Previous Thread
Next Thread
Print Thread
Page 10 of 10 1 2 8 9 10
Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
I'm just impressed that we've found someone who'll defend the x86 instruction set.

Joined: Feb 2004
Posts: 2,608
Likes: 315
Very Senior Member
Very Senior Member
Joined: Feb 2004
Posts: 2,608
Likes: 315
PowerPC is almost always dest,src,src, except in the case of a store, where it's src,dest,dest. OTOH, SPARC is almost always src,src,dest except for compare/exchange. Fun, huh?

I think the argument about order of source and destination is pointless - most people argue for the form they learn with. What does bother me is order of subtractions. I think PowerPC gets this right:

subfic rT,rA,IMM16

You think "subtract from immediate and carry", so the operation is IMM16 - rA => rT. And then there's:

sub rT,rA,rB

Where you think "subtract", and the operation is rA - rB => rT. But PICmicro gets it wrong. Look at this:

sublw IMM8

Now you would think, from the mnemonic, that it's "subtract literal from w", but it actually does the opposite. The operation is IMM8 - w => w. The designer probably thought something like "subtract: literal - w", but it's still ambiguous and/or misleading.

But big-Endian is the one true way - it's far easier to read in a debugger, and you find bugs where a person does things like *(INT32*)ptr8 when they mean (INT32)*ptr8 much faster, as that doesn't come close to working on a big-Endian system.

Joined: Sep 2004
Posts: 392
Likes: 4
A
Senior Member
Senior Member
A Offline
Joined: Sep 2004
Posts: 392
Likes: 4
Originally Posted by R. Belmont
I'm just impressed that we've found someone who'll defend the x86 instruction set.

There's a big difference between defending the operand order in the mnemonics and defending the instruction set!

Joined: Sep 2006
Posts: 89
M
Member
Member
M Offline
Joined: Sep 2006
Posts: 89
Originally Posted by R. Belmont
Have you verified that? The build is newer (I have 5341 on PowerPC and 5367 on Intel) and Dave Dribin said they definitely fixed some reported bugs.

I did a recompile of u1 right after a clean install of Leopard and vr was the second thing I checked (after the mandatory pacman test). I only ran vr for about 10 seconds, but I was definitely looking at a car floating over water.

(Update: I tossed my u1 source directory and am currently compiling a fresh copy. I will post results later.)

Last edited by M. Twitty; 11/04/07 04:01 PM.
Joined: Sep 2000
Posts: 255
W
Senior Member
Senior Member
W Offline
Joined: Sep 2000
Posts: 255
Originally Posted by M. Twitty
I did a recompile of u1 right after a clean install of Leopard and vr was the second thing I checked (after the mandatory pacman test). I only ran vr for about 10 seconds, but I was definitely looking at a car floating over water.

Yes car over water and if you start a game you get a car floating over plain green color. That's been the case for me on both Intel and PPC Macs, compiled under Leopard or Tiger with recent builds of sdlmame. So it doesn't seem as if anything has really changed.

What exactly is the issue with the version of gcc that VR needs to compile and work properly?



=will=
Joined: Sep 2006
Posts: 89
M
Member
Member
M Offline
Joined: Sep 2006
Posts: 89
RB, et al:

Made a fat binary with a clean source tree, untouched XCode 3.0 installation and vr is still a no-go.

It's disappointing that they stuck with gcc 4.0.1, but at least they upgraded to php version 5.

Joined: Aug 2006
Posts: 423
D
Senior Member
Senior Member
D Offline
Joined: Aug 2006
Posts: 423
Leopard definitely has a newer gcc 4.0.1 than Tiger that fixed some bugs. I know it fixed the -mstackrealign bug I ran into.

-Dave


Joined: Aug 2006
Posts: 423
D
Senior Member
Senior Member
D Offline
Joined: Aug 2006
Posts: 423
Originally Posted by R. Belmont
I think it's irritating that Xcode 3 doesn't run on Tiger.

FWIW, Xcode 3 is written in Objective-C 2.0, with garbage-collection, which isn't available in Tiger. Oh, and Xcode 3 can no longer target 10.2 (and it's a 64-bit app).

Xcode 2.5 is also out, which you can install side-by-side with 3 on Leopard as well as Tiger. I'm not sure which gcc that comes with, if it's even any different than 2.4.1. I haven't installed it, yet.

-Dave


Joined: Mar 2001
Posts: 17,239
Likes: 263
R
Very Senior Member
Very Senior Member
R Offline
Joined: Mar 2001
Posts: 17,239
Likes: 263
Yeah, I upgraded both my PPC and Intel dev machines to Leopard now so 2.5 isn't as interesting. (Oddly the install on my G5 locked up with fans a-blazing the first try, but it was fine the second. The Intel Mini was fine).

Still, ObjC 2.0 seems quixotic at best - if I were Apple I'd offer money and engineering time to Novell to get better OS X integration in Mono. The ability to write Linux GUI apps in C# with forms has done wonders for the platform even at this relatively early stage.

Page 10 of 10 1 2 8 9 10

Moderated by  R. Belmont 

Link Copied to Clipboard
Who's Online Now
3 members (Duke, Dorando, yugffuts), 55 guests, and 2 robots.
Key: Admin, Global Mod, Mod
ShoutChat
Comment Guidelines: Do post respectful and insightful comments. Don't flame, hate, spam.
Forum Statistics
Forums9
Topics9,331
Posts122,197
Members5,077
Most Online1,283
Dec 21st, 2022
Our Sponsor
These forums are sponsored by Superior Solitaire, an ad-free card game collection for macOS and iOS. Download it today!

Superior Solitaire
Powered by UBB.threads™ PHP Forum Software 8.0.0