From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Mortimer, Andy" To: 'cygwin' Subject: RE: patch for ash mem fault Date: Thu, 28 Sep 2000 02:38:00 -0000 Message-id: <2E300728EEF9D3119D8000E018C1864F5E427D@gemini> X-SW-Source: 2000-09/msg00933.html Hi, > > I've checked that this fixes the problem I was having; > > however, I haven't done much more testing than that single > > (admittedly many thousand line) script. > > Would you mind to send an example which allows me to test > that? I never saw that problem, so... Unfortunately, I never managed to get it down to a simple reproducible testcase. I managed to get a reproducible case, but unfortunately (as I mentioned above) that's a major script, and it only happens for some inputs anyway. Erm. If there's somebody around who really knows the ash internals it might be possible for them to cook up an example with a more complete description of the symptoms, but I'm afraid that doesn't describe me! :-( All I can think of to say is that the error was occurring in a line: PROBLIST=`cat $PROBFILE` and that, by controlling the size of $PROBFILE (around 504 bytes is the magic number) I could make it either fault or not. Obviously the above statement works most of the time, though; it's something to do with ash's internal memory tracking, and I really can't find a way to decompose it. Sorry, I know it's not very useful! I'll certainly pass it on if I can come up with a test case. Regards, Andy -- Andy Mortimer, CFX-5 Architecture and Infrastructure Team andy.mortimer@aeat.com -- Want to unsubscribe from this list? Send a message to cygwin-unsubscribe@sourceware.cygnus.com