| delorie.com/archives/browse.cgi | search |
On Tue, Aug 06, 2002 at 07:01:37AM +0100, Conrad Scott wrote: >"Christopher Faylor" <cgf AT redhat DOT com> wrote: >> Yep. That was an interesting race. Like most races, it was >really >> obvious once I found it. > >Sometimes I feel deflated after finding bugs: so interesting to >find but boring and trivial once found (and then there's testing). > >> It seems to work ok now but I haven't tested it rigorously. I >was lulled >> into a false sense of security before when I really *had* tested >rxvt. >> I just didn't try it enough times to notice that it failed 50% >of the >> time at least. > >I'll add my 2c of testing in a moment. And thanks for the mount_h >patch: from where I'd got in my vacant gazing at the code it looks >right. I'm going to have to draw a bl**dy big picture of the >whole DLL initialization / fork / spawn system one day; either >that or continue to limp along and hope someone more knowledgable >will fix the problems I find (or create) :-) Hey when you track them down so well, it's easy enough to fix them. The mount_h stuff still isn't "right" now that I think about it but it's 2AM so I should probably stop coding. cgf
| webmaster | delorie software privacy |
| Copyright © 2019 by DJ Delorie | Updated Jul 2019 |