Hello Ali,

The implementation of the MVI instructions (13 of them) and ctpop and
a bugfix for cttz for ALPHA along with sanity testcases is available
here:
http://www.csee.usf.edu/~sroy/techres/m5res/m5_alpha_cix_mvi.tar.bz2

The "gsm" benchmark from Mediabench-I generates the "maxsw4"
instructions unless compiled with something like a -mcpu=ev56 option.

-Soumyaroop

On Thu, Oct 29, 2009 at 4:27 PM, Ali Saidi <[email protected]> wrote:
>
> I'm in no rush...
>
> Here are some references:
> http://www.alphalinux.org/docs/MVI-full.html
> http://www.alphalinux.org/wiki/index.php/Motion_Video_Instructions
>
> libmpeg2 has inverse discrete cosine transform code that makes use of the
> MVI instructions. You should be able to extract an inner loop from one
> there and compare the output of an x86 and the alpha code executed by M5.
>
> As far as CTPOP writing a testcase is pretty trivial. The instruction just
> counts the number of bits set in a 64 bit register.See
> http://en.wikipedia.org/wiki/Hamming_weight for an idea of how to do this
> quickly without native popcount support.
>
>
> Thanks,
> Ali
>
>
>
> On Thu, 29 Oct 2009 16:42:29 -0400, soumyaroop roy <[email protected]>
> wrote:
>> [I am replying to m5-dev instead of m5-users]
>>
>> Hello Ali,
>>
>> Which benchmarks should one use to test the implementation of these
>> instructions?
>>
>> Either I or Vidya Sangkar, a labmate of mine, can work on it. But it
>> has to be after Nov 17th. Is that timeline is ok?
>>
>> regards,
>> Soumyaroop
>>
>> On 10/29/09, Ali Saidi <[email protected]> wrote:
>>>
>>>  ctlz and cttz are available and unless you're doing something crazy I
>>>  don't
>>>  think ctpop would be emitted. MVI isn't an instruction but a class of
>>>  instructions. See
>>>
> http://en.wikipedia.org/wiki/DEC_Alpha#Motion_Video_Instructions_.28MVI.29
>>>
>>>  At this point we've probably spent as much time discussing them as it
>>>  would
>>>  have taken to implement them.
>>>
>>>  Ali
>>>
>>>  On Thu, 29 Oct 2009 15:55:34 -0400, soumyaroop roy <[email protected]>
>>>
>>> wrote:
>>>  > One has to supply the flag, "-mno-cix", during compilation to make
>>>  > sure that the CIX instructions ("ctlz", "cttz", and "ctpop") are not
>>>  > emitted in the code.
>>>  >
>>>  > Here is the list of default flags passed (excerpt from
>>>  > "alphaev67-unknown-linux-gnu -v -Q  dummy.c"):
>>>  >
>>>  > options passed:  -v dummy.c -mcpu=ev67 -mtune=ev67
>>>  > options enabled:  -falign-loops -fargument-alias -fauto-inc-dec
>>>  >  -fbranch-count-reg -fcommon -fearly-inlining
>>>  >  -feliminate-unused-debug-types -ffunction-cse -fgcse-lm -fident
>>>  >  -fivopts
>>>  >  -fkeep-static-consts -fleading-underscore -fmath-errno
>>>  >  -fmerge-debug-strings -fmove-loop-invariants -fpcc-struct-return
>>>  >  -fpeephole -fsched-interblock -fsched-spec -fsched-stalled-insns-dep
>>>  >  -fsigned-zeros -fsplit-ivs-in-unroller -ftoplevel-reorder
>>>  -ftrapping-math
>>>  >  -ftree-cselim -ftree-loop-im -ftree-loop-ivcanon
> -ftree-loop-optimize
>>>  >  -ftree-parallelize-loops= -ftree-reassoc -ftree-scev-cprop
>>>  >  -ftree-vect-loop-version -fvar-tracking -fzero-initialized-in-bss
>>>  >  -mbwx
>>>  >  -mcix -mexplicit-relocs -mfix -mfloat-ieee -mfp-regs -mglibc
>>>  -mlarge-data
>>>  >  -mlarge-text -mlong-double-128 -mmax
>>>  >
>>>  > Note the "-mcix" on the second last line above.
>>>  >
>>>  > I could not find anything about the MVI instruction.
>>>  >
>>>  > regards,
>>>  > Soumyaroop
>>>  >
>>>  > On 10/29/09, Ali Saidi <[email protected]> wrote:
>>>  >>
>>>  >>  I've put a link on the d/l page. The file is downloading to
>>>  >>  m5sim.org
>>>  >>  right
>>>  >>  now and should be available within 5 minutes.
>>>  >>
>>>  >>  Ali
>>>  >>
>>>  >>
>>>  >>
>>>  >>  On Wed, 28 Oct 2009 22:27:11 -0400, soumyaroop roy
>>>  >>  <[email protected]>
>>>  >>
>>>  >> wrote:
>>>  >>  > Nate,
>>>  >>  >
>>>  >>  > It is available for download here (33 MB):
>>>  >>  >
>>>  >>
>>>
> http://72.187.222.34/~sroy/m5_tests/alphaev67-unknown-linux-gnu-x86-32.tar.bz2
>>>  >>  >
>>>  >>  > -Soumyaroop
>>>  >>  >
>>>  >>  > On Wed, Oct 28, 2009 at 9:12 PM, soumyaroop roy
> <[email protected]>
>>>  >>  > wrote:
>>>  >>  >> On Wed, Oct 28, 2009 at 9:04 PM, nathan binkert
>>>  >>  >> <[email protected]>
>>>  >>  wrote:
>>>  >>  >>>>>> My estimate is that the tar-zipped file should between
>>>  >>  >>>>>> 250-300
>>>  >>  >>>>>> MB.
>>>  >>  >>>>> No problem.
>>>  >>  >>>>
>>>  >>  >>>> Let me host it somewhere and send you the link so that you may
>>>  >>  download
>>>  >>  >>>> it.
>>>  >>  >>> Sure.  Once I get it on the m5 site, you can add the link.
>>>  >>  >>
>>>  >>  >> Is there a way for me to upload it on the M5 wiki directly? Is
>>>  >>  >> that
>>>  >>  >> what you were implying all along that I should do?
>>>  >>  >>
>>>  >>  >>>
>>>  >>  >>>> Do you mean providing the link though the wiki? I do not have
>>>  >>  >>>> permissions to edit this page,
>>>  >>  >>>> http://www.m5sim.org/wiki/index.php/Download, which has links
>>>  >>  >>>> to
>>>  >>  >>>> the
>>>  >>  >>>> cross-compilers.
>>>  >>  >>>
>>>  >>  >>> You should be able to, but you have to create an account.
>>>  >>  >>
>>>  >>  >> I already have an account. But it says:
>>>  >>  >>
>>>  >>  >> You do not have permission to edit this page, for the following
>>>  >>  >> reason:
>>>  >>  >> This page has been locked to prevent editing.
>>>  >>  >>
>>>  >>  >> You can view and copy the source of this page:
>>>  >>  >>
>>>  >>  >>>
>>>  >>  >>>  Nate
>>>  >>  >>> _______________________________________________
>>>  >>  >>> m5-users mailing list
>>>  >>  >>> [email protected]
>>>  >>  >>> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
>>>  >>  >>>
>>>  >>  >>
>>>  >>  >> -Soumyaroop
>>>  >>  >>
>>>  >>  >> --
>>>  >>  >> Soumyaroop Roy
>>>  >>  >> Ph.D. Candidate
>>>  >>  >> Department of Computer Science and Engineering
>>>  >>  >> University of South Florida, Tampa
>>>  >>  >> http://www.csee.usf.edu/~sroy
>>>  >>  >>
>>>  >>  _______________________________________________
>>>  >>  m5-users mailing list
>>>  >>  [email protected]
>>>  >>  http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
>>>
>



-- 
Soumyaroop Roy
Ph.D. Candidate
Department of Computer Science and Engineering
University of South Florida, Tampa
http://www.csee.usf.edu/~sroy
_______________________________________________
m5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/m5-dev

Reply via email to