[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [avr-gcc-list] GCC-AVR Register optimisations
|
From: |
Dmitry K. |
|
Subject: |
Re: [avr-gcc-list] GCC-AVR Register optimisations |
|
Date: |
Fri, 11 Jan 2008 08:53:39 +1000 |
|
User-agent: |
KMail/1.5 |
Hi.
For original variant (foo(p++)) the avr-gcc 4.0.4
produces the best:
> > movw r24, r16
> > adiw r24, 1
> > call foo
> > movw r24, r16
> > adiw r24, 2
> > etc..
3.3.6 and 3.4.6 are:
movw r24,r16
subi r16,...
sbci r17,...
And 4.1.2, 4.2.2 and 4.3-20080104 produces the ugly code.
Regards,
Dmitry.
- [avr-gcc-list] Tablejumps - needless run time conversion to byte address, Andrew Hutchinson, 2008/01/04
- RE: [avr-gcc-list] GCC-AVR Register optimisations, Weddington, Eric, 2008/01/10
- [avr-gcc-list] Tip: handling volatile operands, andrewhutchinson, 2008/01/10
- Re: [avr-gcc-list] Tip: handling volatile operands, David Kelly, 2008/01/10
- Re: [avr-gcc-list] Tip: handling volatile operands, Andrew Hutchinson, 2008/01/10
- RE: [avr-gcc-list] Tip: handling volatile operands, Dave Hansen, 2008/01/11
- Re: [avr-gcc-list] Tip: handling volatile operands, Dave N6NZ, 2008/01/11
- Re: [avr-gcc-list] Tip: handling volatile operands, Paulo Marques, 2008/01/11
- RE: [avr-gcc-list] GCC-AVR Register optimisations, John Regehr, 2008/01/10