[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [avr-gcc-list] dev and mod may not be optimized
|
From: |
Weddington, Eric |
|
Subject: |
RE: [avr-gcc-list] dev and mod may not be optimized |
|
Date: |
Wed, 12 Dec 2007 07:04:44 -0700 |
> -----Original Message-----
> From:
> address@hidden
> [mailto:address@hidden
> org] On Behalf Of Nathan Moore
> Sent: Tuesday, December 11, 2007 9:26 AM
> To: address@hidden
> Subject: RE: [avr-gcc-list] dev and mod may not be optimized
>
>
> Thank you for your explination Bjoern.
> I have looked at GCC internals before, but never worked on
> them, so I'm
> likely to say something stupid.
> I would have thought that an optimization pattern for just
> this type of
> thing would have already been in GCC, especially since on the internal
> helper function is not unlike the divide instruction on x86,
> returning 2
> values rather than one. I would think that peephole
> optimization would
> be all over this.
But it requires someone to write that peephole optimization for the AVR.
- [avr-gcc-list] dev and mod may not be optimized, Nathan Moore, 2007/12/11
- AW: [avr-gcc-list] dev and mod may not be optimized, Haase Bjoern (PT/EMM1), 2007/12/11
- Re: [avr-gcc-list] dev and mod may not be optimized, Albert Andras, 2007/12/11
- Re: [avr-gcc-list] dev and mod may not be optimized, Gre7g Luterman, 2007/12/11
- Re: [avr-gcc-list] dev and mod may not be optimized, Paulo Marques, 2007/12/12
- Re: [avr-gcc-list] dev and mod may not be optimized, Paulo Marques, 2007/12/12
- Re: [avr-gcc-list] dev and mod may not be optimized, Gre7g Luterman, 2007/12/12
- Re: [avr-gcc-list] dev and mod may not be optimized, Paulo Marques, 2007/12/12