Re: [syzbot] [hardening?] [mm?] BUG: bad usercopy in vfs_readlink

[Theodore Ts'o]

On Tue, Feb 04, 2025 at 10:25:29PM +0100, Mateusz Guzik wrote:
> >
> > My question is if that's legitimate, I'm guessing not. If not, then
> > ext4 should complain about it.
> >
> > On stock kernel this happens to work because strlen finds the "right" size.
> >
> 
> So it occurred to me to check what fsck thinks about it.
> 
> I ran it twice in a row, it *removed* the problematic symlink.

Can you show me what's in the problematic symlink?  And does the
syzbot reproducer trigger a problem before adding your symlink
caching?

What would be really great if you couldcreate small focused test case
that shows what's going on --- ideally something like a 100k file
system, ala the file systems in the tests directory of the e2fsprogs
sources....

						- Ted