Mail Archives: cygwin-developers/2001/07/26/20:18:55
On 26 Jul 2001 11:16:49 -0400, Jason Tishler wrote:
> Chris,
>
> On Thu, Jul 26, 2001 at 10:45:39AM -0400, Christopher Faylor wrote:
> > I suspect that a simple test case would help unperplex you.
>
> I did write many simple test cases before bothering the list. Attached
> is, wtest2, one that mimics fstat()'s behavior (with regard to the
> Win32 calls). I also stole the the values passed to CreateFile() from
> an strace run of stest (just in case how the file was opened caused
> the problem). For some reason, wtest2 does not affect atime. Hence,
> if CloseHandle() is indeed the culprit, then it isn't one all of the
> time.
My guess would be it's the file header reading logic - looking for
#!/bin/foo
that means that a file access _has occured_. CloseHandle triggering the
actual write might jsut eb a win32 optimisation, to only wr
> If you have any hints for a simple test case that would illuminate me,
> it would be much appreciated.
>
> > I also suspect that CloseHandle is updating the atime.
>
> That was my "conclusion" too from stepping via gdb -- but, I found it
> hard to believe. I searched the MSDN and google for MS atime issues
> but came up empty.
>
> Nevertheless, isn't the the bigger issue about how Cygwin should deal with
> this problem? If Cygwin's stat() affects st_atime (even indirectly),
> then this is going to break Unix apps. I have already found one (i.e.,
> mutt), there are bound to be others.
>
> Thanks,
> Jason
>
> --
> Jason Tishler
> Director, Software Engineering Phone: 732.264.8770 x235
> Dot Hill Systems Corp. Fax: 732.264.8798
> 82 Bethany Road, Suite 7 Email: Jason DOT Tishler AT dothill DOT com
> Hazlet, NJ 07730 USA WWW: http://www.dothill.com
- Raw text -