RE: octave configure on cygwin snapshot

2007-08-30 Thread Dave Korn
On 30 August 2007 01:40, Eric Blake wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > According to Dave Korn on 8/29/2007 11:00 AM: >> >> It would be worth going back through the archives; IIRC, this was a >> recent change made by CV in response to a PR on the list. Any un-cautious

Re: octave configure on cygwin snapshot

2007-08-29 Thread Eric Blake
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Dave Korn on 8/29/2007 11:00 AM: > > It would be worth going back through the archives; IIRC, this was a recent > change made by CV in response to a PR on the list. Any un-cautious solution > would probably just regress back the origin

RE: octave configure on cygwin snapshot

2007-08-29 Thread Dave Korn
On 29 August 2007 17:51, Eric Blake wrote: > Eric Blake byu.net> writes: > >> As your testcase shows, the inclusion of manages to define >> struct timezone, but then subsequent includes trigger the macro that >> changes the spelling to _timezone, and your declarations all end up >> referring to

Re: octave configure on cygwin snapshot

2007-08-29 Thread Eric Blake
Eric Blake byu.net> writes: > As your testcase shows, the inclusion of manages to define struct > timezone, but then subsequent includes trigger the macro that changes the > spelling to _timezone, and your declarations all end up referring to the > incomplete type struct _timezone, hence the

Re: octave configure on cygwin snapshot

2007-08-29 Thread Eric Blake
Marco Atzeri yahoo.it> writes: > Right assumption. This test is performed by configure > to verify is gettimeofday have one or 2 arguments. > As in the snapshot the program fail to compile, > configure wrongly assume that gettimeofday have only > one argument. I'm not sure of any platforms where

RE: octave configure on cygwin snapshot

2007-08-29 Thread Marco Atzeri
--- Dave Korn <[EMAIL PROTECTED]> ha scritto: > On 29 August 2007 13:24, Eric Blake wrote: > > > -BEGIN PGP SIGNED MESSAGE- > > Hash: SHA1 > > > > According to Marco Atzeri on 8/29/2007 1:02 AM: > >> > >> gcc -o conftest.exe -g -O2 conftest.c -ldl > -lwsock32 > >> -lz -lm -lwsock32 >

Re: octave configure on cygwin snapshot

2007-08-29 Thread Eric Blake
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Dave Korn on 8/29/2007 6:33 AM: > Well in the absence of any further information I would assume that's what's > in the attached conftest.c. > > > Eric? I think you haven't had enough caffeine this morning/${TZ}! Oh - there was an a

RE: octave configure on cygwin snapshot

2007-08-29 Thread Dave Korn
On 29 August 2007 13:24, Eric Blake wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > According to Marco Atzeri on 8/29/2007 1:02 AM: >> >> gcc -o conftest.exe -g -O2 conftest.c -ldl -lwsock32 >> -lz -lm -lwsock32 > > Why are you linking against -lwsock32? Well in the absence o

Re: octave configure on cygwin snapshot

2007-08-29 Thread Eric Blake
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Marco Atzeri on 8/29/2007 1:02 AM: > > gcc -o conftest.exe -g -O2 conftest.c -ldl -lwsock32 > -lz -lm -lwsock32 Why are you linking against -lwsock32? That generally doesn't work (you shouldn't mix Windows sockets and cygwin sockets,