Message-ID: <CAF1vpkigHMdKphnNjDm7=rR6TTxViHGGHi3bb64rsHG7KbqYzQ@mail.gmail.com>
Date:   Mon, 4 Oct 2021 10:57:19 +0300
From:   Avi Deitcher <avi@...tcher.net>
To:     Andreas Dilger <adilger@...ger.ca>
Cc:     linux-ext4@...r.kernel.org
Subject: Re: algorithm for half-md4 used in htree directories

Hi Andreas,

I had looked in __ext4fs_dirhash(). Yes, it does reference the seed -
and create a default if none is there at the filesystem level - but it
doesn't appear to use it, in that function. hinfo is populated in the
function - hash, minor-hash, seed - but it never uses the seed to
manipulate the hash.

Are you saying that it is at a higher level? i.e. __ext4fs_dirhash()
is the *first* step, and there is further processing to get the actual
hash? I did walk up the stack, but couldn't figure out.

Thanks for stepping in
Avi

On Sun, Oct 3, 2021 at 7:43 PM Andreas Dilger <adilger@...ger.ca> wrote:
>
> On Oct 3, 2021, at 06:47, Avi Deitcher <avi@...tcher.net> wrote:
> >
> > I can narrow down the question further. In my live sample, one of the
> > entries in the tree is for a directory named "dir155".
> >
> > If I run "dx_hash dir155", I get:
> >
> > # debugfs -R "dx_hash dir155" /var/lib/file.img
> > debugfs 1.46.2 (28-Feb-2021)
> > Hash of dir155 is 0x16279534 (minor 0x0)
> >
> > If I look in the tree with "htree_dump", I get:
> >
> > # debugfs -R "htree_dump /testdir" /var/lib/file.img
> > debugfs 1.46.2 (28-Feb-2021)
> > ....
> > Entry #0: Hash 0x00000000, block 1
> > Reading directory block 1, phys 6459
> > 168 0x00d11d98-b9b6b16b (16) dir155   332 0x009edafe-77de7d72 (16) dir319
> >
> > That hash for dir155 does not match what dx_hash gave. If I try to
> > take the code from fs/ext4/hash.c and build a small program to
> > calculate the hash, I get:
> >
> > $ ./md4 dir155
> > MD4: d90278a1 25182ac7 a02e56be c3f30f04
> > hash: 0x25182ac6
> > minor: 0xa02e56be
> >
> > Clearly that isn't what is in the tree. What basic am I missing?
>
> One important factor is that the directory hash has an initial seed
> to prevent pathological cases where the user can construct thousands
> of directory entries that have a hash collision.
>
> Looking at the code explains this in the comment for __ext4fs_dirhash().
> The seed itself comes from sbi->s_hash_seed and is stored in the
> per-directory hinfo.seed to be used when counting the filename hash.
> In theory there could be a per-directory hash, but it appears to be a
> constant for the whole filesystem.
>
> Cheers, Andreas
>
> >
> >> On Fri, Oct 1, 2021 at 2:49 PM Avi Deitcher <avi@...tcher.net> wrote:
> >>
> >> Hi,
> >>
> >> I have been trying to understand the algorithm used for the "half-md4"
> >> in htree-structured directories. Going through the code (and trying
> >> not to get into reverse engineering), it looks like it is part of md4
> >> but not entirely? Yet any subset I take doesn't quite line up with
> >> what I see in an actual sample.
> >>
> >> What is the algorithm it is using to turn an entry of, e.g., "file125"
> >> into the appropriate hash. I did run a live sample, and try to get
> >> some form of correlation between the actual md4 hash (16 bytes) of the
> >> above to the actual entry (4 bytes) shown by debugfs, without much
> >> luck.
> >>
> >> What basic thing am I missing?
> >>
> >> Separately, how does the seed play into it?
> >>
> >> Thanks
> >> Avi
> >
> >
> >
> > --
> > Avi Deitcher
> > avi@...tcher.net
> > Follow me http://twitter.com/avideitcher
> > Read me http://blog.atomicinc.com



-- 
Avi Deitcher
avi@...tcher.net
Follow me http://twitter.com/avideitcher
Read me http://blog.atomicinc.com

Powered by blists - more mailing lists