There appears to be a 'simple bug in the semaphore implementation of linuxthreads-2.0.6' (I quote this from the glibc bug archive - url is below). I think there are lots of people running this code, so i figure i should conspire with the rest of the industry, to tell the public of this d4nger0uz bug. I did a bugtraq archive search and did n0t see any reportz of this, so I am guessing "no-one knows about this". umm, yah.. my postz to bugtraq were moderated last week, but you can find them here at --> www.big.net.au/~silvio/monkey-Bussines.txt www.big.net.au/~silvio/gg.txt pleaze send me your repliez. # P4RT 1 -------- http://sources.redhat.com/ml/bug-glibc/2001-05/msg00109.html has some information on thiz bug. From the linuxthreads ChangeLog. 2001-05-25 Ulrich Drepper <drepperat_private> * oldsemaphore.c (__old_sem_wait): Clear p_nextwaiting before returning successfully. Patch by Gene Cooperman <geneat_private>. yah.. this bug can crash your code in lotz of places, so I suggest you use some conditional compilation or something.. oh yah, this leads to part two of my advisory --> # P4RT 2 -------- Redhat 7.0 incorrectly sets the glibc minor version in the include filez etc.. It uses a modified glibc 2.1, but marks the minor version as 2, so you might still have this bug even if you try to work around it. Redhat 7.1 moves over to glibc 2.2 for real this time, where the bug is fixed. -- Silvio Communicate in total privacy. Get your free encrypted email at https://www.hushmail.com/?l=2 Looking for a good deal on a domain name? http://www.hush.com/partners/offers.cgi?id=domainpeople
This archive was generated by hypermail 2b30 : Tue Jun 11 2002 - 10:43:56 PDT