Download and Unpacking Tips

Files on the TPF website are packaged in different formats according to the content or the kind of function the file provides.

  • Product documentation files are in Adobe Acrobat (PDF) format.
  • Tools are in compressed ZIP or TAR packages.
  • z/TPF and z/TPFDF APAR packages are in GNU ZIP (gz) files.
  • TPF 4.1 and TPFDF 1.1.3 APAR packages are in compressed USS PAX format files.
  • TPF Operations Server APARs are in Windows zip files.

Note:
Recent changes to some browsers have resulted in some download files being brought into SaveAs dialogs with incorrect names. In most instances the correct name for the download file is displayed nearby the "Download Now" link. When the name is incorrectly displayed in the SaveAs dialog, replace the name shown by SaveAs with the one displayed near the link. For example, in the SaveAs dialog for the MPIF automation driver some browsers display the File name as "mpif.tar.tar" or "PK12345.ascii.tar.htm" but this is incorrect. The file name should be changed to "mpif.tar.gz" before proceeding to Save. The "mpif.tar.gz" name is displayed near the download link on the MPIF automation driver page. All APAR binary files are "apar.ascii.tar.Z", "apar.source.ascii.tar.Z", "apar.listing.ascii.tar.Z", and "apar.binary.tar.Z".

Restricted files (such as those for product maintenance and certain tools) require a special user ID and password provided by the TPF organization. When downloading one of these files, login to the server first, then access the file, right click it and select "Save Target As" (for Internet Explorer) or "Save Link As" (for Mozilla FireFox).

Accessing Product Documentation Files

PDF files are used for white papers, Newsletter issues, TPFUG presentations, and other items. These files are directly linked to our web pages. To download the file, position the mouse pointer on the file link, then right click and select "Save Target As" (Internet Explorer) or "Save Link As" (Netscape).

To view and print PDF files, you must have the Adobe Acrobat Reader software, which is available for free from the Adobe Web site:
http://www.adobe.com/products/acrobat/readstep2.html

Accessing TPF Tools

TPF tools, drivers and automation scripts are available for download. A variety of packing formats are used. In general, you will find unpacking instructions in a Readme file that can be accessed on the same page as each downloadable package.

Some tools, drivers, or automation is restricted to TPF product famly licenseholders. These are indicated as being for customers only. Customers can download them by logging into the download server first, then clicking on the Download Now button. Other tools, drivers, and automation is supplied to anyone "as is."

Working with APAR Packages

Unpacking z/TPF and z/TPFDF APARs

z/TPF and z/TPFDF APAR packages are created on a Linux system using the tar command with Lempel-Ziv (gzip) compression. The file naming convention is Pxyyyyy.tar.gz. There is a single package for each APAR and it can include both text and binary files.

The files in each APAR package have a relative directory structure that includes the APAR number as the highest level directory. For example, the file named base/cp/cpse.cpy that is delivered for an APAR named PJ12345 will have this relative name:

     PJ12345/base/cp/cpse.cpy

This naming convention is used so that if multiple APAR packages are extracted under a common root directory, the files for each APAR will be isolated from one another.

The Pxyyyyy.tar.gz file can be unpacked on a workstation if you have a program, such as WinZip, that supports this format.

The Pxyyyyy.tar.gz file must be unpacked on a Linux system in order to build the APAR. To unpack the APAR package on Linux, do one of the following:

  1. Place the APAR gz file on Linux and use this tar command to unpack the file:

        tar -xzf Pxyyyyy.tar.gz


    2.  
  2. Place the APAR gz file on a Windows drive, extract the contents to the drive and then copy the files to Linux.
    • Use WinZip or another unpack utility to unpack the file on a workstation.
    • If you use FTP to copy extracted files between Windows and Linux platforms, you must ensure that binary or ascii mode is active, as appropriate, for each file copied.
    • If you copy extracted files to an NFS mounted Linux drive, two separate mounts are required, one for text and one for binary. The extracted files must be copied to the appropriate drive by type.
    • If you copy extracted files to an SMB mounted Linux drive, only one mount is required because SMB can determine the type of file when it is copied and set the format accordingly.
  3. Place the APAR gz file on a Windows drive and extract the files directly to a Linux mounted drive.
    • If you use NFS mounted drives, you must have a separate mount for text and binary, and you must extract only the text files to the text mounted drive and only binary files to the binary mounted drive.
    • If you use an SMB mounted drive, a single extract of all files can be used.

z/TPF includes z/OS offline utility programs. Even though these programs must be built and run on z/OS, you should store the source files on Linux and gain access to the code for build purposes by using a Linux mounted drive. This way, all source code can be managed on the Linux platform.


Unpacking TPF 4.1 APARs

TPF 4.1 APAR packages are created on z/OS USS platform using the pax command with Lempel-Ziv compression. Each APAR will deliver a source tar file, and optionally may deliver a listing and binary tar file.

  • The source tar file is named PJyyyyy.source.ascii.tar.Z and contains the changed source code and the APAR description file. The content of this files has been converted to ASCII.
  • The listings tar file is named PJyyyyy.listings.ascii.tar.Z and contains the listings for changed objects that are included in link modules that have been built and delivered with this APAR. The content of this file has been converted to ASCII.
  • The binary tar file is named PJyyyyy.binary.tar.Z and contains "object code only" objects and load modules. The content of this file is NOT converted to ASCII.

Because the source and listings tar files have been converted to ASCII, they can be unpacked for viewing on a workstation, using WinZip or any unpack utility that supports this format. The binary tar file is not in ASCII format and can only be unpacked on z/OS USS platform.

The APAR source code and "object code only" objects (if delivered) must be unpacked on a z/OS system in order to build the APAR. To put the files on z/OS and unpack them, do all of the following steps:

  1. On USS, create a directory where the source and binary tar files will be unpacked. (The listing may also be unpacked here but they are not required to build and install the APAR.) We recommend the directory be the APAR number, such as /u/yourUserid/PJyyyyy.
  2. Use FTP to send the tar file (downloaded from the TPF website) from your workstation to z/OS USS file system. Make sure you transfer the tar file in binary mode. When this step is complete, you will have /u/yourUserid/PJyyyyy/PJyyyyy.source.ascii.tar.Z. If you have a binary tar file to send to USS, doing so will create /u/yourUserid/PJyyyyy/PJyyyyy.binary.tar.Z.
  3. On USS, cd to the directory /u/yourUserid/PJyyyyy and unpack the source tar file using this command:

    pax -rf PJyyyyy.source.ascii.tar.Z -ofrom=ISO8859-1,to=IBM-1047

    The -ofrom and to parameters convert the content from ASCII back to EBCDIC. If you have a binary tar file to unpack, use this command:

    pax -rf PJyyyyy.binary.tar.Z


    4.  
  4. The result of the unpack commands in step 3 will be a directory structure similar to this. The exact structure will vary depending on the APAR content and whether or not a binary tar file was included.

              PJyyyyy/PJyyyyy.txt
              PJyyyyy/source
              PJyyyyy/macro
              PJyyyyy/include
              PJyyyyy/load
              PJyyyyy/load/oco
              PJyyyyy/obj/oco
              PJyyyyy/README/

Load modules that are delivered in the binary tar file are available in two formats. A README file is included in each binary tar file and it explains how to unpack either format. The frist format allows customers to unload the load modules into a partitioned data set using commands that are available on the z/OS system. The second format allows the load modules to be copied from the USS file system to a partitioned data set using the "lmcp" copy program. The LMCP tool (and others) can be found at http://www.ibm.com/software/htp/tpf/download/bldtools.htm.

Unpacking TPFDF 1.1.3 APARs

TPFDF 1.1.3 APAR packages are created on z/OS USS platform. Each APAR will deliver two tar files generated with the pax command with Lempel-Ziv compression, and optionally may deliver a third file if the APAR is applicable to an ALCS system.

  • The "full source" tar file is named PKyyyyy.source.ascii.tar.Z and contains the APAR's changed source code in full source replacement format. The content of this file has been converted to ASCII.
  • The "TPF format delta" tar file is named PKyyyyy.ascii.tar.Z. It contains PKyyyyy.DESCRIBE file which is the APAR description, and PKyyyyy.APARTPF file which is all the source updates for this APAR in MVS delta format.
  • The "ALCS format" file is named PKyyyyy.bin and contains the SMP/E format source updates which can be applied to a TPFDF installation for ALCS.

Since the "full source" and "TPF format delta" tar files have been converted to ASCII, they can be unpacked for viewing on a workstation using WinZip or any unpack utility that supports this format. (The "ALCS format" file is not in ASCII format and can only be processed on z/OS platform.)

To put the two tar files on z/OS and unpack them, do the following steps:

  1. On USS, create a directory where the tar files will be unpacked. We recommend the directory be the APAR number, such as /u/yourUserid/PKyyyyy.
  2. Use FTP to send the tar files (downloaded from the TPF website) from your workstation to z/OS USS file system. Make sure you transfer the tar files in binary mode. When this step is complete, you will have /u/yourUserid/PKyyyyy/PKyyyyy.source.ascii.tar.Z and PKyyyyy.ascii.tar.Z.
  3. On USS, cd to the directory /u/yourUserid/PKyyyyy. Unpack the "full source" tar file using this command:

    pax -rf PKyyyyy.source.ascii.tar.Z -ofrom=ISO8859-1,to=IBM-1047

    The -ofrom and to parameters convert the content from ASCII back to EBCDIC. Unpack the "TPF format delta" tar file using this command:

    pax -rf PKyyyyy.ascii.tar.Z -ofrom=ISO8859-1,to=IBM-1047


    4.  
  4. Once unpacked, the resulting APARTPF or full source files can be moved to partitioned datasets or other systems for further processing.

The "ALCS format" file named PKyyyyy.bin may contain embedded binary object code so it should always be moved in binary mode. This file is also known as the "APARFIX" file and is used by ALCS customers for their TPFDF installation. After downloading PKyyyyy.bin to your workstation, do the following steps to move the file to a partitioned dataset:

  1. On your z/OS system, allocate a partitioned dataset (if one doesn't already exist) to move the APARFIX file into. In this example, the PDS name is "YOUR.APARFIX.TPFDF".
  2. On your workstation in a command window, change to the directory where the PKyyyyy.bin file is, then open an FTP session to your z/OS system and use these commands to move the PKyyyyy.bin file to the PDS.

          cd 'your.aparfix.tpfdf'
          binary
          put PKyyyyy/bin PKyyyyy
          quit


TPF Operations Server APARs

TPF Operations Server APAR packages are available as individual zip files which can be unpacked on a Windows platform. Each APAR zip file contains a copy of the APAR description with instructions on how to install the APAR. Periodically, a group of APARs is consolidated into a patch file which customers can obtain to install multiple APARs at once.