Date: Mon, 14 Feb 2000 16:25:33 -0600 (CST) From: Kevin Day <toasty@dragondata.com> To: hackers@freebsd.org Subject: Off by one error in ulimit? Message-ID: <200002142225.QAA00274@celery.dragondata.com>
next in thread | raw e-mail | index | archive | help
I don't have access to a -current machine to try this on, so this only applies to 3.4. I have my ulimit set to 8MB for data/etc: ulimit -a core file size (blocks) 8192 data seg size (kbytes) 8192 file size (blocks) 8192 max locked memory (kbytes) 4096 max memory size (kbytes) 8192 open files 24 pipe size (512 bytes) 1 stack size (kbytes) 2048 cpu time (seconds) 5400 max user processes 32 virtual memory (kbytes) 8192 I have a program that starts to attempt mallocing 8MB, and decreasing the malloc size by 1 byte every time malloc fails. I end up with a value somewhere between 7 and 8MB when malloc succeeds, which is fine. However, when I actually use this malloc'ed region, I segv when I touch the very last page. Can anyone explain this? :) As long as my malloc doesn't bump into the ulimit, everything works fine. Either malloc() doesn't take into account a fencepost situation somewhere, or I'm losing my mind. :) Kevin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200002142225.QAA00274>