Skip to contentSkip to conversation

3.10.27: BUG on accessing /proc under high load

DISCUSSION wiebittewas 1 participant 1 message Jan 19, 2014

lore
Share & tools
lore
#0 lore raw

https://bugzilla.kernel.org/show_bug.cgi?id=68991

because http://kernelnewbies.org/FoundBug says, a bug-msg should be created in bugzilla as well as send to this list, I just follow this...

currently I don't have the time for this list, so please do a CC to me.

when doing a find /proc -type f -exec sha512sum {} \; x-hundred times parallel, this can lead in a BUG within d_path.

as I can see in the oops and the objdump (filed at bugzilla), the problems occurs in fs/dcache.c:d_path(), testing for existance of a function and then calling it:

if (path->dentry->d_op && path->dentry->d_op->d_dname)
return path->dentry->d_op->d_dname(path->dentry, buf, buflen);

looking at fs/proc/base.c:proc_pid_readlink() "path" is initialized in one of the getlink-functions within proc.c: proc_cwd_link() or proc_root_link()

in both functions first the task-structure is get by get_roc_task() and if 's not NULL, then locked and then the wanted fs_struct is acquired/locked.

now I don't have enough experience, but maybe beetween getting the task-struct and locking it, the task may be already invalidated because it exited, so the lock should be done in get_proc_task() or at least it should be checked after locking, if the task is still alive....???