@(#)$Id: README,v 1.168 2000/11/29 00:08:22 build Exp $
Copyright (C) 1995 - 1999, Hewlett-Packard Company, all rights reserved.
Written by Eric Backus

			=====================
			= E1432 README FILE =
			=====================


	        ======================================
		=   A.06.04 IS AN OFFICIAL RELEASE   =
		======================================


This file serves two purposes.

This file is the README file in the wk_sta anonymous FTP directory for
the E1432/E1433/E1434 VXI module.  The files in this directory contain
various software releases for the E1432/E1433/E1434.  Each file is a
complete release, and will replace all files in the /opt/e1432 and
/opt/vxipnp/hpux/hpe1432 directories.

This file is also the README file for the E1432/E1433 Semaphore
distribution.

This file is divided into the following sections:

    Unpacking an Update or Swinstall File
    Revision Numbering
    Files Available on the FTP Server
    Printing the Function Reference Manual
    Things to Watch Out For
    Known Problems
    Changes Between A.06.00 and A.06.04
    Changes Between A.05.00 and A.06.00
    Changes Between A.04.01 and A.05.00
    Changes Between A.03.00 and A.04.01
    Changes Between A.02.00 and A.03.00
    Changes Between A.02.00 and X.02.00.01
    Changes Between A.01.03 and A.02.00
    Changes Between A.01.01 and A.01.03
    Changes Between A.01.00 and A.01.01
    Changes Between A.00.10 and A.01.00
    Changes Between A.00.00 and A.00.10


		Unpacking an Update or Swinstall File
		=====================================

The files in the FTP directory are compressed "update" files and
compressed "swinstall" files.  These files are the same file format
distributed on tape or CD with each E1432 module.  On an HP-UX 9.x
system, the "update" file is used.  On an HP-UX 10.x system, the
"swinstall" file is used.

The naming convention for HP-UX 9.x "update" files is E1432.X.XX.XX.Z,
where the "X.XX.XX" part is the version number.  The naming convention
for HP-UX 10.x "swinstall" files is E1432.X.XX.XX.depot.Z, where again
the "X.XX.XX" part is the version number.

If you are installing onto an HP-UX 9.x system, you would unpack one
of the update files by doing:

cd /					# Start at root directory
uncompress E1432.X.XX.XX.Z		# Uncompress the update file
/etc/update -s /E1432.X.XX.XX '*'	# Unpack the update file

If you are installing onto an HP-UX 10.x system, you would unpack one
of the swinstall files by doing:

cd /					# Start at root directory
uncompress E1432.X.XX.XX.depot.Z	# Uncompress the swinstall file
/usr/sbin/swinstall -s /E1432.X.XX.XX.depot '*'	# Unpack the swinstall file

Alternatively, if you are not on HP-UX, you can use tar to unpack the
update or swinstall file (both "update" and "swinstall" files are
really just tar files with a few extra files and directories added).
To make this work properly, you may have to move the resulting
directories to get them into the proper location.  In the end, this
may not be very useful because the many of the files are HP-UX
executables and libraries.


			  Revision Numbering
			  ==================

The revision string for each release is of the form "a.nn.nn", where
"a" is a letter and the "n"s are numbers.  If the letter is an "A",
the release is an official release.  If the letter is an "X", the
release is an interim release which may provide new features or fix
bugs from the previous official release.  For each release, the
numerical part of the revision string increases.

For example, the first official release was A.00.00.  The first
interim release after that was X.00.01.  The second interim release
was X.00.02.  Eventually, the features and fixes from the interim
releases get rolled into an official release, whose label starts with
"A".


		  Files Available on the FTP Server
		  =================================

E1432.A.00.00		Nov  9 1995	First official release
E1432.A.00.10		Jan 15 1996	Official release

E1432.A.01.00		Apr 17 1996	Official release
E1432.A.01.01		Apr 30 1996	Official release
E1432.A.01.03		May 14 1996	Official release

E1432.A.02.00		Nov 12 1996	Official release
E1432.X.02.00.01	Mar  4 1997	Patched A.02.00 release

E1432.A.03.00		Jul 18 1997	Official release

E1432.A.04.01		Sep  1 1998	Official release
E1432.X.04.03		Oct 29 1998	Test release (not well tested)
E1432.X.04.04		Nov 24 1998	Test release (not well tested)
E1432.X.04.06		Jan 18 1999	Test release (not well tested)
E1432.X.04.07		Jan 20 1999	Test release (not well tested)
E1432.X.04.08		Jan 20 1999	Test release (not well tested)
E1432.X.04.09		Feb  1 1999	Test release (not well tested)
E1432.X.04.11		Feb  3 1999	Test release (not well tested)
E1432.X.04.12		Feb  8 1999	Test release (not well tested)
E1432.X.04.13		Feb 12 1999	Test release (not well tested)
E1432.X.04.14		Feb 26 1999	Test release (not well tested)
E1432.X.04.16		Mar 10 1999	Test release (not well tested)
E1432.X.04.17		Mar 10 1999	Test release (not well tested)
E1432.X.04.18		Mar 15 1999	Test release (not well tested)
E1432.X.04.19		Mar 16 1999	Test release (not well tested)
E1432.X.04.20		Apr  6 1999	Test release
E1432.X.04.22		Apr 15 1999	Test release
E1432.A.05.00		Apr 21 1999	Official Release
E1432.X.05.01		May 27 1999	Test Build
E1432.X.05.03           Jul  2 1999     Test Build
E1432.A.06.00           Jul  8 1999     Official Release
E1432.X.06.01           Oct  6 1999     Interim Release (not well tested)
E1432.X.06.02		Mar  3 2000	Test Build
E1432.X.06.03		Mar  9 2000	Test Build
E1432.A.06.04		Mar 13 2000	Official Release
E1432.X.06.05		Oct  4 2000	Test Build including new latched overload features.

		Printing the Function Reference Manual
		======================================

See the README file in directory /opt/e1432/man.


		       Things to Watch Out For
		       =======================

When using the local bus for data transfer, doing a D32 bus cycle has
the potential to corrupt the local bus lines.  This is a consequence
of the pin-out of the P2 VXI connector.  Because of this, no D32 bus
cycles should be done while performing a local bus transfer.  The
E1432 interface library locks out D32 bus cycles if a local bus
measurement is running, but the user must ensure that access to other
VXI hardware does not use D32.

When using VXI Plug and Play on HP-UX with the E2091F, revision
G.01.02, of HP I/O Libraries for HP-UX, you must run visacfg
with the settings "Identify all known devices once now." or
"Manually configure devices on this interface." selected.
It must be rerun when the configuration changes.  The setting
"Automatically identify all known devices at run time."does not
work as expected.

Apparently, SICL can occasionally get confused by power-cycling a VXI
mainframe.  Doing "iclear vxi" fixes the problem.

When doing measurements with more than four E1432 modules on HP-UX,
you may run into a problem with memory allocation if the E1432 modules
are old.  E1432 modules with serial number prefixes below 3647 cause
this problem.  This is because these old E1432 modules require more
A24 memory, and the default configuration for the SICL I/O library on
HP-UX does not allocate enough A24 memory.  To fix this, two lines in
the /usr/pil/etc/iproc.cf file must be modified.  Change:

	boot ivxirm -p -I vxi

to:

	boot ivxirm -p -M -I vxi

and change:

	sysreset vxi ivxirm -t &

to:

	sysreset vxi ivxirm -M -t &

After making this change, the iproc daemon must be killed and
restarted.


			    Known Problems
			    ==============

Older E1432 (but not E1433 or E1434) modules have a bug in the VXI
interface which shows itself in one specific configuration.  The
configuration with the problem is a multi-mainframe system where an
embedded V/743 controls the first mainframe, and MXI cables connect
the first mainframe to subsequent mainframes.  Data read from E1432
modules in the non-root mainframes is occasionally corrupted.  HP LSID
has a new VXI interface ROM that can be installed in an old E1432 to
fix this problem.

Older E1432 modules with the optional tach board can occasionally lock
up.  This problem is also present on early E1433 modules (those
shipped before October 1996).  The problem is due to a hardware bug in
U740, which part of the interface between the tach board and the main
substrate board.  This problem is usually quite rare - only one out of
hundreds of millions of tach board accesses will result in the module
locking up.

When using the local bus for data transfer, the block available bit in
the E1432/E1433 status register can go away when there is still data to be
read from a scan.  This is due to a FIFO of 256 points that is not
accounted for when the block available bit is cleared.

When doing RPM arming or order tracking, the second tach channel
sporadically gets a bogus tach edge.  To work around this, always use
the first tach channel as the arming channel.

When using e1432_set_span, the resulting span doesn't always get
rounded the way you want.


		 Changes Between A.06.04 -> X.06.05
		 ===================================

Features added
--------------
Added 3 new functions to support latched overload feature.  These functions are e1432_check_latched_overloads, e1432_check_reset_latched_overloads, and e1432_clear_latched_overloads.  These functions are not officially supported until the next A.xx.xx release. (YujiK)


		 Changes Between A.06.00 -> A.06.04
		 ===================================

Features added
--------------
Changed e1432_reset_measure to allow multiple source modules to ramp down
  at the same time rather than one after another(DonM).

Bugs Fixed
----------

Fixed trigger jitter from one start to the next when source trigger from
  an E1434(DonM).
  
Fixed phase mismatch problem between E1433A and E1433B by adding 1 sample
SYNC delay for the E1433B(DonH).

Fixed tach overflow error reported by outside customers(Yuji)


Other Changes
-------------

Documentation clarification for encryption on 1D1 option(DonH)

Various documentation changes, including this document and changes in the doc
directory to describe how to do a build(DaveN) 


		 Changes Between A.05.00 and A.06.00
		 ===================================

Features added
--------------

Full support of E1433B.
Added vee-examples/e1562eoct.vee.


Bugs Fixed
----------

Fixed interrupt blocking.
Tach holdoff of 0 is now set to TACH_TACH_HOLDOFF_MIN to agree with help text.
Minor improvements, clarifications to examples, pnp-examples, test programs.
Man page, helptext updates.
Modified octave dither generator to have stable amplitude.
Fixed int_meas timeout on low spans bug.
Fixed problem of very infrequent octave garbage data.


		 Changes Between A.04.01 and A.05.00
		 ===================================

Features added
--------------

Updated C-library (non Plug&Play API) to work with VISA.
Updated e1432_block_available manual page.
Changed thruput/eavesdrop with RPM arm/order track to not stop thruput at end.
Added support for "B" versions of Charge and Microphone BoB.
Added support for setting new "B" BoB high-pass filter.
Added "scope" pnp-example program.
Added small speedup of e1432_set_range & e1432_set_input_high for big systems.
Added HPE1432_CURRENT_VAL_FIFO_AVAIL to hpe1432.h.
Added hpe1432_setLockMode and hpe1432getLockMode.
Added hpe1432_initB2 and hpe1432_initB3, to avoid hardcoded sema.bin path.
Added hpe1432_errorDetails2, easier to use from VisualBasic.
Added hpe1432_getUserDecimation, hpe1432_setUserDecimation.
Added hpe1432_getUserFramesPerBlock, hpe1432_setUserFramesPerBlock.
Added E1432_MEAS_WATCHDOG_REG.
Worked around HP-UX Matlab 5.2 compatibility issue.


Bugs Fixed
----------

Restrict trigger delay when doing RPM arm, just like when doing order track.
Fixed error -1360 from e1432_init_measure_to_booted when using interrupts.
Fixed incorrect zero RPM problem when tach_max_time is set to a small value.
Fixed multi-mod source trigger problem with several sources enabled to trigger.
Fixed Win32 byte-swapping in e1432_set_window when doing lbus thruput.
Fixed Win32 byte-swapping in e1432_get_raw_tachs when doing lbus thruput.
Fixed Plug&Play VISA D32 lbus corruption.
Fixed Plug&Play library to block interrupts properly.
Fixed Plug&Play 50-module limit.
Fixed Plug&Play input offset min/max definitions.
Fixed Plug&Play library bug that broke Matlab on HP-UX.
Fixed hpe1432_arrayReadChan declaration.
Fixed hpe1432_installCallback to properly uninstall when given a NULL func ptr.
Fixed hpe1432_tachDecimate for values > 32767.
Fixed tach trigger level rounding to never exceed the min/max levels.
Fixed corner case enable_none/block_available bug in host library.
Fixed occasional bogus, zero, and/or unsynchronized RPM from slave tachs.
Fixed data RPM with no active inputs and only second tach channel is active.
Fixed gap=0 bug with rpm arm with fast runup, causing RPM1/RPM2 mismatch.
Fixed wrong data RPM with no active input channels and non-rpm-arm mode.
Fixed pnp-example/find.c bug with >50 modules (now allows 64).
Fixed multi-module e1432_send_trigger bug.
Fixed trace printf output on Win16 platforms.
Fixed bug that occasionally caused module to interrupt too many times.
Fixed bug in average weighting.
Fixed bug in aux fifo synchronization.
Fixed obscure I2C error from multi-module E1432 measurements.
Fixed HPE1432_COUPLING_FREQ_MAX - was 51200, now 100.
Fixed E1432_SPAN_MAX to be 78125, not 80000.
Fixed bad data from tiny user-specified fifo size when order tracking.
Fixed octave bug.
Fixed another minor octave bug.
Changed hpe1432.sl to better handle interrupt-driven measurements.


		 Changes Between A.03.00 and A.04.01
		 ===================================

Features added
--------------

Updated source firmware to 0x3035, 19980727.  Please re-zap your sources.

Changed source underrun to cause signal to turn off rather than chan-swap.
Changed e1432_write_srcbuffer_data to take a LONGSIZ32 for numwords.
Changed e1432_anti_alias_digital to abort a measurement if one is running.
Changed e1432_check_src_xxx to support group IDs.
Lowered E1433 Fs/123 spur in 16-bit mode, by scaling up the digital filters.
Added second MEAS_ERROR interrupt after fifo emptied after fifo overflow.
Added E1432_CURRENT_VAL_FIFO_AVAIL to e1432_get_current_value.
Added E1432_ANTI_ALIAS_DIGITAL_BES, for time-domain Bessel decimation filters.
Added E1432_SRCBUFFER_ONESHOT for e1432_set_srcbuffer_mode.
Added DRAM support for arb source ONESHOT and PERIODIC_A modes.
Added better srcbuffer_mode error checking.
Added standalone.Z to the distribution.  It's just too useful to leave out.
Added e1432_octave man page.
Added octave for E1433.
Added zoom for E1433.
Added ability to talk to production Charge BoBs and Microphone BoBs.
Added ability to disable digital filters for input channels.
Added e1432_set_decimation_undersamp function.
Added channel and data type to trailer info.
Added E1432_CLOCK_SOURCE_INT_VXI10 to force use of backplane 10 MHz.
Added E1432_RMS_MODE_FILT and E1432_PEAK_MODE_FILT for E1433 Peak/RMS data.
Added manual page for e1432_install_file.
Added e1432_get_octave_meas and e1432_set_octave_meas.
Added e1432_sys_info as a link to e1432supp script.
Updated e1432_auto_zero manual page to better explain how to auto-zero sources.
Updated the s700 host library to work better with large-module-count systems.
Updated arbsrc_custom stuff.
Updated E1433 aux fifo stuff.
Compressed the PCL, postscript, and text versions of the manual, to save space.
Lock out 32-bit xfer modes above 165 kHz with 8 active chans.
Cleaned up semascope3 demo program.
Removed no-longer-needed semascope2 demo program.
Removed e1432_set_AC_settling, use e1432_set_filter_settling_time instead.


Bugs Fixed
----------

Worked around 96002 bug.  Yes, really.
Fixed averaging gap-too-big problem.
Fixed multi-module OVERLAP_FREERUN mode when changing enable settings.
Fixed hpe1432_init to allow reset=zero to work when no sema.bin is present.
Fixed not-a-number filter_freq setting on non-BoB input channels.
Fixed trailer gap when disabling channels while a measurement is running.
Fixed overlap when zooming.
Fixed fifo overflow with FFTs enabled to properly empty out the fifo.
Fixed Peak and RMS modes to return zero rather than bogus data on E1432.
Fixed sourceaz test program.
Fixed two bugs which could result in RPM=0.
Fixed disabling all channels on one module of multi-module measurement.
Fixed BoB I2C glitch causing hardware I2C error.
Fixed several rollover bugs at 2^31 samples into a measurement.
Fixed arb source hang when parms change between srcbuffer init and meas init.
Fixed External Trigger LED on non-Source non-Tach modules.
Fixed obscure autorange-after-install bug.
Fixed tiny blocksize DMA bug, mostly seen with octave.
Fixed e1432_set_analog_input to use correct range function for Charge & Mike.
Fixed occasional SCA bus error after fifo overflow.
Fixed multi-module auto-zero causing source output under unusual circumstances.
Fixed E1433 input trigger bug on 5 mV range.
Fixed data scaling when resampling, when not all channels are active.
Fixed some problems with external trigger in resample-at-tach-times mode.
Fixed an arb source bug at top span causing last point in burst to be wrong.
Fixed "SCA bus error" problem.
Fixed E1433 ICP turn-off problem.
Fixed tach RPM freeze after e1432_init_measure_to_booted+e1432_reset_measure.
Fixed occasional tach RPM frozen at zero at start of measurement.
Fixed tach delay error when no input channels are active.
Fixed tach delay error on E1432 at clock frequencies above 51200 Hz.
Fixed restriction on number of channels for order tracking at 10 kHz for E1433.
Fixed order tracking "5dB drop" problem on E1432.
Fixed e1432_set_span man page: E1433 can decimate at clock freqs up to 128 kHz.
Fixed obscure multi-module clock bug which could hang the firmware.  Ick.
Fixed init_measure_to_booted to lock the PLL so next init_measure is fast.
Fixed long timeout delay (80 sec?) for e1432_init_measure with large systems.
Fixed bug in E1433 trigger level upper parameter limits.
Fixed 500 pC range setting for charge amp when used on E1432.
Fixed E1432 autozero to not glitch the AC coupling capacitors.
Fixed E1432 "divide-by-two" gain error at lowest span.
Fixed thruput hang when e1432_init_measure is done with the local bus in reset.
Fixed multi-mainframe thruput to pick the correct module for TTLTRG driver.
Fixed e1432supp to work properly on HP-UX 10.20.
Fixed bug related to RMS/Peak causing the module to temporarily lock up.
Fixed bug when only second tach channel was active.
Fixed glitch in time data.


		 Changes Between A.02.00 and A.03.00
		 ===================================

Features added
--------------

Moved entire E1432 tree to /opt/e1432.
    (You may need to do "rm -rf /usr/e1432; ln -s /opt/e1432 /usr/e1432".)

Updated source firmware to 0x302d, 19970711.  Please re-zap your sources.
Updated E1434 rom recovery procedure.

Added realtime source restrictions lockout.
Added support for E1434.
Added ability to change trigger delay while a measurement is running.
Added OVERLAP_FREERUN data mode.
Added eavesdrop with order tracking.
Added ability to use larger arb buffer for the arb source.
Added weighting functions (A, B, C?) to E1433.
Added accurate absolute triggering input and external trigger.
Added "fast span change" capability.
Added time arming.
Added averaging for resampling measurements (so it works in all modes now).
Added trailer info bits to encode what span the data is from.
Added "custom signal generation" mode for optional source board.
Added source mooz, see e1432_set_source_mode.
Added ability to continuously resample data.
Added auto-range.
Added much more info to e1432_display_state function.
Added serial number to hwconfig structure.
Added e1432_arm_measure_master_finish and related functions.
Added e1432_assign_channels function.
Added e1432_channel_group_add and e1432_channel_group_remove functions.
Added e1432_check_src_underrun function.
Added e1432_get_auto_range_mode and e1432_set_auto_range_mode functions.
Added e1432_get_center_freq and e1432_set_center_freq function.
Added e1432_get_current_value function.
Added e1432_get_fwrev function.
Added e1432_get_group_info function.
Added e1432_get_mmf_delay and e1432_set_mmf_delay functions.
Added e1432_get_peak_decay_time and e1432_set_peak_decay_time functions.
Added e1432_get_peak_hold_time and e1432_set_peak_hold_time functions.
Added e1432_get_peak_mode and e1432_set_peak_mode functions.
Added e1432_get_rms_avg_time and e1432_set_rms_avg_time functions.
Added e1432_get_rms_decay_time and e1432_set_rms_decay_time functions.
Added e1432_get_rms_mode and e1432_set_rms_mode functions.
Added e1432_get_source_centerfreq and e1432_set_source_centerfreq functions.
Added e1432_get_weighting and e1432_set_weighting functions.
Added e1432_get_zoom and e1432_set_zoom functions.
Added e1432_sca_dsp_download and e1432_dsp_exec_query functions.
Added e1432_src_get_fwrev and e1432_src_get_rev functions.
Added e1432_src_prog_romimage and e1432_src_rxfr functions.
Added e1432_set_tach_ppr and e1432_set_tach_max_time to llib-l1432.c.
Added autorange, gate3, sca_dnld, spans, and src_dnld test programs.
Added test/e1485 subdirectory with e1485 test code.
Added semascope3 demo program.
Added /opt/e1432/arbsrc/srcutil for programming and reading source roms.
Added ability to deal with E1433 Sample0 and AC coupling integrator fix.
Added some hostdiag tests of SCA boards.
Added ability to use new "rev 4" factory cal fields.
Added more documentation on multi-module clock setup.
Updated e1432_multimain manual page.
Removed the requirement that all modules have the same data_port setting.
Removed ident.sema script, no longer needed since we have a what string.
Changed e1432_finish_measure to also reset interrupt mask and priority.
Changed E1433 autozero to do a better job of not charging the input capacitor.
Changed xplot.a to enable it to open large numbers of windows.
Changed semascope demo to plot all windows even if there are lots of them.
Changed semascope demo to have an option to plot a subset of channels.
Changed e1432supp to print even more info.

Added ability for E1433 to change input_high, input_offset, input_mode, range,
    and trigger_level while a measurement is running.

Added "what" strings to:
  core.rev, e1432supp, ident.sema, ptman, e1432mon, hostdiag,
  mandb, macros.ptman, e1432.h, err1432.h, sema.bin

Moved the anonymous FTP path used to get update files.  The new path
is:
ftp://hpls01.lsid.hp.com/dsp/products/e1432/software/wk_sta/E1432.x.xx.xx.Z.
This was done to make the E1432 distribution fit in with other LSID
VXI products.


Bugs fixed
----------

Fixed tach2 trigger to start at same time as external trigger.
Fixed input digital filters to reduce DC offsets.
Fixed local bus transfer to work better with E1485.
Fixed tach board problem with high tach rates.
Fixed order tracking raw time data to be non-oversampled.
Fixed thruput example program to allow for E1562D and E1562E modules.
Fixed E1433 autozero failure caused by input signal coupling into cal line.
Fixed multi-module resampling problem with low memory.
Fixed input bound triggering, which was sometimes flaky at 51200 Hz.
Fixed array overflow bug in examples/throughput/pbck.
Fixed hang problem when doing e1432_reset_measure, mostly seen with zoom?.
Fixed E1432 autozero problem causing occasional DMA errors.
Fixed zoom center frequency to not round to nearest 1 Hz.
Fixed eavesdrop with no enabled channels to correctly report FIFO overflow.
Fixed e1432_init_measure to force the same sample0 freq in all modules.
Fixed so "oversample on" is illegal at top span, since that can't be done.
Fixed so "multipass" is illegal at top span, since that can't be done.
Fixed E1433 autozero limits to allow for large ADC limit cycles when cold.
Fixed source span to switch to a valid value when clock_freq is changed.
Fixed source triggering jitter relative to decimated input samples -
    but this doesn't seem to work under all circumstances.
Fixed e1432_set_interrupt_mask and e1432_set_interrupt_priority to
    reset interrupt mask before enabling interrupts.
Fixed interrupt window in e1432_set_interrupt_mask.
Fixed e1432_preset and e1432_assign_channel_numbers to
    reset interrupt mask and priority.
Removed references to Vibrato/Sonata in manual pages, replace with E1432/E1433.


		Changes Between A.02.00 and X.02.00.01
		======================================

Added support for E1433 AC filtering hardware changes.
Added support for E1433 Sample0 IRQ hardware changes.
Added ability to use new "rev 4" factory cal fields.


		 Changes Between A.01.03 and A.02.00
		 ===================================

Features added
--------------

Added support for 256K A24 VXI interface Xilinx.
Added support for Charge BoB and Mike BoB.
Added support for E1433.
Added external pre-arm mode.
Added averaging code - e1432_set_avg_mode and related functions.
Added FFT capability.
Added fft and orderfft demo programs.
Added eavesdrop monitoring.  See e1432_set_data_port manual page.
Added eavesdrop example program to demonstrate this new capability.
Added a24_used field to struct e1432_hwconfig.
Added e1432_bob manual page, explaining BoBs a little bit.
Added e1432_get_coupling_freq and e1432_set_coupling_freq functions.
Added e1432_get_fifo_size_current_max function.
Added e1432_get_input_low and e1432_set_input_low functions.
Added e1432_get_input_offset and e1432_set_input_offset functions.
Added e1432_get_ramp function.
Added e1432_get_range_charge and e1432_set_range_charge functions.
Added e1432_get_range_mike and e1432_set_range_mike functions.
Added e1432_get_samples_to_pre_arm function.
Added e1432_get_srcparm_mode and e1432_set_srcparm_mode functions.
Added e1432_get_tach_delay function.
Added e1432_get_warning_string function.
Added e1432_get_window and e1432_set_window functions.
Added e1432_get_xfer_size and e1432_set_xfer_size functions.
Added e1432_update_srcparm function.
Added e1432_read_i2c and e1432_write_i2c functions.
Added pnp-examples directory.
Added i2c, realtime, and srcphasetest test programs.
Added auto-detection of break-out boxes.
Added bob_id field to struct e1432_hwconfig, to identify break-out boxes.
Added better title page on printed manual pages.
Added better on-the-fly parameter changes for source channels.
Added support for source trigger at zero crossing.
Added srcrandb and srcsineb example programs.
Added srcspanbs, src_zeroxing and srcmultichg example programs.
Added separate model codes for E1433 and E1434.
Added calibration for E1433.
Added use of internal 10MHz clock, if present.
Added new flash rom format, contains E1433 cal and substrate board number.
Added /usr/e1432/lib/vximodel.cf.
De-coupled source span from input span.
De-coupled source blocksize from input blocksize.
Improved handling of raw tach buffer overflow.
Speeded up e1432_install function.
Speeded up e1432_reenable_interrupt function.
Speeded up parameter set functions if the parameter is unchanged.
Changed E1432_CLOCK_FREQ_MAX to 200000, since 204800 was never tested.
Changed most floating-point constants in e1432.h to float rather than double.
Changed PLL setup to ignore non-active channels.
No longer installs compiled versions of the test programs, to save space.

Moved the anonymous FTP path used to get update files.  The new path
is: ftp://hpls01.lsid.hp.com/E1432/s700/E1432.x.xx.xx.Z.  This was
done to allow a "pc" subdirectory in place of "s700", for PC files.


Bugs fixed
----------

Fixed E1485 library to not bus error, by restricting it to D16.
Fixed e1432_auto_zero to work for input channels.
Fixed e1432_set_fifo_size to not cause errors if set too big.
Fixed source sine phase clipping to +-180 degrees.
Fixed clock setup synchronization PLL settling problem.
Fixed clock setup meas reset bug.
Fixed a case where e1432_check_overloads didn't catch an overload.


		 Changes Between A.01.01 and A.01.03
		 ===================================

Bugs fixed
----------

Fixed order-tracking bus-error problem.
Fixed 1D4 source board firmware causing a glitch at start of freerun output.
Fixed measurement hang after 20-30 minutes.


		 Changes Between A.01.00 and A.01.01
		 ===================================

Features added
--------------

Added e1432_uninit_io_driver function.
Added README file for /usr/e1432/man.
Added PCL version of the printable manual pages.


Bugs fixed
----------

Fixed multi-module RPM arm hang problem.
Fixed interrupt-driven RPM armed measurements to run more smoothly.


		 Changes Between A.00.10 and A.01.00
		 ===================================

Features added
--------------

Added e1432_cached_parm_update function.
Added e1432_check_src_arbrdy function.
Added e1432_check_src_overload function.
Added e1432_check_src_overread function.
Added e1432_check_src_shutdown function.
Added e1432_get_calc_data and e1432_set_calc_data functions.
Added e1432_get_delta_order and e1432_set_delta_order functions.
Added e1432_get_delta_order_limits function.
Added e1432_get_enable and e1432_set_enable functions.
Added e1432_get_max_order and e1432_set_max_order functions.
Added e1432_get_max_order_limits function.
Added e1432_get_meas_warning function.
Added e1432_get_raw_tachs function.
Added e1432_get_source_seed and e1432_set_source_seed functions.
Added e1432_get_source_seed_limits function.
Added e1432_get_src_arbstates function.
Added e1432_get_tach_clock_freq function.
Added e1432_get_trig_corr function.
Added e1432_send_tachs function.
Added e1432_set_interface_addr function.
Added e1432_install ability to install from memory instead of file, for e1485.
Added ability to talk to non-servant E1432 modules, for e1485.
Added logical address to error printout.
Added compensation for offset voltages when monitoring tach channels.
Added E1432_CALIN_SUMBUS_TACH for monitoring tach channels from other modules.
Added more firmware markers initialization, to help board diagnosis.
Added manual page for e1432supp script.
Added manual pages for arb source functions.
Added overload/halfrange-change interrupt.
Added order, order2, and semascope2 demo programs.
Added srcarbtest, intr2, ovld, and multisrc test programs.
Added disabling of D32 during local-bus measurements.
Added /usr/e1432/vee-examples, for VEE example programs.
Added /opt/vxipnp/hpux/hpe1432, for Plug&Play.
Added /usr/e1432/arbsrc, for source firmware zapping files.
Improved internal DMA speed.
Improved e1432supp script.
Improved data transfer speed.
Improved e1432_check_overloads speed.
Improved measurement startup speed.
Updated various README files and manual pages.
Modified e1432_check_overloads function to return halfrange information.
Moved printable manuals to /usr/e1432/man from /usr/e1432/lib.
Removed unused lost_trigs field (now called "reserved1") from e1432_trailer.
Removed e1432_get_srcbuffer_rdy4data function.


Bugs fixed
----------

Fixed timeout problem on slow or heavily-loaded hosts.
Fixed obscure error when printing host error messages.
Fixed local bus detection bug when installing into multiple modules.
Fixed local bus data shift bug.
Fixed ttltrg corruption bug.
Fixed the firmware to treat board with unprogrammed flash as rev B board.
Fixed bogus-overload bug at start of measurement.
Fixed lack-of-error-printout when e1432_install fails due to lack of module.
Fixed manual pages to show that Tach channels default to being active.
Fixed underrange/overload to work correctly when monitoring tach channels.
Fixed various order-tracking bugs, especially for multiple modules.
Fixed arbitrary source so it works much better.
Fixed FIFO overflow interrupt bug.
Fixed channel swap bug at >51200 sample rate.
Fixed resample data scrambling when using 12 channels.
Fixed resample data scrambling caused by stopping the FIFO.
Fixed source interrupts.
Fixed e1432_init_measure to force the same fifo size for all channels.
Fixed manual trigger with interrupt-driven measurement bug.
Fixed resampling bug with losing every 16th point on lowest resampling span.
Fixed ramp-rate/trigger-delay problem.
Fixed DMA Error problem.
Fixed tach monitoring scale factor bug.
Fixed bug in manual pre-arm where host didn't see state change to pre-arm.
Fixed bug in multi-module RPM pre-arm where measurement would hang.
Fixed slow multi-module startup after e1432_finish_measure.
Fixed multiple sources to start synchronously, both free-run and triggered.
Fixed RPM-armed measurements to stop filling FIFO after run-up finishes.
Fixed bug in Burst Arb source mode.


		 Changes Between A.00.00 and A.00.10
		 ===================================

Features added
--------------

Added e1432_get_blocksize_current_max function.
Added ability to read raw tach times from the module.
Added source interrupts.
Added trigger interrupts.
Improved input calibration to track down spurious cal failures.
Improved data transfer speed improvement for 16-bit data.
Improved local bus transfer speed, to 10 MB/sec with 8K blocksize.

The (struct e1432_trailer)->reserved1 field renamed to lost_trigs.
Parameter value defines in e1432.h are more alphabetical.

Added temporary firmware workaround for a hardware bus error problem.
(The hardware problem was later fixed.)


Bugs fixed
----------

Fixed trailer data corruption bug, and trailer bus-error bug.
Fixed local bus fifo overflow hang problem.
Fixed bug in local bus IRQ which was causing FIFO errors and maybe bad data.
Fixed various RPM arming, pre-arming, and tach channel bugs.
Fixed manual pre-arm mode.
Fixed Clock/Trigger programming on Rev B substrates.
Fixed bug in overlap block mode which sometimes caused a bus error.
Fixed bugs in resampling and triggering.
Fixed potential VXLink bus error.
Fixed polling of inactive sources.

E1432 underrange bit fixed, except for hw problem on fourth
channel.  (The hardware problem was later fixed.)
