basilisk-devel — Basilisk II development-related discussion

[B2-devel] Re: SheepShaver

[Carsten K. <ck...@ku...>]

Howard,

Thank you for your e-mails.

hs> I have seen at least two other people having problems like this on Suse distributions.

This darned, obscure autoconf/automake/m4 system! Are there more than
10 people on this planet who actually know what's going on inside it?

hs> As far is I know there is an issue with the naming of the
hs> floppy devices between suse and, for instance, mandrake linux. 
hs>
hs> Something like using FD0 instead of fd0
hs>
hs> On my machine that leads to unacceptable waiting times before SheepShaver starts.

I don't worry about the floppy and the little delay it creates in the
boot process. Haven't used floppy disks in a year or two anyway.

hs> The gtk 1.2 issue is still open, also on my machine. I
hs> haven't been able to succesfully compile sheepshaver on Suse 9.0 

(repeat rant about auto-*)

hs> Some remarks to you possible explanations below
hs> Here are a few possible explanations:
hs> - The ROM Image on the MacOS8.5 CD is not suitable for SheepShaver

hs> This isn't the case

Thank you for confirming this.

hs> - MacOS8.5 doesn't run under SheepShaver
hs> MacOS 8.5 does run under SheepShaver

Thank you for confirming this. I didn't really expect it but didn't
want to rule it out either.

hs> - The MacOS CD has a physical defect and SheepShaver tries to execute a
hs> program segment located in a defective area

hs> You have to get a new medium

Seems like I have to go shopping on E-Bay...

hs> - SheepShaver works the CD ROM Driver in such an uncommon way that my
hs> PlexWriter crashes

hs> I haven't seen any reports on issues like this.

I'd be surprised, too, as I regularly use it to write CDs

hs> - SheepShaver works the CD ROM Driver in such an uncommon way that a
hs> programming error in the IDE driver shows up

hs> I couldn't tell

hs> - SheepShaver works the CD ROM Driver in such an uncommon way, so that a
hs> difference between a CD Writer and an ordinary CD Drive shows up

hs> I haven't had any problems concerning this

I still can't rule this out. What else would explain that the CD drive
is locked and I can't remoce the MacOS8.5 CD after SheepShaver
crashes?! As far as I know, it is Linux Kernel semantics to free all
resources a process has kept after the process is killed. All files
get closed, etc., so there should be no reason for the device driver
in the kernel to stop me from removing the CD, as it is not mounted or
anything.

I'll try to get another CD and see if the problem persists.

Thank you for your help,

Carsten

[B2-devel] SheepShaver 2.2 from RPM + MacOS8.5 CD on PlexWriter = strange problems

[Carsten K. <ck...@ku...>]

I discovered SheepShaver a few days ago, and have since tried to get it 
running on my Linux machine. I have SuSE 8.1, SuSE 9.0 and LFS on my 
disposal, as well as my wife's PowerMac 8600/300 with MacOS9.1 and OS X 
installed on it.

First I tried to build SheepShaver from the sources, but as hard as I tried, I 
always failed at the configuration stage (some macro with GTK 1.2 in it). So 
then I tried to get the pre-built RPMs running and first succeeded with them 
on Suse 9.0, and after having "upgraded" my LFS system to gcc 3.2.2 in order 
to resolve the libstdc++ 3.2.2 dependency, also on my LFS system. I took my 
wife's Mac OS 8.5 CD and tried to mount it -t hfs, but that failed. So I used 
her Mac, and ftped the Mac OS ROM file to my Linux box. That succeeded. Put 
the MacOS CD back into my Linux box, and started SheepShaver. I created a 1GB 
Mac volume from the GUI, entered the path to the ROM file, and started 
emulation. From a technical point of view, it's very successfull: I get the 
grey screen with the smiling Mac on it, so the PPC emulator works, and the 
ROM is ok, too. Then I get the MacOS Bitmap, so reading from the CD works, 
too, and the majority of the initialization works well, too. Then I see a few 
extension icons showing up -- and then the whole emulation window vanishes. 
This behaviour is the same on SuSE 9.0 and LFS, with 2.4 or 2.6 kernels. The 
severity of this emulator crash depends on if the JIT is turned on or not. If 
the JIT was turned on, then the whole machine freezes: I can't move the 
mouse, can't enter a key, can't switch to a virtual console, and even 
caps-lock doesn't toggle the LED anymore. Strangely, the HD LED blinks every 
now and then, so not everything is dead. If I turn the JIT off, the system 
stays alive, but I can't eject the CD anymore, so I'm forced to reboot, too.
I also tried booting with the extensions off (Shift key pressed during boot) 
but to no avail. In the Non-JIT case these here are the last few lines of 
console output:

>>>>>
bash-2.05a# ./SheepShaver
SheepShaver V2.2 by Christian Bauer and Mar"c" Hellwig
Reading ROM file...
WARNING: Cannot open /dev/fd1u1440 (No such device or address)
PowerPC CPU emulator by Gwenole Beauchesne
Illegal instruction at 40810000, opcode = 00000000
bash-2.05a#
<<<<<

(Is 40810000 RAM or ROM on a Mac?!)

Here are a few possible explanations:
- The ROM Image on the MacOS8.5 CD is not suitable for SheepShaver
- MacOS8.5 doesn't run under SheepShaver
- The MacOS CD has a physical defect and SheepShaver tries to execute a 
program segment located in a defective area
- SheepShaver works the CD ROM Driver in such an uncommon way that my 
PlexWriter crashes
- SheepShaver works the CD ROM Driver in such an uncommon way that a 
programming error in the IDE driver shows up
- SheepShaver works the CD ROM Driver in such an uncommon way, so that a 
difference between a CD Writer and an ordinary CD Drive shows up

I'm an experienced Windows C++ Programmer, but I have no experience with Mac's 
whatsoever (except as a user) so could one of you please point me in the 
right direction?

TIA

Carsten

[B2-devel] 256 colours (SheepShaver)

[John G. <joh...@bi...>]

I am currently using a Darwin port of SheepShaver that I compiled from 
the latest CVS code. Just wondering if there is currently a way to 
switch from millions of colours to 256 colours.

thanks,

John

[B2-devel] SheepShaver snapshot (2004/02/24)

[Gwenole B. <gb...@di...>]

Hi,

Here is the announcement I sent to comp.emulators.misc. Remaining bits
prior to a formal release would include runtime depth switching from B2
and fixes for DGA displays + other ports. Possibly ethernet too, but I am
still not satisfied with the current fixes.

Note that you no longer needs the -hacks patch, as I finally committed
another variant for now. That should make it more user-friendly to build &
get something working with MacOS 8.6 out-of-the-box. ;-)

* Official website:
<http://www.uni-mainz.de/~bauec002/SheepShaver.html>

* Sources (CVS snapshots) + precompiled binaries as RPM:
<http://gwenole.beauchesne.free.fr/sheepshaver/files/>
Files: SheepShaver-2.2-20040224.tar.bz2, SheepShaver-2.2-10.*.rpm

* What's new from previous test release?
- Improved generic JIT, also now translates FPU instructions
- AltiVec emulation, thus enabling the PVR to match a 7400 (G4)
- Partial port to Darwin/PPC, i.e. it runs but some features may be
missing

Re: [B2-devel] OS 9 killed SheepShaver.

[Gwenole B. <gb...@di...>]

Hi,

> Well now no matter what os 7.5.5 to 8.5 it get to mac os screen where 
> it about to load extensions
> nathen just bomb out and quite on me with no error code or anything.

If you have just built from sources, make sure sure you applied -hacks 
patch until someone comes up with a better SCSI Manager/Expert 
replacement. Alternatively, also try to get rid of your nvram file.

Bye,
Gwenole.

[B2-devel] OS 9 killed SheepShaver.

[Robert S. <ro...@xb...>]

Built for sources from what ever was up on Saturday.I setup my usualy hfv
file had 8.5 on there and running fine.  Thenout of no where at alli wanted
to pull some stuff from the os 9 cd. And sheep shaver picked up on the cd
and started to load os9 ( using hardware rom) then sheep shaver quit. Normal
response for something that is not supported.  Well now no matter what os
7.5.5 to 8.5 it get to mac os screen where it about to load extensions
nathen just bomb out and quite on me with no error code or anything.
Replaced the rom wth a new clean copy fr9om a cd I had burn that I put all
the harware rom images on and still.  Same problem.  What in the heck happen
there.
Also could it be a problem that part of sheepshaver was compiled with gcc
3.2.3 to the point where it got up to gwenoles new altivec code then gcc
3.3.3 from there.  

-----Original Message-----
From: bas...@li...
[mailto:bas...@li...] On Behalf Of Gwenole
Beauchesne
Sent: Sunday, February 22, 2004 5:48 AM
To: bas...@li...
Subject: Re: [B2-devel] Lack of config files ;)

Hi,

> I just checked out the CVS of SheepShaver and it lacks config.sub and 
> config.guess, and without these it is not configurable and then not 
> compilable.

You probably missed the "make links" part as config.* files are in the
B2 tree. Besides, once you have checked out trees, always cvs update -d in
both B2 and SheepShaver dirs. Then, if you see a change to configure.in, rm
-f config.cache & rerun autogen.sh to recreate the configure script.

Bye,
Gwenole



-------------------------------------------------------
SF.Net is sponsored by: Speed Start Your Linux Apps Now.
Build and deploy apps & Web services for Linux with a free DVD software kit
from IBM. Click Now!
http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
_______________________________________________
basilisk-devel mailing list
bas...@li...
https://lists.sourceforge.net/lists/listinfo/basilisk-devel

RE: [B2-devel] Optimizations to AltiVec code now in CVS

[Robert S. <ro...@xb...>]

I will just update gcc to what ever the current version is.  I check the
headers and the seem to be fine.  This was a fresh install of slack Called I
had got into tweaking the 2.6 kernel a little to much and well I kinda
killed linux.  So I had to format and reinstall everything.  But I will just
update toe the latest and check slack ftp for updates not that ever really
been a a lot of updates for it but I will check try again and report. 

-----Original Message-----
From: bas...@li...
[mailto:bas...@li...] On Behalf Of Gwenole
Beauchesne
Sent: Sunday, February 22, 2004 5:48 AM
To: bas...@li...
Subject: Re: [B2-devel] Optimizations to AltiVec code now in CVS

Hi,

> In file included from ../kpx_cpu/src/cpu/ppc/ppc-dyngen-ops.cpp:1493:
> /usr/lib/gcc-lib/i486-slackware-linux/3.2.3/include/xmmintrin.h: In 
> function
>    `void _mm_stream_pi(vector int*, vector int)':
> /usr/lib/gcc-lib/i486-slackware-linux/3.2.3/include/xmmintrin.h:1036: 
> cannot
>    convert `vector int*' to `long long unsigned int*' for argument `1' 
> to
> `void
>    __builtin_ia32_movntq(long long unsigned int*, long long unsigned 
> int)'

AFAIK, there is no "vector" definition in any released gcc for x86. SSE
types are generally defined as __attribute__((__mode__(__V4SI__))) &
friends.

I tried a build with 3.2.2, 3.3.1, 3.3.2 and 3.4, they are all fine. 
Note that on AMD64, you would need to arrange <xmmintrin.h> a little if you
were to use 3.3.1-4mdk. You will have to check your headers and possibly
report to your distributor. Look around
ppc-dyngen-ops.cpp:1493 and try to excercise a testcase. I simply know that
MDK compilers are fine. ;-)

Bye,
Gwenole



-------------------------------------------------------
SF.Net is sponsored by: Speed Start Your Linux Apps Now.
Build and deploy apps & Web services for Linux with a free DVD software kit
from IBM. Click Now!
http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
_______________________________________________
basilisk-devel mailing list
bas...@li...
https://lists.sourceforge.net/lists/listinfo/basilisk-devel

Re: [B2-devel] Lack of config files ;)

[Gwenole B. <gb...@di...>]

Hi,

> I just checked out the CVS of SheepShaver and it lacks config.sub and 
> config.guess, and without these it is not configurable and then not 
> compilable.

You probably missed the "make links" part as config.* files are in the 
B2 tree. Besides, once you have checked out trees, always cvs update -d 
in both B2 and SheepShaver dirs. Then, if you see a change to 
configure.in, rm -f config.cache & rerun autogen.sh to recreate the 
configure script.

Bye,
Gwenole

Re: [B2-devel] Optimizations to AltiVec code now in CVS

[Gwenole B. <gb...@di...>]

Hi,

> The only problem is that for some reason ICC 7 does not like your sse 
> &mmx
> optimizations and now crashes the compiler.

Even icc v8 never actually compiled ppc-dyngen-ops.cpp correctly. 
That's why I explictly make use of gcc for the synthetic opcodes. 
Reconfigure your build with an updated configure script.

> to see the speed boost really needed on the x86 side just maybe it 
> time that
> we found an assembly programmer to optimize as much ppc to x86 code as
> possible in asm.

This is silly. There is nothing to asm optimize. The only bits 
asm-optimized are byte-swapping and locking mechanisms. I tried to asm 
optimize subfe/adde/addeo/etc. instructions. gain was nearly zero.

> Jit is amazing technology but it don't think it handles
> the complexity of ppc emulation nearly as well as it did on the 68k 
> core.

The current JIT is designed to be portable. The next JIT is aimed at 
producing optimized code.

> Also can anyone explain why even with fresh install of 7.6.1 with 
> extensions
> off and virutal memeory off why I am still getting bits of ms mac 
> office
> code or something in my rom images on Alchemy based powermacs.

That's indeed unsual. ;-) Maybe try to boot off a CD to get a non 
polluted system folder?

Bye,
Gwenole.

Re: [B2-devel] Optimizations to AltiVec code now in CVS

[Gwenole B. <gb...@di...>]

Hi,

> In file included from ../kpx_cpu/src/cpu/ppc/ppc-dyngen-ops.cpp:1493:
> /usr/lib/gcc-lib/i486-slackware-linux/3.2.3/include/xmmintrin.h: In 
> function
>    `void _mm_stream_pi(vector int*, vector int)':
> /usr/lib/gcc-lib/i486-slackware-linux/3.2.3/include/xmmintrin.h:1036: 
> cannot
>    convert `vector int*' to `long long unsigned int*' for argument `1' 
> to
> `void
>    __builtin_ia32_movntq(long long unsigned int*, long long unsigned 
> int)'

AFAIK, there is no "vector" definition in any released gcc for x86. SSE 
types are generally defined as __attribute__((__mode__(__V4SI__))) & 
friends.

I tried a build with 3.2.2, 3.3.1, 3.3.2 and 3.4, they are all fine. 
Note that on AMD64, you would need to arrange <xmmintrin.h> a little if 
you were to use 3.3.1-4mdk. You will have to check your headers and 
possibly report to your distributor. Look around 
ppc-dyngen-ops.cpp:1493 and try to excercise a testcase. I simply know 
that MDK compilers are fine. ;-)

Bye,
Gwenole

RE: [B2-devel] Optimizations to AltiVec code now in CVS

[Robert S. <ro...@xb...>]

Gwenole I think I spoke to soon.  When I compile the binary last time some
how I got caught up in a old directory as I have a tendency to keep old
source floating around in and just rename folders well when I realized what
I did and I went to fix it.  So I do the usual steps and this is what I get
with gcc 3.2.3

g++ -I../kpx_cpu/include -I../kpx_cpu/src -DUSE_JIT -I../include -I. 
g++ -DHAVE_CONFIG_H -D_REENTRANT 
g++ -DDATADIR=\"/usr/local/share/SheepShaver\" -g -O2  
g++ -I/usr/X11R6/include -I/usr/include/gtk-1.2 -I/usr/include/glib-1.2 
g++ -I/usr/lib/glib/include -I/usr/X11R6/include  -fomit-frame-pointer 
g++ -mpreferred-stack-boundary=2 -falign-functions=0 -mmmx -msse 
g++ -finline-limit=10000 -fno-reorder-blocks -fno-optimize-sibling-calls 
g++ -c ../kpx_cpu/src/cpu/ppc/ppc-dyngen-ops.cpp -o obj/ppc-dyngen-ops.o
In file included from ../kpx_cpu/src/cpu/ppc/ppc-dyngen-ops.cpp:1493:
/usr/lib/gcc-lib/i486-slackware-linux/3.2.3/include/xmmintrin.h: In function
   `void _mm_stream_pi(vector int*, vector int)':
/usr/lib/gcc-lib/i486-slackware-linux/3.2.3/include/xmmintrin.h:1036: cannot
   convert `vector int*' to `long long unsigned int*' for argument `1' to
`void
   __builtin_ia32_movntq(long long unsigned int*, long long unsigned int)'
make: *** [obj/ppc-dyngen-ops.o] Error 1 
root@slackrules:/home/rob/SheepShaver/src/Unix# 

-----Original Message-----
From: bas...@li...
[mailto:bas...@li...] On Behalf Of Gwenole
Beauchesne
Sent: Friday, February 20, 2004 12:57 PM
To: bas...@li...
Subject: [B2-devel] Optimizations to AltiVec code now in CVS

Hi,

I have just committed some tentative SSE & MMX optimizations for AltiVec
emulation. You need gcc recent enough to take benefit. e.g. gcc 3.2.2 with
the respective intrinsics in <mmintrin.h> & <xmmintrin.h>.

This is generic code that can be improved even more. Still, the 1.8 GHz
Opteron here performs at around 783 MegFlops on AltiVec Fractal Carbon, i.e.
the exact half performance than my PBG4/400. I really would like to reach 1+
GigaFlops in emulation. ;-) However, I doubt this will be reached with the
current "JIT1" dynamic translation engine.

Obviously, we can't translate all AltiVec instructions to SSE code simply
because that's not practical, especially saturating variants as we have to
update the SAT bit and there is not extra flag in x86 land to know that a
value saturated. i.e. we only have approx 30 native code templates for key
VMX instructions used e.g. in AltiVec Fractal Carbon.

I hope I haven't broken anything. One possible next step is to use the
run-time assembler I wrote last year for the new JIT infrastructure. That
way, we can remove some useless instructions in the process.

I don't know yet what you can expect next, but at least the CPU emulation is
now in an interesting shape. ;-)

Bye,
Gwenole.


-------------------------------------------------------
SF.Net is sponsored by: Speed Start Your Linux Apps Now.
Build and deploy apps & Web services for Linux with a free DVD software kit
from IBM. Click Now!
http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
_______________________________________________
basilisk-devel mailing list
bas...@li...
https://lists.sourceforge.net/lists/listinfo/basilisk-devel

RE: [B2-devel] Optimizations to AltiVec code now in CVS

[Robert S. <ro...@xb...>]

The only problem is that for some reason ICC 7 does not like your sse &mmx
optimizations and now crashes the compiler.  Which is a shame as that is the
only way I was able to extra speed boost out of the program was using icc
and there -03 optimaztions on the sheepshaver code.  Optimizing for the p4.
But Compiled fine out of the gate for my generic build I use to make the
slackware install packages.  It gets back to what I had wrote you in another
email.  It is great that we have altivec instruction support but that should
be the farest thing from completeing in the whole big picture. Example maybe
to see the speed boost really needed on the x86 side just maybe it time that
we found an assembly programmer to optimize as much ppc to x86 code as
possible in asm.  Jit is amazing technology but it don't think it handles
the complexity of ppc emulation nearly as well as it did on the 68k core.

Also can anyone explain why even with fresh install of 7.6.1 with extensions
off and virutal memeory off why I am still getting bits of ms mac office
code or something in my rom images on Alchemy based powermacs.  If anyone
with a on this list with a 54/55/64/6500/TAM with a rom dump that works with
sheepshaver let me know as this has gotten to be very annoying.  I also have
a problem with a lot of freezing using a Gossamar rom image pulled from a
Beige G3 DT.
 

-----Original Message-----
From: bas...@li...
[mailto:bas...@li...] On Behalf Of Gwenole
Beauchesne
Sent: Friday, February 20, 2004 12:57 PM
To: bas...@li...
Subject: [B2-devel] Optimizations to AltiVec code now in CVS

Hi,

I have just committed some tentative SSE & MMX optimizations for AltiVec
emulation. You need gcc recent enough to take benefit. e.g. gcc 3.2.2 with
the respective intrinsics in <mmintrin.h> & <xmmintrin.h>.

This is generic code that can be improved even more. Still, the 1.8 GHz
Opteron here performs at around 783 MegFlops on AltiVec Fractal Carbon, i.e.
the exact half performance than my PBG4/400. I really would like to reach 1+
GigaFlops in emulation. ;-) However, I doubt this will be reached with the
current "JIT1" dynamic translation engine.

Obviously, we can't translate all AltiVec instructions to SSE code simply
because that's not practical, especially saturating variants as we have to
update the SAT bit and there is not extra flag in x86 land to know that a
value saturated. i.e. we only have approx 30 native code templates for key
VMX instructions used e.g. in AltiVec Fractal Carbon.

I hope I haven't broken anything. One possible next step is to use the
run-time assembler I wrote last year for the new JIT infrastructure. That
way, we can remove some useless instructions in the process.

I don't know yet what you can expect next, but at least the CPU emulation is
now in an interesting shape. ;-)

Bye,
Gwenole.


-------------------------------------------------------
SF.Net is sponsored by: Speed Start Your Linux Apps Now.
Build and deploy apps & Web services for Linux with a free DVD software kit
from IBM. Click Now!
http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
_______________________________________________
basilisk-devel mailing list
bas...@li...
https://lists.sourceforge.net/lists/listinfo/basilisk-devel

[B2-devel] Optimizations to AltiVec code now in CVS

[Gwenole B. <gb...@di...>]

Hi,

I have just committed some tentative SSE & MMX optimizations for AltiVec
emulation. You need gcc recent enough to take benefit. e.g. gcc 3.2.2 with
the respective intrinsics in <mmintrin.h> & <xmmintrin.h>.

This is generic code that can be improved even more. Still, the 1.8 GHz
Opteron here performs at around 783 MegFlops on AltiVec Fractal Carbon,
i.e. the exact half performance than my PBG4/400. I really would like to
reach 1+ GigaFlops in emulation. ;-) However, I doubt this will be reached
with the current "JIT1" dynamic translation engine.

Obviously, we can't translate all AltiVec instructions to SSE code simply
because that's not practical, especially saturating variants as we have to
update the SAT bit and there is not extra flag in x86 land to know that a
value saturated. i.e. we only have approx 30 native code templates for key 
VMX instructions used e.g. in AltiVec Fractal Carbon.

I hope I haven't broken anything. One possible next step is to use the
run-time assembler I wrote last year for the new JIT infrastructure. That
way, we can remove some useless instructions in the process.

I don't know yet what you can expect next, but at least the CPU emulation
is now in an interesting shape. ;-)

Bye,
Gwenole.

[B2-devel] Lack of config files ;)

[Natalia P. <cl...@te...>]

Hi

I just checked out the CVS of SheepShaver and it lacks config.sub and 
config.guess, and without these it is not configurable and then not 
compilable.

Regards,
Natalia Portillo
http://museum.claunia.com

[B2-devel] Re: Darwin port GUI

[Toshimitsu T. <t_t...@db...>]

Hi.

On 2004/02/16, at 23:17, Gwenole Beauchesne wrote:
> Ah, probably extra GTK libs are loaded at the same conflicting region?  
> Try
> e.g. to make RAM_BASE around 0x10000000.

OK, It works.
Thank you.
http://www006.upp.so-net.ne.jp/t_tanaka/images/settingpanel.jpg
------------------------------------------------------------------------ 
-----------------------------------------------
--- main_unix.org	Sat Jan 31 20:10:49 2004
+++ main_unix.cpp	Mon Feb 16 23:42:25 2004
@@ -149,7 +149,7 @@
  const char ROM_FILE_NAME[] = "ROM";
  const char ROM_FILE_NAME2[] = "Mac OS ROM";

-const uintptr RAM_BASE = 0x20000000;		// Base address of RAM
+const uintptr RAM_BASE = 0x10000000;		// Base address of RAM
  const uint32 SIG_STACK_SIZE = 0x10000;		// Size of signal stack


@@ -493,8 +493,10 @@
  		}
  		fclose(proc_file);
  	} else {
+#if defined(__linux__)
  		sprintf(str, GetString(STR_PROC_CPUINFO_WARN), strerror(errno));
  		WarningAlert(str);
+#endif
  	}
  #endif
  	D(bug("PVR: %08x (assumed)\n", PVR));

--- prefs_editor_gtk.org	Tue Jan 13 00:37:22 2004
+++ prefs_editor_gtk.cpp	Sun Feb  1 21:57:58 2004
@@ -815,7 +815,11 @@
  	if (d) {
  		struct dirent *de;
  		while ((de = readdir(d)) != NULL) {
+#if defined(__linux__)
  			if (strncmp(de->d_name, "ttyS", 4) == 0 || strncmp(de->d_name,  
"lp", 2) == 0) {
+#else
+			if (false) {
+#endif
  				char *str = new char[64];
  				sprintf(str, "/dev/%s", de->d_name);
  				glist = g_list_append(glist, str);
@@ -846,7 +850,13 @@
  			struct ifreq req, *ifr = ifc.ifc_req;
  			for (int i=0; i<ifc.ifc_len; i+=sizeof(ifreq), ifr++) {
  				req = *ifr;
+#if defined(__linux__)
  				if (ioctl(s, SIOCGIFHWADDR, &req) == 0 && req.ifr_hwaddr.sa_family  
== ARPHRD_ETHER) {
+#elif defined(__APPLE__) && defined(__MACH__)
+				if (ioctl(s, SIOCGIFADDR, &req) == 0 && (req.ifr_addr.sa_family ==  
ARPHRD_ETHER || req.ifr_addr.sa_family == ARPHRD_ETHER+1)) {
+#else
+				if (false) {
+#endif
  					char *str = new char[64];
  					strncpy(str, ifr->ifr_name, 63);
  					glist = g_list_append(glist, str);

--- user_strings_unix.org	Tue Jan 13 00:37:22 2004
+++ user_strings_unix.cpp	Sun Feb  1 21:51:46 2004
@@ -25,10 +25,10 @@
  // Platform-specific string definitions
  user_string_def platform_strings[] = {
  	// Common strings that have a platform-specific variant
-	{STR_VOLUME_IS_MOUNTED_WARN, "The volume '%s' is mounted under Linux.  
Basilisk II will try to unmount it."},
-	{STR_EXTFS_CTRL, "Linux Root"},
-	{STR_EXTFS_NAME, "Linux Directory Tree"},
-	{STR_EXTFS_VOLUME_NAME, "Linux"},
+	{STR_VOLUME_IS_MOUNTED_WARN, "The volume '%s' is mounted under Unix.  
SheepShaver will try to unmount it."},
+	{STR_EXTFS_CTRL, "Unix Root"},
+	{STR_EXTFS_NAME, "Unix Directory Tree"},
+	{STR_EXTFS_VOLUME_NAME, "Unix"},

  	// Purely platform-specific strings
  	{STR_NO_DEV_ZERO_ERR, "Cannot open /dev/zero: %s."},
---
Toshimitsu Tanaka
t_t...@db...
http://homepage3.nifty.com/toshi3/index.html

Re: [B2-devel] Re: Darwin port GUI

[Gwenole B. <gb...@di...>]

On Mon, 16 Feb 2004, Toshimitsu Tanaka wrote:

> The error has occurred here.
> main_unix.cpp:line 587
> 
> 	if (vm_acquire_fixed((char *)RAM_BASE, RAMSize) < 0) {

Ah, probably extra GTK libs are loaded at the same conflicting region? Try 
e.g. to make RAM_BASE around 0x10000000.

[B2-devel] Re: Darwin port GUI

[Toshimitsu T. <t_t...@db...>]

Hello.

On 2004/02/16, at 2:50, Gwenole Beauchesne wrote:
>> However, after clicking the "Start" button,
>> SheepShaver says "Cannot map RAM:Unknown error:0".
>
> Does it work without the GUI with those settings? It could be that the 
> GUI hasn't saved the RAM prefs thus defaulting to 0 RAM size and 
> failing?

  The GTK+ is enabled and when "--nogui true" option is specified,  same 
error occurs.
But, when GTK+ is disabled, SheepShaver works  with same prefs file.

The error has occurred here.
main_unix.cpp:line 587

	if (vm_acquire_fixed((char *)RAM_BASE, RAMSize) < 0) {
		sprintf(str, GetString(STR_RAM_MMAP_ERR), strerror(errno));
		ErrorAlert(str);
		goto quit;
	}
---
Toshimitsu Tanaka
t_t...@db...
http://homepage3.nifty.com/toshi3/index.html

[B2-devel] Re: Darwin port GUI

[Toshimitsu T. <t_t...@db...>]

Hi,Gwenole.

on 04.2.16 2:50, Gwenole Beauchesne at gb...@di... wrote:
> Since PVR is an OEA register, is there a way in Darwin (or NetBSD) to
> get the current processor ID from a userspace application?

In Darwin, we can get CPU information by the following commands.

ioreg -l | grep cpu-version

---
Toshimitsu Tanaka
t_t...@db...
http://homepage3.nifty.com/toshi3/index.html

Re: [B2-devel] Sheepshaver for W32

[Gwenole B. <gb...@di...>]

Hi,

> I understand you could build and run sheepshaver on Windows/x86. Is  
> there any
> information or tutorial how to do this?

There is no Windows port of SheepShaver yet. As neither Christian nor I  
are Windows developers, we would appreciate any contribution to bring  
SheepShaver there.

I have tried to detail prerequesties for that port:
<http://sourceforge.net/mailarchive/ 
forum.php?thread_id=3784763&forum_id=7009>

Bye,
Gwenole.

Re: [B2-devel] Darwin port GUI

[Gwenole B. <gb...@di...>]

Hi,

> However, after clicking the "Start" button,
> SheepShaver says "Cannot map RAM:Unknown error:0".

Does it work without the GUI with those settings? It could be that the  
GUI hasn't saved the RAM prefs thus defaulting to 0 RAM size and  
failing?

> ----------------------------------------------------------------------- 
> ------------------------------------------------
> --- main_unix.org	Sat Jan 31 20:10:49 2004
> +++ main_unix.cpp	Sun Feb  1 22:11:54 2004
> +#if defined(__linux__)
>  		sprintf(str, GetString(STR_PROC_CPUINFO_WARN), strerror(errno));
>  		WarningAlert(str);
> +#endif

Since PVR is an OEA register, is there a way in Darwin (or NetBSD) to  
get the current processor ID from a userspace application?

Bye,
Gwenole.

[B2-devel] SheepShaver with AltiVec emulation in CVS

[Gwenole B. <gb...@di...>]

Hi,

I have just committed AltiVec emulation to CVS. It is *not* optimized 
yet but provides already a speed up factor of 1.8x on AltiVec Fractal 
Carbon v1.2. Now that we have AltiVec emulation, I also set the PVR to 
match a 7400 (G4) processor. Tell me if this causes problems.

I have not tried a SheepShaver build on x86 yet, but the Kheperix 
testsuite was expanded to include more than 800K testcases, and they 
all pass on both ppc & x86. So, it should just work.

Note that I make intensive use of templates. i.e. only 15 templates 
expand to the whole 150+ instruction handlers. That said, you will need 
a C++ compiler good enough to cope with that, both ISO C++ conformance 
wise and optimization wise. FYI, I used gcc 3.2.2.

Bye,
Gwenole.

[B2-devel] Re: Sheepshaver for W32

[Toshimitsu T. <t_t...@db...>]

Guten Tag.

On 2004/02/04, at 22:34, B=E4ssem Salem wrote:
> I understand you could build and run sheepshaver on Windows/x86. Is=20
> there any
> information or tutorial how to do this?

No.
Currently,SheepShaver does not run on Windows.

I use VMware, and I am running SheepShaver on Linux/x86 on Windows.
http://www006.upp.so-net.ne.jp/t_tanaka/images/sswin.jpg

VMware is the virtual machine software for Windows.
http://www.vmware.com/

Best regards,
---
Toshimitsu Tanaka
t_t...@db...
http://member.nifty.ne.jp/poseidon/index.html

[B2-devel] Sheepshaver for W32

[<bas...@we...>]

Hello Mr. Tananka,

I understand you could build and run sheepshaver on Windows/x86. Is 
there any
information or tutorial how to do this? I would appreciate any 
instructions or helpful
propositions. Though I found a website covering this aspect, it was only 
available
in japanese, which i don't understand well enough.
I would be glad for any hints or replies,

Thank you very much,
Yours,

Bassem Salem
Germany

[B2-devel] Darwin port GUI

[Toshimitsu T. <t_t...@db...>]

Hi.

The GUI cannot be enabled in Darwin port of SheepShaver.

I installed GTK+ and esound using Fink.
I got source code of SheepShaver from CVS and modified some files.

Make was successful.
The Panel is displayed.
http://www006.upp.so-net.ne.jp/t_tanaka/images/settingpanel.jpg

However, after clicking the "Start" button,
SheepShaver says "Cannot map RAM:Unknown error:0".

What's wrong?
------------------------------------------------------------------------ 
-----------------------------------------------
--- main_unix.org	Sat Jan 31 20:10:49 2004
+++ main_unix.cpp	Sun Feb  1 22:11:54 2004
@@ -493,8 +493,10 @@
  		}
  		fclose(proc_file);
  	} else {
+#if defined(__linux__)
  		sprintf(str, GetString(STR_PROC_CPUINFO_WARN), strerror(errno));
  		WarningAlert(str);
+#endif
  	}
  #endif
  	D(bug("PVR: %08x (assumed)¥n", PVR));

--- prefs_editor_gtk.org	Tue Jan 13 00:37:22 2004
+++ prefs_editor_gtk.cpp	Sun Feb  1 21:57:58 2004
@@ -815,7 +815,11 @@
  	if (d) {
  		struct dirent *de;
  		while ((de = readdir(d)) != NULL) {
+#if defined(__linux__)
  			if (strncmp(de->d_name, "ttyS", 4) == 0 || strncmp(de->d_name,  
"lp", 2) == 0) {
+#else
+			if (false) {
+#endif
  				char *str = new char[64];
  				sprintf(str, "/dev/%s", de->d_name);
  				glist = g_list_append(glist, str);
@@ -846,7 +850,13 @@
  			struct ifreq req, *ifr = ifc.ifc_req;
  			for (int i=0; i<ifc.ifc_len; i+=sizeof(ifreq), ifr++) {
  				req = *ifr;
+#if defined(__linux__)
  				if (ioctl(s, SIOCGIFHWADDR, &req) == 0 && req.ifr_hwaddr.sa_family  
== ARPHRD_ETHER) {
+#elif defined(__APPLE__) && defined(__MACH__)
+				if (ioctl(s, SIOCGIFADDR, &req) == 0 && (req.ifr_addr.sa_family ==  
ARPHRD_ETHER || req.ifr_addr.sa_family == ARPHRD_ETHER+1)) {
+#else
+				if (false) {
+#endif
  					char *str = new char[64];
  					strncpy(str, ifr->ifr_name, 63);
  					glist = g_list_append(glist, str);

--- user_strings_unix.org	Tue Jan 13 00:37:22 2004
+++ user_strings_unix.cpp	Sun Feb  1 21:51:46 2004
@@ -25,10 +25,10 @@
  // Platform-specific string definitions
  user_string_def platform_strings[] = {
  	// Common strings that have a platform-specific variant
-	{STR_VOLUME_IS_MOUNTED_WARN, "The volume '%s' is mounted under Linux.  
Basilisk II will try to unmount it."},
-	{STR_EXTFS_CTRL, "Linux Root"},
-	{STR_EXTFS_NAME, "Linux Directory Tree"},
-	{STR_EXTFS_VOLUME_NAME, "Linux"},
+	{STR_VOLUME_IS_MOUNTED_WARN, "The volume '%s' is mounted under Unix.  
SheepShaver will try to unmount it."},
+	{STR_EXTFS_CTRL, "Unix Root"},
+	{STR_EXTFS_NAME, "Unix Directory Tree"},
+	{STR_EXTFS_VOLUME_NAME, "Unix"},

  	// Purely platform-specific strings
  	{STR_NO_DEV_ZERO_ERR, "Cannot open /dev/zero: %s."},
---
Toshimitsu Tanaka
t_t...@db...


---
田中 俊光
QYT...@ni...
t_t...@db...
http://member.nifty.ne.jp/poseidon/index.html

Re: [B2-devel] Archive List

[Christian B. <Chr...@un...>]

Hi!

On Fri, Jan 30, 2004 at 02:04:08PM -0500, Mark Hart wrote:
> Is there an archive that I can go back and read what I've missed so far?

http://sourceforge.net/mailarchive/forum.php?forum_id=7009

Bye,
Christian

-- 
  / Physics is an algorithm
\/ http://www.uni-mainz.de/~bauec002/