Re: Stupid Intel question
Re: Stupid Intel question
- Subject: Re: Stupid Intel question
- From: "Frederik Slijkerman" <email@hidden>
- Date: Sat, 18 Mar 2006 11:27:53 +0100
Hi Urs,
I started the thread on Xcode-users about strict aliasing. You should know
that there is a problem in AUElement::RestoreState that means that you can't
use strict aliasing (and therefore also -ftree-vectorize) for now, unless
you're willing to modify the CoreAudio SDK. I'll file a bug on that.
Regarding the optimization, I've noticed too in the disassembler that GCC 4
seems to produce rather sub-optimal code for Intel, even with -O3 and all
the other optimization stuff on. Luckily the processor is quite fast so it
doesn't matter that much at the moment, but it would be nice if this is
improved in the future, or if it turns out to be due to a particular
compiler switch.
Cheers,
Frederik Slijkerman
FabFilter
----- Original Message -----
From: "Urs Heckmann" <email@hidden>
To: "Steve Checkoway" <email@hidden>
Cc: "Core Audio" <email@hidden>
Sent: Friday, 17 March 2006 23:27
Subject: Re: Stupid Intel question
Hi Steve,
schrieb Steve Checkoway:
As was just discussed on the Xcode-users list, -ftree-vectorize
enables -fstrict-aliasing which is breaking code in the CoreAudio SDK.
To quote Eric Albert:
We turn off strict aliasing by default because, as you've noticed, it
breaks a lot of code (even though it's more correct and results in
faster code). But we explicitly turn it on for -O2 and higher
when -ftree-vectorize is passed to the compiler, as you're doing here.
I'm not sure why, but it doesn't seem right to me, particularly
when -fno-strict-aliasing is passed.
Thanks for the info! I was almost going nuts about it.
It's interesting to note that you're having problems on Intel but not
PowerPC as it was the other way around in the thread "GCC 4 and strict
aliasing" in the Xcode list.
I know this wasn't much help, but it seems that for now you should
probably not use -ftree-vectorize nor -fstrict-aliasing, at least until
the code and/or the compiler gets fixed.
I've used it on Panther/gcc3 without problems. Havn't tried the UBs on
the PowerPC yet, so I don't know if it's corrosive there too.
My code is noticably faster with strict-aliasing, so I hope it'll be
fixed soon!
Cheers,
;) Urs
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden