Quartus II and DSE

Altera includes the Design Space Explorer (DSE) along with Quartus II.   I
have found this tcl script to be very helpful when I can get it to work for
me.  But lately, it has been very frustrating because it fails to explore
the space I set it up to explore.

Has anyone else seen this behavior?
  - I set up the exploration space in the DSE GUI, for example, to do a seed
sweep.
  - DSE starts by running a compilation flow on the base.
  - DSE finishes this and archives the results.
  - DSE stops and displays a message that there are no errors or warnings.
It doesn't go on to the next point in the exploration space.

I have also tried this with the distributed computing option.  DSE will
create archives, one for each exploration point.  It then submits one to a
remote computer and fails to actually compile it, and the process ends.

If you have seen this behavior, do you know what could be causing it and how
to correct it.

I have not found any references to this problem in the archives and haven't
received much help yet from the Altera support group, but I am hoping this
is forthcoming.

I am running with the Windows XP operating system (also tried Windows 2000)
and using the latest version of Quartus II (4.2 SP1).

I have used DSE successfully, so it doesn't always display this behavior.

Thanks in Advance,
Doug

Hi Doug,

there are versions of DSE that supports seeds like 1-10 20-30, other 
versions need that you write 1 2 3 4 5 6 7 8 9 10 20 21 22 23 24 25 26 27 28 
29 30. Maybe you are writing "1-10" in a version that does not support this? 
Beside that DSE worked always fine for me (not tested distributed usage 
yet).

Thomas

www.entner-electronics.com

"Douglas Sykora" <djsykoraNOSPAMM@execpc.com> schrieb im Newsbeitrag 
news:113cqgdrss51rbf@corp.supernews.com...
> Altera includes the Design Space Explorer (DSE) along with Quartus II.   I
> have found this tcl script to be very helpful when I can get it to work 
> for
> me.  But lately, it has been very frustrating because it fails to explore
> the space I set it up to explore.
>
> Has anyone else seen this behavior?
>  - I set up the exploration space in the DSE GUI, for example, to do a 
> seed
> sweep.
>  - DSE starts by running a compilation flow on the base.
>  - DSE finishes this and archives the results.
>  - DSE stops and displays a message that there are no errors or warnings.
> It doesn't go on to the next point in the exploration space.
>
> I have also tried this with the distributed computing option.  DSE will
> create archives, one for each exploration point.  It then submits one to a
> remote computer and fails to actually compile it, and the process ends.
>
> If you have seen this behavior, do you know what could be causing it and 
> how
> to correct it.
>
> I have not found any references to this problem in the archives and 
> haven't
> received much help yet from the Altera support group, but I am hoping this
> is forthcoming.
>
> I am running with the Windows XP operating system (also tried Windows 
> 2000)
> and using the latest version of Quartus II (4.2 SP1).
>
> I have used DSE successfully, so it doesn't always display this behavior.
>
> Thanks in Advance,
> Doug
>
> 

Thomas,
  Thanks for the tip.  I will try the enumerated seed specification as soon
as I get into work.   I have been using the range specification such as
2-10.
Thanks
Doug
"Thomas Entner" <aon.912710880@aon.at> wrote in message
news:42369e19$0$15878$91cee783@newsreader02.highway.telekom.at...
> Hi Doug,
>
> there are versions of DSE that supports seeds like 1-10 20-30, other
> versions need that you write 1 2 3 4 5 6 7 8 9 10 20 21 22 23 24 25 26 27
28
> 29 30. Maybe you are writing "1-10" in a version that does not support
this?
> Beside that DSE worked always fine for me (not tested distributed usage
> yet).
>
> Thomas
>
> www.entner-electronics.com
>
> "Douglas Sykora" <djsykoraNOSPAMM@execpc.com> schrieb im Newsbeitrag
> news:113cqgdrss51rbf@corp.supernews.com...
> > Altera includes the Design Space Explorer (DSE) along with Quartus II.
I
> > have found this tcl script to be very helpful when I can get it to work
> > for
> > me.  But lately, it has been very frustrating because it fails to
explore
> > the space I set it up to explore.
> >
> > Has anyone else seen this behavior?
> >  - I set up the exploration space in the DSE GUI, for example, to do a
> > seed
> > sweep.
> >  - DSE starts by running a compilation flow on the base.
> >  - DSE finishes this and archives the results.
> >  - DSE stops and displays a message that there are no errors or
warnings.
> > It doesn't go on to the next point in the exploration space.
> >
> > I have also tried this with the distributed computing option.  DSE will
> > create archives, one for each exploration point.  It then submits one to
a
> > remote computer and fails to actually compile it, and the process ends.
> >
> > If you have seen this behavior, do you know what could be causing it and
> > how
> > to correct it.
> >
> > I have not found any references to this problem in the archives and
> > haven't
> > received much help yet from the Altera support group, but I am hoping
this
> > is forthcoming.
> >
> > I am running with the Windows XP operating system (also tried Windows
> > 2000)
> > and using the latest version of Quartus II (4.2 SP1).
> >
> > I have used DSE successfully, so it doesn't always display this
behavior.
> >
> > Thanks in Advance,
> > Doug
> >
> >
>
>

"Douglas Sykora" <djsykoraNOSPAMM@execpc.com> wrote in message
news:113djevp68pa56d@corp.supernews.com...
> Thomas,
>   Thanks for the tip.  I will try the enumerated seed specification as
soon
> as I get into work.   I have been using the range specification such as
> 2-10.
> Thanks
> Doug
> "Thomas Entner" <aon.912710880@aon.at> wrote in message
> news:42369e19$0$15878$91cee783@newsreader02.highway.telekom.at...
> > Hi Doug,
> >
> > there are versions of DSE that supports seeds like 1-10 20-30, other
> > versions need that you write 1 2 3 4 5 6 7 8 9 10 20 21 22 23 24 25 26
27
> 28
> > 29 30. Maybe you are writing "1-10" in a version that does not support
> this?
> > Beside that DSE worked always fine for me (not tested distributed usage
> > yet).
> >
> > Thomas
> >
> > www.entner-electronics.com

Thomas,
  I tried the typing in the seeds as you suggested, but it did not help.
The version I am using (4.2 SP1) allows entering seeds in either format.
Thanks,
Doug


> >
> > "Douglas Sykora" <djsykoraNOSPAMM@execpc.com> schrieb im Newsbeitrag
> > news:113cqgdrss51rbf@corp.supernews.com...
> > > Altera includes the Design Space Explorer (DSE) along with Quartus II.
> I
> > > have found this tcl script to be very helpful when I can get it to
work
> > > for
> > > me.  But lately, it has been very frustrating because it fails to
> explore
> > > the space I set it up to explore.
> > >
> > > Has anyone else seen this behavior?
> > >  - I set up the exploration space in the DSE GUI, for example, to do a
> > > seed
> > > sweep.
> > >  - DSE starts by running a compilation flow on the base.
> > >  - DSE finishes this and archives the results.
> > >  - DSE stops and displays a message that there are no errors or
> warnings.
> > > It doesn't go on to the next point in the exploration space.
> > >
> > > I have also tried this with the distributed computing option.  DSE
will
> > > create archives, one for each exploration point.  It then submits one
to
> a
> > > remote computer and fails to actually compile it, and the process
ends.
> > >
> > > If you have seen this behavior, do you know what could be causing it
and
> > > how
> > > to correct it.
> > >
> > > I have not found any references to this problem in the archives and
> > > haven't
> > > received much help yet from the Altera support group, but I am hoping
> this
> > > is forthcoming.
> > >
> > > I am running with the Windows XP operating system (also tried Windows
> > > 2000)
> > > and using the latest version of Quartus II (4.2 SP1).
> > >
> > > I have used DSE successfully, so it doesn't always display this
> behavior.
> > >
> > > Thanks in Advance,
> > > Doug
> > >
> > >
> >
> >
>
>

Sorry no help here, just a confirmation that I see the same thing.  I've 
successfully run it previously and never touch the seed setting, but it 
doesn't work for me anymore -- just stops after the first iteration 
regardless of how many point there are.

Strange.

Tommy
PS: 4.2.SP1 also


Douglas Sykora wrote:
> "Douglas Sykora" <djsykoraNOSPAMM@execpc.com> wrote in message
> news:113djevp68pa56d@corp.supernews.com...
> 
>>Thomas,
>>  Thanks for the tip.  I will try the enumerated seed specification as
> 
> soon
> 
>>as I get into work.   I have been using the range specification such as
>>2-10.
>>Thanks
>>Doug
>>"Thomas Entner" <aon.912710880@aon.at> wrote in message
>>news:42369e19$0$15878$91cee783@newsreader02.highway.telekom.at...
>>
>>>Hi Doug,
>>>
>>>there are versions of DSE that supports seeds like 1-10 20-30, other
>>>versions need that you write 1 2 3 4 5 6 7 8 9 10 20 21 22 23 24 25 26
> 
> 27
> 
>>28
>>
>>>29 30. Maybe you are writing "1-10" in a version that does not support
>>
>>this?
>>
>>>Beside that DSE worked always fine for me (not tested distributed usage
>>>yet).
>>>
>>>Thomas
>>>
>>>www.entner-electronics.com
> 
> 
> Thomas,
>   I tried the typing in the seeds as you suggested, but it did not help.
> The version I am using (4.2 SP1) allows entering seeds in either format.
> Thanks,
> Doug
> 
> 
> 
>>>"Douglas Sykora" <djsykoraNOSPAMM@execpc.com> schrieb im Newsbeitrag
>>>news:113cqgdrss51rbf@corp.supernews.com...
>>>
>>>>Altera includes the Design Space Explorer (DSE) along with Quartus II.
>>
>>I
>>
>>>>have found this tcl script to be very helpful when I can get it to
> 
> work
> 
>>>>for
>>>>me.  But lately, it has been very frustrating because it fails to
>>
>>explore
>>
>>>>the space I set it up to explore.
>>>>
>>>>Has anyone else seen this behavior?
>>>> - I set up the exploration space in the DSE GUI, for example, to do a
>>>>seed
>>>>sweep.
>>>> - DSE starts by running a compilation flow on the base.
>>>> - DSE finishes this and archives the results.
>>>> - DSE stops and displays a message that there are no errors or
>>
>>warnings.
>>
>>>>It doesn't go on to the next point in the exploration space.
>>>>
>>>>I have also tried this with the distributed computing option.  DSE
> 
> will
> 
>>>>create archives, one for each exploration point.  It then submits one
> 
> to
> 
>>a
>>
>>>>remote computer and fails to actually compile it, and the process
> 
> ends.
> 
>>>>If you have seen this behavior, do you know what could be causing it
> 
> and
> 
>>>>how
>>>>to correct it.
>>>>
>>>>I have not found any references to this problem in the archives and
>>>>haven't
>>>>received much help yet from the Altera support group, but I am hoping
>>
>>this
>>
>>>>is forthcoming.
>>>>
>>>>I am running with the Windows XP operating system (also tried Windows
>>>>2000)
>>>>and using the latest version of Quartus II (4.2 SP1).
>>>>
>>>>I have used DSE successfully, so it doesn't always display this
>>
>>behavior.
>>
>>>>Thanks in Advance,
>>>>Doug
>>>>
>>>>
>>>
>>>
>>
> 
> 

Hi Doug, Tommy:

> Strange.

Our DSE guys are equally confused.  Things are working fine for us...

For us to be able to completely debug your issues we need see your
design, or at least the full DSE output logs (*.dse.rpt) with the
-debug flag on.

We tried using seeds in the form n1-n2 and it seemed to work fine for
us on 4.2 SP1.  Maybe you can try separating your seeds with commas in
case there is something weird there.  Another possibility is that your
base compile has an error or no-fit, which causes compilation to halt
in a similar way to what you describe.

Regards,

Paul Leventis
Altera Corp.

Paul Leventis wrote:
 > Our DSE guys are equally confused.  Things are working fine for us...
 >
 > For us to be able to completely debug your issues we need see your
 > design, or at least the full DSE output logs (*.dse.rpt) with the
 > -debug flag on.

I'd be happy to, but pardon my ignorence, can I set the -debug though 
the gui?

Here's how the it looks when it stops: http://numba-tu.com/DSE.png

 > We tried using seeds in the form n1-n2 and it seemed to work fine for
 > us on 4.2 SP1.  Maybe you can try separating your seeds with commas in
 > case there is something weird there.

I my case I've never touched it and it's at the default value: 3 5 7 11

 > Another possibility is that your
 > base compile has an error or no-fit, which causes compilation to halt
 > in a similar way to what you describe.

Not the case AFAICT.  I compile the project before launching DSE (it 
appears I have too).

I'll send you (personally) a link to the whole directory in case it's 
helpful (any comments or questions most welcome).

Thanks,
Tommy

"Paul Leventis" <paul.leventis@utoronto.ca> wrote in message
news:1111083250.701659.304200@o13g2000cwo.googlegroups.com...
> Hi Doug, Tommy:
>
> > Strange.
>
> Our DSE guys are equally confused.  Things are working fine for us...
>
> For us to be able to completely debug your issues we need see your
> design, or at least the full DSE output logs (*.dse.rpt) with the
> -debug flag on.
>
> We tried using seeds in the form n1-n2 and it seemed to work fine for
> us on 4.2 SP1.  Maybe you can try separating your seeds with commas in
> case there is something weird there.  Another possibility is that your
> base compile has an error or no-fit, which causes compilation to halt
> in a similar way to what you describe.
>
> Regards,
>
> Paul Leventis
> Altera Corp.
>
Paul,
  I submitted a mysupport SR two weeks ago and I am waiting for some support
to
help me out here.  (SR#10490208) I attached lots of reports and my design.
I talked to one of the engineers at mysupport late yesterday and they said
they are
now bumping it up to the higher level engineers. Any help you can provide
will be greatly
appreciated.

I was not aware of the -debug flag.  I will attempt to run DSE with this
flag on and
get another update to the mysupport engineers.

Also, my design does fit.  There is no "no-fit" error.  That would at least
produce an error message.  In my case there is no error message.

Thanks for your help Paul,

Doug

"Tommy Thorn" <foobar@nowhere.void> wrote in message
news:423A707D.1030808@nowhere.void...
> Paul Leventis wrote:
>  > Our DSE guys are equally confused.  Things are working fine for us...
>  >
>  > For us to be able to completely debug your issues we need see your
>  > design, or at least the full DSE output logs (*.dse.rpt) with the
>  > -debug flag on.
>
> I'd be happy to, but pardon my ignorence, can I set the -debug though
> the gui?
>
> Here's how the it looks when it stops: http://numba-tu.com/DSE.png
>
>  > We tried using seeds in the form n1-n2 and it seemed to work fine for
>  > us on 4.2 SP1.  Maybe you can try separating your seeds with commas in
>  > case there is something weird there.
>
> I my case I've never touched it and it's at the default value: 3 5 7 11
>
>  > Another possibility is that your
>  > base compile has an error or no-fit, which causes compilation to halt
>  > in a similar way to what you describe.
>
> Not the case AFAICT.  I compile the project before launching DSE (it
> appears I have too).
>
> I'll send you (personally) a link to the whole directory in case it's
> helpful (any comments or questions most welcome).
>
> Thanks,
> Tommy

Tommy,
  I took a look at how it looks when it stops for you, and that is the same
thing
that I see, except my compile time is more like 2-4 hours.

 Doug

"Douglas Sykora" <djsykoraNOSPAMM@execpc.com> wrote in message
news:113lae56ejqj0f5@corp.supernews.com...
>
> "Paul Leventis" <paul.leventis@utoronto.ca> wrote in message
> news:1111083250.701659.304200@o13g2000cwo.googlegroups.com...
> > Hi Doug, Tommy:
> >
> > > Strange.
> >
> > Our DSE guys are equally confused.  Things are working fine for us...
> >
> > For us to be able to completely debug your issues we need see your
> > design, or at least the full DSE output logs (*.dse.rpt) with the
> > -debug flag on.
> >
> > We tried using seeds in the form n1-n2 and it seemed to work fine for
> > us on 4.2 SP1.  Maybe you can try separating your seeds with commas in
> > case there is something weird there.  Another possibility is that your
> > base compile has an error or no-fit, which causes compilation to halt
> > in a similar way to what you describe.
> >
> > Regards,
> >
> > Paul Leventis
> > Altera Corp.
> >
> Paul,
>   I submitted a mysupport SR two weeks ago and I am waiting for some
support
> to
> help me out here.  (SR#10490208) I attached lots of reports and my design.
> I talked to one of the engineers at mysupport late yesterday and they said
> they are
> now bumping it up to the higher level engineers. Any help you can provide
> will be greatly
> appreciated.
>
> I was not aware of the -debug flag.  I will attempt to run DSE with this
> flag on and
> get another update to the mysupport engineers.
>
> Also, my design does fit.  There is no "no-fit" error.  That would at
least
> produce an error message.  In my case there is no error message.
>
> Thanks for your help Paul,
>
> Doug
>
>

I looked through all the DSE documentation and could find nothing about
a -debug flag
for DSE.  I also tried running DSE from the command line, which I never
tried before,  and it ran
without the stopping after compiling the base.  I tried it in both the local
and in the distributed
computing mode.  Both worked!  This weekend I will try running it with five
or six distributed computers
okay.  I'll let you know how it goes.
Doug