home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.os2.misc
- Path: sparky!uunet!haven.umd.edu!darwin.sura.net!wupost!gumby!destroyer!ubc-cs!unixg.ubc.ca!kakwa.ucs.ualberta.ca!ee.ualberta.ca!kurtk
- From: kurtk@ee.ualberta.ca (Kurt Klingbeil)
- Subject: Specific Version: File/Device IO & EMS problems
- Message-ID: <kurtk.715390556@ee.ualberta.ca>
- Keywords: EMS specific
- Sender: news@kakwa.ucs.ualberta.ca
- Nntp-Posting-Host: bode.ee.ualberta.ca
- Organization: University Of Alberta, Edmonton Canada
- Date: Tue, 1 Sep 1992 23:35:56 GMT
- Lines: 35
-
-
- Having booted a DOS specific version, try 'DIR >NUL' and 'DIR >> NUL'.
- Find any 'Invalid handle' or 'Not OS2 Device' messages... ?
- Note that redirect-append works to files, to LPTx, to CON, etc. but for
- some reason, >>NUL is broken. (\dev\nul too... natch) [Works in VDM's]
-
- Try running your favorite 'debug < script' under specific version... NOTE
- that linefeeds (of CRLF-as-EOL fame) are no longer quietly ignored, but are
- taken to be command-invalidators. Strip them out, and everything's OK...
-
- Next off, a couple of specific version EMS-related questions...:
- 1. Does OS2\MDOS\EMM386.SYS accept (and follow) command-line switches?
- 2. Is 'frame=e000' a valid switch? If so, why would the EMS frame end up
- at cc00 (first non-ROM location) instead?
- 3. Is 'x=d000-dfff' (LAN Card) still a valid switch ? If so why wouldn't
- that prevent the memory range from being used ? And even more strange,
- why would the frame buffer be (partly) located in this excluded range?
-
- Further to 1., 2., &3., I've observed the following odd ems-related behaviour:
- upon bootup, the HIMEM and EMM386 messages both indicate the drivers were
- successfully installed. If EMS is accessed in any way (including Dos5's
- mem command) it is reported as available. Subsequently examining the memory
- range D000:0-DCFF:0 with debug will reveal mirrored copies of part of
- the SETUP region of the (AMI) ROM BIOS (F000:0-F3FF:0).
- What's going on??? At this point no EMS has actually been used, yet the
- LAN-card/cum/page-frame region of memory, which was specifically excluded,
- appears to contain ROM-oid stuff usually never accessed post POST...
-
- Conversely... on a freshly rebooted spec.ver.session, If one examines the
- D000-DCFF region using debug, one will find the expected LAN-card RAM.
- No ROM BIOS, just 00's (and the odd status byte and misc dribbles...).
- Having done this, >>PooF<<, EMS is gone. Somehow, merely accessing the
- page-frame region makes it unuseable...
-
- Any Ideas Anyone??? thx, kk
-