Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Mon, 8 Jul 2013 22:45:28 -0400
From: Yaniv Sapir <>
Subject: Re: Parallella: bcrypt

> I tried this and for me it doesn't work, I don't get correct results. I
> took code I sent you (
> 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.

> And one more question - can core halt itself and if can, how?

Technically, it can when issuing the IDLE instruction.

> Since writing to core's local memory and than starting cores using e_start
> works, I can have infinite loop in which first instruction is halt and
> after all writes are done, host starts the cores with e_start. After one
> iteration core halts itself and gets new data. If this is possible
> scenario, what would happen if core is halted and data transfers aren't
> executed completely (by data transfer I mean transfer of result from local
> memory to shared buffer)?

If the core itself is the initiator of the halt condition (either by IDLE,
or just by polling a signal mailbox ), I don't see how data transfers are
not executed completely? Do I miss something? Anyway if your data size per
round is relatively small, why not use the core's memory directly as a
mailbox? This way, you save the effort of writing to DRAM. Host reading
from a core location is just as easy as reading from external mem.

> Thank you,
> Katja

Yaniv Sapir
Adapteva Inc.
1666 Massachusetts Ave, Suite 14
Lexington, MA 02420
Phone: (781)-328-0513 (x104)
CONFIDENTIALITY NOTICE: This e-mail may contain information
that is confidential and proprietary to Adapteva, and Adapteva hereby
designates the information in this e-mail as confidential. The information
 intended only for the use of the individual or entity named above. If you
not the intended recipient, you are hereby notified that any disclosure,
distribution or use of any of the information contained in this
transmission is
strictly prohibited and that you should immediately destroy this e-mail and
contents and notify Adapteva.

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.