"stat" system call problem

[Hsuan Chang]

It escaped me last when I had a glance at the mention of the "fstat"
problem a while back on SunSpots.  Now, I am having this terrible problem
with "stat", which is supposed to be bullet-proof.  I hope my problem is
related.

It goes like this:
		stat(file, &buf)
	returns a buffer via buf which has all the info concerning
	file.  buf.st_size is said to have the size of the file
	in bytes.  It worked fine until lately when I recompiled my
	program. It started to return 1 to me no matter what
	the size of the file is!  To make the problem even more
	annoying, it DID return a few times the correct size!
	No, I am not kidding! (frequency of getting it right: <1%)

I checked the latest version of TFM (dated: Last Change: 22 Nov '87
Release 4.0), and failed to find any trace of changes to the structure of
the buffer or to the specification of the function.  Well, does any one of
you folk have the same ugly experience?

Hsuan Chang  				hsc at vuse.vanderbilt.edu
Image Processing Lab
Computer Science Dept    
Vanderbilt University   
Nashville, Tn 37235    
(615) 322-[2949|0962]