Read this first !
=================

More PCI news:
--------------

There have been more reports about PCI problems, some of them
were solved by upgrading the (flash) BIOS.
Other rumours are that it has to do with the FDC being on the
PCI bus, but that is not the case with the Intel Premiere boards.

Here is a list of systems and the BIOS versions known to work:

board:                          bios revision:

Intel Premiere PCI (Revenge)    1.00.09.AF2

Intel Premiere PCI II (Plato)   1.00.08.AX1 (disable GAT in BIOS!)
                                1.00.10.AX1

To see if you're having the GAT problem, try making a backup
under DOS. If it's very slow and often repositions you're
probably having this problem.

PCI news:
---------
There have been some problem reports from people using PCI-bus based
systems getting overrun errors.
I wasn't able to reproduce these until I ran ftape on a Intel Plato
(Premiere PCI II) motherboard with bios version 1.00.08AX1.
It turned out that if GTA (Guaranteed Access Timing) is enabled (?)
ftape gets a lot of overrun errors.
The problem disappears when disabling GAT in the bios.
Note that Intel removed this setting (permanently disabled) from the
1.00.10AX1 bios !

It looks like that if GAT is enabled there are often large periods
(greater than 120 us !??) on the ISA bus that the DMA controller cannot
service the floppy disk controller.
I cannot imagine this being acceptable in a decent PCI implementation.
Maybe this is a `feature' of the chipset. I can only speculate why
Intel choose to remove the option from the latest Bios...

The lesson of this all is that there may be other motherboard
implementations having the same of similar problems.
If you experience a lot of overrun errors during a backup to tape,
see if there is some setting in the Bios that may influence the
bus timing.

I judge this a hardware problem and not a limitation of ftape ;-)
My DOS backup software seems to be suffering from the same problems
and even refuses to run at 1 Mbps !
Ftape will reduce the datarate from 1 Mbps to 500 Kbps if the number
of overrun errors on a track exceeds a thresshold.

General:
--------
This version should work with all kernel versions from 1.0.9 up to
1.1.66 (and probably earlier and later versions too).
The Install-guide describes an older version of ftape but will work
for this version too.

When compiling the driver make sure that /usr/src/linux points to the
sourcecode of the kernel that is actually running !

I've included the file `mt-patch' that can patch the mt found in gnu
cpio-2.3 and will display usefull information when using eof marks.

Normally there is no need to modify the Makefile anymore, except when:
- you're not using the tape drive connected to the standard floppy controller
  or you're using a FC-10.
- you're compiling with gcc 2.4.5.
- you don't want kernel messages or you want a more compact driver.
In any of these cases see the Makefile.

For all the people that send me mail with bug fixes, suggestions and
complaints: As you've noticed I'm unable to answer everyone as that
would leave me no time at all time to work on the driver (and those
moments are already very rare).
Note however that I do read all your comments and will keep improving
the driver with your feedback !


        Bas.


=== Release notes for ftape-1.14d ===

Bug fixes:

- removed reference to do_floppy when not used.
- fixed using wrong header segment when not two successive segments.

Added features:

- added support for Mountain MACH-2 controller.


=== Release notes for ftape-1.14c ===

Bug fixes:

- Fixed rare irq & dma not being released situation if open failed.
- request_dma changed in 1.1.48 and not in 1.1.43: Fixed Makefile.

Added features:

- Now able to handle tapes with bad sectors in the header segment(s).


=== Release notes for ftape-1.14b ===

Bug fixes:

- Fixed recovery of overrun errors during read operations.

New Features:

- Reducing datarate if experiencing excessive overrun errors.
- Improved startup code to reduce number of no-data errors.
- Added code that allows two fdc's to share the same dma channel
  (not tested yet).
- Looks like the cooperation with the floppy driver in recent
  (1.1.59) kernels is ok now.


=== Release notes for ftape-1.14a ===

Bug fixes:

- Fixed bug that caused fdc's that didn't support register dump
  command not being recognized.


=== Release notes for ftape-1.14 ===

Bug fixes:

- Fixed problem with sequential writes to non-rewinding device that caused
  the first file always to be overwritten.
- Fixed infinite recursion causing stack overflow and panic if fdc is
  not responding to reset signal.
- Ioctl status command no longer references tape.
- Oops: fixed wrong polarity of fifo enable bit.
- Fixed 'putting drive offline forever' problem.
- Fixes for Conner bugs allways enabled, eliminated CONNER_BUG conditions.
- Fixed (?) dirty write buffers being sticky after error.
- Changed registered device name from "mt" into "ftape"
- Swapped segment and sector number in bad sector log, old format version 1
  will be upgraded to new version 2 when written.

New Features:

- Added patch for mt command (found in GNU cpio-2.3) to correctly
  display status information.
- IMPORTANT !!! The non-rewinding device will not update the header
  segment anymore, take care and may sure to rewind the tape before
  taking the tape out !!!
- Auto detect of kernel version prevents configuration problem with
  FLOPPY_HACK causing kernel panics.
- Included support for more than one wake-up method per vendor-id.
- Ftape driver now unlocks fdc and restores fifo state at exit.
  Leaves fdc in reset state but new fdc code doesn't notice :-(
- (unsupported) lseek operation now returns an error.

Known Bugs:

- Floppy access may fail after ftape use. This is because of recent changes
  to the floppy driver and must be fixed in the floppy driver.
  Version 1.1.59 seems ok :-)

--------------------

