From: sophana <jobarjo78_at_yahoo.fr>
Date: Wed, 26 May 2004 09:25:38 +0200

----- Original Message -----
From: "Linus Nielsen Feltzing" <linus_at_haxx.se>
To: "Rockbox development" <rockbox_at_cool.haxx.se>
Sent: Wednesday, May 26, 2004 7:35 AM
Subject: Re: found a bug in power_off()


> sophana wrote:
> > I found out that when I remove the while(1); at the end of the
> > routine, it works... :-)
>
> I had the same experience once, but as soon as I started debugging it,
> the problem disappeared. And now I see it again. Let's hope that I find
> it this time...
>
the -g flag disable some optimizations I think.

> > This must be caused by the optimizer optmizing too much, even if the
> > and_b and or_b routine are in assembly inline. (the optimizer is
> > generally working at the assembler level I think)
>
> What makes you think that? Did you disassemble it? I just did, and I
> can't see anything wrong with it.
No, I did not disassemble. But it seems probable, or should I say I have no
other explaination.
Are you sure you have the same optimization flags when you disassemble?
do you use obj dump?
I don't know about sh1 assembly but I'll take a look

>
> Linus
> _______________________________________________
> http://cool.haxx.se/mailman/listinfo/rockbox

_______________________________________________
http://cool.haxx.se/mailman/listinfo/rockbox
Received on 2004-05-26