Re: [avr-gcc-list] [VOTE] objdump patch

[Keith Gudger]

I still vote for the option D, your first example.  It matches other
versions of machine specific GCC!  

IMHO it is much more readable than your preference.

Keith

On Thu, 25 Nov 2004, Svein E. Seldal wrote:

> On Monday 13 September 2004 17:55, Svein E. Seldal wrote:
> > Hi
> >
> > It seems like there are several preferences on how the new objdump with
> > symbols output should look like.
> >
> > To make a decision more fair (and to do a little experiment), I will now
> > arrange a vote on the syntax of the objdump patch. Here are the choices:
> 
> I have totally forgotten this vote, and will take action now. It seems that 
> the general interest is to have it on the "D" format.
> 
> I have changed the patch to use this format, and I have to say that I do not 
> think it looks too nice. In fact, I think it looks messy when you have a lot 
> of call's with branch instructions, since the symbol is presented in 
> different locations.
> 
> So I would like everybody to reconsider their decision.
> 
> 
> This is an example dump using the voted format:
> 
>      52c:       0e 94 4b 16     call    0x2c96 <slave_test>
>      530:       88 23           and     r24, r24
>      532:       89 f0           breq    .+34            ; 0x556 <main+0xcc>
>      534:       0e 94 28 17     call    0x2e50 <led_status>
>      538:       88 23           and     r24, r24
>      53a:       21 f4           brne    .+8             ; 0x544 <main+0xba>
>      53c:       0e 94 30 17     call    0x2e60 <led_set_enable>
>      540:       88 23           and     r24, r24
>      542:       19 f0           breq    .+6             ; 0x54a <main+0xc0>
>      544:       0e 94 96 0e     call    0x1d2c <serial_rx>
>      548:       06 c0           rjmp    .+12            ; 0x556 <main+0xcc>
>      54a:       0e 94 99 0e     call    0x1d32 <spi_trans>
>      54e:       80 e4           ldi     r24, 0x40       ; 64
>      550:       90 e0           ldi     r25, 0x00       ; 0
>      552:       0e 94 3e 16     call    0x2c7c <set_bit>
> 
> While this format IMHO looks better:
> 
>      52c:       0e 94 4b 16     call    0x2c96 <slave_test>
>      530:       88 23           and     r24, r24
>      532:       89 f0           breq    0x556 <main+0xcc>       ; .+34
>      534:       0e 94 28 17     call    0x2e50 <led_status>
>      538:       88 23           and     r24, r24
>      53a:       21 f4           brne    0x544 <main+0xba>       ; .+8
>      53c:       0e 94 30 17     call    0x2e60 <led_set_enable>
>      540:       88 23           and     r24, r24
>      542:       19 f0           breq    0x54a <main+0xc0>       ; .+6
>      544:       0e 94 96 0e     call    0x1d2c <serial_rx>
>      548:       06 c0           rjmp    0x556 <main+0xcc>       ; .+12
>      54a:       0e 94 99 0e     call    0x1d32 <spi_trans>
>      54e:       80 e4           ldi     r24, 0x40       ; 64
>      550:       90 e0           ldi     r25, 0x00       ; 0
>      552:       0e 94 3e 16     call    0x2c7c <set_bit>
> 
> 
> Svein
> 
> _______________________________________________
> avr-gcc-list mailing list
> address@hidden
> http://www.avr1.org/mailman/listinfo/avr-gcc-list
>