[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CA+EaD-abbQ1KextHmWdBtcKkB7E0xtVpWJGXNtJyQHc9fsLYrg@mail.gmail.com>
Date: Sun, 10 Nov 2013 16:58:56 +0100
From: Katja Malvoni <kmalvoni@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: bcrypt-parallella on 64-core (was: Katja's weekly
report #13)
Hi Alexander,
On Sun, Nov 10, 2013 at 6:31 AM, Solar Designer <solar@...nwall.com> wrote:
> On Sat, Nov 09, 2013 at 10:09:03PM +0100, Katja Malvoni wrote:
> > But if I run self test again, it fails on get_hash[0](1).
>
> I think you need to look for possible uninitialized variables in your
> Epiphany code.
>
The problem was that I only point to a specific memory location so until
host writes data, whole inputs structure is uninitialized. I added writing
zero to start flag before entering do-while loop so that computation
doesn't start before data is ready. However, this code is unreliable:
linaro@...aro-ubuntu-desktop:~/GitHub/JohnTheRipper/run$ while :; do rm -f
john.log john.pot; seq 0 49999 | sudo -E LD_LIBRARY_PATH=$LD_LIBRARY_PATH
./john -stdin -form=bcrypt-parallella ~/pw-bcrypt-2salt-50k 2>&1 >
/dev/null | fgrep 0:00; done
30670g 0:00:00:13 2290g/s 3734p/s 6764c/s 83484KC/s 49920..49999
37577g 0:00:00:17 2143g/s 2852p/s 5692c/s 87933KC/s 49920..49999
37770g 0:00:00:12 2990g/s 3958p/s 7917c/s 123279KC/s 49920..49999
37770g 0:00:00:12 2985g/s 3952p/s 7905c/s 123084KC/s 49920..49999
37770g 0:00:00:12 2955g/s 3912p/s 7824c/s 121832KC/s 49920..49999
37770g 0:00:00:12 3026g/s 4006p/s 8012c/s 124760KC/s 49920..49999
37770g 0:00:00:12 2950g/s 3906p/s 7812c/s 121641KC/s 49920..49999
37770g 0:00:00:12 2985g/s 3952p/s 7905c/s 123084KC/s 49920..49999
37770g 0:00:00:12 2955g/s 3912p/s 7824c/s 121832KC/s 49920..49999
37770g 0:00:00:12 2978g/s 3943p/s 7886c/s 122792KC/s 49920..49999
And on zed with E16:
kmalvoni@...aro-ubuntu-desktop:~/E64/JohnTheRipper/run$ while :; do rm -f
john.log john.pot; seq 0 49999 | sudo -E LD_LIBRARY_PATH=$LD_LIBRARY_PATH
./john -stdin -form=bcrypt-parallella ~/pw-bcrypt-2salt-50k 2>&1 >
/dev/null | fgrep 0:00; done
46903g 0:00:00:45 1020g/s 1087p/s 2175c/s 28906KC/s 49984..49999
46903g 0:00:00:44 1056g/s 1126p/s 2252c/s 29934KC/s 49984..49999
46900g 0:00:00:44 1056g/s 1125p/s 2251c/s 29924KC/s 49984..49999
46903g 0:00:00:44 1055g/s 1124p/s 2249c/s 29894KC/s 49984..49999
46902g 0:00:00:44 1053g/s 1123p/s 2246c/s 29855KC/s 49984..49999
46902g 0:00:00:44 1056g/s 1125p/s 2251c/s 29922KC/s 49984..49999
46902g 0:00:00:44 1056g/s 1126p/s 2252c/s 29929KC/s 49984..49999
46902g 0:00:00:44 1056g/s 1125p/s 2251c/s 29922KC/s 49984..49999
46902g 0:00:00:44 1056g/s 1126p/s 2252c/s 29929KC/s 49984..49999
46900g 0:00:00:44 1055g/s 1125p/s 2250c/s 29902KC/s 49984..49999
> In fact, where is this version of code, the one you're doing for E64?
> A separate GitHub branch?
>
Now it's on GitHub: git clone https://github.com/kmalvoni/JohnTheRipper -b
master-E64
Katja
Content of type "text/html" skipped
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
