Hi, I have double news:

First, there is a strange bug / incompatibility in our kernel:
My DRL 1.01 always works as expected, but while not being much
different, DRL 2.03 shows \ instead of . when given no arguments!
This does only happen on real disks and DOSEMU diskimage disks,
not on DOSEMU redirector disks (Linux directories that simulate
disks). And I have no clue why. Both versions are available, for
example, somewhere at:
http://www.coli.uni-sb.de/~eric/stuff/soft/ (or in subdirectories)

The next issue is with DOS32A:
I have now turned off CPUID -> PCTEST now works (otherwise, it had
failed to notice that CPUID of my CPU tells "no RDTSC" and just assumed
RDTSC would exist in any 586, leading to an invalid opcode abort.
But I have other troubles: While CPUID on/off does not seem to affect ANY
other program of my collection (I use mostly older software), apart from
a CPUID program (displays "486" or "586" depending on the setting),
I get an effect with DR EMM386.
-> WITH DR emm386, Descent works with its builtin DOS4/G(W), but the
   keyboard/mouse and file loading all act weird with DOS32A.
-> WITHOUT DR emm386, Descent works with DOS32A, but aborts with a GPF
   when I try to use the builtin DOS4/G(W).
Both tests done with FDXXMS 0.9 loaded as well. By the way, DR emm386 with
DPMI=ON does not work at all, no matter if MULTI=ON or OFF and no matter
which XMS drivers I do or do not load.

By the way, Terminal Velocity aborts with "sound driver not loaded" and
the DOS32A builtin debugger tells me "int 3 breakpoint" when I try to run
s3drun dos32a game, while it first works but aborts as soon as 3d is turned
on when I let it use its builtin DOS4/G(W) (the sound ... abort happens no
matter if I do or do not load DR emm386). The error message is "Virge not
on board" - I have a Virge DX, but s3drun is supposed to simulate a Virge.

PS: PCTEST reports without DR emm386 220-240k protected<->real switches per
second with DOS32A and PMW, but only 110k with DR emm386. With DOS4GW, both
cases allow 86k switches / second. Surprisingly, all cases report CPL=IOPL=0,
and the memory speed does not differ either (move: 80-88/43-45 M/s in 16/32bit,
store: 112-145/44-48 M/s, 66 MHz SDRAM and 60ns EDO RAM mixed, Cyrix M1 6x86
133 MHz, 123 MIPS, 33 MFLOPS (FADD)...). And since I bought an UDMA cable, at
least Linux now allows UDMA3 (UDMA4 gives errors, but my disk would do it: the
VIA Apollo VPX 35 (82C585) plus 82C586 6 IDE can only handle UDMA33.

Eric

----------
list options/archives/etc.: http://www.topica.com/lists/fd-dev
unsubscribe: send blank email to: [EMAIL PROTECTED]

==^================================================================
This email was sent to: [email protected]

EASY UNSUBSCRIBE click here: http://topica.com/u/?bz8Rv5.bbRv4l.YXJjaGl2
Or send an email to: [EMAIL PROTECTED]

T O P I C A -- Register now to manage your mail!
http://www.topica.com/partner/tag02/register
==^================================================================

Reply via email to