This is the mail archive of the
mailing list for the Cygwin project.
Re: Weird behavior when using "-I" with gcc-2.95.2
I've just seen a problem in cygwin that could account for this behavior.
It would be interesting to see if tonight's snapshot improves this
On Wed, Apr 26, 2000 at 02:56:21PM -0500, Mumit Khan wrote:
>On Tue, 25 Apr 2000, Fontenot Larry A DLVA wrote:
>> Attached is the requested bug.txt. It looks as though the default search
>> paths are still valid. And everything works just fine until I add a "-I
>> anypath". The only way to get the compiler to work properly at that point
>> is to provide both "-I /usr/i686-pc-cygwin" and "-I /usr/include/g++-3".
>Thanks for the information. However, I don't see anything wrong with
>your setup, so we'll just have to do some experiments and track it
>I noticed that you've set up the mount table to mount the various
>drives as /<drive_letter>, which should be fine. Also, PWD is set
>to /i/CPP/hello, which means you're working on a remote drive at
>this point, and your "home" directory is also on a server ie remote
>Could you please do the following -- run the same testcase on a local
>drive and see if the problem is repeatable. I'd like to remove as many
>variables as possible, and that may help tracking this down. As you
>can imagine, lots of use GCC and we don't see this problem.
>I think the `invalid argument' is simply a misleading error message,
>and the real error is still lurking in there. The final experiment
>will be to run the compilation under `strace', but that's going to
>generate a *huge* log file and I'm trying to avoid that.
>Want to unsubscribe from this list?
>Send a message to firstname.lastname@example.org
email@example.com Cygnus Solutions, a Red Hat company
Want to unsubscribe from this list?
Send a message to firstname.lastname@example.org