCM command (step 740). The request block of the Configure SCM command is obtained by the channel subsystem (step 742) and the channel subsystem attempts to execute the command (step 744). If the execution attempt of the command yields a response code that does not indicate success (query 746), the response code is placed in the response block of the Configure SCM command (step 748) and the response block is returned (step 750).

If a successful response code is displayed (query 746), the response code is placed in the response block (step 752) and the response block is returned (step 754). In this example, a successful response code indicates that the length field of the request block is valid; This command is available in the system; The command request block has a valid format; The channel subsystem is capable of executing commands (i.e., not busy); The requested total size does not exceed the limit of the maximum configurable SCM increments of the requested configuration; Indicating that the requested total size does not exceed the number of SCM increments in the initialized state.

Then, an asynchronous process for completing the configuration is initiated (step 756). Further details relating to this process are described with reference to Figure 7d.

In one embodiment, the asynchronous process performs the configuration and assigns one or more increments (step 760). For example, for each configured SCM increment, the internal controls are changed to allow the newly configured increment to access the I / O move requests for that partition. Specifically, in response to the channel subsystem receiving the CHSC Configure command, the firmware of the channel subsystem examines the internal tables, in order to verify that there are sufficient increments to meet the request and that the request is for configuration So as not to exceed the maximum configurable SCM increments. If the request is valid, the firmware updates one or more tables, in order to allocate the increment (s) to the configuration and place the increment (s) in an operable state for the configuration. The increments are then accessible to the I / O transfer requests (as described above) from the configuration. Completion of the asynchronous process is indicated in the notification response (step 762).

The notification response data for the Configure Storage Class Memory command is returned to the response block of the SEI (Store Event Information) command. One embodiment of the format of the notification response block used for the Configure Storage Class Memory command is described with reference to FIG.

In one embodiment, the Notify Reply block 800 of the Configure Storage Class Memory command includes:

Length 802: A value that indicates the length of the command response block.

Response Code 804: Value describing the results of an attempt to execute the Store Event Information CHSC command.

Format (FMT) 806: A value indicating the format of the command response block.

Notification Type (808): This is a value indicating that this is an EADM related notification.

P (810): When set, specifies that the channel subsystem has pending event information in addition to the information provided in response to this CHSC command.

V (812): When set, it indicates that the channel subsystem has acknowledged an overflow condition and event information has been lost.

Content code 814: A value indicating that this is a response to the termination of the execution of the asynchronous process initiated by the Configure Storage Class Memory command request.

Secondary Response Code (816): A value that further describes the results of an attempt to execute the Configure SCM command.

When the secondary response code is the specified value, the configuration of the full amount of storage class memory is completed as originally requested. Otherwise, for example, response codes may be provided indicating: Invalid Length field, No Configure SCM command installed, Configure SCM command block has invalid format, Total requested size will exceed MCI limit, Number of SCM increments in which the requested total size is initialized , And the channel subsystem is busy.

Asynchronous Completion Corrector (ACC) 818: The value originally specified in the corresponding command request block.

One embodiment of the Store Event Information command used to return the notification response block for the Configure Storage Class Memory command is described with reference to Figures 8b-8c.

The Store Event Information command is used to store event information pending by the channel subsystem. Normally, this command is executed as a result of the program receiving the event information pending channel report.

Execution of the Store Event Information command can change the information contained in the channel subsystem. The Store Event Information command is executed synchronously.

One embodiment of a command request block for the Store Event Information command is described with reference to FIG. In one example, the request block 830 includes:

Length 832: This field specifies the length of the command request block.

Command code 834: This field specifies the Store Event Information command.

Format (FMT) 836: Value specifying the format of the command request block.

Notification Type Selection Mask 838: A mask in which each bit position corresponds to a logical processor selector (LPS) value of the same numeric value. In one example, bit 0 is assumed to be ignored. When the bit position is zero in the specified range starting at 1, the notification type corresponding to that bit position is not stored in the response block and is discarded if it is found to be pending. When this bit is 1, the notification type corresponding to that bit position may be stored in the response block.

In one embodiment, with reference to FIG. 8C, the response block 850 for the Store Event Information command is described below as: < RTI ID = 0.0 >

Length 852: A value that specifies the initial length of the command response block. The completion length depends on the response code, and the response code is stored as a result of an attempt to execute the Store Event Information command.

If a response code other than the code indicating success is stored in the response code field, no information is stored in the response data area.

If a response code indicating success is stored in the response code field, the event information is stored in the response data area.

Response Code 854: Value describing the results of an attempt to execute the Store Event Information command.

For the Store Event Information command, the response data area includes a fixed length portion and a variable length portion.

For a specified format response, when NT is not zero, the format is determined according to the specific notification type, and the format of the content code dependent field is determined together in accordance with the specific notification type and the content code (CC) field.

Format (FMT) 856: Value specifying the format of the command response block.

Notification Type (NT) 858: A value indicating the notification type (NT). A specific value is provided for the Configure SCM command.

P flag 860: When set, specifies that the channel subsystem has pending event information in addition to the information provided in response to this CHSC command. The program can obtain additional information by executing the Store Event Information command again. When not set, this flag specifies that the channel subsystem does not have additional pending event information.

V flag 862: When set, it indicates that the channel subsystem has acknowledged an overflow condition and event information has been lost. An overflow condition was recognized while event information not included in the response data area was the latest pending information. The overflow does not affect the information contained in the response data area.

Content code (CC) 864: A value that describes the type of information contained in the response data area. In one example, this value indicates a storage class memory change notification in which one or more SCM increments changed the state or status.

Content Code Dependent Field 866: This field may contain additional information about the event.

Successful notification of a configuration change may cause the program to issue a Store SCM Information command to obtain details about the configuration.

The storage class memory may be increased, but the storage class memory may be decreased. The Deconfigure Storage Class Memory command requests that a certain amount of storage class memory be removed from the SCM address space of the requesting configuration. The SCM increment to be deconfigured must be in a configured state.

The SCM increments to be deconfigured are specified in the SCM increment request list described herein. One or more adjacent SCM increments may be specified in the SCM increment request list entry (SIRLE). A separate SIRLE may be specified for a list of other increments and a separate list (also known as an extent) of non-contiguous increments.

Regardless of the persistence rules, a successful deconfiguration of the SCM increment places the increment in an uninitialized state. When zeroing is completed, the SCM increment is switched from the unitialized state to the initialized state.

One embodiment of a command request block for a Deconfigure Storage Class Memory command is shown in FIG. 9A. The command request block 900 for the Deconfigure Storage Class Memory command includes, for example, the following:

Length 902: A value that specifies the length of the command request block. In one example, the length is 32+ (Nxl6) bytes, where N is a count of SCM incremental request list entries (SIRLEs). In this example, the valid length is evenly divisible to 16 and (32 + 1x16) ≤ L1 ≤ (32 + 253x16).

Command code (904): Deconfigure Storage Class A value that specifies the command code for the Memory command.

Format (FMT) 906: A value specifying the format of the command request block.

Asynchronous Completed Correlator (ACC) 908: The value returned in the asynchronous completion notification described above.

SCM incremental request list 910: This field contains the SCM incremental request list (SIRL). The SCM increment request list includes one or more entries (SIRLEs). The length of the SIRL is determined from the value of the length field.

The SCM incremental request list entry (SIRLE) specifies the size and location of a specified area of storage class memory (e.g., a list of increments). The extent or SCM extent is the explicit size of the storage class memory.

Referring to Figure 9B, in one example, SIRLE 920 includes:

Total Size (TS) (922): A value specifying the size of the storage class memory to be deconfigured, encoded as a count of SCM increments.

Start SCM address (SA) (924): This is the SCM address, which is the location within the SCM address space, from which the first or only SCM increment unconfigured by SIRLE is removed. In this example, it is assumed that the least significant bit positions that make up the offset in the first SCM increment are ignored and become zero.

When the total size is greater than one, each additional SCM increment beyond the first increment is located at an SCM address that is equally divisible into the SCM incremental size, including the configured SCM increment, And its position is adjacent to the last byte of the prior SCM increment. In other words, it refers to the next continuous position.

If the space described is not fully filled with the configured SCM increments, depending on the starting address and the total size, the specified response code is provided, the SCM increment is not unconfigured, and the asynchronous process is not started And no subsequent notification will occur.

Upon successful completion, each unconfigured SCM increment enters a reserved state and then becomes zero before being placed into a standby state.

A command response block for the Deconfigure Storage Class Memory command is shown in Figure 9c. In one embodiment, the command response block 950 includes:

Length 952: A value indicating the length of the command response block.

Response Code (954): Value describing the results of an attempt to execute a Deconfigure Storage Class Memory command.

If the response code of the specified value is stored in the response code field, an asynchronous process is started to complete the processing of the command. If a response code other than the specified value is stored in the response code field, the SCM increment is not unconfigured, the asynchronous process is not started, and no subsequent notification occurs. Completion of the asynchronous process is indicated in the notification response.

All entries in the SCM increment request list are examined before the synchronous part of the Deconfigure Storage Class Memory command is completed with the specified response code, so that all specified SCM increments are in a configured state.

Format (FMT) 956: Value indicating the format of the command response block.

One embodiment of the logic associated with the Deconfigure SCM command is described with reference to Figures 9d-9e.

Initially, the program issues a Channel Subsystem Call command containing a Deconfigure SCM command (step 970). The request block of the Deconfigure SCM command is obtained by the channel subsystem (step 972) and the channel subsystem attempts to execute the command (step 974). If the execution attempt of the command yields a response code that does not indicate success (query 976), the response code is placed in the response block of the Deconfigure SCM command (step 978) and the response block is returned (step 980) .

If a successful response code is displayed (query 976), the response code is placed in the response block (step 982) and the response block is returned (step 984). In this example, the successful response code is: the length field of the request block is valid; Commands are available in the system; The command request block has a valid format; The channel subsystem is able to perform (i. E., Not busy) the command; And SCM increments were originally in the configured state.

An asynchronous process is then initiated to complete the deconfiguration (step 986). More details regarding this process are described with reference to FIG. 9E.

In one embodiment, the asynchronous process performs an unconfiguration (step 990). For example, one or more increments may be deallocated. The SCM increment moves from the configured state to the held state. Upon entry into the reservation state, the zeroing process continues, and upon completion, the SCM increment is switched to the wait state. Completion of the asynchronous process is indicated in the notification response (step 992).

Notification response data for the Deconfigure Storage Class Memory command is returned to the response block of the Store Event Information (SEI) CHSC command. One example of this response block is described with reference to FIG. However, in this example, the content code indicates that it is a response to the termination of the execution of the asynchronous process initiated by the Deconfigure Storage Class Memory command request. Similarly, the secondary response code further describes the results of the attempted execution of the Deconfigure Storage Class Memory command.

In a further embodiment, allocation and deallocation of storage increments may be requested through a panel submitted to the user. For example, one service element is used to provide a graphical interface through which a user can specify parameters in the system. For storage class memory, a panel called the storage class memory allocation panel allows the user to perform the following operations:

1. Specify maximum configurable increments (MCI) for a given configuration.

2. Assign increments to the configuration.

3. Unassign the increments from the configuration.

The panel also allows you to view configuration incremental allocations and MCIs, and to view the number of available pools, unavailable pools, and increments in uninitialized pools. A notification is sent to the configurations when the allocation of increments is changed due to an action at the SE, or when the size of one of the pools changes.

What is described in detail above is about facilities for managing storage class memory. Facility provides abstraction that allows a program to access the memory without any specific knowledge of the memory. In accordance with one or more embodiments of the present invention, for carrying data between the main storage and the SCM; For clearing or clearing the SCM; For configuring or unconfiguring SCM; And the capability to discover the configuration of the SCM is provided. Other capabilities are also provided.

In one embodiment, the storage class memory is provided as a flat memory space in user level programs, regardless of its physical implementation across multiple devices and I / O adapters.

Details regarding the channel subsystem and / or ADM facility are disclosed in "Method and Means for Enabling Virtual Addressing Control by a Software User via a Hardware Page Transmission Control Entity" A Hardware Page Transfer Control Entity, U.S. Patent No. 5,377,337 entitled "Antognini et al., Issued December 27, 1994; U.S. Patent Serial No. 5,442,802 entitled " Asynchronous Co-Processor Data Mover Method and Means, "(Brent et al., Issued Aug. 15, 1995); U.S. Patent No. 5,526,484 entitled " Method and System for Pipelining the Processing of Channel Command Words, "(Casper et al., Issued June 11, 1996) Each of which is incorporated herein by reference in its entirety.

Then, the additional information (for the specific implementation of the z / Architecture ^®) in the channel subsystem and the command associated with it is provided below:

Input / output (I / O)

The terms "input" and "output" are used to describe the transfer of data between I / O devices and main storage. Operations related to this kind of transmission are called I / O operations. The facilities used to control I / O operations are collectively referred to as the channel subsystem. (I / O devices and their control units are connected to the channel subsystem.

Channel subsystem ( The Channel Subsystem )

The channel subsystem directs the flow of information between the I / O devices and the main storage. The channel subsystem alleviates the task of communicating directly with I / O devices from CPUs and enables data processing to proceed concurrently with I / O processing. The channel subsystem uses one or more channel paths as communication links to manage the flow of information into / out of I / O devices. As part of the I / O processing, the channel subsystem also performs path management operations by testing channel path availability, selecting available channel paths, and initiating performance of I / O operations by the device.

There are sub-channels in the channel subsystem. One dedicated subchannel is provided for each I / O device that the program can access through the channel subsystem.

The multiple-subchannel-set facility is an optional facility. When it is installed, the subchannels are partitioned into multiple subchannel sets, and each subchannel set can provide one dedicated subchannel to the I / O device. Depending on the model and interface used, some I / O devices may only be allowed access through certain sub-channel sets.

Each subchannel provides information about the device's attachment to its associated I / O device and channel subsystem. The subchannel also provides information about I / O operations and other functions associated with the associated I / O devices. Subchannels are the means by which the channel subsystem provides information about the associated I / O devices to the CPUs and the CPUs obtain this information by executing I / O commands. The actual number of subchannels provided depends on the model and configuration; The maximum addressability is 0-65,535 in each subchannel set.

The I / O devices are connected to the channel subsystem using channel paths through the control units. The control units can be connected to the channel subsystem using more channel paths than one channel path, and the I / O device can be connected to more than one control unit. In all, an individual I / O device is accessible to the channel subsystem using up to eight different channel paths, via subchannels, depending on the model and configuration. The total number of channel paths provided by the channel subsystem depends on the model and configuration; The maximum addressing capability is 0-255.

The performance of the channel subsystem depends on its use and the system model it is implemented on. An I / O device (e.g., a magnetic tape unit or disk storage) designed to provide channel paths with different data-transfer capabilities and to transfer data only at a particular rate, It can only work in the path.

The channel subsystem includes common facilities for control of I / O operations. When these facilities are provided in the form of separate autonomous devices specifically designed to control I / O devices, the I / O operations are completely redundant with activities in the CPUs do. The only main storage cycles required by the channel subsystem during I / O operations are when the sub-channels are implemented as part of the unaddressable main storage, the channel subsystem may be required to access the sub-channels Cycles along with the cycles needed to transfer data and control information to / from the final locations in main storage. These cycles do not delay CPU programs, except when the CPU and the channel subsystem attempt to reference the same main storage area at the same time.

The subchannel set ( Subchannel Sets )

When multiple-subchannel-set facilities are installed, the subchannels are divided into multiple subchannel sets. These can be up to a maximum of four subchannel sets, each identified by a subchannel-set identifier (SSID). When the majority-subchannel-set facility is not installed, there is only one set of subchannels with SSID of 0 (zero). When the majority-subchannel-set facility is not enabled, only the subchannel set 0 (zero) appears in the program.

Subchannel ( Subchannels )

The subchannel contains the information required to provide the logical appearance of the device to the program and to sustain a single I / O operation. The subchannel includes information in the form of a channel-program designation, channel-path identifier, device number, count, status indications, and I / O-interruption-subclass code, And internal storage that also includes information about availability and pending or performing functions. I / O operations are initiated by the execution of I / O commands that specify a subchannel associated with the device with the device.

Each device is accessible using one subchannel in each channel subsystem to which the device is assigned while being configured at installation. The device may be a physically identifiable unit or may be contained within a control unit. For example, in certain disk-storage devices, each actuator used to retrieve data is considered a device. In all cases, in terms of the channel subsystem, the device is uniquely associated with one subchannel, and the channel path is accessible via the channel path. It is the entity that responds to the selection by the system.

In some models, subchannels are provided as blocks. In these models, more sub-channels may be provided than connected devices. Subchannels - that is, subchannels that are provided but do not have the devices assigned to them - are not used by the channel subsystem to perform a function, and have an associated device-number-valid bit Information is stored in the sub-channel-information block of the sub-channel.

The number of subchannels provided by the channel subsystem is independent of the number of channel paths to the associated devices. For example, devices still accessible through alternate channel paths are represented by a single sub-channel. Each subchannel is addressed using a 16 bit binary subchannel number and a 2 bit SSID when the subchannel-set facility is installed.

After the I / O processing in the subchannel is requested by the execution of START SUBCHANNEL, the CPU is released for other work and the channel subsystem assemble or disassemble the data to create I / O Synchronize the transfer of data bytes between the device and main storage. To achieve this, the channel subsystem maintains and updates the addresses and counts describing the destination or source of data in the main storage. Similarly, when an I / O device provides signals to invoke the attention of the program, the channel subsystem converts the signals into status information and sends the information to the subchannel Where the information can be retrieved by the program.

Input / Output Devices  connect( Attachment of Input / Output Devices )

Channel path ( Channel Paths )

The channel subsystem communicates with the I / O devices using channel paths between the channel subsystem and the control units. The control unit can be accessed by the channel subsystem using more than one channel path. Similarly, I / O devices can be accessed by the channel subsystem through more than one control unit, each with one or more channel paths to the channel subsystem.

Devices connected to the channel subsystem by multiple channel paths configured to go to a subchannel may be accessed by the channel subsystem using any of the available channel paths. Similarly, a device with a dynamic-reconnection capability and operating in a multiple path mode, when logically reconnecting to a channel subsystem to continue a series of I / O operations, To < / RTI > select any of the available channel paths that are configured to go to < RTI ID = 0.0 >

The channel subsystem may include one or more channel path types. Examples of channel-path types used by the channel subsystem are an ESCON I / O interface, a FICON I / O interface, a FICON-converted I / O interface, and an IBM System / 360 and System / 370 I / O interface. The term "serial-I / O interface" is used to refer to the ESCON I / O interface, the FICON I / O interface, and the FICON-converted I / O interface. The term "parallel-I / O interface" is used to refer to the IBM System / 360 and System / 370 I / O interfaces.

The ESCON I / O interface is described in the system library publication "IBM Enterprise Systems Architecture / 390 ESCON I / O Interface, SA22-7202" Are included by reference.

The FICON I / O interface is described in the ANSI standard document "Fiber Channel-Single-Byte Command Code Sets-2 (FC-SB-2)".

The IBM System / 360 and System / 370 I / O interfaces are described in the System Library Publication "IBM System / 360 and System / 370 I / O Interface Channel to Control Unit OEMI, GA22-6974" Are included by reference.

Depending on the type of channel path, the facilities provided by that channel path, and the I / O device, the I / O operation may be performed in a frame-multiplex mode, a burst mode, Multiplex mode can occur in one of these three modes.

In frame-multiplex mode, the I / O device remains logically connected to the channel path for the duration of the execution of the channel program. The facilities of the channel path operable in the frame-multiplex mode can be shared by a number of concurrently operating I / O devices. In this mode, the information required to complete an I / O operation may be divided into frames and interleaved with frames of I / O operations for other I / O devices. During this period, multiple I / O devices are considered to be logically connected to the channel path.

In burst mode, the I / O device remains logically connected to the channel path to monopolize the channel path and transmit a burst of information. During the time that a burst is transmitted, no other device can communicate over the channel path. A burst may consist of a number of bytes, a whole block of data, a series of blocks with associated control and status information (block lengths may be zero), or status information that monopolizes the channel path. The facilities of the channel paths operable in the burst mode may be shared by a number of concurrently operating I / O devices.

Some channel paths may allow an absence of data transfer for 30 seconds during a burst mode operation, such as when a long gap is read on a magnetic tape. A device malfunction may be indicated when the absence of data transmission exceeds a specified limit.

In byte-multiplex mode, the I / O device remains logically connected to the channel path only for a short interval of time. The facilities of the channel path operable in the byte-multiplex mode may be shared by a plurality of simultaneously operating I / O devices. In this mode, all I / O operations are divided into short time intervals and only one segment of information is transmitted over that channel path during the time intervals. During this interval, only one device and its associated subchannels are logically connected to that channel path. The intervals associated with the concurrent operation of the multiple I / O devices are arranged in order in response to requests from the devices. The facility of the channel subsystem associated with the subchannel exercises its control over any one operation for the time required to transmit one segment of information. A segment may consist of a single byte of data, a few bytes of data, a status report from the device, or a control sequence used to start a new operation.

Typically, devices with high data-transfer-rate requirements operate in this channel path and frame-multiplex mode, slower devices operate in burst mode, and the slowest devices are byte- Mode. Some control units have a manual switch to set the desired operation mode.

An I / O operation occurring on a channel path of a parallel-I / O-interface type may occur in either a burst mode or a byte-multiplex mode depending on the channel path and the facilities provided by the I / O device. For improved performance, some channel paths and control units are provided with facilities for high-speed transfer and data streaming. A description of these two facilities can be found in the system library publications "IBM System / 360 and System / 370 I / O Interface Channel to Control Unit OEMI, GA22-6974 ", which is incorporated herein by reference in its entirety.

I / O operations that occur on a serial-I / O-interface-type channel path may occur in either frame-multiplex mode or burst mode. For the sake of performance enhancement, some control units connected to the serial-I / O-interface provide sense data to the program with a presentation of the unit- And the like.

Depending on the control unit or channel subsystem, access to the device via subchannels may be limited to a single channel-path type.

The modes and features described above only affect the protocol and transmission rate used to transmit information over the channel path. There is no impact that can be observed by the CPU or channel programs as to how these programs are executed.

Control unit ( Control Units )

The control unit provides the logical capabilities needed to operate and control the I / O devices and configures the characteristics of each device so that each device can respond to the standard form of control provided by the channel subsystem ).

Communication between the control unit and the channel subsystem takes place via the channel path. A control unit receives control signals from the channel subsystem, controls the timing of data transmission over the channel path, and provides indications of the status of the device.

The I / O device connected to the control unit may be designed to perform only certain limited operations, or may perform many other operations. A typical operation moves the recording medium and the data to be recorded. To accomplish those operations, the device needs specific signal sequences that are specific to the type of device. The control unit decodes the received commands from the channel subsystem, interprets them for the particular type of device, and provides the signal sequence required to perform the operation.

The control units may be housed separately in a housing, or may be physically and logically integrated with an I / O device, a channel subsystem, or a CPU. In the case of most electromechanical devices, there is a clear interface between the control unit and the control unit as there is a difference in the type of device the device requires. These electromechanical devices are often of the type requiring only one device in the group to be connected to the control unit to transfer data at one time (for example, a self-tape unit or a disk-access mechanism) Shared among the I / O devices. Conversely, in some electronic I / O devices such as channel-to-channel adapters, the control unit does not have its own identity.

From the programmer's point of view, most of the functions performed by the control unit can be integrated with the functions performed by the I / O device. Therefore, this specification does not usually make any specific reference to the function of the control unit; Implementation of I / O operations is described as if the I / O device is in direct communication with the channel subsystem. A reference to a control unit is made only when emphasizing the function performed by the control unit or when sharing the control unit among multiple devices affects the performance of the I / O operations.

I / O devices (I / O devices) Devices )

An input / output (I / O) device provides external storage that is a means of communication between data processing systems or a means of communication between the system and the environment of the system. I / O devices include devices such as magnetic tape units, direct-access-storage devices (e.g. disks), display units, typewriter-keyboard devices, printers, teleprocessing devices, and sensor- . An I / O device may be a physically distinct device, or it may share a device with other I / O devices.

Most types of I / O devices, such as printers and tape devices, use external media, which are physically distinguishable and identifiable. Other types are electronic only and do not directly handle physical recording media. A channel-to-channel adapter, for example, provides data transfer between two channel paths, and the data never reaches a physical recording medium outside the main storage. Similarly, communication controllers can handle transmissions between a data-processing system and a remote station, the inputs and outputs of which are signals on a transmission line.

In the simplest case, the I / O device is connected to one control unit and is accessible from one channel path. Switching equipment may be used to make certain devices accessible from two or more channel paths by switching devices between control units and switching control units between channel paths. This switching device provides multiple paths through which I / O devices can be accessed. Multiple channel paths to the I / O device within the system are provided for improved performance or I / O availability, or both. The management of the multiple channel paths to the devices is under the control of the channel subsystem and the device, but the channel paths may be indirectly controlled by the program.

I / O addressing (I / O) Addressing )

Four different types of I / O addressing require the necessary addressing of the various components: channel-path identifiers, subchannel numbers, device numbers, and addresses that are not visible to programs but are channel- Is provided by the channel subsystem. When a multiple-subchannel-set facility is installed, a subchannel-set identifier (SSID) may also be used for I / O addressing.

Subchannel-set identifier ( Subchannel - Set Identifier )

The subchannel-set identifier (SSID) is a 2-bit value assigned to each of the provided subchannel sets.

Channel-path identifier ( Channel - Path Identifier )

The channel-path identifier (CHPID) is a system-unique 8-bit value assigned to each installed channel path in the system. CHPID is used to specify an address in the channel path. The CHPID is used to specify the channel path to be specified and reset by the second-operand address of the RESET CHANNEL PATH (reset the channel path). Channel paths, i.e., channel paths accessible by the device by them, are identified in the subchannel-information block (SCHIB) by their associated CHPIDs when the STORE SUBCHANNEL is executed. The CHPID may also be used in operator messages when it is necessary to identify a particular channel path. The system model can provide up to 256 channel paths. It is up to the system model to allocate the maximum number of channel paths and CHPIDs to the channel paths.

Sub channel number ( Subchannel Number )

The subchannel number is a system-specific 16-bit value used to address the subchannel. This value is unique within the sub-channel set of the channel subsystem. The subchannel is addressed by eight I / O commands: CANCEL SUBCHANNEL, CLEAR SUBCHANNEL, HALT SUBCHANNEL, MODIFY SUBCHANNEL, RESUME SUBCHANNEL, START SUBCHANNEL, STORE SUBCHANNEL, and TEST SUBCHANNEL. All I / O functions for a particular I / O device are specified by specifying the subchannel assigned to that I / O device by the program. The subchannels in each subchannel set are always the subchannel numbers assigned within the single range of adjacent numbers. The lowest numbered subchannel is subchannel 0. The highest numbered subchannel of the channel subsystem has an equal subchannel number that is one less than the number of offered subchannels. A maximum of 65,536 sub-channels may be provided in each sub-channel set. Normally, subchannel numbers are used only for communications between the CPU program and the channel subsystem.

device  number( Device Number )

Each subchannel with an assigned I / O device also includes a parameter called a device number. The device number is a 16-bit value that is assigned as one of the subchannel parameters at the time the device is assigned to the subchannel. The device number uniquely identifies the device to the program.

The device number provides a means of identifying the device, independent of the constraints imposed by the system model, configuration, or channel-path protocols. The device number is used in communications about the device that occur between the system and the system operator. For example, the device number is entered by the system operator, which is entered to specify the input device to be used for initial program loading.

Programming annotations (Programming Note : Device numbers are assigned at device-install time and can have any value. However, the user must adhere to any restrictions on the device-number assignment that may be required by the control program, supporting programs, or specific control units or I / O devices.

device  Identifier ( Device Identifier )

The device identifier is an address that is not apparent to the program, i.e., used to communicate with the I / O devices by the channel subsystem. The type of device identifier used is determined according to the specific channel-path type and protocols provided. Each subchannel contains one or more device identifiers.

In the channel path of the parallel-I / O-interface type, the device identifier is called the device address and is composed of 8-bit values. For the ESCON I / O interface, the device identifier consists of a 4-bit control-unit address and an 8-bit device address. For the FICON I / O interface, the device identifier consists of an 8-bit control-unit-image ID and an 8-bit device address. For the FICON-translated I / O interface, the device identifier consists of a 4-bit control-unit address and an 8-bit device address.

The device address identifies the particular I / O device (and, on the parallel-I / O interface, the control unit) associated with the subchannel. The device address may, for example, identify a particular self-tape drive, disk-access mechanism, or transmission line. Any number in the range 0-255 can be assigned to the device address.

Fiber-channel expansion ( Fiber - Channel Extensions )

The formation of a channel program consisting of a transmission-control-block (TCCB) and a transmission-control-word (TCW) designating a transmission-state block (TSB) is an optional facility that provides for formation. The TCCB includes a transmission-command area (TCA), which has a list of up to 30 I / O commands that exist in the form of device-command words (DCWs). The TCW and its TCCB can specify a read or write operation. In addition to the IRB, the TSB contains the completion status and other information associated with the TCW channel program.

The FCX facility provides the ability to specify either directly or indirectly any or all of the TCCB, the input data storage area, and the output data storage area. If the storage area is specified directly, the TCW specifies the location of a single contiguous storage block. If the storage area is indirectly specified, the TCW specifies the location of the list of one or more transfer-indirect-data-address words (TIDAWs). The TIDAW lists and storage areas specified by each TIDAW in the list are limited to crossing 4K-byte ranges.

The FCX facility also provides an interrogate operation that can be initiated by the CANCEL SUBCHANNEL command to determine the state of the I / O operation.

I / O- Command  Word (I / O- Command Words )

The I / O-command word specifies the command and contains the information associated with the command. When the FCX facility is installed, there are two basic types of I / O command words, the channel-command word (CCW) and the device-command word (DCW).

The CCW is 8 bytes in length and specifies the command to be executed. For commands that initiate certain operations, the CCW may also specify a storage area associated with the operation, a count of data bytes, an action to be taken when the command is completed, and other options. All I / O devices recognize CCWs.

The DCW specifies 8 bytes in length, a command to be executed, a count of data bytes, and other options. I / O devices that support FCX recognize DCWs.

send Command  word( TCW )

The TCW specifies a transmit-command-control block (TCCB) and the TCCB contains a list of commands to be sent to and executed by the I / O device. The TCW also specifies the storage areas for commands in the TCCB and also specifies the transfer-status block (TSB) to have the status of the I / O operation.

Channel programming ( Channel Program Organization )

When the FCX facility is not installed, there is a single type of channel program, which is a CCW channel program. When the FCX facility is installed, there is another type of channel program, which is the TCW channel program. Both types of channel programs are described below.

CCW  Channel program ( CCW Channel Program )

A channel program consisting of one or more CCWs is called a CCW channel program (CCP). These channel programs include one or more CCWs that are logically linked and arranged for sequential execution by the channel subsystem.

TCW  Channel program ( TCW Channel Program )

A channel program consisting of a single TCW is called a TCW channel program. The TCW specifies a transmit-command-control block (TCCB) that contains DCWs from 1 to 30. The DCWs in the TCCB are logically linked and arranged for sequential execution. For DCWs that specify control information, the TCCB also contains control information for those commands. The TCW also specifies storage areas or areas of DCWs that specify the location of the transfer-status block (TSB) for the transfer and completion of data to / from the device. The TCCB and storage areas for the transmission of data may be specified as contiguous or non contiguous storage.

The TCW may also specify a TSB for the completion status.

Execution of I / O operations ( Execution of  I / O Operations )

The I / O operation is initiated and controlled by four types of formats: command START SUBCHANNEL, transfer-command words, I / O-command words, and orders. The START SUBCHANNEL command is executed by the CPU and supervises the flow of requests for I / O operations from other programs that manage or process I / O data as part of the CPU program.

When START SUBCHANNEL is executed, parameters are sent to the target subchannel to request that the channel subsystem perform a start function with an I / O device associated with the subchannel. The channel subsystem starts the start function by using the information in the subchannel including information sent during the execution of the START SUBCHANNEL command to find an accessible channel path to the device. Once the device is selected, the execution of the I / O operation decodes and executes the CCW for the CCW channel programs or for the TCW channel programs, by the channel subsystem and the I / O device, To the TCCB to the I / O device, and decoding and executing the DCW by the device. The I / O-command words and the transfer-command words are fetched from the main storage, although the modifier bits in the command code of the CCW DCW specify device-dependent conditions for the execution of the operation in the device .

Operations unique to the device, such as rewinding the tape or placing the access mechanism on the disk drive, are specified by orders that are decoded and executed by the I / O devices. Orders may be sent to the device with modifier bits in the command code of the control command, transferred to the device as data during a control or write operation, or may be used by the device using other means.

start- Function  Start( Start - Function Initiation )

CPU programs initiate I / O operations with the command START SUBCHANNEL. This command sends the contents of the operation-request block (ORB) to the subchannel.

If the ORB specifies a CCW channel program, the contents of the ORB include the subchannel key, the address of the first CCW to be executed, and the specification of the format of the CCWs. The CCW specifies the command to be executed and, if any, the storage area to be used. If the ORB specifies a TCW channel program, the contents of the ORB contain the subchannel key and the address of the TCW to be executed. The TCW specifies the commands to be sent to the device for execution, the storage area or areas to be used for data transfer, if any, and the TCCB to have the TSB to contain the state of the I / O operation.

When the sending of ORB contents to the subchannel is complete, execution of START SUBCHANNEL is complete. The results of the execution of the instruction are indicated by a condition code set in the program-status word.

When facilities are available and the ORB specifies a CCW channel program, the channel subsystem fetches the first CCW and decodes it according to the format bits specified in the ORB. If the format bit is 0, format-0 CCWs are specified. If the format bit is 1, format-1 CCWs are specified. Format-0 and Format-1 CCWs contain the same information, but the fields are arranged differently in Format-1 CCW so that 31-bit addresses can be specified directly in the CCW. When facilities are available and the ORB specifies a TCW channel program, the channel subsystem fetches the specified TCW and transfers the specified TCCB to the device. Storage areas designated by the TCW for the transfer of data to / from the device are 64-bit addresses.

Subchannel operation mode ( Subchannel Operation Modes )

There are two modes of subchannel operation. The subchannel enters the transport mode when the start function is set to that subchannel as a result of the execution of the START SUBCHANNEL command with the FCX facility installed and specifying the TCW channel program. The subchannel remains in transmit mode until the start function is reset on the subchannel. In all other cases, the subchannel is in command mode.

Path management ( Path Management )

If the ORB specifies a CCW channel program and the first CCW passes certain validity tests and does not have a suspend flag specified as 1, or if the ORB specifies a TCW channel program and the TCW The channel subsystem attempts device selection by selecting the channel path from the group of selectable channel paths. A control unit recognizing the device identifier logically connects itself to the channel path and responds to the selection.

If the ORB specifies a CCW channel program, the channel subsystem sends the command-code portion of the CCW over the channel path, and the device responds with a status byte indicating whether the command can be executed. The control unit can in this case logically disconnect from the channel path or remain connected to initiate data transmission.

If the ORB specifies a TCW channel program, the channel subsystem uses the information in the specified TCW to transfer the TCCB to the control unit. The contents of the TCCB are ignored by the channel subsystem and have meaning only for the control unit and the I / O device.

If the attempted choice does not occur as a result of either a busy indication or a path-operation-impossible condition, the channel subsystem attempts to select the device by alternate channel path, if available . When the selection attempt is completed on all selectable paths and the busy condition persists, the operation remains pending until one path is free. If a path-operation-impossible condition is detected on one or more channel paths on which device selection was attempted, the program is alerted by subsequent I / O interruption. I / O interruption occurs as a result of a channel program being executed (assuming that the device has been selected on an alternate channel path) or aborted because of path-not-operational conditions ) Was detected on all channel paths on which the device selection was attempted.

Channel - Program execution ( Channel - Program Execution )

If the command is initiated at the device and the command execution does not require any data to be transferred into or out of the device, the device can signal the end of the operation upon receipt of the command code. In operations involving the transfer of data, the subchannel is set to assume that the channel subsystem responds to service requests from the device and assumes additional control of the operation.

The I / O operations may involve the transfer of data into or out of one storage area, as specified by a single CCW or TCW, or the transfer of data to / within a plurality of non-contiguous storage areas. In the latter case, a list of CCWs is typically used for the execution of I / O operations, each CCW designates an adjacent storage area, and the CCWs are combined by data chaining. Data chaining is specified by a flag in the CCW and causes the channel subsystem to fetch another CCW when the storage area designated by the current CCW is exhausted or filled. The storage area specified by the CCW fetched in the data chaining relates to I / O operations already in progress at the I / O device, and the I / O device is not notified when the new CCW is fetched.

When the CCW is decoded, the channel subsystem requests I / O interruption as quickly as possible, thereby allowing the CPU program to notify the CPU program that the chaining has progressed from its channel program to at least its CCW. do.

To complement dynamic address translation in CPUs, CCW indirect data addressing and modified CCW indirect data addressing are provided.

When the ORB specifies a CCW channel program and CCW-indirect-data addressing is used, the flag in the CCW specifies that the indirect-data-address list will be used to specify the storage areas for that CCW. Each time a range of blocks of storage is reached, the list is consulted to determine the next block of storage to be used. The ORB specifies whether the size of each block of storage is 2K bytes or 4K bytes.

When the ORB specifies a CCW channel program and the modified-CCW-indirect-data addressing is used, the flag in the ORB and the flag in the CCW specify that the modified-indirect-data-address list will be used to specify the storage areas for that CCW do. Each time a count of bytes specified for a block of storage is reached, the list is consulted to determine the next block of storage to be used. Unless when indirect data addressing is used, the block can be specified in arbitrary ranges and lengths up to 4K under conditions where no data transmission over a 4K-byte range is specified.

When the ORB specifies a TCW channel program and transport-indirect-data addressing is used, the flags in the TCW indicate whether the transfer-indirect-data-address list is used to specify the storage areas containing the TCCB and the transfer- Specifies that the address list is to be used to specify data storage areas associated with the DCWs in the TCCB. Each time a count of bytes specified for a block of storage is reached, the next storage block to be used is determined by referring to the corresponding transfer-indirect-data-address list.

The CCW indirect data addressing and the modified CCW indirect data addressing are the same to be used for program execution when dynamic address translation is active on the CPU, as if the CPU were operating with equivalent adjacent real storage CCW sequences are essentially allowed. CCW indirect data addressing allows the program to specify blocks of data with absolute storage addresses of up to 2 ⁶⁴ -l, regardless of whether format-0 or format-1 CCWs are specified in the ORB. The modified CCW indirect data addressing allows the program to specify blocks of data with absolute storage addresses of up to 2 ⁶⁴ -l, regardless of whether format-0 or format-1 CCWs are specified in the ORB.

In general, the execution of I / O operations or chain operations involves three levels of participation:

1. The CPU is busy during the execution duration of the START SUBCHANNEL except for the effects due to the integration of the CPU and the channel-subsystem devices, which causes the ORB Until the content is completely sent.

2. The subchannel is busy with the new START SUBCHANNEL from receiving the ORB contents until the primary interruption condition is cleared in the subchannel.

3. The I / O device is busy until the subchannel from the start of the first operation in the I / O device is interrupted or a secondary interruption condition is placed in the subchannel. When the subchannel is interrupted, the device is busy again when the suspended channel program resumes execution.

Closing the I / O operation ( Conclusion of  I / O Operations )

The termination of an I / O operation is typically indicated by two state conditions: channel end and device end. The channel-termination condition indicates that the I / O device has received or provided all data associated with the operation and no more channel-subsystem facilities are needed. This condition is called a primary interruption condition, and in this case the channel termination is a primary state. In general, the primary interruption condition is an interruption condition associated with an I / O operation and signaling an end to a subchannel of I / O operations or chain I / O operations.

The device-termination signal indicates that the I / O device has completed execution and is ready to perform another operation. This condition is called a secondary interruption condition. In this case, the device termination is a secondary state. In general, a secondary interruption condition is an interruption condition associated with an I / O operation and signaling a termination to a device of an I / O operation or a chain operation. The secondary interruption condition may occur simultaneously with or after the primary interruption condition.

Simultaneously with the primary interruption condition or the secondary interruption condition, both the channel subsystem and the I / O device can provide indications of unusual situations.

Conditions that signal the end of an I / O operation may be triggered by I / O interrupts, or by programmed queries of the channel subsystem when CPUs are disabled due to I / O interrupts, . In the former case, these conditions cause the I / O-interruption code to be stored, which contains information related to the interrupting source. In the latter case, the interruption code is stored as a result of execution of the TEST PENDING INTERRUPTION.

If the primary interruption condition is recognized, the channel subsystem attempts to notify the program using the interruption request that the subchannel contains information describing the termination of the I / O operation in that subchannel. In command-mode interrupts, the information can identify the last CCW used and provide its residual byte count, thereby describing the extent in which main storage is used. For transmission-mode interrupts, the information identifies the current TCW and the TSB associated with the channel program, including the additional state regarding the I / O operation, such as the residual byte count. In addition to information about the channel program, the channel subsystem and the I / O device may provide additional indications of unusual conditions as part of the primary or secondary interruption condition. The information contained in the subchannel may be stored by execution of a test subchannel or execution of a store subchannel. This information, when stored, is called a subchannel-status word (SCSW).

CCW  Chaining when using channel programs ( Chaining When Using  a CCW Channel Program )

When the ORB specifies a CCW channel program, facilities are provided for the program to initiate the execution of chain I / O operations with a single START SUBCHANNEL command. Upon receipt of a device-termination signal when the current CCW specifies command chaining and no abnormal conditions are detected during the operation, it causes the channel subsystem to fetch a new CCW. If the CCW sends certain validity tests and the abort flag is not specified as 1 in the new CCW, execution of a new command on that device is initiated. If the CCW fails to send validity tests, no new command is initiated, command chaining is suppressed, and the state associated with the new CCW causes the generation of an interruption condition. If the suspend flag is specified as 1 and this value is valid because it is a value of 1 in the suspend control, which is bit 4 of word 1 of the associated ORB, the execution of the new command is not initiated, Command chaining is terminated.

Execution of the new command is initiated by the channel subsystem in the same manner as in the previous operation. Termination signals that occur at the end of an operation caused by the CCW specifying command chaining are not available to the program. When another I / O operation is initiated by command chaining, the channel subsystem continues execution of the channel program. However, if an abnormal condition is detected, command chaining is suppressed, the channel program is terminated, an interruption condition is generated, and termination signals causing termination are made available to the program.

The suspend-and-resume function provides the program with control over the execution of the channel program. The start of the break function is controlled by the setting of the break-control bit in the ORB. The interrupt function is passed to the channel subsystem during channel-program execution when the interrupt-control bit in the ORB is 1 and the interrupt flag in the CCW fetched during the first CCW or command chaining is one.

Suspension occurs when the channel subsystem fetches a CCW with an abort flag that is effectively 1 (due to a value of 1 in the break-control bit in the ORB). This CCW command is not sent to the I / O device, and the device receives a signal that chain commands are to be terminated. A subsequent RESUME SUBCHANNEL command notifies the channel subsystem that the CCW causing the interrupt has changed and the channel subsystem should re-fetch the CCW to examine the current set of interrupt flags. If the interrupt flag is found to be 0 in the CCW, the channel subsystem resumes execution of the chain commands with the I / O device.

TCW  Chaining when using channel programs ( Chaining When Using  a TCW Channel Program )

When the ORB specifies a TCW channel program, facilities are also provided for the program to initiate the execution of chain device operations with a single START SUBCHANNEL command. Command chaining can be specified for DCWs specified by a single TCW. When the current DCW specifies command chaining and no abnormal conditions are detected during the operation, if successful execution of the DCW is recognized, the next DCW is processed in the current TCCB.

If the DCW then passes certain validity tests, execution of a new command is initiated on the device and the DCW is now the DCW. If the DCW fails to pass validity tests, no new command is initiated, command chaining is suppressed, the channel program is terminated, and the state associated with the new DCW causes the generation of an interruption condition.

Execution of the new command is initiated in the same manner as in the previous operation. Termination signals that occur at the end of an operation caused by the DCW other than the last specified DCW will be unavailable to the program. When another I / O operation is initiated by command chaining, the execution of the channel program continues. However, if an abnormal condition is detected, the command chaining is suppressed, the channel program is terminated, an interruption condition is generated, and the program becomes available to identify the abnormal condition.

Early termination of I / O operations ( Premature Conclusion of  I / O Operations )

The execution of the channel-program can be terminated early by the CANCEL SUBCHANNEL, HALT SUBCHANNEL or CLEAR SUBCHANNEL. Running CANCEL SUBCHANNEL causes the channel subsystem to terminate the start function on the subchannel if the channel program is not started on the device. If the start function is terminated by the execution of the CANCEL SUBCHANNEL, the channel subsystem sets the condition code to 0 in response to the CANCEL SUBCHANNEL command. Running HALT SUBCHANNEL causes the channel subsystem to issue a halt signal to the I / O device to terminate the channel-program execution on the subchannel. When the execution of the channel-program is terminated by the execution of the HALT SUBCHANNEL, the program is notified of the end through the I / O-interruption request. If the subchannel is in command mode, an interruption request is generated when the device submits a status for the operation that was terminated. If the subchannel is in transmit mode, an interruption request is immediately generated. However, if a halt signal is issued to the device after receiving a device termination or during the command chaining before the next command is sent to the device, an interruption request is generated after the device receives the signal . In the latter case, the device-status field of the SCSW will contain zeros. Executing CLEAR SUBCHANNEL clears the indications of the channel program running on the subchannel, causes the channel subsystem to issue a clear signal to the I / O device, and the channel subsystem generates an I / O- And notify completion of the clear function.

I / O interruption (I / O Interruptions )

The conditions that cause I / O interruption requests are asynchronous with the activity in the CPUs, and more than one condition can occur at the same time. These conditions are stored in subchannels until cleared by SUBCHANNEL or CLEAR SUBCHANNEL or reset by I / O-system reset.

When an I / O-interruption condition is acknowledged by the channel subsystem and displayed on a subchannel, an I / O-interruption request is pending for the specified I / O-interruption subclass in the subchannel. I / O-interruption subclasses, where interruption is pending, are under programmed control through the use of MODIFY SUBCHANNEL. Pending I / O interruptions can be accommodated by all CPUs enabled for interrupts from their I / O-interruption subclasses. Each CPU has eight mask bits in the control register 6, which are the eight I / O-bus masks when the I / O mask, or bit 6 in the PSW, is the master I / O- Controls the enablement of the CPU for each of the interruption subclasses.

When an I / O interruption occurs in the CPU, the I / O-interruption code is stored in the I / O-communication area of the CPU, and the I / O-interruption request is cleared. The I / O-interruption code identifies the subchannel to which the interruption was pending. The conditions that cause the generation of the interruption request may then be retrieved explicitly by the TEST SUBCHANNEL or by the STORE SUBCHANNEL from the subchannel.

The pending I / O-interruption request is also cleared by the TEST PENDING INTERRUPTION when the corresponding I / O-interruption subclass is enabled but the PSW is cleared by the TEST PENDING INTERRUPTION when the I / O interrupts are disabled, It can be cleared by the TEST SUBCHANNEL when it is disabled for I / O interrupts from the interruption subclass. Pending I / O-interruption requests can also be cleared by the CLEAR SUBCHANNEL. Both CLEAR SUBCHANNEL and TEST SUBCHANNEL also clear the reserved interruption condition of the subchannel.

Normally, if the interruption request is not cleared by CLEAR SUBCHANNEL, the program issues a TEST SUBCHANNEL to obtain information about the execution of the operation.

CLEAR SUBCHANNEL  Command

The designated subchannel is cleared and the current start or hold function is terminated at the specified subchannel, if any, and the channel subsystem is signaled to asynchronously perform the clear function at the designated subchannel and the designated device.

General purpose register 1 contains a subsystem-identification word (SID) that specifies the subchannel to be cleared.

If the start or hold function is in progress, the function ends at the subchannel.

The subchannel is no longer pending state. All activity is cleared in the subchannels, except that the subchannels are clear pending, as indicated in the activity-control field of the SCSW. The ongoing functions are all cleared on the subchannels, except for the clear function to be performed due to the execution of this command, as indicated in the function control field of the SCSW.

When the subchannel is operating in the transmit mode and the condition code 2 is set, the CPU signals the channel subsystem to perform the interrogate function asynchronously and can terminate the command.

The channel subsystem is signaled to perform the clear function asynchronously. The clear function is summarized below in the section entitled " Associated Functions "and is described in detail later.

The condition code 0 is set to indicate that the above-described actions have been taken.

The associated function ( Associated Functions )

Following the execution of the CLEAR SUBCHANNEL, the channel subsystem performs the clear function asynchronously. If the conditions allow, the channel subsystem selects the channel path and issues a clear signal to the device to attempt the I / O operation, if there is one. The subchannel then becomes state pending. Conditions that prevent a clear signal from being issued to the device, where the channel subsystem meets, do not prevent the subchannel from being state pending.

If the subchannel is state pending as a result of performing a clear function, the data transmission is terminated, if any, with the associated device. If the resulting status is cleared by the TEST SUBCHANNEL, the stored SCSW has clear-function bits stored as one. If the channel subsystem can determine that a clear signal has been issued to the device, the clear-pending bit is stored as zero in the SCSW. Otherwise, the clear-pending bits are stored as 1, and other indications are provided describing the conditions that they met in great detail.

For the start function terminated by CLEAR SUBCHANNEL, the measurement data is not accumulated in the extension-status word for the subchannel, and the device-connection time is also not stored.

Special conditions ( Special Conditions )

Condition code 3 (Condition code 3) is set and no other action is taken, in which case the subchannel is not operable against the CLEAR SUBCHANNEL. A subchannel is not operable for a CLEAR SUBCHANNEL if the subchannel is not provided in the channel subsystem, does not have an assigned valid device number, or is not enabled.

The CLEAR SUBCHANNEL can encounter program exceptions listed or listed below.

If multiple-subchannel-set facilities are not provided, bits 32-47 of general purpose register 1 must contain 0001 hex; Otherwise, an operand exception is recognized.

When a multiple-subchannel-set facility is provided, bits 32-44 of general register 1 must contain zeros, bits 45-46 must contain a valid value, and bit 47 must contain a value of 1 ; Otherwise, an operand exception is recognized.

Resulting Condition Code:

0 function is started

One -

2 -

3 Operation not possible

Program exceptions:

오퍼랜드

Operand

특권 오퍼레이션(privileged operation)

Privileged operation

clear  Function ( Clear Function )

Following the execution of the CLEAR SUBCHANNEL, the channel subsystem performs a clear function. Performing clear functions may include (1) performing a path-management operation, (2) changing fields in a subchannel, (3) issuing a clear signal to the associated device, (4) , And displaying the completion of the clear function.

clear - Function  Path management ( Clear - Function Path Management )

The path-management operation is performed as part of the clear function, which examines the channel-path conditions for the associated subchannel to select the available channel path that the clear signal can be issued to the associated device.

The channel-path conditions are examined in the following order:

1. If the channel subsystem is actively communicating or attempting to establish active communication with the device receiving the signal, the channel path in use is selected.

2. If the channel subsystem is in the process of accepting a no-an-busy-no-indication (which will not cause the interruption condition to be recognized) from the device receiving the signal and the associated sub- If there is no allegiance, the channel path in use is selected.

3. If the associated subchannel has a dedicated allegiance for a channel path, that channel path is selected.

4. If the associated subchannel has a working allegiance for one or more channel paths, then one of those channel paths is selected.

5. If the associated subchannel has no loyalty to any channel path, then if the last-used channel path is indicated, and if the channel path is selectable, then that path is selected. If the channel path is not selectable, no channel path is selected, or if there is a channel path is selected from the set of selectable channel paths (as if the last-used channel path was not indicated).

6. If the associated subchannel does not have loyalty to any channel path, if the last-used channel path is not indicated, and if there are one or more selectable channel paths, then one of the channel paths is selected do.

If none of the channel-path conditions listed above apply, no channel path is selected.

For item 4, item 5 under specified conditions, and item 6, the channel subsystem selects a channel path from a set of channel paths. In these cases, the channel subsystem may attempt to select a channel path, provided that the following conditions do not apply:

1. A channel-path-terminal condition exists for that channel path.

2. In a parallel or ESCON channel path: Another subchannel has an active allegiance to its channel path.

In the FICON channel path: a channel path is currently in use and is used to actively communicate with the maximum number of subchannels that may have simultaneously active communications.

3. The subchannel for another device to which the device to receive the signal is connected to the Type-1 control unit and is connected to the same control unit is not loyal to work loyalty and the primary state is not accommodated by that subchannel One, they have loyalty to the same channel path.

4. The device to receive the signal is connected to the Type-3 control unit, and the subchannels for another device connected to the same control unit have a dedicated allegiance for the same channel path.

clear - Function  SubChannel Change ( Clear - Function Subchannel Modification )

The path-management-control indications in the subchannels are changed while the clear function is performed. Substantially, this change occurs after an attempt to select a channel path, but before an attempt to select a device to issue a clear signal. The modified path-management-control displays are as follows:

1. The state of all eight possible channel paths in the subchannel is set to be operable for the subchannel.

2. The used-last-path indication is reset to indicate that there is no last-used channel path.

3. Path-operation-impossible conditions, if any, are reset.

clear - Function Signaling  And completion ( Clear - Function Signaling and Completion )

Following an attempt to select a channel path and a change in path-management-control fields, the channel subsystem attempts to select a device to issue a clear signal if conditions permit. The conditions associated with the subchannel and the selected channel path, if any, affect whether (1) an attempt is made to issue a clear signal and (2) the attempt to issue a clear signal is successful. Regardless of these conditions, the subchannel is subsequently set to state pending and the execution of the clear function is complete. These conditions and their effect on the clear function are described as follows:

No attempt to issue a clear signal: the channel subsystem does not attempt to issue a clear signal to the device if any of the following conditions exist:

1. Channel path not selected.

2. The selected channel path is no longer selectable.

3. Channel-Path-The terminal condition exists for the selected channel path.

4. For parallel and ESCON channel paths: The selected path is currently used and is used to actively communicate with other devices. For FICON channel paths: The selected channel path is currently used and is used to actively communicate with the maximum number of devices that may have simultaneous active communications.

5. The device to receive the signal is connected to the Type-1 control unit, and the subchannel for another device connected to the same control unit is not loyal to work loyalty and the primary state is not accommodated by that subchannel One, they have loyalty to the same channel path.

6. The device to receive the signal is connected to a Type-3 control unit, and the subchannels for another device connected to the same control unit have a dedicated allegiance to the same channel path.

If any of the above conditions are present, the subchannel is maintained in clear pending and set to state pending, and the performance of the clear function is complete.

An attempt to issue a clear signal is unsuccessful: When the channel subsystem attempts to issue a clear signal to the device, the attempt may not be successful due to the following conditions:

1. When the channel subsystem attempts to select a device to issue a clear signal, the control unit or device signals a busy condition.

2. When a channel subsystem attempts to select a device to issue a clear signal, a path-operation-impossible condition is recognized.

3. The channel subsystem encounters an error condition when attempting to select a device to issue a clear signal.

If any of the above conditions exist and the channel subsystem determines that the clear signal issue attempt was unsuccessful or can not determine whether the attempt was successful, then the subchannel remains clear pending and set to state pending, and the clear function Execution is complete.

Clear Signal Issuance Successful: If the channel subsystem determines that the clear signal issuance attempt was successful, the subchannel is no longer clear pending and is set to state pending, and the clear function execution is complete. When the subchannel is state pending, the I / O operation is terminated, if any, with the associated device.

Programming Notes: Following the execution of the clear function, all non-zero states submitted to the channel subsystem by the device, except for the control unit end, are sent to the program as an unsolicited alert status. Loses. An unsolicited status or zero status consisting of control unit termination only is not submitted to the program.

MODIFY SUBCHANNEL  Command

The information contained in the subchannel-information block SCHIB is placed in the program-changeable fields of the subchannel. As a result, the program is able to determine, for that subchannel, specific aspects of I / O processing, such as clear, halt, resume and start functions, and specific I / O support functions .

The general purpose register 1 contains a subsystem-identification word (SID) that specifies the subchannel to be changed as specified by the specific fields of the SCHIB. The second-operand address is the logical address of the SCHIB and should be specified in the word range; Otherwise, a specification exception is recognized.

The channel-subsystem operations that can be affected by placing SCHIB information in the subchannels are:

I/O 처리 (E 필드)

I / O processing (E field)

인터럽션 처리 (인터럽션 파라미터 및 ISC 필드)

Interruption processing (interruption parameter and ISC field)

경로 관리 (D, LPM, 및 POM 필드들)

Path management (D, LPM, and POM fields)

모니터링 및 주소-한계 검사 (측정-블록 인덱스, LM, 및 MM 필드들)

Monitoring and address-limit checking (measurement-block index, LM, and MM fields)

측정-블록-포맷 컨트롤 (F 필드)

Measurement - Block - Format Control (F field)

확장-측정-워드-모드 인에이블 (X 필드)

Expand-Measure-Word-Mode Enable (X field)

동시-센스 퍼실리티 (S 필드)

Simultaneous Sensibility (S field)

측정-블록 주소 (MBA)

Measurement - Block Address (MBA)

Bits 0, 1, 6, and 7 of word 1 of SCHIB operand and bits 0-28 of word 6 must be 0, and bits 9 and 10 of word 1 should not be all ones. If the expansion-I / O-measure-block facility is installed and the format-1 measurement block is specified, bits 26-31 of word 11 must be specified as zeros. If the Extension-I / O-Measure-Block Facility is not installed, bit 29 of word 6 should be specified as 0; Otherwise, an operand exception is recognized. If the expansion-I / O-measure-word facility is not installed or is installed but not enabled, bit 30 of word 6 should be specified as 0; Otherwise, an operand exception is recognized. The remaining fields of SCHIB are ignored and do not affect the processing of MODIFY SUBCHANNEL.

Condition code 0 is set to indicate that information from the SCHIB has been placed in the program-changeable fields of the subchannel, provided that the device-number-valid bit (V) in the designated subchannel is 0, then the condition code 0 is set Except that information from that SCHIB is not placed in program-modifiable fields.

Special conditions ( Special Conditions )

When the subchannel is pending state, the condition code 1 is set, and no other action is taken.

If the clear, hold, or start function is in progress on the subchannel, condition code 2 is set and no other action is taken.

If the subchannel is not operable for MODIFY SUBCHANNEL, condition code 3 is set and no other action is taken. If a subchannel is not provided in the channel subsystem, that subchannel is not operable for MODIFY SUBCHANNEL.

MODIFY SUBCHANNEL can meet with program exceptions listed or listed below.

In word 1 of SCHIB, bits 0, 1, 6, and 7 must be zeros, and if address-limit-checking facilities are installed, bits 9 and 10 should not be all ones. In word 6 of SCHIB, bits 0-28 must be zero. Otherwise, an operand exception is recognized.

If the Extension-I / O-Measure-Block Facility is installed and the Format-1 measurement block is specified, bits 26-31 of word 11 must be specified as zeros; Otherwise, an operand exception is recognized. If the Extension-I / O-Measure-Block Facility is not installed, bit 29 of word 6 should be specified as 0; Otherwise, an operand exception is recognized. If the expansion-I / O-measure-word facility is not installed or is installed but not enabled, bit 30 of word 6 should be specified as 0; Otherwise, an operand exception is recognized.

If multiple-subchannel-set facilities are not provided, bits 32-47 of general purpose register 1 must contain 0001 hex; Otherwise, an operand exception is recognized.

When a multiple-subchannel-set facility is provided, bits 32-44 of general register 1 must contain zeros, bits 45-46 must contain a valid value, and bit 47 must contain a value of 1 ; Otherwise, an operand exception is recognized.

The second operand must be specified in the word range; Otherwise, the specification exception is recognized. The execution of MODIFY SUBCHANNEL is suppressed in all addressing and protection exceptions.

Result Condition code:

0 function completed

1 state pending

2 bezie

3 Operation not possible

Program exceptions:

액세스 (페치, 오퍼랜드 2)

Access (fetch, operand 2)

오퍼랜드

Operand

특권 오퍼레이션

Privileged operation

명세

details

Programming notes:

1. When the associated subchannel is disabled and the device signals an I / O-error alarm, the channel subsystem issues a clear signal to the device and does not generate an I / O-error condition, Discard the alarm indication.

2. If the device submits an unsolicited status while the associated subchannel is disabled, the state is discarded by the channel subsystem without generating an I / O-interruption condition. However, if the submitted status includes a unit check, the channel subsystem issues a clear signal for the associated subchannel and does not generate an I / O-interruption condition. This should be considered when the program uses the MODIFY SUBCHANNEL to enable subchannels. For example, the media on the associated device that existed when the subchannel was disabled may have been replaced, and therefore the program must verify the integrity of the media.

3. It is recommended that the program examine the contents of the subchannel by subsequently issuing a STORE SUBCHANNEL when MODIFY SUBCHANNEL sets condition code 0. Using STORE SUBCHANNEL is a way to determine if a given subchannel has changed or not. Failure to check the subchannel following the set of condition code 0 by MODIFY SUBCHANNEL may result in conditions that the program may not be expected to occur.

START SUBCHANNEL  Command

The channel subsystem is signaled to perform the start function asynchronously for the associated device, and execution parameters contained in the specified ORB are placed in the designated subchannel.

General purpose register 1 contains a subsystem-identification word that specifies the subchannel to be started. The second-operand address is the ORB's logical address and must be specified in the word range; Otherwise, the specification exception is recognized.

The execution parameters contained in the ORB are placed in the subchannel.

When START SUBCHANNEL is executed, the subchannel is state pending only in the secondary state, the extended-state-word-format bit (L) is 0, and the state-pending condition is discarded in the subchannel.

The subchannel is pending starting, and the start function is displayed on the subchannel. If the second operand specifies a command-mode ORB, the subchannels remain in command mode. If the second operand specifies a transmit-mode ORB, the subchannel enters a transmit mode. When the subchannel enters the transmission mode, LPUM is set to 0 if there is no previous loyalty; Otherwise, the LPUM is not changed.

Logically, before the set of condition code 0, the path-operation-impossible conditions of the subchannels are cleared, if any.

The channel subsystem is signaled to perform the start function asynchronously. The start function is summarized in the "Associated function" section below and is described in detail later.

Condition code 0 is set to indicate that the actions described above have been taken.

The associated function ( Associated Functions )

Following the execution of START SUBCHANNEL, the channel subsystem performs the start function asynchronously.

The contents of the ORB are validated, not all fields that need to contain zeros. In some models, the fields of the ORB that need to contain zeros are checked asynchronously instead of during execution of the instruction. When invalid fields are detected asynchronously, the subchannels are state pending with primary, secondary, and alarm conditions, and delay condition code 1 and program check are indicated. In this situation, the I / O operation or chain I / O operations are not initiated at the device, and the condition is marked as the start-pending bit is stored as 1 when the SCSW is cleared by the execution of the TEST SUBCHANNEL.

In some models, path availability is tested asynchronously instead of during command execution. When there is no selectable channel path, the subchannel is state pending to the primary and secondary states and the delay condition code 3 is displayed. I / O operations or chain I / O operations are not initiated at the device, but when the SCSW is cleared by execution of the TEST SUBCHANNEL, this condition is marked as the start-pending bit being stored as one.

If the conditions permit, the channel path is selected and execution of the channel program specified in the ORB is started.

Special conditions ( Special Conditions )

If the subchannel is state pending when START SUBCHANNEL is executed, condition code 1 is set and no other action is taken. In some models, the condition code 1 is not set when the subchannel is state pending only in the secondary state; Instead, the state-pending condition is discarded.

When the start, hold, or clear function is currently in progress on the subchannel, condition code 2 is set and no other action is taken.

When the subchannel is inoperable for START SUBCHANNEL, condition code 3 is set and no other action is taken. If the subchannel is not provided in the channel subsystem, does not have an effective device number associated with it, or is not enabled, then the subchannel is not operable for START SUBCHANNEL.

In some models, when there is no selectable channel path, the subchannel is also inoperable for START SUBCHANNEL. In these models, the lack of available channel paths is detected as part of the START SUBCHANNEL execution. In other models, channel-path availability is tested only as part of the asynchronous start function.

START SUBCHANNEL can meet any of the program exceptions listed or listed below.

In word 1 of the command-mode ORB, bits 26-30 must be 0, and in word 2 of the command-mode ORB, bit 0 must be zero. Otherwise, in some models, an operand exception is recognized. In other models, an I / O-interruption condition is generated to indicate a program check as part of the asynchronous start function.

START SUBCHANNEL can also encounter program exceptions listed below.

If multiple-subchannel-set facilities are not provided, bits 32-47 of general purpose register 1 must contain 0001 hex; Otherwise, an operand exception is recognized.

If a multiple-subchannel-set facility is provided, bits 32-44 of general purpose register 1 must contain zeros, bits 45-46 must contain a valid value, and bit 47 must contain a value of 1 ; Otherwise, an operand exception is recognized.

The second operand must be specified in the word range; Otherwise, the specification exception is recognized. Execution of START SUBCHANNEL is suppressed in all addressing and protection exceptions.

Result Condition code:

0 function is started

1 state pending

2 bezie

3 Operation not possible

Program exceptions:

액세스 (페치, 오퍼랜드 2)

Access (fetch, operand 2)

오퍼랜드

Operand

특권 오퍼레이션

Privileged operation

명세

details

start Function  And resume function ( Start Function and Resume Function )

The start and resume functions initiate I / O operations as described below. The start function is applied to the sub-channels operating in the command mode or the transmission mode. The resume function applies only to sub-channels operating in command mode.

Following the execution of START SUBCHANNEL and RESUME SUBCHANNEL, the channel subsystem performs a start function and a resume function, respectively, to initiate an I / O operation with the associated device. Performing the start or resume function may include: (1) performing a path-management operation, (2) performing I / O operations or chain I / O operations with the associated device, and (3) To indicate completion of the start function. In contrast to starting the execution of an aborted channel program, starting from the CCW that caused the abort, if any, the resume function, if any, is specified in the ORB and assigned to the operand of the next START SUBCHANNEL Start execution of the channel program; Otherwise, the resume function is performed as if it were a start function.

start- Function  And Resume - Function  Path management ( Start - Function and Resume - Function  Path Management )

The path-management operation is performed while the start or resume function is performed by the channel subsystem to select an available channel path that can be used for device selection to initiate an I / O operation with the device. The actions to be taken are:

1. If the subchannel is currently starting pending and device active, the start function remains pending in the subchannel, where the secondary state for the previous start function is accepted from the associated device, It remains pending until it is full. When the state is accepted and no alarm interruption condition is described, the subchannel does not become state pending and the execution of the pending start function is subsequently initiated. If the state describes an alarm interruption condition, the subchannel is state pending in the secondary and alert states, the pending start function is not started, the delay condition code 1 is set, and the start-pending bit is set to 1 . If the subchannel is currently the only starting pending, the performance of the start function is initiated as described below.

2. If a dedicated allegiance exists in a subchannel for a channel path, the channel subsystem selects that path for device selection. If a busy condition is encountered during a device selection attempt and full loyalty is present in the subchannel, the start function maintains pending until the internal indication of busy is reset for that channel path. When the internal representation of the busy is reset, the performance of the pending start function is initiated in its channel path.

3. If there is no selectable channel path and no full loyalty to the channel path exists in the subchannel, the channel path is not selected.

4. If all selectable channel paths have been tried and one or more of them are being used to actively communicate with other devices, or, alternatively, if the channel subsystem is one or more of its channel paths If the channel path, control unit or device meets the control-unit-busy or device-busy condition in one or more of the channel paths or meets a combination of the conditions in one or more of the channel paths, And remains pending in the subchannel until it is made < RTI ID = 0.0 >

5. If (1) the start function requires that a device be started on any channel path connected to a Type-1 control unit and (2) no other device is connected to the same control unit - The subchannel has a dedicated alliance for the same channel path or a working allegiance for the same channel path and a primary status is not received for that subchannel - if selected, the channel path is selected; Otherwise, the channel path is not selected. However, if another path to the device is selectable and there is no loyalty as described above, then the other path is selected. If there are no other selectable channel paths, the start or resume function remains pending appropriately until the channel path is available.

6. If the device is connected to a Type-3 control unit and if at least one other device is connected to the same control unit and the subchannel of that unit has full loyalty to that same channel path, The start function remains pending until another channel path can be selected, or until full loyalty to the other device is cleared.

7. If a busy indication is received while the channel path is selected and the selection of the device to transmit the TCCB of the pending TCW channel program to initiate the execution of the first command of the pending CCW channel program is received, It is not used again for the device or control unit until the internal indication of busy has been reset (depending on whether a device busy or control-unit busy indication has been received).

8. To initiate the execution of the first command specified for the start function or implied for the restart function for the CCW channel program, or to initiate the transmission of the TCCB for the start function for the TCW channel program While attempting to select a device (as described in action 7 of FIG. 5), if the channel subsystem receives a busy indication, it performs one of the following actions:

a. If the device is specified to operate in multiple path mode and the received busy indication is device busy, the start or resume function remains pending until the internal busy indication is reset.

b. If the device is specified to operate in a multiple path mode and the received busy indication is a control unit busy or if the device is specified to operate in a single-path mode, then the channel subsystem is initiated or restarted, Until a selection is attempted for all possible paths, a selection of the device is attempted by selecting a selectable alternate channel path and the path-management operation continues. If the start or resume function is not initiated by the channel subsystem even after all selectable channel paths have been selected, the start or resume function remains pending until the internal busy indication is reset.

c. If the subchannel has full loyalty, action 2 is applied.

9. While attempting to make a selection to transmit the first command for the CCW channel program or to transmit the TCCB for the TCW channel program, the device appears to be inoperable and, when its corresponding channel path is operable for the subchannel, Path-operation-impossible condition is recognized, and the state of the channel path in the subchannel is changed from operable to subchannel to non-operable to subchannel. Path-operation-incapable conditions in the subchannels are maintained until the subchannel becomes clear pending, start pending, or resumed pending (if the subchannel is stopped), if any, - Impossible conditions are cleared. However, if the corresponding channel path is not operable with respect to the subchannel, the path-operation-impossible condition is not recognized. When a device appears to be inoperable while attempting to make a selection to transmit a first command or TCCB on a selectable channel path, one of the following actions occurs:

a. If there is full loyalty for the channel path, then the channel path is the only channel path that is selectable; Thus, further attempts to initiate a start or resume function are abandoned, and an interruption condition is acknowledged.

b. If there is no alternative loyalty and there are alternative channel paths that can be selected that are not attempted, then one of the channel paths to attempt a device selection and transmit a first command for the CCW channel program or a TCCB for the TCW channel program Is selected.

c. If no full loyalty exists, no alternative channel paths are attempted that are not attempted, and the device appears to be operable in at least one of the channel paths attempted, the start or resume function may be a channel path, a control unit, The device remains pending in the subchannels until appropriately available.

d. If no full loyalty exists, no alternate selectable channel paths are attempted, and the device appears to be inoperable on all channel paths attempted, additional attempts to initiate a start or resume function are abandoned, The interruption condition is recognized.

10. When a subchannel is active and an I / O operation is to be initiated with the device, all device selection occurs in accordance with the LPUM indication unless a multiple-path mode is specified in the subchannel. For example, if command chaining is specified for a CCW channel program, the channel subsystem transmits the first and all subsequent commands describing the chain I / O operations on the same channel path.

Execution of I / O operations ( Execution of  I / O Operations )

After the channel path is selected, the channel subsystem initiates execution of the I / O operation with the associated device, if the conditions permit. The execution of additional I / O operations may follow the initiation and execution of the first I / O operation.

For subchannels operating in command mode, the channel subsystem can execute the following seven types of commands: write, read, read backward, control, Sense, sense ID, and transfer in channel. Each command excludes the transfer command from the channel and initiates a corresponding I / O operation. Except for periods when channel-program execution is interrupted in the subchannel, the subchannel becomes active from the acceptance of the first command until the first-order interruption condition is acknowledged in the subchannel. If the primary interruption condition is recognized before the acceptance of the first command, the subchannel is not activated. Usually, the primary interruption condition is caused by the channel-termination signal, or, in the case of command chaining, by the channel-termination signal for the last CCW of the chain. The device is active until the primary interruption condition is acknowledged in the subchannel. Normally, the secondary interruption condition is caused by a device-termination signal, or, in the case of command chaining, by a device-termination signal for the last CCW of the chain.

For subchannels operating in the transmit mode, the channel subsystem may transmit six types of commands: write, read, control, sense, sense ID, and interrogate for execution. Each command starts a corresponding device operation. If one or more commands are sent to the I / O device TCCB in the TCCB, the subchannels remain on pending until the primary state is submitted.

Programming notes:

In single-pass mode, all transfers of commands, data, and status to I / O operations or chain I / O operations occur on the channel path over which the first command is sent to the device.

When a device has a dynamic-reconnect feature installed, I / O operations or chain I / O operations can be performed in multiple-path mode. In order to operate in the multiple path mode, MODIFY SUBCHANNEL must be executed for the subchannel with bit 13 of word 1 of SCHIB specified as 1. In addition, the device must be set to multiple path mode by the execution of specific model-dependent commands appropriate for the type of device. The general procedures for handling multi-path operations are as follows:

1. Setup

a. Set-multiple-path-mode type command must be successfully executed by the device on each channel path to be set up as a member of multiple path groups; Otherwise, operations in the multiple path mode may result in unpredictable results in the subchannel. If, for some reason, one or more physically available channel paths to the device are not included in the multiple path group, then these channel paths should not be selectable while the subchannel is operating in multiple path mode. The channel path may not be selectable by setting the corresponding LPM bit in the ORB to 0 before execution of MODIFY SUBCHANNEL or before execution of START SUBCHANNEL in SCHIB.

b. When a set-multiple-path-mode type command is sent to the device, only a single channel path must be logically available to avoid alternative channel-path selection for the execution of the start function; Otherwise, device-busy conditions may be detected by the channel subsystem over one channel path, which may result in unpredictable results for subsequent multiple-path-mode operations. This type of set-up procedure should be used whenever the membership of a multiple-path group changes.

2. Leaving the Multipath Mode

To release multiple path mode and continue processing in single-path mode, you must use one of the following two procedures:

a. A disassembly-multipath-mode type command may be executed for all channel paths in the multiple path group. This command must be followed by either (1) execution of MODIFY SUBCHANNEL with bit 13 of SCHIB word 1 set to 0, or (2) only single channel path is logically available in LPM. The start function should not be performed on a subchannel operating in a multiple path mode in which the device is operating in a single-path mode and can select multiple channel paths; Otherwise, unpredictable results for the start or subsequent start functions may occur in the subchannel.

b. An abort-multipath-mode type of command is executed on each channel path of the multiple path group (the inverse of the setup). This command should be followed by either (1) execution of MODIFY SUBCHANNEL with bit 13 of SCHIB word 1 set to 0, or (2) only one channel path is logically available in LPM. The start function can not be performed in a subchannel operating in a multiple path mode in which the device is operating in a single-path mode and can select multiple channel paths; Otherwise, unpredictable results for the start or subsequent start functions may occur in the subchannel.

Blocking of data ( Blocking of Data )

The data written by the I / O device is divided into blocks. The length of the block depends on the device. For example, a block may be a card, a line of printing, or information recorded between two consecutive gaps on a magnetic tape.

The maximum amount of information that can be transmitted in one I / O operation is one block. The I / O operation is terminated when the associated main-storage area is exhausted or reaches the end of the block, whichever occurs first. In certain operations, such as a write to a self-test unit or an inquiry station, the blocks are not defined and the amount of information transmitted is only controlled by the program.

Operation-request block ( Operation - Request Block )

The operation-request block (ORB) is an operand of START SUBCHANNEL. The ORB specifies the parameters to be used to control a specific start function. These parameters include an interruption parameter, a subchannel key, the address of the first CCW or TCW, operation-control bits, priority-control numbers, and a description of the logical availability of channel paths to the specified device.

The contents of the ORB are placed in the subchannel specified during the execution of START SUBCHANNEL before the set of condition code 0. If the result of the execution is expected to result in a non-zero condition code, the contents of the ORB are not placed in the designated subchannel.

The rightmost two bits of the ORB address are zeros, and the ORB must be placed in the word range; Otherwise, the specification exception is recognized. When the Fiber Channel Extended (FCX) facility is installed, the ORB's channel-program-type control (B) (word 1, bit 3) specifies the type of channel program specified by the ORB. If B is 0, the ORB specifies the CCW channel program. If B is 1, the ORB specifies the TCW channel program. Only I / O-devices that support FCX recognize TCW channel programs.

If the contents of the ORB specifying the CCW channel program are placed in the subchannel designated during execution of the START SUBCHANNEL, the subchannels remain in command mode. Therefore, such an ORB is also known as a command-mode ORB. If the contents of the ORB designating the TCW channel program are placed in the subchannel designated during execution of the START SUBCHANNEL, the subchannel enters the transfer mode. Therefore, these ORBs are also known as transfer-mode ORBs.

TEST PENDING INTERRUPTION  Command

The I / O-interruption code for the pending I / O interruption of the subchannel is stored at the location specified by the second-operand address, and the pending I / O-interruption request is cleared.

The second-operand address is the logical address of the location where the two-word I / O-interruption code, consisting of words 0 and 1, is to be stored, when not 0. The second-operand address must be specified in the word range; Otherwise, the specification exception is recognized.

If the second-operand address is 0, a three-word I / O-interruption code consisting of words 0-2 is stored in real locations 184-195. In this case, low-address protection and key-controlled protection do not apply.

In the access-register mode when the second-operand address is zero, it is unpredictable whether an access-register translation will occur for the access register B2. If a translation occurs, the resulting address-space-control element is not used; That is, the interruption code is still stored in actual positions 184-195.

Pending I / O-interruption requests are accepted only for the I / O-interruption subclasses allowed by the I / O-interruption subclass mask in the control register 6 of the CPU executing the instruction. If there are no I / O-interruption requests allowed by control register 6, the I / O-interruption code is not stored, the second-operand location is not modified, Respectively.

If a pending I / O-interruption request is accepted, the I / O-interruption code is stored, the pending I / O-interruption request is cleared, and condition code 1 is set. The stored I / O-interruption code is the same as that stored when an I / O interruption occurs. However, PSWs are not swapped as when I / O-interruption occurs. The execution of the command is defined as follows:

Subsystem-Identification Word (SID):

Bits 32-63 of the SID are placed in word 0.

Interruption parameters: word 1, was specified by the program is sent to the ORB or word 0 of the sub-channel PMCW, and a 4-byte parameter. This field contains zeros when the device has submitted an alarm condition and its interruption parameter was not previously sent to the subchannel by the execution of START SUBCHANNEL or MODIFY SUBCHANNEL.

Interruption -Identification Word: Word 2, when stored, contains an interruption-identifying word, which further identifies the source of the I / O-interruption. Word 2 is stored only when the second-operand address is zero.

The interruption-identification word is defined as:

A bit (A): Bit 0 of the interruption - identification word specifies the type of pending I / O - interruption request cleared. When bit 0 is zero, an I / O-interruption request was associated with a subchannel.

I / O -Interruption Subclass ( ISC ): The bit positions 2-4 of the interruption-identification word indicate the I / O-interruption subclass associated with the subchannel for which the pending I / O- Contains an unsigned binary integer, in the range 0-7, to specify. The remaining bit positions are reserved and stored as zeros.

Special conditions ( Special Conditions )

TEST PENDING INTERRUPTION can meet with program exceptions listed or listed below.

The second operand must be specified in the word range; Otherwise, the specification exception is recognized.

Execution of the TEST PENDING INTERRUPTION is suppressed in all addressing and protection exceptions.

Result Condition code:

0 Interruption code not saved

1 Interruption code stored

2 -

3 -

Program exceptions:

액세스 (저장, 오퍼랜드 2, 제2-오퍼랜드 주소 0이 아닌 것만)

Access (store, operand 2, only non-zero second-operand address)

특권 오퍼레이션

Privileged operation

명세

details

Programming notes:

1. TEST PENDING INTERRUPTION should only be executed with a second operand address of 0 when I / O interrupts are masked off. Otherwise, the I / O-interruption code stored by the instruction may be lost if an I / O interruption occurs. An I / O-interruption code that identifies the source of an I / O interruption to be taken subsequent to the TEST PENDING INTERRUPTION is also stored in actual locations 184-195, and the I / O- Replace the deflation code.

2. In the access-register mode when the second-operand address is zero, an access-exception is recognized if an access-register conversion occurs and the access register encounters an error. This exception can be prevented by making the B2 field zero or by placing 00000000 hex, 00000001 hex, or any other valid content in the access register.

STORE SUBCHANNEL  Command

Control and status information for the specified subchannel is stored in the specified SCHIB.

General purpose register 1 contains a subsystem-identification word that specifies a subchannel, and information about that subchannel will be stored. The second-operand address is the logical address of the SCHIB and should be specified in the word range; Otherwise, the specification exception is recognized.

When the Extension-I / O-Measure-Block Facility is not installed, the information stored in the SCHIB consists of the path-management-control word, the SCSW, and the model-dependent information of the three words. When the Extension-I / O-Measure-Block Facility is installed, the information stored in the SCHIB consists of the path-management-control word, the SCSW, the measurement-block-address field and the model-dependent information of a word .

The execution of the STORE SUBCHANNEL does not change any information on the subchannel.

Condition code 0 is set to indicate that control and status information for the specified subchannel is stored in SCHIB. If the result of the execution of STORE SUBCHANNEL results in a set of condition code 0, the information in SCHIB indicates a consistent state of the subchannel.

Special conditions ( Special Conditions )

When the specified subchannel is not operable for STORE SUBCHANNEL, condition code 3 is set and no other action is taken. If a subchannel is not provided in the channel subsystem, that subchannel is not operable for the STORE SUBCHANNEL.

STORE SUBCHANNEL may meet with the program exceptions listed or listed below.

When no multi-subchannel-set facility is installed, bits 32-47 of general purpose register 1 must contain 0001 hex; Otherwise, an operand exception is recognized.

When multi-subchannel-set facilities are installed, bits 32-44 of general register 1 must contain zeros, bits 45-46 must contain a valid value, and bit 47 contains a value of 1 Should be; Otherwise, an operand exception is recognized.

The second operand must be specified in the word range; Otherwise, the specification exception is recognized.

Result Condition code:

0 SCHIB stored

One -

2 -

3 Operation not possible

Program exceptions:

액세스 (저장, 오퍼랜드 2)

Access (store, operand 2)

오퍼랜드

Operand

특권 오퍼레이션

Privileged operation

명세

details

Programming notes:

1. The device state stored in the SCSW may include a device-busy indication, a control-unit busy indication, or a control-unit-termination indication.

2. Information stored in the SCHIB is obtained from the subchannel. The STORE SUBCHANNEL command does not cause the channel subsystem to interrogate the addressed device.

3. The STORE SUBCHANNEL can sample the conditions that exist in the subchannel at any time without causing any pending state conditions to be cleared.

4. Repeated execution of a STORE SUBCHANNEL without an intervening delay (for example, to determine when a subchannel changes state) should be avoided because iterative access of the subchannel by the CPU is performed by updating the subchannel Because the channel subsystem can delay or prevent access to the subchannel.

TEST SUBCHANNEL  Command

The control and status information for the subchannel is stored in the specified IRB.

The general purpose register 1 includes a subsystem-identifying word that specifies a subchannel, and information will be stored for this subchannel. The second-operand address is the logical address of the IRB and should be specified in a word range; Otherwise, the specification exception is recognized.

The information stored in the IRB consists of SCSW, an extended-status word, and an extended-control word.

If the subchannel is state pending, the state-pending bit of the state-control field is stored as one. Whether the subchannel is state pending or not affects the functions performed when the TEST SUBCHANNEL is executed.

When the subchannel is state pending and the TEST SUBCHANNEL is executed, the information is stored in the IRB, as described above, and then the specific conditions and indications present in the subchannel are cleared. If the subchannel is in the transmit mode, clearing these conditions, specifically clearing the start function, places the subchannel in the command mode. If the I / O-interruption request is pending for the subchannel, the request is cleared. Condition code 0 is set to indicate that these actions have been taken.

When the subchannel is not state pending and the TEST SUBCHANNEL is executed, the information is stored in the IRB, and the conditions or indications are not cleared. Condition code 1 is set to indicate that these actions have been taken.

Special conditions ( Special Conditions )

When the subchannel is not operable for TEST SUBCHANNEL, condition code 3 is set, and no other action is taken. If the subchannel is not provided, has no valid device number associated with it, or is not enabled, then the subchannel is not operable for the TEST SUBCHANNEL.

The TEST SUBCHANNEL can meet the program exceptions listed or listed below.

When no multi-subchannel-set facility is installed, bits 32-47 of general purpose register 1 must contain 0001 hex; Otherwise, an operand exception is recognized.

If multiple-subchannel-set facilities are provided, bits 32-44 of general register 1 must contain zeros, bits 45-46 must contain a valid value, and bit 47 must contain a value of 1 do; Otherwise, an operand exception is recognized.

The second operand must be specified in the word range; Otherwise, the specification exception is recognized.

When the execution of the TEST SUBCHANNEL terminates in addressing and protection exceptions, the state of the subchannel is not changed.

Result Condition code:

0 IRB stored; Subchannel state pending

1 IRB stored; Not subchannel state pending

2 -

3 Operation not possible

Program exceptions:

액세스 (저장, 오퍼랜드 2)

Access (store, operand 2)

오퍼랜드

Operand

특권 오퍼레이션

Privileged operation

명세

details

Programming notes:

1. The device state stored in the SCSW may include a device-busy indication, a control-unit busy indication, or a control-unit-termination indication.

2. Information stored in the IRB is obtained from the subchannel. The TEST SUBCHANNEL command does not cause the channel subsystem to query the addressed device.

3. When an I / O interruption occurs, it is the result of a state-pending condition on the subchannel, and typically a TEST SUBCHANNEL is executed to clear the state. TEST SUBCHANNEL can also be executed at any time to sample conditions that exist in a subchannel.

4. Repeated execution of the test subchannel to determine when the start function is completed should be avoided because there are conditions under which the completion of the start function may or may not be displayed. For example, if the channel subsystem is interrupted by an interface-control-check (IFCC) condition (for any subchannel), because another subchannel is already state pending, If the start function being tested by SUBCHANNEL has a channel path with an IFCC condition as the only selectable path, its start function allows the state-pending condition in the other subchannel to be cleared and displayed on the subchannel to which the IFCC condition is applied May not be initiated until such time.

5. For example, to determine when a subchannel changes state, iterative execution of a TEST SUBCHANNEL without an intervening delay should be avoided because the iterative access of the subchannels by the CPU is prevented by the channel subsystem The access of the sub-channel can be delayed or prohibited. The execution of the TEST SUBCHANNEL by multiple CPUs for the same subchannel at approximately the same time can and should also have the same effect.

6. Priority of I / O-interruption handling by the CPU can be changed by execution of TEST SUBCHANNEL. When the TEST SUBCHANNEL is executed and the specified subchannel has an I / O-interruption request pending, its I / O-interruption request is cleared and the SCSW is stored regardless of the previously established priority. The relative priority of the remaining I / O-interruption requests is not changed.

As one of ordinary skill in the art will appreciate, embodiments of the invention may be implemented as a system, method, or computer program product. Thus, embodiments of the present invention may take the form of entirely hardware embodiments, entirely software embodiments (including firmware, resident software, micro-code, etc.) or embodiments combining software and hardware embodiments, all of which are generic Quot; circuit ","module",or" system ". Further, embodiments of the invention may take the form of a computer program product embodied in one or more computer readable medium (s) in which the computer readable program code is embodied.

Any combination of one or more computer readable media (s) may be used. The computer readable medium may be a computer readable storage medium. The computer readable storage medium may be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (non-exhaustive list) of computer readable storage media include: electrical connection wiring having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read only memory ROM), erasable and programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read only memory (CD-ROM), optical storage device, magnetic storage device, or any combination of the foregoing. In the context of this document, a computer readable storage medium can be any type of medium that can contain or store a program for use by or in connection with an instruction execution system, device, or device.

Referring now to FIG. 10, in one example, a computer program product 1000 includes computer readable program code means or logic 1004, for example, to provide and facilitate one or more embodiments of the present invention. One or more non-transitory computer readable storage media 1002 for storing data thereon.

The program code embodied on the computer readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for performing operations on embodiments of the present invention may be implemented using object-oriented programming languages such as Java, Smalltalk, C ++, or the like and a "C" programming language, an assembler, Including one or more conventional procedural programming languages, such as those described in U.S. Pat. The program code may be executed entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on the remote computer or entirely on the remote computer or server have. In the last case above, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN) Through the Internet using a provider).

The present invention is described herein with reference to flow and / or block diagrams of methods, apparatus (systems) and computer program products in accordance with embodiments of the present invention. It will be appreciated that each block of flow diagrams and / or block diagrams and combinations of blocks within flow diagrams and / or block diagrams may be implemented by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, or other programmable data processing apparatus to create a machine so that the instructions may be executed via a processor of the computer or other programmable data processing apparatus Thereby creating means for implementing the functions / operations specified in the blocks and / or blocks of the flowchart and / or block diagrams.

These computer program instructions may also be stored on a computer readable medium, and instructions to a computer, other programmable data processing apparatus, or other devices may be used to cause the instructions stored on the computer readable medium to be transferred to a block or block May be made to function in a particular manner to create an article of manufacture comprising instructions that implement the function / action specified in the specification.

The computer program instructions may also be loaded into a computer, other programmable data processing apparatus, or other devices to cause a series of operation steps to be performed on the computer, other programmable data processing apparatus, or other devices, Capable device to generate computer-implemented processes to provide processes for implementing the functions / operations specified in the blocks and / or blocks of the flowchart and / or block diagrams.

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products in accordance with various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a portion of a module, segment, or code that includes one or more executable instructions for implementing the specified logical function (s). It should also be noted that in some alternative implementations, the functions mentioned in the blocks may occur differently from the order mentioned in the figures. For example, two blocks shown in succession may in fact be executed substantially concurrently, or the blocks may sometimes be executed in reverse order, depending on the functionality involved. Each block of the block diagrams and / or flow diagrams, and combinations of block diagrams in the block diagrams and / or flow diagrams, may be implemented with special purpose hardware and / or software that performs the specified functions or operations of the special purpose hardware and computer instructions, - based systems. ≪ RTI ID = 0.0 >

In addition to the above, one or more embodiments of the invention may be provided, provisioned, managed, serviced, etc. by a service provider servicing the management of the customer environment. For example, a service provider may create, maintain, support, etc. computer code and / or computer infrastructure to perform one or more embodiments of the present invention for one or more customers. In return, the service provider may receive payment from the customer under a subscription and / or fee arrangement, which is an example. Additionally or alternatively, a service provider may sell advertising content and receive costs to one or more third parties.

In one embodiment of the present invention, an application may be deployed to perform one or more embodiments of the present invention. By way of example, the deployment of an application includes providing a computer infrastructure operable to perform one or more embodiments of the present invention.

As a further embodiment of the present invention, a computing infrastructure may be deployed that includes integrating computer readable code into a computing system in which the code is coupled to one or more implementations of the present invention It is possible to perform the examples.

As a further embodiment of the present invention, a process for computing infrastructure integration may be provided that includes integrating computer readable code into a computer system. The computer system includes a computer readable medium, wherein the computer medium comprises one or more embodiments of the present invention. The code may be combined with a computer system to perform one or more embodiments of the invention.

While various embodiments have been described above, these are only examples. For example, computing environments of different architectures can include and use one or more embodiments of the present invention. By way of example, servers other than z196 servers may also include, use and / or benefit from one or more embodiments of the present invention. Other commands and / or commands may also be used, and the commands / commands may include additional, less, and / or other information than those described herein. Many variations are possible.

Other types of computing environments may also benefit from one or more embodiments of the present invention. By way of example, a data processing system suitable for storing and / or executing program code may be used, the system comprising at least two processors directly or indirectly coupled to memory elements via a system bus. Memory elements include, for example, local memory used during actual execution of the program code; Bulk storage; And a cache memory that provides temporary storage of at least some program code to reduce the number of times code must be retrieved from mass storage during execution.

Input / output or I / O devices (including but not limited to keyboards, displays, pointing devices, DASDs, tapes, CD, DVDs, thumb drives and other memory media) lt; RTI ID = 0.0 > intervening < / RTI > I / O controllers. A network adapter may also be coupled to the system to enable the data processing system to be coupled to other data processing systems or remote printers or storage devices via an intermediary private or public network. Modems, cable modems, and ethernet cards are some examples of available network adapters.

Referring to FIG. 11, representative components of a host computer system 5000 for implementing one or more embodiments of the present invention are shown. An exemplary host computer 5000 includes one or more CPUs 5001 that communicate with computer memory (i.e., central storage) 5002 and may also include other storage devices such as storage media devices 5011 and other computers or SANs 0.0 > 5010 < / RTI > The CPU 5001 conforms to an architecture having architected functionality with an architected instruction set. The CPU 5001 stores program addresses (virtual addresses) (DAT) 5003. DATs typically have a translation index buffer (not shown) for caching translations so that there is no delay in address translation when accessing a block of computer memory 5002 later, 5005. A cache 5009 is typically used between a computer memory 5002 and a processor 5001. The cache 5009 may be a large cache (lower level) caches between the large cache and each CPU, in some implementations. In some implementations, the lower level caches may be called instruction fetches, place In one embodiment, one instruction is fetched from the memory 5002 via the cache 5009 by the instruction fetch unit 5004. The instructions are fetched from the instruction fetch unit 5004 via the cache 5009. In one embodiment, Is decoded (in conjunction with other instructions in some embodiments) in an instruction decode unit 5006 and dispatched to an instruction execution unit or units 5008. Typically, several execution units For example, an arithmetic execution unit, a floating point execution unit, and a branch instruction execution unit. The instruction is executed by an execution unit And accesses the operands as needed from the registers or memo that the instruction specified. When an operand is accessed (loaded or stored) from memory 5002, a load / store unit 5005 typically handles access under control of the executed instruction. The instructions may be executed in hardware circuits or in an internal microcode (firmware) or a combination of the two.

As described above, the computer system includes information in local (or main) storage and also includes addressing, protection, and reference and change recording. Some examples of addressing are: the format of addresses, the concept of address spaces, various types of addresses, and the type of address that another type of address There is a way to be transformed. Some of the main storage includes permanently assigned storage locations. Main storage provides direct addressable, fast-access storage of data to the system. Both data and programs can be processed after being loaded into main storage (from input devices).

The main storage includes one or more smaller and faster access buffer storages, sometimes called a cache. The cache is typically physically associated with a CPU or I / O processor. The effects of the physical construction and use of a distinct storage medium are generally not observed by the program, except in practice.

Separate caches can be reserved for commands and for data operands. Information in the cache is stored in contiguous bytes on an integral boundary called a cache block or cache line (or shortened line). Some models can provide an EXTRACT CACHE ATTRIBUTE command that returns the size of the cache line in bytes. Some models may also provide a PREFETCH DATA command and a PREFETCH DATA RELATIVE LONG command that realize storage prefetching to data or instruction cache or release of data from cache.

Storage appears as a long horizontal string of bits. For most operations, access to storage proceeds in a left-to-right order. The sequence of bits is subdivided into 8-bit units. An 8-bit unit is called a byte, which is the basic building block of all information formats. Each byte location in the storage is identified by a nonnegative unique integer, which is the address of that byte location, or simply the byte address. Adjacent byte positions are consecutive addresses starting from left 0 and proceeding from left to right. Addresses are unsigned binary integers and are 24, 31, or 64 bits.

Information is transferred between the storage and the CPU or channel subsystem in one byte or group of bytes at a time. Unless otherwise specified, for example, in z / Architecture ^® , the group of bytes in the storage is addressed by the leftmost byte of the group. The number of bytes in a group is implied or explicitly specified by the operation to be performed. When used in a CPU operation, the group of bytes is called a field. Within each group of bytes, for example z / Architecture ^® , the bits are numbered from left to right. In z / Architecture ^® , the leftmost bits are sometimes referred to as "high-order" bits and the rightmost bits are referred to as "low-order" bits. However, the bit number is not a storage address. Only bytes can be addressed. To operate on individual bits of one byte in the storage, the entire byte is accessed. The bits within a byte (for example, z / Architecture ^® ) are numbered from left to right, from 0 to 7. The bits within an address may be numbered 8-31 or 40-63 for a 24-bit address, or 1-31 or 33-63 for a 31-bit address; 64-bit addresses are numbered 0-63. Within the multiple bytes of another fixed-length format, the bits that make up the format are numbered consecutively starting at zero. For purposes of error detection, and preferably for calibration, one or more check bits are transmitted with each byte or group of bytes. These check bits are automatically generated by the machine and can not be directly controlled by the program. Storage capacity is expressed in bytes. When the length of the storage-operand field is implied by the instruction's operation code, the field is said to have a fixed length, which can be 1, 2, 4, 8, or 16 bytes in length. Larger fields can be implied in certain commands. When the length of the storage-operand field is explicitly mentioned without being implied, it is said that the field has a variable length. Variable-length operands can vary in length by increments of one byte (or in some instructions, in multiples of two bytes, or in different multiples). When information is placed in storage, only the contents of that byte location contained in that designated field are replaced, although the width of the physical path to storage may be greater than the length of the field being stored.

Units of information will be in the integral boundary of storage. A boundary is said to be integral to a unit of information when its storage address is a multiple of the length of that unit. The fields of 2, 4, 8, and 16 bytes in the integral range are given special names. A halfword is a group of two consecutive bytes in a two-byte range and is the basic building block of instructions. A word is a group of four consecutive bytes in a 4-byte range. A doubleword is a group of eight consecutive bytes in the 8-byte range. A quadword is a group of 16 consecutive bytes in a 16-byte range. When storage addresses specify halfwords, words, double words, and quad words, the binary representation of the address includes 1, 2, 3, or 4 most significant right 0 bits, respectively. The instructions will be in 2-byte integral ranges. The storage operands of most of the instructions do not have boundary-alignment requirements.

On devices that implement separate caches for instructions and data operands, if the program stores in a cache line and instructions are subsequently fetched from the cache line, Whether or not you will be experiencing significant delays.

In one embodiment, the invention may be implemented in software (the software is sometimes referred to as licensed internal code, firmware, micro-code, milli- code, pico-code, etc., something to do). Referring to Figure 11, the software program code embodying the present invention is typically stored on the host system 5000 from long-term storage media devices 5011, such as a CD-ROM drive, tape drive or hard drive. And is accessed by the processor 5001. The software program code may be embodied in any of a variety of media known for use in a data processing system such as a diskette, hard drive, or CD-ROM. The code may be distributed on such media, or distributed to other computer systems via the network 5010 from users of the computer memory 5002 or storage of one computer system, for use by a user of such other systems have.

The software program code includes an operating system that controls the functions and interactions of various computer components and one or more application programs. The program code is usually paged from storage media device 5011 to relatively faster computer storage 5002, which is available for processing by processor 5001. [ Techniques and methods for implementing in-memory software program code on physical media and / or techniques and methods for distributing software code over networks are well known and will not be further discussed herein. The program code is often created and stored on a type of medium (including but not limited to electronic memory modules (RAM), flash memory, compact discs (CDs), DVDs, magnetic tape, etc.) It is called "computer program product". The computer program product medium is typically readable by a processing circuit and is preferably readable for execution by a processing circuit in a computer system.

Figure 12 illustrates a representative workstation or server hardware system in which the invention may be practiced. System 5020 of FIG. 12 includes an exemplary base computer system 5021, such as a personal computer, workstation, or server, including optional peripheral devices. Base computer system 5021 includes one or more processors 5026 and a bus that interconnects processor (s) 5026 and other components of system 5021 in accordance with known techniques for communication In order to make it possible. The bus connects the processor 5026 to the memory 5025 and the long term storage 5027 and the long term storage may be stored in a hard drive such as a hard drive (e.g., including magnetic media, CD, DVD and flash memory) Drive. The system 5021 may also include a user interface adapter that connects the microprocessor 5026 via a bus to a keyboard 5024, a mouse 5023, a printer / scanner 5030, and / Lt; / RTI > devices, such as devices. The other interface devices may be user interface devices such as a touch sensitive screen, a digitized entry pad, and the like. The bus also connects a display device 5022, such as an LCD screen or monitor, to the microprocessor 5026 via a display adapter.

System 5021 may communicate with networks of other computers or computers via a network adapter capable of communication 5028 with network 5029. [ Examples of network adapters include communications channels, token ring, Ethernet, or modems. Alternatively, the system 5021 may communicate using a wireless interface, such as a cellular digital packet data (CDPD) card. The system 5021 may be associated with other computers in a local area network (LAN) or a wide area network (WAN), or the system 5021 may be a client in another computer and client / server arrangement . All of these configurations and appropriate communication hardware and software are known in the art.

Figure 13 illustrates a data processing network 5040 in which the invention may be practiced. Data processing network 5040 may include a plurality of separate networks, such as a wireless network and a wired network, each of which may include a plurality of individual workstations 5041, 5042, 5043, and 5044. Also, as those skilled in the art will appreciate, one or more LANs may be included, wherein the LAN includes a plurality of intelligent workstations coupled to the host processor .

13, the networks may also include mainframe computers, such as a gateway computer (client server 5046) or an application server (a remote server 5048 that can access the data store and is directly accessible from the workstation 5045) Servers. The gateway computer 5046 serves as a point of entry to each individual network. A gateway is needed to connect one networking protocol to another networking protocol. The gateway 5046 may preferably be coupled to another network (e.g., the Internet 5047) over a communication link. The gateway 5046 may also be coupled directly to one or more workstations 5041, 5042, 5043, 5044 using a communication link. The gateway computer may be implemented utilizing an IBM eServer ™ System z ^® server available from International Business Machines Corporation.

12 and 13, software programming code that may implement the present invention may be accessed by a processor 5026 of system 5020 from a long term storage medium 5027, such as a CD-ROM drive or a hard drive have. The software programming code may be embodied in any of a variety of media known for use with a data processing system such as a diskette, hard drive, or CD-ROM. The code may be distributed on such a medium or distributed to other computer systems through the network from users 5050, 5051 of memory or storage of one computer system, for use by a user of such other systems.

Alternatively, the programming code may be implemented in memory 5025 and accessed by processor 5026 using a processor bus. Such programming code includes an operating system that controls the functions and interactions of various computer components and one or more application programs 5032. The program code is typically paged from the storage medium 5027 to a high-speed memory 5025, which is available for processing by the processor 5026. [ Techniques and methods for implementing in-memory software programming code on physical media and / or techniques and methods for distributing software code over networks are well known and will not be further discussed herein. The program code is often created and stored on a type of medium (including but not limited to electronic memory modules (RAM), flash memory, compact discs (CDs), DVDs, magnetic tape, etc.) It is called "computer program product". The computer program product medium is typically readable by a processing circuit and is preferably readable for execution by a processing circuit in a computer system.

The most readily available cache (usually faster and smaller than the other caches in the processor) is the lowest (L1 or level 1) cache and the main storage (main memory) is the highest level cache L3 if it is. The lowest level cache is often divided into an instruction cache (I-cache) which holds machine instructions to be executed and a data cache (D-cache) which holds data operands.

Referring to FIG. 14, an exemplary processor embodiment is shown for processor 5026. One or more levels of cache 5053 are typically employed to buffer the memory blocks to improve processor performance. Cache 5053 is a high-speed buffer that holds cache lines of memory data that may be used. Typical cache lines are 64, 128 or 256 bytes of memory data. Separate caches are often employed to cache instructions rather than to cache data. Cache coherence (synchronization of copies of lines in memory and caches) is often provided by "snoop" algorithms well known in the art. The main memory storage 5025 of the processor system is often referred to as a cache. In a processor system with a four-level cache 5053, the main storage 5025 is sometimes referred to as a level 5 (L5) cache because it is typically faster and non-volatile storage (DASD, tape, etc.) As shown in FIG. Main storage 5025 "caches " pages of data paged in and out of main storage 5025 by the operating system.

A program counter (command counter) 5061 tracks the address of the current instruction to be executed. Program counters in the z / Architecture ^® processor are 64 bits and can be truncated to 31 or 24 bits to support previous addressing limits. The program counter is typically implemented in the computer's PSW (Program Status Word), allowing it to persist during context switching. Thus, an ongoing program having a program counter value can be interrupted, for example, by the operating system (context switching from the program environment to the operating system environment). When the program is not active, the PSW of the program retains the program counter value, and the operating system's (in PSW) program counter is used when the operating system is running. Typically, the program counter is incremented by an amount equal to the number of bytes of the current instruction. Reduced Instruction Set Computing (RISC) instructions are typically fixed-length, while Complex Instruction Set Computing (CISC) instructions are typically variable length. The IBM z / Architecture ^® instructions are CISC instructions with lengths of 2, 4, or 6 bytes. The program counter 5061 is changed by, for example, a branch taken operation of the branch instruction or a context switching operation. In the context switching operation, the current program counter value is saved in the program status word with status information (e.g., condition codes) about the program being executed, and a command of the new program module to be executed The new program counter value is pointed to. A branch taken operation is performed to allow the program to make a decision or to loop in the program by loading the result of the branch instruction in the program counter 5061. [

The instruction fetch unit 5055 is typically employed to fetch instructions on behalf of the processor 5026. [ The fetch unit fetches the "next sequential instructions", the target instructions of the branch adoption instructions, or the first instructions of the program following the context switch. Modern Instruction The fetch unit often employs prefetch techniques that speculatively prefetch instructions based on the likelihood that prefetched instructions may be used. For example, the fetch unit may fetch a 16-byte instruction, which includes the next sequential instruction and the subsequent sequential instructions.

The fetched instructions are then executed by the processor 5026. [ In one embodiment, the fetched instruction (s) is sent to the dispatch unit 5056 of the fetch unit. The dispatch unit decodes the instruction (s) and passes information about the decoded instruction (s) to the appropriate units (5057, 5058, 5060). Execution unit 5057 will typically receive information about decoded arithmetic instructions from instruction fetch unit 5055 and perform arithmetic operations on the operands in accordance with the opcode of the instruction arithmetic operations. The operands are preferably provided to the execution unit 5057 from the memory 5025, the architectural registers 5059, or from the immediate field of the instruction being executed. When stored, the results of execution are stored in memory 5025, registers 5059, or in other machine hardware (e.g., control registers, PSW registers, etc.).

Typically, the processor 5026 has one or more units 5057, 5058, 5060 for executing the function of the instruction. 15A, an execution unit 5057 includes arithmetic generalized registers 5059, a decode / dispatch unit 5056, a load store unit 5060, and other 5065 processors via interfacing logic 5071, Lt; / RTI > units. The execution unit 5057 may employ some register circuits 5067, 5068, 5069 to hold the information to be operated by an arithmetic logic unit (ALU) 5066. The ALU performs arithmetic operations such as addition, subtraction, multiplication, and division, as well as logical functions such as AND, OR, and XOR, rotate, and shift. Preferably, the ALU supports special operations dependent on the design. Other circuits may provide other architectured facilities 5072, including, for example, condition codes and recovery assistance logic. Typically, the result of the ALU operation is held in the output register circuit 5070, which can pass the results to various other processing functions. The arrangement of many processor units may vary, and the description is only intended to provide a representative understanding of one embodiment of the present invention.

For example, the ADD instruction will be executed in an execution unit 5057 having arithmetic and logic functions, while a floating point instruction, for example, will be executed in floating point execution with special floating point capability. Preferably, the execution unit operates on the operands identified by the instruction by performing an opcode defined function on the operands. For example, an ADD instruction may be executed by execution unit 5057 with respect to the operands found in two registers 5059 identified by the register fields of the instruction.

Execution unit 5057 performs arithmetic addition on the two operands and stores the result in a third operand where the third operand can be either the third register or one of the two source registers have. Preferably, the execution unit utilizes an arithmetic logic unit (ALU) 5066, which not only includes various algebraic functions including addition, subtraction, multiplication, division, Various logic functions such as Shift, Rotate, AND, Or, and XOR can be performed. Some ALUs 5066 are designed for scalar operations and some are designed for floating point. Depending on the architecture, the data may be in the Big Endian (where the least significant byte is at the highest byte address) or Little Endian (where the least significant byte is at the lowest byte address) . IBM z / Architecture ^® is a big endian. Signed fields may be sign and size, 1's complement, or 2's complement, depending on the architecture. A two's complement is advantageous in that a negative or positive value in a two's complement requires only the addition in the ALU, so that the ALU does not need to design the subtracting ability. Numbers are generally described as shorthand, in which a 12-bit field defines the address of a 4,096 byte block and is typically described as a 4 Kbyte (Kilo-byte) block.

Referring to FIG. 15B, branch instruction information for executing a branch instruction is typically sent to branch unit 5058 which, before the completion of other conditional operations, A branch prediction algorithm such as branch history table 5082 is often employed. The target of the current branch instruction will be fetched and executed speculatively before the conditional operations are completed. When the conditional operations are completed, the speculatively executed branch instructions are completed or discarded based on the conditions of the conditional operation and the inferred result. A typical branch instruction may test the condition codes and branch to the target address if the condition codes meet the branch requirement of the branch instruction, and the target address may be, for example, a register field or an immediate field of the instruction ≪ / RTI > may be computed based on a number of numbers including those found in < RTI ID = 0.0 > The branch unit 5058 may employ an ALU 5074 having a plurality of input register circuits 5075, 5075, 5077 and an output register circuit 5080. The branch unit 5058 may communicate with, for example, general purpose registers 5059, a decode dispatch unit 5056, or other circuits 5073.

Execution of a group of instructions can be interrupted for various reasons, including, for example, operating system-initiated context switching, program exceptions or errors resulting in context switching, context switching, I / O interruption signals that result in multi-threading activity of a plurality of programs. Preferably, the context switching action saves state information about the currently running program, and then loads state information about another program being called. The state information may be stored in, for example, hardware registers or memory. Preferably, the state information includes a next instruction to be executed, condition codes, memory conversion information, and a program counter value indicating the architectural register content. The context switching activity may be executed by hardware circuits, application programs, operating system programs, or firmware code (microcode, pico-code or licensed internal code (LIC)) alone or in combination.

The processor accesses the operands according to instruction defined methods. The instruction may provide an immediate operand upon use of the value of a portion of the instruction and may include one or more registers that explicitly point to general purpose registers or special purpose registers (e.g., floating point registers) Fields. The instruction may use implied registers identified as operands by the opcode field. The instruction may use memory locations for the operands. Memory locations of the operand registers, immediate and fields may be provided by a combination of (immediate field), or registers and immediate field, which facility is a typical example Displacement Field ^{® z / Architecture (long displacement)} , wherein command base Registers, index registers, and immediate fields (displacement fields) -these are added together to provide the address of the operand in memory, for example. If not indicated otherwise, the location here typically implies a location in main memory (main storage).

Referring to FIG. 15C, the processor accesses the storage using a load / store unit 5060. The load / store unit 5060 may perform load operations by obtaining the address of the target operand in the memory 5053 and loading the operand to the location of the register 5059 or another memory 5053, To perform the store operation by obtaining the address of the target operand in the memory 5053 and storing the data obtained from the register 5059 or another memory 5053 location in the memory 5053 at the target operand location. The load / store unit 5060 may be speculative and may access memory in an out-of-order order relative to the instruction order, And will maintain the appearance of the programs as they are executed in sequence. The load / store unit 5060 can communicate with the general purpose registers 5059, the decode / dispatch unit 5056, the cache / memory interface 5053 or other elements 5083, ALUs 5085 and control logic 5090 to provide pipeline sequencing to maintain operations in order. Some operations may be reversed in order, but as is well known in the art, the load / store unit provides the ability to cause out-of-order operations to appear as if they were performed sequentially on the program.

Preferably, the addresses " sees "of the application are often referred to as virtual addresses. Virtual addresses are sometimes referred to as "logical addresses" and "effective addresses ". These virtual addresses are virtual in that they are sent back to the physical memory location by one of several dynamic address translation (DAT) techniques. The various dynamic address translation (DAT) techniques described above may include simply prefixing the virtual address with an offset value, translating the virtual address through one or more translation tables, It is not. Preferably, the conversion tables include at least a segment table and a page table or a combination thereof, and preferably, the segment table has an entry pointing to a page table. In the z / Architecture ^®, there is provided a layer (hierarchy of translation) of the conversion, the conversion of the layer includes a region first table, region second table, region third table, a segment table and an optional page table. Implementation of address translation is often enhanced using a translation lookaside buffer (TLB), which contains entries that map a virtual address to an associated physical memory location. Entries are created when the DAT translates virtual addresses using translation tables. You can then use the entries in the faster TLB rather than the slow consecutive conversion table accesses when you subsequently use the virtual address. The TLB content may be managed by various replacement algorithms including LRU (Least Recently Used).

When a processor is a processor of a multi-processor system, each processor maintains interlocks for coherency with shared resources such as I / O, caches, TLBs, and memory. It plays a role. Typically, "snoop" techniques will be used to maintain cache coherency. In a snoop environment, each cache line is either in a shared state, an exclusive state, a changed state, or an invalid state to facilitate sharing Can be displayed.

I / O units 5054 (FIG. 14) provide a means for connecting the processor to peripheral devices, for example, tape, disk, printer, display, and network. I / O units are often provided to a computer program by soft drivers. In mainframes such as IBM ^® System z ^® , channel adapters and open system adapters are mainframe I / O units that enable communication between the operating system and peripheral devices.

Further, other types of computing environments may benefit from one or more embodiments of the present invention. In one example, the environment may include an emulator (e.g., software or other emulation mechanisms) in which a specific architecture (e.g., instruction execution, architectural functions such as address translation, (E. G., On a native computer system with a processor and memory) is < / RTI > emulated. In such an environment, one or more emulation functions of the emulator may implement one or more embodiments of the invention, although the computer running the emulator may have an architecture different from the emulating capabilities. As an example, in the emulation mode, the particular command or operation being emulated is decoded and the appropriate emulation function is made to implement the individual command or operation.

In an emulation environment, the host computer may include, for example, a memory that stores instructions and data, an instruction fetch unit that fetches instructions from memory and optionally also provides local buffering for the fetched instruction, And an instruction decode unit for determining the type of instructions fetched, and an instruction execution unit for executing instructions. Execution may include loading data from a memory into a register; Storing the data back into the memory from the register; Or performing some type of arithmetic or logical operation, as determined by the decode unit. In one example, each unit is implemented in software. For example, the operations being performed by the units are implemented as one or more subroutines within the emulator software.

More specifically, in the mainframe, architecturized machine instructions are often used by programmers, often by today's "C" programmers, often through compiler applications. These commands, which are stored on the storage media, can be executed natively on machines running z / Architecture ^® IBM ^® servers or alternatively different architectures. They can be emulated on existing and future IBM ^® mainframe servers and on other IBM ^® machines (eg, Power System Servers and System x ^® servers). They can run on machines running Linux on a wide range of machines using hardware manufactured by IBM ^® , Intel ^® , AMD ™ and others. In addition to running on the hardware under z / Architecture ^® , Hercules, UMX, or FSI (Fundamental Software, Inc) - where execution is usually in emulation mode - as well as machines using emulation, as well as Linux . In the emulation mode, the emulation software is executed by the native processor to emulate the architecture of the emulated processor.

A native processor typically executes emulation software, including firmware or a native operating system, to perform emulation of the emulated processor. The emulation software is responsible for fetching and executing instructions of the emulated processor architecture. The emulation software maintains an emulated program counter to track instruction boundaries. The emulation software may fetch one or more emulated machine instructions at a time to convert one or more of the emulated machine instructions into a corresponding group of native machine instructions for execution by the native processor. These translated instructions may be cached so that faster conversion may be performed. Nonetheless, the emulation software must maintain the architectural rules of its emulated processor architecture to ensure that the operating systems and applications written for the emulated processor are operating correctly. Further, the emulation software must provide the resources identified by the emulated processor architecture-these resources include control registers, general purpose registers, floating point registers, such as segment tables and page tables Architectural interfaces to dynamic address translation functions, interrupt mechanisms, context switching mechanisms, Time of Day clocks, and I / O subsystems are included so that applications designated to run on an operating system or an emulated processor Allowing the program to run on a native processor with emulation software.

The particular command being emulated is decoded and the subroutine is called to perform the function of the individual command. Emulation software functions that emulate the functions of the emulated processor are implemented, for example, as "C" subroutines or drivers, or some other method of providing the drive for specific hardware, It will be possible for those skilled in the art to derive those skilled in the art. Various software and hardware emulation patents are described in, for example, Beausoleil et al., Letters Patent No. 5,551,013, "Multiprocessor for Hardware Emulation "; Scalzi et al., U.S. Patent No. 6,009,261, entitled " Preprocessing of Stored Target Routines for Emulating Uncompatible Instructions on a Target Processor, " Davidian et al., U.S. Patent No. 5,574,873 entitled " Decoding Guest Instruction to Direct Access Emulation Routines That Emulate the Guest Instructions " for Direct Access Emulation Routines Emulating Guest Instructions; Gorishek et al., US Patent 6,308,255 "Symmetric Multiprocessing Bus and Chipset Used for Coprocessor Support to Enable Non-Native Code Execution in a System (Symmetrical Multiprocessing Bus and Chipset Used for Coprocessor Support Allow Non-Native Code to Run in a System) "; U.S. Patent No. 6,463,582 to Lethin et al., Entitled " Dynamic Optimization Object Code Translator for Architectural Emulation and Dynamic Optimizing Object Code Translation Method "; U. S. Patent No. 5,790, 825 to Eric Traut, entitled " Method for Emulating Guest Commands on a Host Computer Through Dynamic Recompilation of Host Commands " But are not limited to, those known to those of ordinary skill in the art for achieving emulation of an architectural command format for other machines for a target machine .

16, an example of an emulated host computer system 5092 that emulates a host computer system 5000 'of the host architecture is provided. In an emulated host computer system 5092, a host processor (CPU) 5091 is an emulated host processor (or virtual host processor) and is a native instruction set of the processor 5091 of the host computer 5000 ' architecture having a native instruction set architecture different from the native instruction set architecture. The emulated host computer system 5092 has a memory 5094 accessible by the emulation processor 5093. In an embodiment, memory 5094 is divided into portions of host computer memory 5096 and portions of emulation routines 5097. The host computer memory 5096 may be utilized by programs of the emulated host computer 5092 according to the host computer architecture. The emulation processor 5093 executes native instructions obtained from the native instructions of the architectural instruction set of the architecture other than the instruction of the emulated processor 5091, namely the emulation routines memory 5097, and the sequence & By employing one or more instructions (s) obtained in a routine - which can decode host instruction (s) accessed to determine a native instruction execution routine to emulate a function of the host command being accessed - A host command for execution from the program in the memory 5096 can be accessed. Other facilities defined for the host computer system 5000 'architecture may be emulated by architected facilities routines, such as, for example, general purpose registers, control registers registers, dynamic address translation and I / O subsystem support, and processor cache. The emulation routines can also improve the performance of the emulation routines using functions available in the emulation processor 5093 (e.g., dynamic translation of general registers and virtual addresses). Also, special hardware and off-load engines may be provided to assist the processor 5093 in emulating the functions of the host computer 5000 '.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms "a,"" one, "and" one ", and the like are intended to also include the plural forms unless the context clearly dictates otherwise. Also, the terms "comprises" and / or "comprising " when used herein specify the presence of stated features, integers, steps, operations, elements, and / Steps, operations, elements, components, and / or groups of elements, integers, steps, operations, elements, components, and / or groups thereof.

In the following claims, equivalents of structures, materials, operations, and all means or step-plus functional components are to be construed in a manner that is consistent with the structures, materials, Or operations described herein. The description of the present invention is presented for purposes of illustration and description, and is not intended to be exhaustive or to limit the invention to the full scope of the embodiments of the present invention in the form disclosed. It will be apparent to those skilled in the art that many modifications and variations may be made without departing from the scope of the present invention. The embodiments are described in order to best explain the principles and practical applications of the invention and to enable those of ordinary skill in the art to understand the invention for various embodiments with various modifications as are suited to the particular use contemplated In order to be able to do so.

Claims (9)

CLAIMS What is claimed is: 1. A method of executing a Start Subchannel command in a computing environment comprising main storage and storage class memory, the method comprising:
In response to determining that the subchannel identified by the Start Subchannel command is an Asynchronous Data Mover (ADM) subchannel, perform the following steps:
Obtaining an operation request block from the main storage, the operation request block including an address of an operation block;
Obtaining an operation block from the main storage based on an address of the operation block, the operation block comprising a request block, a response block, and one or more first movement specification blocks (MSBs) Wherein the request block comprises an MSB count field, the MSB count field having a value indicating the number of one or more MSBs referenced by it included in the operation block, the response block being configured to hold exception conditions Each move specification block is configured to include an operation code field, a block count field, a main storage address field for main storage addressing, a storage class memory address field for storage class memory addressing, a block size field and flags field, ;
Obtaining a move specification block of the one or more move specification blocks;
Determining an operation configured to be performed on blocks of the storage class memory based on an operation code in the obtained move specification block;
Initiating the operation on blocks of storage class memory associated with a storage class memory address of the storage class memory address field based on the move specification block; And
Storing information in the response block in response to at least starting the operation and encountering an exception condition, the response block including an exception flags field, a field validity field, a failed MSB address field, a failed AIDAW (indirect data address word) An address field, a failure main storage address field, and a failure storage class memory address field, said information being associated with one or more exceptions resulting from said operation, said one or more exceptions being: A storage class memory associated with a storage class memory address of the storage class memory address field, or the obtained move specification block, the information being associated with an indication of an exception and an exception, Related Including those that perform - and that includes the information, the additional information should specify one or more address exceptions
Way. The method according to claim 1,
In response to the field validity field comprising a first value, the response block indicating that a failed MSB address is stored;
Responsive to the field validity field including a second value, the response block indicates that a failed main storage address is to be stored; And
In response to the field validity field including a third value, the response block indicates that the failed storage class memory address is stored
Way. The method according to claim 1,
In response to the exception flags field containing a first value, the response block indicates a program check;
In response to the exception flags field containing a second value, the response block indicates a storage protection exception; And
In response to the exception flags field containing a third value, the response block indicates an unrecoverable storage class memory data error
Way. 4. The method of claim 1, wherein the response block further comprises an exception modifier code that provides additional information related to the exception
Way. 5. The method of claim 4,
In response to the exception modifier code having a first value, the response block indicates an MSB count error;
In response to the exception modifier code having a second value, the response block indicates a block size error;
In response to the exception qualifier code having a third value, the response block indicates a block count error;
Responsive to the exception qualifier code having a fourth value, the response block indicates a main storage address specification error;
In response to the exception qualifier code having a fifth value, the response block indicates a storage class memory address specification error;
In response to the exception qualifier code having a sixth value, the response block indicates a main storage address exception; And
In response to the exception qualifier code having a seventh value, the response block indicates a storage class memory address exception
Way. 5. The method of claim 4,
In response to the exception qualifier code having an eighth value, the response block indicates a main storage error;
In response to the exception qualifier code having a ninth value, the response block indicates an MSB list error;
Responsive to the exception modifier code having a tenth value, the response block indicates an MSB branch error;
In response to the exception modifier code having an eleventh value, the response block indicates an AIDAW list error; And
In response to the exception modifier code having a twelve value, the response block indicates an AIDAW branch error
Way. 5. The method of claim 4, wherein in response to the exception qualifier code having a thirteen value, the response block displays one or more unlock errors
Way. A computer system for executing a Start Subchannel command in a computing environment including main storage and storage class memory, the computer system comprising:
Memory; And
A processor in communication with the memory, the computer system being configured to perform the method according to any one of claims 1 to 7
Computer system. A computer-readable medium comprising:
The computer readable medium comprising a computer program for executing a Start Subchannel command in a computing environment comprising main storage and storage class memory,
The computer program comprising instructions executable by a processing circuit readable by a processing circuit and for performing a method according to any one of claims 1 to 7
Computer readable medium.

Images