Messages in this thread | | | Date | Tue, 18 Mar 2003 14:00:19 -0800 | From | "Randy.Dunlap" <> | Subject | Re: bug in kernel/sysctl.c (SYSIPC) ? |
| |
On Tue, 18 Mar 2003 19:55:36 +0100 (CET) Der Herr Hofrat <der.herr@hofr.at> wrote:
| > On Tue, 18 Mar 2003 08:51:17 +0100 (CET) Der Herr Hofrat <der.herr@hofr.at> wrote: | > | > | atleast in kernel 2.4.19 and 2.4.20 in kernel/sysctl.c shmmax and shmall use | > | the proc_dointvec_minmax callback without passing a min/max value - is there | > | a reson for this or is this a simple bug ? | > | | > | linux/kenrel/sysctl.c (line 221 for 2.4.19/20) | > | | > | #ifdef CONFIG_SYSVIPC | > | {KERN_SHMMAX, "shmmax", &shm_ctlmax, sizeof (size_t), | > | 0644, NULL, &proc_doulongvec_minmax}, | > | {KERN_SHMALL, "shmall", &shm_ctlall, sizeof (size_t), | > | 0644, NULL, &proc_doulongvec_minmax}, | > | ... | > | #endif | > | > The min and max values default to 0 if not specified (initialized), | > and the _minmax functions have code to handle those cases. | > | > so as long as the intended min/max values were 0, I don't see a | > problem. Are you seeing a problem? | > | | not directly a problem - my assumption was that shmmax would be bounded in some | way - currently you can write any value into /proc/sys/kernel/shmmax - and I | don't think that the inteded behavior of this is to permit any value - or if | it is - what is the point of using the _minimax and not simple proc_doulongvec ?
Good observation. Yes, shmmax is unbounded, root can write any value to it. I don't know why proc_doulongvec_minmax is used there, except that I don't see a proc_doulongvec() without _minmax on it.
-- ~Randy - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |