Peter Teoh wrote:I think my code snippet was wrong. Actually, it should be something like, $ cat /space/stap/inode-watch.stp # input device number and inode number, this prints out who accesses the file # probe kernel.function ("vfs_read"), kernel.function("vfs_write") { inode = $file->f_path->dentry->d_inode->i_ino dev_nr = $file->f_path->dentry->d_inode->i_sb->s_dev if(inode == $1 && (dev_nr == ($2 << 20 | $3))){ printf("%s(%d) -> %s\n", execname(), pid(), probefunc()) } } When I took out the f_dentry, (which is #defined in fs.h) and replaced it with f_path->dentry (in C, it would be f_path.dentry), it started working fine. So, it may not be a bug with systemtap after all. But if f_dentry is supposed to work, it would be a bug. I don't know about this. Thanks, Om. -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@nl.linux.org Please read the FAQ at http://kernelnewbies.org/FAQ
| Amit K. Arora | [RFC] Heads up on sys_fallocate() |
| James Bottomley | Re: Integration of SCST in the mainstream Linux kernel |
| Stephen Rothwell | Re: Announce: Linux-next (Or Andrew's dream :-)) |
| Greg KH | [GIT PATCH] driver core patches against 2.6.24 |
git: | |
| Gerrit Renker | [PATCH 27/37] dccp: Integration of dynamic feature activation - part 2 (server side) |
| Jarek Poplawski | [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock(). |
| Patrick McHardy | Re: [GIT]: Networking |
| Natalie Protasevich | [BUG] New Kernel Bugs |
