[prev in list] [next in list] [prev in thread] [next in thread] 

List:       pcc-list
Subject:    Re: initializer regresssion
From:       Anders Magnusson <ragge () ludd ! ltu ! se>
Date:       2010-06-03 19:02:15
Message-ID: 4C07FC37.7060703 () ludd ! ltu ! se
[Download RAW message or body]

Jesus Sanchez wrote:
> El 03/06/2010 17:05, Anders Magnusson escribió:
>> Iain Hibbert wrote:
>>> Hi,
>>>
>>> I see a recent regression (in the last few days, I think .. sorry no 
>>> time
>>> to investigate more just now as I have to go out for the rest of the 
>>> day)
>>>
>> Things ended up in the wrong order after recent changes :-/  Thanks, 
>> I've fixed it now.
>> There are still some pecularities that I may deal with in the future.
>>
>> -- Ragge
>
> man, you're just awesome, this list is like "I have this problem
> with PCC ..." and hours later ragge: "Thanks, I've fixed it now."
>
> seriously amazing.
>
Heh, thanks.  Actually, the compiler is not so much code so it's quite 
simple
to get into it.  ...And I have been hacking it quite much last years... :-)

-- Ragge

> -J
>
>>
>>> The following code fails with pcc/NetBSD/i386
>>>
>>> struct foo {
>>>     struct foo *f;
>>> };
>>>
>>> struct bar {
>>>     int i;
>>>     struct foo f;
>>> };
>>>
>>> struct bar b1 = {
>>>     1,
>>>     { &b1.f },    /* this is ok */
>>> };
>>>
>>> struct bar b2 = {
>>>     .i = 2,
>>>     .f = { &b2.f },    /* this is ok */
>>> };
>>>
>>> struct bar b3 = {
>>>     .f = { &b3.f },    /* "illegal combination of pointer and 
>>> integer" */
>>> };
>>>
>>> iain
>>>
>>>
>>>
>>
>>
>>
>
>

[prev in list] [next in list] [prev in thread] [next in thread]