Previously, when running a guest that uses the HMC for communicating with the user, the text for each reply was always logged to the HMC as-is:
As this was deemed to be a security exposure when the reply contained sensitive information such as a user's password, a new SCP command prefix was created to prevent the echoing of the SCP reply. Beginning with this release of SDL Hercules Hyperion, the SCP reply prefix '\' (backslash) should be used for replies that you do not want to be written to the screen or logfile:08:35:19.528 Ubuntu 18.04.6 LTS s390x sclp_line0 08:35:19.638 s390x login: 08:56:15.453 HHC00160I SCP command: hercules 08:56:15.556 hercules 08:56:15.776 Password: 08:56:17.357 HHC00160I SCP command: hercules 08:56:18.367 Last login: Sat Jan 28 11:31:16 EST 2023 on sclp_line0 08:56:48.327 Welcome to Ubuntu 18.04.6 LTS (GNU/Linux 4.15.0-200-generic s390x)
As you can see, the reply is not echoed when the '\' (backslash) SCP reply prefix is used instead. Rather, the reply text is instead replaced with the message text "(suppressed)". This prevents the logging of sensitive information (such as passwords) to the screen and logfile.17:50:53.980 Ubuntu 18.04.6 LTS s390x sclp_line0 17:50:54.089 s390x login: 17:51:39.281 HHC00160I SCP command: hercules 17:51:39.389 hercules 17:51:39.689 Password: 17:51:43.265 HHC00160I SCP command: (suppressed) 17:52:29.270 Welcome to Ubuntu 18.04.6 LTS (GNU/Linux 4.15.0-200-generic s390x)
Previously, when running a s390x Linux guest, boot progress messages would appear garbled (badly formatted) when logged to the HMC:
Beginning with this release of SDL Hercules Hyperion, efforts are made to more properly parse the screen messages that Linux is issuing so that they can appear on the HMC in a much more readable format:20:03:20.579 Welcome to 1mUbuntu 18.04.6 LTS 0m| 20:03:20.580 20:03:20.580 77.148546! systemd 1!: Set hostname to. 20:03:22.494 HHC00004I Control program identification: type LINUX, name UBUNTU, sysplex 18 04 6, level 40F00 20:03:26.059 0;32m OK 0m! Reached target User and Group Name Lookups. 20:03:26.059 0;32m OK 0m! Created slice System Slice. 20:03:26.060 0;32m OK 0m! Listening on fsck to fsckd communication Socket. 20:03:26.060 0;32m OK 0m! Listening on Journal Socket. 20:03:26.060 0;32m OK 0m! Listening on Syslog Socket. 20:03:26.061 82.621379! random: systemd: uninitialized urandom read (16 bytes read) 20:03:26.061 82.621788! systemd 1!: Reached target User and Group Name Lookups.
Please be aware that in order for s390x Linux's messages to be properly parsed by Hercules you should use a Hercules15:17:37.374 Welcome to Ubuntu 18.04 LTS! 15:17:37.374 15:17:37.393 [ 74.041881] systemd: Set hostname to. 15:17:38.286 HHC00004I Control program identification: type LINUX, name UBUNTU, sysplex 18 04, level 40F00 15:17:40.303 [ 76.951718] random: crng init done 15:17:42.653 [ OK ] Started Forward Password Requests to Wall Directory Watch. 15:17:42.653 [ OK ] Created slice User and Session Slice. 15:17:42.654 [ OK ] Created slice System Slice. 15:17:42.654 [ OK ] Listening on fsck to fsckd communication Socket. 15:17:42.655 [ OK ] Listening on Journal Socket (/dev/log). 15:17:42.655 [ OK ] Listening on Journal Audit Socket. 15:17:42.655 [ OK ] Listening on LVM2 metadata daemon socket. 15:17:42.656 [ OK ] Created slice system-serial\x2dgetty.slice. 15:17:42.656 [ OK ] Listening on udev Kernel Socket. 15:17:42.656 [ OK ] Listening on Syslog Socket. 15:17:42.656 [ OK ] Reached target User and Group Name Lookups.
CODEPAGE
mapping that matches the same code page that your s390x Linux guest actually uses.
For (most?) s390x Linux guests this is usually code page "819/500".
(Other code pages may also work, but probably not as well as 819/500.)
Beginning with this release of SDL Hercules Hyperion, you can now have debug tracing (e.g. instruction and/or CCW tracing) written to an external file instead of it always being written to the HMC panel (which is incredibly inefficient). Tracing to an external trace file greatly decreases the Hercules overhead involved with debug tracing, thus allowing your guest to run significantly faster than otherwise when debug tracing is active.
Additionally, a much finer grained selection of information can be printed from the resulting trace file by means of the new 'tfprint' utility. Please refer to the 'tfprint' utility's help display for information regarding the many options available for choosing what trace information you wish to print. Enter "help tf" for more information regarding the "Trace to File" command itself.
SDL Hercules Hyperion now supports socket punch devices in a manner identical to its current support for socket printer devices, thereby allowing remote spooling of card punch data. Simply specify the 'sockdev' option for your Card punch device in exactly the same manner as you would for a socket printer device.
A potentially serious long standing bug in Hercules's compressed CKD (i.e. CCKD/CFBA/CCKD64/CFBA64) dasd emulation logic was recently discovered and quickly fixed. Due to the natue of the bug, Locate Record channel commands (opcodes X'47' or X'E7') would incorrectly prematurely fail in an unusual way causing unexpected guest behavior (e.g. the guest might end up looping or hanging waiting for the I/O to complete for example. See e.g. GitHub Issue #464).
Please note however that due to the nature of this bug, damage to your emulated dasd images could never occur. The bug would only cause the guest's I/O to unexpectedly fail or never finish at all. The integrity of the actual Hercules emulated dasd image file would remain intact. Also note that this bug existed only in the "CCKD" compressed dasd image handling logic, and not in the regular non-compressed "CKD" dasd handling logic. Regular uncompressed CKD dasd images were unaffected by this bug and Locate Record would behave correctly.
During instruction stepping/tracing ('t+' or 's+' Hercules commands),
when the address of the instruction's operand is a virtual address, the corresponding translated
real address is now also shown when possible
(e.g: HHC02326I CP00: V:20012CD8:R:7FFF3CD8:K:06=00000000 00000000 ........)
Hercules now reports its version using the format "V.M.R.nnnnn-SDL-ghhhhhhhh" where V.M.R is the major Version, Modification and Revision level (e.g. 4.4.1), nnnnn is an ever increasing internal build number (to ensure strictly sequentally ascending version strings), and hhhhhhhh is the first eight characters of the hash of the most recent commit made to the git source code repository (which identifies exactly which fixes/changes are present in your build and which are not, which is vitally important for bug reporting and problem analysis purposes).
For between release development builds the format is "V.M.R.nnnnn-SDL-DEV-ghhhhhhhh" which is identical to the new format but with the extra "DEV-" string to indicate the build is not an official release build but rather is a privately/manually built intermediate development build.
-o, --output and --logfile command-line options
New command-line options are now available to make the creation of a log file much easier and less error prone, which is important for bug reporting and problem analysis purposes. Refer to the "Command line arguments" section of the "Installation and Operation" web page for more information.
During development of Hercules Version 4.4 it was noticed Hercules would sometimes experience unreliable operation under certain conditions when built with certain versions of the gcc compiler. The unreliability seems to most typically occur (but may not be limited to) when a z/Architecture or ESA/390 guest is IPLed under a z/Architecture or ESA/390 VM (i.e. z/VM or VM/ESA). We believe the problem has since been resolved, but are not 100% certain about that.
Therefore, if you plan to run an SIE operating system (e.g. VM/ESA or z/VM) as your primary guest under Hercules, we recommend, purely as a precaution, that you build Hercules using the clang compiler and not gcc.
Please refer to GitHub Issue #437 for the latest status regarding this particular issue.
Significant efforts have been made to make building Hercules on non-Windows systems (e.g. Linux) easier and more reliable. Please refer to the README for Bill Lewis's highly recommended "Hercules-Helper" repository for more information.
The PFPO (Perform Floating Point Operation) instruction contains a (minor?) unresolved bug in its handling of the EXTENDED DFP TO LONG HFP operation that seems to only occur when the value being converted has more than 31 digits in its fractional component. Refer to GitHub Issue #407 for more information.
This bug is currently the only one known to exist in the PFPO instruction, but given that a comprehensive test suite still does not exist yet for the PFPO instruction (such a test suite is still in development), other undiscovered bugs may also exist. Until such time as a thorough PFPO test suite can be completed, the PFPO instruction should be considered experimental.
While still considered experimental (since it is not yet 100% complete), the Hercules implementation of the z/Architecture "Transactional-Execution Facility" feature is largely complete and considered to be reliable. There are still a few items that have not been implemented yet (such as tracing (i.e. PER) support), but on the whole, extensive testing has demonstrated it to be reliable enough for most everyday purposes.
Also note that while it is still considered experimental and incomplete, any
"FACILITY ENABLE 073_TRANSACT_EXEC" and/or
"FACILITY ENABLE 050_CONSTR_TRANSACT" configuration file statements
that you may currently have in your Hercules configuration file, are now no longer
needed. Beginning with Hercules version 4.4, both facilities are now enabled by default.
Significant enhancements have been made to LCS device functionality to support the SNA protocol. Preliminary documentation regarding its configuration and use can be found in our README.SNA readme document.
The CTCE device driver has been enhanced and improved, including changes to the syntax of the device statement parameters, Please refer to the Hercules Networking README as well as to the actual Hercules Configuration File web page itself for more information.
A new MSGCOLOR option has been added to the existing
PANOPT
configuration file statement
that will colorize non-informational Error, Warning and Debug panel messages to make them more
noticable to the user (less likely to be missed). Users are strongly encouraged to enable
this new option in their configuration files.
Previously, Hercules had an internal "watchdog" monitoring thread that would monitor all running CPUs to make sure they were functioning properly, and if any of them were determined to be malfunctioning, a Machine Check Interruption for the malfunctioning CPU would be presented to the guest in a vain attempt to allow it to continue running. This functionality was removed in version 4.2 however, due to it masking/hiding otherwise serious bugs in the Hercules emulator.
Version 4.3 now re-introduces watchdog monitoring for malfunctioning (hung) CPUs and other serious internal emulator errors (such as deadlock detection), but does not otherwise notify the guest of the problem (i.e. no Machine Check Interruption is presented), nor is any attempt made to allow the emulator to continue. Instead, the problem is reported to the Hercules user via fatal emulator error messages HHC00822S and/or HHC90024S, and the emulator is then immediately terminated with a crash dump to be forwarded to the Hercules development team for analysis.
For more information regarding crash dumps ("core files" on Linux), such as how to enable their creation and how to analyze them when they occur, please refer to GitHub Issue #199: "How to analyze a Linux "core" file (i.e. crash dump)".
This release of SDL Hyperion includes preliminary experimental support for the z/Architecture "Transactional-Execution Facility" (bits 73 and 50).
This support is known to be incomplete and not 100% functional yet (it still has some problems, especially under SIE), so it is still considered to be highly experimental at this stage. It seems to work well enough to allow certain guest operating systems requiring it (such as z/OS 2.4 for example) to successfully IPL and run though.
Because it is still considered to be experimental and under active development
however, it is not enabled by default. Therefore users wishing to take advantage
of it must first manually enable it in their Hercules configuration file via the
following two FACILITY
configuration file statements:
FACILITY ENABLE 073_TRANSACT_EXEC
FACILITY ENABLE 050_CONSTR_TRANSACT
Only when the feature has been fully completed and thoroughly tested however, will it then be enabled by default. For the time being however, since it is still under active development and considered to be experimental, it must be manually enabled as described above.
If you decide to use it, Bug Reports would be greatly appreciated by creating a NEW GitHub Issue describing the problem along with all supporting documentation. Thank you!
Version 4.2.1 of SDL Hercules Hyperion introduces support for assigning serial numbers to CKD dasd devices. This is usually only a concern for guests which care about the serial number field in the device's Node Element Descriptor (NED) information returned by the RCD (Read Configuration Data) CCW, such as z/VM SSI (Single System Image) configurations. For most other guests the change is transparent.
The dasdinit utility now assigns a randomly generated serial
number for all CKD dasd images. For existing CKD dasds created with earlier
(older) versions of Hercules without serial number support, the CKD dasd's
serial number will always be zero.
However, if an all zero serial number is detected during Hercules's CKD device initialization (indicating the dasd does not have a permanent serial number), a random serial number value is generated and used instead. Because of this, all older dasd images (created with an older version of Hercules and thus always having an all zero serial number) will always appear to have a completely different (unstable ever changing) serial number each time Hercules is started.
To assign a permanent (stable never changing) serial number to a CKD dasd
image (including those created with older versions of Hercules) a new
dasd utility has been created called dasdser. Start
dasdser without any arguments for usage information.
The serial number that Hercules actually uses for a given CKD dasd
(whether internally generated or obtained from the actual dasd image file)
can also be overridden via a new
ser=nnnnnnnnnnnn
device configuration file statement option. This new option is supported
only for CKD dasd devices and not FBA dasd devices since FBA devices
don't support the RCD (Read Configuration Data) CCW opcode and thus do not
have any serial number.
Version 4.2 of SDL Hercules Hyperion introduces support for very large Compressed CKD (CCKD) dasd image files, called CCKD64, which can be much larger than 4GB in size. More information about CCKD64 can be found in the README.CCKD64 document in the source code distribution. Detailed CCKD64 file format information can be found on the "Compressed Dasd Emulation" web page.
The original designed "SIGABEND" handler on Linux builds that intercepted Hercules abends (e.g. SIGSEGV, i.e. Hercules crashes, etc) and presented a Machine Check Interruption to the guest in an attempt to allow Hercules and the guest to continue running (in the hope that the guest would be able to either recover from the Machine Check or be able to gracefully shutdown) has been removed since it served no valid purpose other than to hide/mask otherwise very serious bugs in the Hercules emulator making it nearly impossible to find or fix them.
Now, if Hercules crashes (due to a SIGSEGV, etc), the guest is no longer notified of the event (a Machine Check Interruption does not occur) and Hercules makes no attempt to recover or otherwise continue. Instead, it attempts to issue a message to the Hercules user informing them of the crash and then requests that the host perform default handling of the event (which is usually to create a "core" dump file which can then be analyzed by gdb or by some other means to determine the cause of the crash).
For more information regarding Linux core files, such as how to enable their creation and how to analyze them when they occur, please refer to the following GitHub Issue: "How to analyze a Linux "core" file (i.e. crash dump)" (#199).
For those who want integrated Regina rexx support in Hercules, you will
now have to explicitly request such support on your ./configure
command by specifying the '--enable-regina-rexx' option. For
everyone else, integrated support for Regina rexx will no longer be
automatically provided by Hercules, even if it is installed in their system.
In other words, you can still have Regina rexx installed on your system and use it as you normally would, but simply having it installed on your system will now no longer cause Hercules to also provide integrated support for it.
For more information refer to the README.REXX document in the source code distribution.
Thanks to hard work of one of our developers, Hercules now has a complete implementation of the PFPO (Perform Floating Point Operation) instruction. (previously we had no implementation at all!)
Efforts are currently under way to develop a formal runtest PFPO instruction verification test for the current implementation to prove that it fully complies with the published architecture, but the effort has not been completed yet. Nevertheless, the current implementation is believed to fully comply with the Principles of Operation even though we can't yet prove that.
The VMFPLC2 tape utility has been greatly enhanced and now supports the scan and load functions, making it much easier to transfer files between the host system and z/VM or VM/370 systems. Refer to the README.VMFPLC2 document for usage details.
It was a mistake to enable the System/370 and System/390 Vector Facility set of instructions in release 4.1. Hercules's implementation of the facility only implements 26 of the required 180 instructions the facility provides and thus is so incredibly incomplete as to be virtually unusable.
Given that virtually no one has ever complained about its unavailability beforehand (and the interference it causes internally with the desired proper functioning of the Hercules S/370 Extension pseudo-facility), it has once again been disabled.
This is very likely to be a permanent setting given the extreme unlikelihood of the facility ever being fully coded due to its unpopularity and the extreme challenges posed by trying to code all of the missing instructions for a fully compliant facility.
This instruction was developed by Jason Winter to allow the guest OS access to the IP stack of the host OS. It is of particular importance to older operating systems that do not support their own IP stack. It provides full IPv4 networking capabilities to those systems.
Note that the TCPIP instruction doesn't fully comply with IBM's S/3x0, ESA/390 or z/Architecture specifications. For this reason it is not enabled by default. File README.TCPIP provides the information needed to enable the instruction.
The b (breakpoint) and s (stepping) commands
now support an optional asid (Address Space
IDentifier) parameter. If specified, the breakpoint or stepping
will only occur if the instructions being executed are both within
the specified address range and are being executed by the
specified address space. If specified, the breakpoint or stepping will
not occur for any other address spaces even if the instructions are
within the specified address range. If the optional ASID is not specified,
breakpoints or steppings occurs normally as before (i.e. if the instruction
being executed is within the specified range).
For more information refer to the help command display
for the b and/or s commands (i.e. "help b", "help s", etc).
Release 4.2 introduces a new configuration file option called
PANOPT
that controls how emulated device files are displayed on the screen. The
NAMEONLY option will display only the emulated device file's
base filename, whereas FULLPATH displays the device file's
full path like before.
Release 4.2 introduces a new Diagnose F09 instruction that is essentially an enhanced version of the existing Diagnose F08 instruction providing access to Hercules's internal instruction counter. For detailed usage information please refer to the final comment in GitHub Issue #211: "Diagnose F08 Improvement".
Select source code and associated functionality has been moved out of the Hercules repository and into separately maintained External Package repositories. Refer to the README.EXTPKG document for more information.
The new minimum supported Windows platform is now Windows Vista. All Hercules users still running older versions of Windows should upgrade to at least Windows Vista or greater, with Windows 7 being preferred.
Previously, Hercules supported various commands/statements that allowed the tweaking of Hercules's various internal thread priorities (HERCPRIO, CPUPRIO, TODPRIO, etc). This ability has been removed. Such statements in your configuration file will now cause a "deprecated and ignored" warning message instead. You should remove all such statements from your configuration file.
The FACILITY command was completely
redesigned and rewritten in order to more properly support and enforce the concept
of prerequisite facilities (i.e. a facility presuming/requiring some other facility).
As part of this rewrite the ARCHLVL command functionality to enable/disable/query
facilities was moved into a new separate FACILITY command such that the ARCHLVL command
now only deals with setting/querying the architecture mode. To enable/disable/query
facilities, use the new FACILITY command.
Additionally, the facility names used in the FACILITY command to
enable, disable or query a facility have also been changed. All defined facilities
now begin with a 3-digit facility bit number in addition to their abbreviated
facility name. This was done to ensure the correct facility was being enabled or
disabled due to the similarity of some facility names. For example, the previously
named "ASN_LX_REUSE" facility is now named "006_ASN_LX_REUSE". Similarly, the
previously named "PFPO" facility is now named "044_PFPO".
For a detailed list of the new abbreviated (short) facility names,
use the "FACILITY QUERY ..." command.
The "ASN and LX Reuse" facility (STFL bit 6) is now enabled by default for the z/Architecture mode, as it should have been all along. It is still disabled by default for the ESA/390 architecture mode however.
In earlier versions of Hercules it defaulted to disabled requiring those
running z/Architecture guest operating systems to have to manually add
an FACILITY ENABLE 006_ASN_LX_REUSE statement to their
configuration file. Such statements are now no longer necessary.
The combination of NUMCPU and MAXCPU controls the behavior of how many CPU engines will be configured online upon startup and how many can be configured online later. In previous versions this was controlled via the NUMCPU statement and the compile-time constant 'MAX_CPU_ENGS'.
For compatibility with previous versions of Hercules, if MAXCPU is not specified its value defaults to NUMCPU. If neither is specified it defaults to 1.
Any httproot or httpport statements are now
rejected as invalid statements. All users still using either statement
format in their configuration files must now replace all occurrences
with http root ... and http port ... instead.
The default implementation for the CMPSC Compression Call instruction is now the new cmpsc_2012 implementation. The previous legacy implementation no longer exists. Refer to the README.CMPSC document of the source code distribution for more information.
The ARCHLVL
statement (previously called ARCHMODE for example, defines
the initial architectural mode of the system) should generally precede
(come before) any FACILITY statements that enable or disable
a given architectural feature.
For example: on certain IBM operating systems the z/Architecture "PFPO
Facility" must be enabled or the system will not IPL. While at the time
of this writing Hercules does not currently support the Perform Floating-Point
Operation (PFPO) Facility, the STFLE facility bit can nonetheless be
forcibly enabled anyway via the
FACILITY ENABLE BIT44
configuration file statement. Since the facility is a z/Architecture-only
feature however, your initial architecture must either be set to "z/Arch"
beforehand, or else you need to specify the optional
[archlvl] parameter on your FACILITY ENABLE BIT44 statement.
This usually means your "ARCHLVL z/Arch" statement must come
before your "FACILITY ENABLE BIT44" statement in your
configuration file. Otherwise the PFPO Facility would not get enabled.
With the introduction of limited automatic LPAR-mode/BASIC-mode switching
(see next item below), LPARNUM
statements should follow ARCHLVL
statements if LPAR mode is truly desired for ARCHLVL S/370
(which is usually not the case).
The ARCHLVL and LPARNUM statements are currently
the only known configuration file statements whose sequence matters. The
order of all other configuration file statements should not currently matter.
This may change in the future however, so be sure to always carefully read
through the RELEASE NOTES (this document) with each new update to Hercules.
When ARCHLVL S/370 is set,
LPARNUM is now automatically
changed to BASIC (which also changes
CPUIDFMT to BASIC as well),
and when ARCHLVL z/Arch is set, it is then changed back to the
default LPARNUM 1 again (which also automatically changes the
CPUIDFMT back to 0 as well), but only when needed.
That is to say, the automatic switching to and from LPARNUM BASIC
(and CPUIDFMT BASIC) and LPARNUM 1
(and CPUIDFMT 0) only occurs when needed (and only when switching
between S/370 and z/Arch). If LPARNUM (and
CPUIDFMT) are already set to the expected values they will not
be changed. When ARCHLVL ESA/390 is set however, LPARNUM
and CPUIDFMT are never changed from their current values.
This new behavior was introduced to eliminate the "surprise" factor that would
otherwise occur when, if the LPARNUM remained set to 1
(or some other number), not doing so causes the STIDP (Store CPU ID) instruction
to be stored in an otherwise unexpected format. In most all situations when
a Hercules user sets ARCHLVL S/370, they are truly expecting the
LPARNUM and CPUIDFMT to be set to BASIC
such that the STIDP instruction then stores the CPU ID in the expected format.
For the unusual case where users actually do want to run a System/370
Operating System within an LPAR, you will need to manually re-set your
LPARNUM and CPUIDFMT values back to their numeric
values after setting ARCHLVL S/370. That is to say,
one should always follow their ARCHLVL S/370 statement with a
LPARNUM n|nn statement (and CPUIDFMT statement
too if needed) if LPAR mode for S/370 is truly desired.
Hercules 4.x Hyperion's channel subsystem implementation now more closely adheres to published and unpublished architectural specifications. It does not precisely adhere to the complete specification but it does adhere much more closely than ever before (and definitely more closely than the legacy implementation still used in the older spinhawk 3.xx series of Hercules).
If you experience any anomalies in the behavior of your guest operating system, verify that you are using architecurally valid/correct configuration settings that your guest operating system expects.
Pay particular attention to your ARCHLVL, LPARNUM, CPUIDFMT, CPUMODEL, CPUSERIAL, CPUVERID, MODEL, PLANT and MANUFACTURER values since they typically have a direct impact on how certain guest operating systems behave.
Most configuration file statements are now available as and processed as panel commands. Most valid configuration file statements may now also be entered as a panel command. Most panel commands are now also allowed as configuration file statements. See the documentation for full details.
Hyperion 4.x now provides a mostly complete ECPSVM CP/VM Assist implementation. While still not 100% complete, support for many new assists has been added and several bugs have been fixed. Refer to the README.ECPSVM document for more detailed information.
Rexx support was first added to Hercules by Jan Jaeger in 2010 and has been gradually enhanced over the years by both Enrico Sorichetti and Jan Jaeger as well as a few other people too.
If you have Rexx installed on your
host and Hercules is built with the Rexx build option (the default for
the SDL version of Hyperion), then Rexx scripts
can be run directly from within the Hercules environment (i.e. directly
from the Hercules HMC command line via the Hercules exec command).
Rexx scripts, when run within Hercules via the 'exec' command, can Address the HERCULES enviroment allowing you to issue Hercules commands and retrieve the results via the builtin 'AWSCMD' Rexx function call.
For more information please refer to the new Hercules Integrated Rexx Support web page.
The defaults for the SHCMDOPT have changed.
The old defaults were ENABLE DIAG8.
The new defaults are DISABLE NODIAG8.
This was done for security reasons.
The ENABLE and DISABLE option of the
SHCMDOPT command now also controls
the Rexx exec command as well as the sh command.
That is to say, the Rexx exec command is now properly treated
as a host shell command. Thus, enabling or disabling the execution of host
shell commands via the SHCMDOPT option
now also enables or disables execution of the exec command as well.
This was done for security reasons.
The behavior of the DIAG8CMD command's
ECHO and NOECHO options has changed. Previosuly
the echo/noecho option had a a direct impact on the data that was placed into
the Diagnose 8 instruction's response buffer. This has been fixed so that now
it does not.
Previously, specifying the ECHO option caused audit trail messages HHC01950I and HHC01603I to also appear in the instruction's response buffer, whereas now only the command's actual output is placed into the response buffer. The ECHO option now properly controls only the issuing of the HHC01950I and HHC01603I audit messages to both the panel and hardcopy logfile, but does not otherwise impact in any way what is placed into the instruction's response buffer, as was the original intent for this option.
Support for 3088, CTCI-W32, CTCT and VMNET devices has been dropped. 3088 Channel-To-Channel Adapter support (which was the original intent of CTCT) has instead been replaced with Peter Jansen's vastly improved Enhanced Channel-To-Channel (CTCE) device support, which provides true Extended mode Channel-To-Channel Adapter capabilities, allowing you to use GRS to connect multiple Hercules instances together in a type of Sysplex.
Previous versions of Hercules supported a "CAPPING" configuration statement designed to purposely reduce performance to ensure that your MIPS rate never exceeded the specified value. This functionality has now been removed.
A new NETDEV configuration file statement is now supported
to allow you to specify which host network adapter should be used by default
for Hercules communications devices. It only represents a default and can be
overridden on the device statements themselves. Refer to the documentation
for the "NETDEV" statement
for more information.
A new "t+-" (tee-plus-minus) Automatic Tracing command has been
created that may prove to be helpful in capturing guest issues that occur
very early in the guest's IPL sequence. Refer to the "help t+-"
help display for more information.
The AUTOINIT option
controls whether device files for emulated tape volumes should be
automatically created or not if they do not exist. The previous default
was OFF. The new default is ON.
Hercules's Diagnose X'F14' functionality was designed to allow a Hercules guest to call into any host system DLL (Dynamic Link Library). Since its use (dubius to begin with) was determined to be a potential security risk (and the code didn't appear to be used anywhere anyway), support for it was removed.
Hercules has always supported the System/370 and System/390 Vector Facility set of instructions but support for them was never enabled by default. This has now been corrected.
A new Hercules S/370 Instruction Extension Facility has been implemented to replace the functionality previously provided by the "s37x.dll" dynamically loadable module. Please refer to the README.S37X document for more information.
Starting with Hercules version 3.06 a new AUTOMOUNT option is available that allows guest operating systems to directly mount, unmount and query tape device filenames for themselves, without any intervention on the part of the Hercules operator.
Automount support is enabled via the AUTOMOUNT configuration file statement.
An example guest automount program for VSE called "TMOUNT" is provided in the util subdirectory of the Hercules source code distribution.
Briefly, the 0x4B (Set Diagnose) CCW is used to mount (or unmount) a file onto a tape drive, and the 0xE4 (Sense Id) CCW opcode is used to query the name of the currently mounted file.
For mounts, the 0x4B CCW specifies the filename of the file to be mounted onto the drive. The file MUST reside in the specified AUTOMOUNT directory or the automount request will be rejected. To unmount the currently mounted file, simply do a mount of the special filename "OFFLINE".
To query the name of the currently mounted file, the 0xE4 CCW is used. Note however that the 0xE4 (Sense Id) CCW opcode cannot be used by itself since the drive may also already natively support the Sense Id CCW opcode. Instead, it must be preceded by (command-chained from) a 0x4B CCW with a data transfer length of one byte. The following 0xE4 command is the one that then specifies the i/o buffer and buffer length of where the query function is to place the device's currently mounted host filename.
In summary:
MOUNT: X'4B', filename, X'20', length
UNMOUNT: (same thing but use filename "OFFLINE" instead)
QUERY: X'4B', buffer, X'60', 1
X'E4', buffer, X'20', buffersize
Again please refer to the provided TMOUNT program for a simple example of how automount support might be implmented on a guest operating system.
The 'conspawn' utility used to process 'sh' commands now recognizes a specially designed keyword "startgui" to accomodate automatic starting of Windows GUI applications via the .RC file or panel command-line.
If the first word following 'sh' is "startgui", then the "ShellExecute" API is used to start the requested program rather than the 'system()' API as otherwise.
The "startgui" keyword must always be used to start any Windows program that is not a command-line program. Hercules, itself being a command-line program, monitors the 'stderr' and 'stdout' pipes so it can log messages received from either pipe directly to the Hercules console log. Programs such as notepad however, because they are not command-line programs, do not use stdout/stderr thus causing Hercules to hang if "start" is used instead.
This rule applies regardless of how Hercules itself is started (i.e. via HercGUI or directly via the command-line) and regardless of whether the "start" command is wrapped in a batch file or not. That is to say, using the Hercules command "sh batchfile foobar" to start your batch file which then does "start notepad %1" still causes Hercules to hang until notepad first exits. Instead, you should ask Hercules to "sh startgui batchfile", and let the batchfile start notepad however it wants.
Real SCSI tape drives used with Hercules must provide a certain minimum set is "IBM compatible" support in their SCSI command set/behavior in order to work properly with Hercules. Furthermore, the Hercules device-type used on your device statement in your Hercules configuration file should match the the level of support/behavior that they provide.
For example, all SCSI tape drives used with Hercules must provide the ability to set variable-length blocks as well as long erase-gaps (long erase-gaps allows new data to be appended to the end of existing data without having to write a tape-mark to separate the new data from the old existing data first).
Another example would be using a model of SCSI tape drive that happens to report physical block-id values in a format different from the way real IBM mainframe tape drives report them. 3480/3490 tape drives for example report their block-ids (used in Read Block Id and Locate CCWs) in a very specific format wherein bits 1-7 of the high-order byte of the reported 4-byte block- id indicates the tape's physical "segment" location of where the lower 22- bit block number is physically located on the tape. (The block-id segment is used to allow the tape drive to quickly position itself to the approximate location where the desired block acually resides on the tape and thus allows high-speed positioning for the Locate CCW).
If the model of SCSI tape drive you are actually using with Hercules does not use this same block-id format however, then it cannot be used with Hercules as a 3480 or 3490 model tape drive with specially defined options.
If the SCSI tape drive you are using reports its block-ids using a 32-bit block-id value (the same way a 3590 model tape drive does), then similarly, it should be defined to Hercules as a model 3590 device-type as well (since that is how it is behaving with respect the format of the returned blockid values). It you wish to define it in Hercules as a model 3480 or 3490, then you will need to use specially defined options before it will work properly as the model drive you wish it to emulate.
With all that being said, it should be noted that PARTIAL support for 3590 device emulation is possible with judicious use the aforementioned special options, but full/complete 3590 support is unlikely due to lack of publicly available documentation. Details regarding 3590 CCW handling is restricted (confidential) IBM proprietary information, and is not normally available outside of IBM. Not long ago IBM was required by US law to publish such information, but unfortunately for Hercules, such is no longer the case.
For further information regarding use of SCSI attached tape drives with Hercules and their associated specially defined options, please refer to the section on SCSI tape drives in the Hercules's Device Configuration documentation.
In order to ensure proper functioning of the TOD clock with older versions of guest operating systems, the default values of Hercules's internal thread priorities for the Windows version of Hercules were changed to be identical to those used by all other supported platforms. Originally, the default thread priority values for the Windows version of Hercules were:
*** 3.04 (and prior) Default Priorities ***
Thread Priority Meaning
------- -------- ------------------------
HERCPRIO 0 Normal Process priority
DEVPRIO -8 Above Normal Thread priority
TODPRIO 0 Normal Thread priority
CPUPRIO 0 Normal Thread priority
which caused acceptable performance/functioning on most, but not all, guest operating systems. Beginning with version 3.05 however, the prioriries now default to:
*** 3.05 (and later) Default Priorities ***
Thread Priority Meaning
------- -------- ------------------------
HERCPRIO 0 Normal Process priority
TODPRIO -20 Time Critical Thread priority
DEVPRIO 8 Below Normal Thread priority
CPUPRIO 15 Lowest Thread priority
which may on more modern guest operating systems (which handle the TOD clock differently than do older less sophticated versions) cause a slight decrease in overall performance. If such is the case, the original default priorities (and thus the original behavior) can be obtained via addition of appropriate HERCPRIO, TODPRIO, DEVPRIO and CPUPRIO control file statements with values identical to the original version 3.04 default values.
Additional configuration file usability enhancements have been implemented in the form of a new 'INCLUDE' (and associated 'IGNORE') statement, allowing configuration files to "include" statements from a different named file.
Additonally, a new "enhanced" symbolic substitution syntax is now also supported. Refer to the Hercules "Configuration File" documentation for further information and details.
A rather nifty "Automatic Operator" facility has also been implemented in the current release as well. While not exactly a "configuration file usability enhancement", it is nevertheless something we hope might prove to be more useful/helpful to our many users. See the "README.HAO" document for more information.
Release date: 20 December 2005
The new integrated console printer-keyboard is emulated on the hercules console. Commands are sent to the console by means of a command character. (default '/', thus a logon command is sent by /logon)
Starting from release 3.03 the glibcrypt library is no longer needed.
Release date: 11 December 2004
This is a new feature of z/Architecture which can cause problems with
certain versions of operating systems running in ARCHLVL=2 mode without
the so-called "Driver 55" fixes. To avoid such problems, specify
ASN_AND_LX_REUSE DISABLE in the configuration file.
Release date: 30 November 2003
An error in the 3.00 configuration script caused many users to have to override the default modules and HTTP documents directory in the Hercules configuration file, or by setting an environment variable. This error has been corrected. Hercules also now reports the actual directory that it uses by default for these files at startup time. If you specified the MODPATH or HTTPROOT configuration file statements because you encountered problems, you should examine the messages printed at startup to see if the default directories are now correct, and remove the statements if so.
In general, MODPATH and HTTPROOT should not have to be specified except in unusual circumstances.
In conjunction with the fix above, the default directories of the Windows distributed binaries have been changed. The new directories are under C:\cygwin\usr\local (which is the same as /usr/local under the Cygwin environment). No action is needed unless you have specified the MODPATH or HTTPROOT configuration file entries; if so, see the previous note.
Support for z990 crypto instructions is conditional on the presence of the glibcrypt library. When Hercules is BUILT, the development files for glibcrypt should be available. When hercules is RUN, the runtime files for glibcrypt should be installed.
Depending on the level of glibcrypt used to *build* hercules, the associated level of glibcrypt should also be present on the target machine. On systems supporting shared library versioning, multiple levels of the glibcrypt runtime libraries can be installed simultaneously, ensuring availability of the z990 crypto instructions, regardless of the level of glibcrypt with which hercules was initially built.
CTC and LCS devices now MUST specify ALL addresses on the configuration statement. Previously (i.e. with version 3.00), only the first (even numbered) device needed to be defined and Hercules would automatically define the odd numbered device for you. Starting with Hercules version 3.01 however, you now need to define BOTH devices, just like you did with versions prior to 3.00. Once again, starting with version 3.01, you **MUST** define BOTH DEVICES.
Release date: 3 October 2003
Both of these will go away in a future release.
In addition, you must not define both even/odd CTCI device pairs in your configuration file. You should only define the first even numbered device. Hercules will automatically define the odd numbered device for you. If you define the odd numbered device by mistake, an open error will occur on that device. This is by design. See the README.NETWORKING document for further details.
Starting with version 3.00, Hercules now contains support for the dynamic loading of certain modules upon startup on Linux, Windows, and Mac OS X. This support should also work on any platform supported by GNU libtool. As a result of this new feature, Hercules itself now no longer consists of just the 'hercules.exe' module by itself, but rather consists of both the 'hercules.exe' program as well as whatever dynamic modules (DLLs) that accompany it.
As a result of this change, whenever you install a new version of Hercules, you must ensure that you ALSO install the accompanying new versions of the new dynamic modules as well. Attempting to use a version of Hercules with a dynamic module that was not specifically built for that version will cause loading of that dynamic module to fail.
You cannot mix versions of Hercules with differing versions of dynamically loaded modules.
Ensure that your library path (set by the environment variable
LD_LIBRARY_PATH) set correctly such that it includes the directory of your
Hercules dynamic load libraries. If you see message HHCCF042E,
which indicates that the system is unable to locate necessary loadable
modules, this is likely your problem. This should not be necessary if you
have a binary download, but if you're building from source, especially if
you've previously installed a binary package, this should be the first thing
you do.
Do not use ECPS:VM (See README.ECPSVM) in an AP or MP environment
in VM/370. If AP=YES or MP=YES is coded in DMKSYS
and the AP/MP control file is used to build the CP nucleus and
NUMCPU is set to more than 1 in the hercules.cnf
file, any of LOK001, LOK003 or other abends will occur. This occurs because
the Hercules ECPS:VM CP Assist implementation is not MP safe, and
particularly, attempts VM dispatching without holding necessary AP or MP
locks.
Due to the change in the "mainstor" memory allocation technique used by Hercules to
address a "double memory consumption" bug in Cygwin's malloc implementation,
some Windows Hercules users may experience an "out of memory" error whenever
Hercules is started with a large MAINSIZE configuration file
value:
HHCCF031S Cannot obtain nnnMB main storage
This error will most likely occur (if it does at all) for those users who
have manually adjusted their Cygwin heap_chunk_in_mb Windows
registry setting value (in order to allow them to specify a large
MAINSIZE value when running Hercules). If this problem does occur
(i.e. if you do happen to experience the above mentioned error with
this new release of Hercules), then either reduce your
heap_chunk_in_mb value (yes, that's correct: reduce it,
as in change it to a smaller value) or else remove it altogether (so
as to let it default).
A complete discussion of this issue is in the RELEASE.NOTES file in the source distribution.
There is a known problem with thread priority handling under Mac OS X. The OS X threading model is different from the one classically used in Linux. This causes failures to set the timer thread priority, and slow performance as all of Hercules is set to a low execution priority. This will be fixed in a future release. A workaround, for now, for slow performance is to add the statement
CPUPRIO 0
to your Hercules configuration file.
A possibly related problem is that Hercules fails in random ways when using the NPTL (new POSIX threads library) library under Linux. This library is used by default in Red Hat 9, and possibly other systems. If problems are encountered on a very recent version of Linux, try issuing the command
export LD_ASSUME_KERNEL=2.4.1
before starting Hercules.
If you have a question about Hercules, see the Hercules Frequently-Asked Questions page.