-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Christopher Faylor on 9/26/2009 8:57 AM:
>> But how does it look now?
>
> It looks good. Thanks. Please check in.
I missed one corner case in my testing; how about this followup?
2009-09-29 Eric Blake
* syscalls.cc (rename
On Sep 29 13:18, Eric Blake wrote:
> I missed one corner case in my testing; how about this followup?
>
> 2009-09-29 Eric Blake
>
> * syscalls.cc (rename): Fix regression on rename("dir","d/").
Looks ok to me. Isn't that partly covered by the next if, though?
YA piece of code lacking c
Compiling a file which #include's in C89 mode fails:
$ echo "#include " > test.c
$ gcc -c test.c
$ gcc -c -std=c89 test.c
In file included from /usr/include/sys/signal.h:107,
from test.c:1:
/usr/include/cygwin/signal.h:74: error: expected
specifier-qualifier-list before 'pthrea
On Tue, Sep 29, 2009 at 03:50:53PM -0500, Yaakov (Cygwin/X) wrote:
>Compiling a file which #include's in C89 mode fails:
>
>$ echo "#include " > test.c
>$ gcc -c test.c
>$ gcc -c -std=c89 test.c
>In file included from /usr/include/sys/signal.h:107,
> from test.c:1:
>/usr/include/c
On 29/09/2009 17:33, Christopher Faylor wrote:
I see two possible solutions:
1) Unconditionally #include in (newlib), OR
2) #include in.
Since this appears to be Cygwin specific, I went for the latter. Patch
attached.
WDLD?
On IRC:
cygwinports: cgf: wtf WDLD?
cgf: What Does Linux Do?
c
On 29/09/2009 19:35, Yaakov (Cygwin/X) wrote:
Anyway, to answer the question, AFAICS in glibc, #include
unconditionally[1]. ( is just one line:
#include [2])
So should I take the first route, patching newlib instead?
OTOH, this comment in the offending hunk of (which is
only for Cygwin an
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Corinna Vinschen on 9/29/2009 1:35 PM:
> On Sep 29 13:18, Eric Blake wrote:
>> I missed one corner case in my testing; how about this followup?
>>
>> 2009-09-29 Eric Blake
>>
>> * syscalls.cc (rename): Fix regression on rename("dir
