Re: WARNING: CPU: 1 PID: 14735 at fs/dcache.c:365 dentry_free+0x100/0x128

[Al Viro <viro@zeniv.linux.org.uk>]

On Wed, Jul 20, 2022 at 03:31:10AM +0100, Al Viro wrote:
> On Sat, Jul 16, 2022 at 07:27:30AM +0200, Helge Deller wrote:
> > [108564.845444] dentry = 000000031624e6c0
> 
> *blink*
> Is that really a plausible address of a kernel object on
> parisc?  Or is that "mangle pointers for security fetish^W
> purposes, lest somebody manages to get a useful information
> out of dmesg" in action?
> 
> > [108564.889437] spin_is_locked(&dentry->d_lock) = 0
> > [108564.945436] dname_external(dentry) = 0
> > [108564.993436] dentry->d_flags & DCACHE_NORCU = 0
> > [108565.045446] dentry->d_name.len = 3
> > [108565.089435] dentry->d_name.hash = 89116695
> > [108565.137435] dentry->d_lockref.count = -128
> > [108565.189434] dentry->d_flags = 32776
> 
> 0x8008, i.e. DCACHE_OP_DELETE | DCACHE_DENTRY_KILLED.  No
> DCACHE_PAR_LOOKUP in sight...
> 
> > [108569.801407] dentry = 000000016d7d0000
> > [108569.845407] spin_is_locked(&dentry->d_lock) = 0
> > [108569.901422] dname_external(dentry) = 0
> > [108569.949405] dentry->d_flags & DCACHE_NORCU = 0
> > [108570.001405] dentry->d_name.len = 10
> > [108570.045421] dentry->d_name.hash = e6582e53
> > [108570.093476] dentry->d_lockref.count = -128
> > [108570.145404] dentry->d_flags = 32776
> > [108570.189420] dentry->d_inode = 0000000000000000
> > [108570.241404] dentry->d_parent = 00000002332c2780
> > [108570.297403] dentry->d_u.d_rcu = 0x416be770
> 
> Ditto.  Pointers look really weird...

PS: I think I've caught what could be found on l-k; could you
please forward the rest (parisc list? private thread?) to
fsdevel or directly my way.