[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CA+EaD-ZQvVz9k-b_byehQYuvrRmJrG6kLbR4e6Rz=7neK6xTRg@mail.gmail.com>
Date: Tue, 9 Jul 2013 18:43:09 +0200
From: Katja Malvoni <kmalvoni@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: Parallella: bcrypt
On Tue, Jul 9, 2013 at 4:45 AM, Yaniv Sapir <yaniv@...pteva.com> wrote:
>
> I tried this and for me it doesn't work, I don't get correct results. I
>> took code I sent you (http://www.openwall.com/lists/john-dev/2013/07/06/3)
>> and did only that, commented out "outbuf.start[corenum] = 0" and some cores
>> return wrong results. Even worse, cores that return wrong results are
>> different for every run and wrong results are also different. Output that I
>> get is attached. Could you please check that you changed only that?
>>
>>
> Sorry if it was not clear, but in addition to commenting out the "start[]"
> signal, I also added a dummy place-holder after the "done" member of the
> structure, to make sure both sides have the same member alignment.
>
> I did the same and I still get wrong results. I also changed BufSize to
0x1000 but it didn't help. Again, some data is not transferred correctly:
eCore 0x808 (0, 0): 1bb69143, f9a8d304, c8d23d99, ab049a77, a68e2ccc,
74420600
eCore 0x809 (0, 1): 0, 0, 0, 0, 0, 0
eCore 0x80a (0, 2): 6f15130a, 8d681e90, 5cfec901, f4493e64, ebb352f8,
35ce3600
eCore 0x80b (0, 3): 2bfd15b2, 71622fde, b982676d, 22550521, 45957d0e,
1ae35200
eCore 0x848 (1, 0): 1bb69143, f9a8d304, c8d23d99, ab049a77, a68e2ccc,
74420600
eCore 0x849 (1, 1): 1bb69143, f9a8d304, c8d23d99, ab049a77, a68e2ccc,
74420600
eCore 0x84a (1, 2): 1bb69143, f9a8d304, c8d23d99, ab049a77, a68e2ccc,
74420600
eCore 0x84b (1, 3): 1bb69143, f9a8d304, c8d23d99, ab049a77, a68e2ccc,
74420600
eCore 0x888 (2, 0): 1bb69143, f9a8d304, c8d23d99, ab049a77, a68e2ccc,
74420600
eCore 0x889 (2, 1): 1bb69143, f9a8d304, c8d23d99, ab049a77, a68e2ccc,
74420600
eCore 0x88a (2, 2): 1bb69143, f9a8d304, c8d23d99, ab049a77, a68e2ccc,
74420600
eCore 0x88b (2, 3): 1bb69143, f9a8d304, c8d23d99, ab049a77, a68e2ccc,
74420600
eCore 0x8c8 (3, 0): 1bb69143, f9a8d304, c8d23d99, ab049a77, a68e2ccc,
74420600
eCore 0x8c9 (3, 1): 1bb69143, f9a8d304, c8d23d99, ab049a77, a68e2ccc,
74420600
eCore 0x8ca (3, 2): 1bb69143, f9a8d304, c8d23d99, ab049a77, a68e2ccc,
74420600
eCore 0x8cb (3, 3): 1bb69143, f9a8d304, c8d23d99, ab049a77, a68e2ccc,
74420600
Execution time - Epiphany: 19.926000 ms
done = 16
core_done[ 0] = 1 test[ 0] = 16 ciphertext[0] = P
core_done[ 1] = 1 test[ 1] = 16 ciphertext[1] = P
core_done[ 2] = 1 test[ 2] = 16 ciphertext[2] = P
core_done[ 3] = 1 test[ 3] = 16 ciphertext[3] = P
core_done[ 4] = 1 test[ 4] = 16 ciphertext[4] = P
core_done[ 5] = 1 test[ 5] = 16 ciphertext[5] = P
core_done[ 6] = 1 test[ 6] = 16 ciphertext[6] = P
core_done[ 7] = 1 test[ 7] = 16 ciphertext[7] = P
core_done[ 8] = 1 test[ 8] = 16 ciphertext[8] =
core_done[ 9] = 1 test[ 9] = 16 ciphertext[9] = P
core_done[10] = 1 test[10] = 16 ciphertext[10] = P
core_done[11] = 1 test[11] = 16 ciphertext[11] = P
core_done[12] = 1 test[12] = 16 ciphertext[12] = P
core_done[13] = 1 test[13] = 16 ciphertext[13] = P
core_done[14] = 1 test[14] = 16 ciphertext[14] =
core_done[15] = 1 test[15] = 16 ciphertext[15] =
ciphertext = P
If I pass e_false as parameter to e_load_group() and than start cores with
e_start(), than it works. I guess that when I pass e_true as argument to
e_load_group(), cores start bcrypt computation but not all data is
transferred and that is why some cores return incorrect results. Normally I
would use e_start() to start cores because it works but there are problems
with JtR integration.
When cracking passwords, after 60 to 65 passwords are cracked I get
following error: john: e_alloc(): mmap failure
I set verbosity for host and loader to H_D4 and L_D4 (this was overhead,
probably H_D1 and L_D1 would be enough) and this happens while allocating
memory for loading *.srec file. Can this be a consequence of calling
e_init(), e_reset_system(), e_open(), e_alloc(), e_load_group(), e_close(),
e_free() and e_finalize() every time Epiphany chip is used?
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.
