Dear All,
> The new code (some time before 2.6.3) is more pessimistic and always
> generates banksels unless it is sure that a banksel is not required.
> Thus 2.6.3 emits more banksels than 2.6.0.
> The patch contributed by Alex identifies another class of redundant
> BANKSELs and removes them.
I
Thomas,
We're all waiting for the SourceForge Compile Farm to
come back online. They do not have a schedule when it's
expected to be back.
Maarten
> Dear all,
>
> where can I download an snapshot-File of SDCC build in 2007 for windows?
>
> Best regards
>
> Thomas
>
>
> ---
On Thursday 25 January 2007 12:10, Raphael Neider wrote:
> Am Donnerstag, den 25.01.2007, 15:57 +0530 schrieb george john:
> > But the problem is with the compiler if we compile the above code
> > with sdcc 2.6.0 it will work but with sdcc 2.6.3 it will not work.
>
And I'd say its not as much a p
Am Donnerstag, den 25.01.2007, 15:57 +0530 schrieb george john:
> But the problem is with the compiler if we compile the above code
> with sdcc 2.6.0 it will work but with sdcc 2.6.3 it will not work.
After 2.6.0 was released I meddled with the banksel generation code in
the compiler because the
But the problem is with the compiler if we compile the above code with sdcc
2.6.0 it will work but with sdcc 2.6.3 it will not work.
On 1/25/07, Ernst Bachmann <[EMAIL PROTECTED]> wrote:
On Thursday 25 January 2007 08:50, george john wrote:
> > EECON2 = 0x55; // Wr
> Latest snapshots only available trough SVN ?
Seemingly. According to the SF Compile Farm Status Page
http://sourceforge.net/docman/display_doc.php?docid=10472&group_id=1#top, the
compile farm is more or less down since 2007-01-05.
SVN works, however.
Regards,
Raphael
--
Hi Raphael,
Latest snapshots only available trough SVN ?
Regards,
KTy
Quoting Raphael Neider <[EMAIL PROTECTED]>:
> Hi,
>
> > Depending on the compiler optimization, redundand banksels etc, your EECON2
> > writes won't be translated into the exact reqired ASM sequence.
> >
> > Please check th
> Several days ago, I have created a patch what can suppress most of the
> banksel instructions only in the pic14 port.
Sorry, I got your mail (2007-01-15) on the devel list, but simply forgot
to act...Thanks for the reminder.
Patch applied in SDCC r4595, thanks for it. It just solved an issue
re
Hi,
> Depending on the compiler optimization, redundand banksels etc, your EECON2
> writes won't be translated into the exact reqired ASM sequence.
>
> Please check the intermediate ASM code generated by SDCC if that is the case.
Ernst is right, SDCC inserts a redundant BANKSEL before updating
On Thursday 25 January 2007 08:50, george john wrote:
> > EECON2 = 0x55; // Write "password" to EECON2
> > EECON2 = 0xAA;
> > WR = 1; // Initiate a write cycle
> >
Well, the PIC Datasheet claims that an exact timed sequence is ne
10 matches
Mail list logo
