| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <817d11390712080343v54bff28j8a08299f445b6905@mail.gmail.com> Date: Sat, 8 Dec 2007 13:43:37 +0200 From: "T T" <gentuu.call@...il.com> To: linux-kernel@...r.kernel.org Subject: Quotaon livelock Hi lkml, Enabling FS quotas can take up to several minutes to finish and depends on other users disk activity. This makes it a livelock (as it can be finished anyway). The problem described at: http://bugzilla.kernel.org/show_bug.cgi?id=9520 Yes, if we need to enable quotas once in the system start - no problem. But if we have to disable/enable it periodically - this become a nightmare. Please don't advice to use edquota or so (quotactl(Q_SETQUOTA...) based things). What if it's really needed to use quotactl(Q_QUOTAOFF...)/quotactl(Q_QUOTAON...) The problem is at fs/dquot.c:add_dquot_ref() It's restarting to loop over the inodes list once it inits quota on an inode. And this happens over and over if something is modifying inodes list during quotas enabling. It looks like the action must take some time depending on FS size, but not on other users disk activity... Any advices how the algorithm at fs/dquot.c:add_dquot_ref() can be enhanced? Thanks Nick -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@...r.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists