Mail Archives: cygwin-developers/2001/03/09/20:07:32
On Sat, Mar 10, 2001 at 08:17:12AM +0930, Trevor Forbes wrote:
>
>----- Original Message -----
>From: "Christopher Faylor" <cgf AT redhat DOT com>
>To: <cygwin-developers AT cygwin DOT com>
>Sent: Friday, 9 March 2001 6:56
>Subject: Re: Outstanding issues with current DLL?
>
>
>> On Fri, Mar 09, 2001 at 06:17:32AM +0930, Trevor Forbes wrote:
>> >Anyway, did you do a make check on winsup...
>> >I did and it is failing in two areas
>> >
>> > X_OK is defined as a constant unsigned in file.h cygwin_S_IEXEC.cc
>> >and /src/cygwin/src/winsup/testsuite/winsup.api/ltp/access01.c:147:
>> >gives the error: initializer element is not constant
>> >
>> >file -- access01.c
>> >..
>> >static struct test_case_t {
>> > char *file;
>> > int mode;
>> > char *string;
>> > int experrno;
>> >} Test_cases[] = {
>> > { Fname, F_OK, "F_OK", 0 },
>> > { Fname, X_OK, "X_OK", 0 }, <------ access01.c:147
>> > { Fname, W_OK, "W_OK", 0 },
>> > { Fname, R_OK, "R_OK", 0 },
>> >};
>> >
>> >I think X_OK has to be an int in your "interesting" changes
>>
>> Actually, to circumvent this, it has to be a constant. That would
>> blow my change out of the water.
>
>Good, but can we fix the test. I have found the tests are very good
>indicator into the health of my cygwin1.dll build
It looks like something jumped to 0.
If you
set CYGWIN=error_start=x:\path\to\gdb
you'll get a gdb started rather than a stackdump. Then you may be
able to get a backtrace from various threads to see what the program
is up to.
If you have other settings in your CYGWIN, you might also try unsetting
them to see if it makes a difference.
cgf
- Raw text -