reassign 504030 flex thanks On Fri, Oct 31, 2008 at 10:59:16AM -0500, Manoj Srivastava wrote: > Hi, > > Thinking about it, if the test was done with the same build > chain, and two versions of flex, and one fails, and the pther does not, > then the bug is really with flex. Please assign back if that is the > case.
Thanks Manoj. I tested in an i386 lenny chroot. I observed no failure when building w/ flex_2.5.35-2. I then made a single change - upgrading flex to 2.5.35-3, and the failure was reproducible. Reassigning back to flex as requested. On Fri, 31 Oct 2008 08:50:14 -0500, Manoj asked: > aaargh. Is this related to the buffer size increase, you > think? I can't say - I didn't look into the scope of changes between the two releases. I don't see a difference in redirect vs. pipe, I didn't see a smoking gun when looking at strace output, and its an issue observed equally on x86 and ia64. That leads me to believe its unrelated to the kernel change that uncovered the previous issue. However, it is possible that this is another symptom of whatever the root cause issue is behind #501960. -- dann frazier -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
