|
»
« Return to Thread: [bug #35467] Not sure this the right plce to submit this but...
Re: [bug #35467] Not sure this the right plce to submit this but...
by David Brown-4
::
Rate this Message:
On 08/02/2012 17:12, Daved wrote:
> Thanks for the responses guys. It's not always clear where to post
> problems with this stuff.
>
> I have since done a general web search for the problem and did find a
> code correction for w5100.h. I was about to try it when my Arduino
> ATmega328P chip went bad and I am now waiting for another one in the
> mail. It's still puzzling why an earlier complier version works while a
> newer one does not.
>
Personally, I would not be happy without knowing what has changed, why
the old w5100.h did not work with the newer compiler, and why the new
w5100.h does work. These things are usually a case of errors in the
source code (such as missing volatiles, or strict aliasing issues) that
play badly with the newer optimiser. But I would want to know exactly
what the problem is, so that I could be sure it won't occur elsewhere in
the same program, and so that I wouldn't make the same mistake myself in
the future.
Since you have a new version of the header file that fixes the issue (at
least, that's the idea), then the first thing to do is compare the old
and new headers and find the differences. The chances are the it is
only a small part that has been changed.
mvh.,
David
> Anyway, I have already spent way too much time tracking this down so
> other than testing the code correction when I get a new chip, I am going
> to let it rest
>
> Dave
>
> ----- Original Message ----- From: "Joerg Wunsch" <j@...>
> To: "David Brown" <david@...>
> Cc: "David DeHaven" <dehaveda@...>; <avr-libc-dev@...>
> Sent: Wednesday, February 08, 2012 5:46 AM
> Subject: Re: [avr-libc-dev] [bug #35467] Not sure this the right plce to
> submit this but...
>
>
>> As David Brown wrote:
>>
>>> For future reference, remember that when you make a bug report, the
>>> "summary" is critical - it lets people see at a glance what the problem
>>> is, it is used for searching for similar problems, and is used in
>>> reports of outstanding issues.
>>
>> Thanks for the reminder, I changed the subject in the bug tracker even
>> though it's closed, for the benefit of those who are searching old bug
>> reports.
>> --
>> cheers, J"org .-.-. --... ...-- -.. . DL8DTL
>>
>> http://www.sax.de/~joerg/ NIC: JW11-RIPE
>> Never trust an operating system you don't have sources for. ;-)
>
_______________________________________________
AVR-libc-dev mailing list
AVR-libc-dev@...
https://lists.nongnu.org/mailman/listinfo/avr-libc-dev
> Thanks for the responses guys. It's not always clear where to post
> problems with this stuff.
>
> I have since done a general web search for the problem and did find a
> code correction for w5100.h. I was about to try it when my Arduino
> ATmega328P chip went bad and I am now waiting for another one in the
> mail. It's still puzzling why an earlier complier version works while a
> newer one does not.
>
Personally, I would not be happy without knowing what has changed, why
the old w5100.h did not work with the newer compiler, and why the new
w5100.h does work. These things are usually a case of errors in the
source code (such as missing volatiles, or strict aliasing issues) that
play badly with the newer optimiser. But I would want to know exactly
what the problem is, so that I could be sure it won't occur elsewhere in
the same program, and so that I wouldn't make the same mistake myself in
the future.
Since you have a new version of the header file that fixes the issue (at
least, that's the idea), then the first thing to do is compare the old
and new headers and find the differences. The chances are the it is
only a small part that has been changed.
mvh.,
David
> Anyway, I have already spent way too much time tracking this down so
> other than testing the code correction when I get a new chip, I am going
> to let it rest
>
> Dave
>
> ----- Original Message ----- From: "Joerg Wunsch" <j@...>
> To: "David Brown" <david@...>
> Cc: "David DeHaven" <dehaveda@...>; <avr-libc-dev@...>
> Sent: Wednesday, February 08, 2012 5:46 AM
> Subject: Re: [avr-libc-dev] [bug #35467] Not sure this the right plce to
> submit this but...
>
>
>> As David Brown wrote:
>>
>>> For future reference, remember that when you make a bug report, the
>>> "summary" is critical - it lets people see at a glance what the problem
>>> is, it is used for searching for similar problems, and is used in
>>> reports of outstanding issues.
>>
>> Thanks for the reminder, I changed the subject in the bug tracker even
>> though it's closed, for the benefit of those who are searching old bug
>> reports.
>> --
>> cheers, J"org .-.-. --... ...-- -.. . DL8DTL
>>
>> http://www.sax.de/~joerg/ NIC: JW11-RIPE
>> Never trust an operating system you don't have sources for. ;-)
>
_______________________________________________
AVR-libc-dev mailing list
AVR-libc-dev@...
https://lists.nongnu.org/mailman/listinfo/avr-libc-dev
« Return to Thread: [bug #35467] Not sure this the right plce to submit this but...
| Free embeddable forum powered by Nabble | Forum Help |
