| From: | "Simon 'corecode' Schubert" <corecode@xxxxxxxxxxxx> |
| Date: | Tue, 12 Jul 2005 19:49:10 +0200 |
Lately YONETANI Tomokazu <qhwt+dfly@xxxxxxxxxx> said: > This is very weird; if I add `return False' BEFORE FD_ZERO() in > open_sockets_smbd(), it terminates with SIGTERM, but if I put the > `return False' AFTER FD_ZERO() and stack smash protector detects > the stack overflow and it catches SIGABRT. okay, it seems that gcc with -fstack-protector (default) and -O2 places the variables wrongly on the stack. I can give more insight if desired. Basically listen_set is at -72(%ebp), whereas the canary is already at -40(%ebp). sizeoF(fd_set) == 128, so figure :) this regression should be pushed upstream. joerg, do you handle this? cheers simon -- Serve - BSD +++ RENT this banner advert +++ ASCII Ribbon /"\ Work - Mac +++ space for low $$$ NOW!1 +++ Campaign \ / Party Enjoy Relax | http://dragonflybsd.org Against HTML \ Dude 2c 2 the max ! http://golden-apple.biz Mail + News / \
Attachment:
pgp00005.pgp
Description: PGP signature