Removed the recently added SET CPU IDLE=SYSV since any SET CPU IDLE command will work for SysV. Existing configurations probably did a SET CPU IDLE=32V which works fine now.
Changed things based on the realization that ANY branch instruction which tests memory and then branches to itself is an idle loop if the branch is taken. This is without regard to address space, access mode, If interrupts are disabled, then it is a hung system and the simulation should halt.
Find the header and library paths with findpaths.
We need -lnetwork for sockets.
We do support pthread, it's in libroot.
We do support dlopen, it's in libroot.
- Added checks for dependencies before execution
- Added setup and update support for local Win32-Development-Binaries repo init
- Added support for operation under both a CMD.EXE invoked from either a git bash process or a natural windows CMD.EXE
- Fixed proper operation for first build operation after the local build repo has been cloned
- Make sure the build is of the latest code in the master branch from the origin repository
Writing to the TXCS register always updated the transmit enable mask, even when the "Write Mask Now" bit was not set. That bit needs to be set for the write to the register to actually update the transmit enable mask.
CPU Idle detection for this OS is now supported and the combination of SET CPU IDLE=ULTRIX-1.X and explicitly using a DEQNA device (SET XQ TYPE=DEQNA) will enable the automatic enabling of device interrupt generation.
Avoid permanent network network hangs when a network transport starts to return errors for various reasons.
These hangs also result in at least one thread in a CPU bound loop attempting to read on a pcap connection which will never be useful again.
When transmit or receive errors occur, the invoking thread sleeps for 1 second and subsequent operations proceed. When the total of read + write errors reaches a multiple of ETH_ERROR_REOPEN_THRESHOLD the current transport connection (usually pcap) is closed and re-opened after a ETH_ERROR_REOPEN_PAUSE second delay. If the open succeeds, we're happy. If it fails, then the link behaves as if it were never attached until some subsequent operator intervention is performed.
Activities which are known to induce this problems include (but are not limited to) when a simulator is running in a Windows Virtual Machine on a Hyper-V host system and the Hyper-V host system performs a SAVE and a subsequent Restart of the Guest Windows Virtual Machine. This operation can occur due to specific operator requests or merely when the Hyper-V host system reboots.
Migrated the XQ help to the hierarchical help model. This is a work in progress which will eventually merge much from 0readme_ethernet.txt into the device help.
