From ks4kk at virginia.edu  Thu May  2 10:52:27 2013
From: ks4kk at virginia.edu (Karthik Sankaranarayanan)
Date: Thu, 2 May 2013 13:52:27 -0400
Subject: [Hotspot] DTM in hotspot
In-Reply-To: <CAKLY5EmLKs+-2V4DFX4DtyME1jF2g7uXCgqSq=fAL2+hpqo1yQ@mail.gmail.com>
References: <CAKLY5EmLKs+-2V4DFX4DtyME1jF2g7uXCgqSq=fAL2+hpqo1yQ@mail.gmail.com>
Message-ID: <CAPE+kwBt1u1tCqjPRQQmpEkUpVUqjNJ3eDJjrh1xgSPWhhe6TQ@mail.gmail.com>

Hi Mushfique,

Sorry about the delayed response - yes, the dtm_used flag just changes the
initial temperatures - it has no other impact. The rationale is, when DTM
is employed and the initial temperatures are set to be higher than the
thermal threshold, then they are all scaled down by a constant factor so
that the max initial temperature = thermal threshold.

Hope this clarifies. Please let me know if you have further questions.

Thanks,
-karthik


On Mon, Apr 22, 2013 at 6:04 PM, Mushfique Khurshid <khurshid at wisc.edu>wrote:

>
> My question is what does hotspot do with regards to the option DTM_USED in
> the *.config? So far as I see in the code, this option only is applied for
> the INITIAL TEMPERATURE? Is that true? What about during the simulation?
> And if it is during the simulation, then is there any way for us to check
> how DTM affected performance/exec. time?
>
> _______________________________________________
> HotSpot mailing list
> HotSpot at mail.cs.virginia.edu
> http://www.cs.virginia.edu/mailman/listinfo/hotspot
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130502/1b34010d/attachment.html 

From anurag.umbarkar at stonybrook.edu  Sat May  4 15:54:19 2013
From: anurag.umbarkar at stonybrook.edu (Anurag Umbarkar)
Date: Sat, 4 May 2013 18:54:19 -0400
Subject: [Hotspot] Cannot download Hotspot!
Message-ID: <CAHQsE=QGxm1R-TmVtsnVjyEbDtYEaZ5M2-rrKQK_A6sNrJRxMw@mail.gmail.com>

I filled out the form on the download page and submitted the information.
 The subsequent webpage said I'll receive the download link via email but I
did not get any such mail (I checked the spam folder as well). I tried
again with a different email address, still no success. Any suggestions?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130504/e810c3f2/attachment.html 

From grandhi.rahul at gmail.com  Mon May  6 08:35:15 2013
From: grandhi.rahul at gmail.com (Rahul Grandhi)
Date: Mon, 6 May 2013 08:35:15 -0700
Subject: [Hotspot] unsubscribe me please. Thanks.
Message-ID: <CAAZh+b5nWM7MK+dzUUGRWzC+t+cu_tm=FurQP+8G+zLb4PmGcA@mail.gmail.com>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130506/40d9190b/attachment.html 

From anurag.umbarkar at stonybrook.edu  Mon May 13 17:03:41 2013
From: anurag.umbarkar at stonybrook.edu (Anurag Umbarkar)
Date: Mon, 13 May 2013 20:03:41 -0400
Subject: [Hotspot] Few questions
Message-ID: <CAHQsE=TRBbifherVs-Tpua_N1q6F6y8uOsEBKvkTa4LH9_OJZQ@mail.gmail.com>

I recently started using Hotspot. I am using the excel GUI (the 'Draw
Thermal Map' option is great!) and my aim is to simulate a 3D (multi-layer,
single die) chip but I have a few queries. I looked through the mailing
list archives and the FAQ page but could not figure out the answers to the
following questions. I would really appreciate a quick response from the
authors and/or members on this list.

1. What time step (millisec, microsec etc.) should be used for the power
traces in the ptrace file? What is the time step used in the example that
is provided?

2. Is there a way to simulate a 3D stack from the excel interface? How do I
include the LCF file in the simulation?

3. Is there a way to simulate the chip (die/3D stack) without the heat
spreader and sink?

4. In the LCF, what is the purpose of including 'floorplan file' as a
parameter for the TIM layers? Is it just to maintain uniformity in the
description of the layers?

Thanks.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130513/cd8e3691/attachment.html 

From skadron at cs.virginia.edu  Tue May 14 08:00:46 2013
From: skadron at cs.virginia.edu (Kevin Skadron)
Date: Tue, 14 May 2013 11:00:46 -0400
Subject: [Hotspot] Few questions
In-Reply-To: <CAHQsE=TRBbifherVs-Tpua_N1q6F6y8uOsEBKvkTa4LH9_OJZQ@mail.gmail.com>
References: <CAHQsE=TRBbifherVs-Tpua_N1q6F6y8uOsEBKvkTa4LH9_OJZQ@mail.gmail.com>
Message-ID: <5192519E.1070605@cs.virginia.edu>

See answers inline below.
/K

On 5/13/13 8:03 PM, Anurag Umbarkar wrote:
> I recently started using Hotspot. I am using the excel GUI (the 'Draw
> Thermal Map' option is great!) and my aim is to simulate a 3D
> (multi-layer, single die) chip but I have a few queries. I looked
> through the mailing list archives and the FAQ page but could not figure
> out the answers to the following questions. I would really appreciate a
> quick response from the authors and/or members on this list.
>
> 1. What time step (millisec, microsec etc.) should be used for the power
> traces in the ptrace file? What is the time step used in the example
> that is provided?

The only requirement is that the time step in the power trace and the 
time step specified in the HotSpot configuration options need to match.

> 2. Is there a way to simulate a 3D stack from the excel interface? How
> do I include the LCF file in the simulation?

Unfortunately not via the Excel interface.

> 3. Is there a way to simulate the chip (die/3D stack) without the heat
> spreader and sink?

Unfortunately not in any part of HotSpot. The closest you can get is to 
make these layers very thin.

> 4. In the LCF, what is the purpose of including 'floorplan file' as a
> parameter for the TIM layers? Is it just to maintain uniformity in the
> description of the layers?

Exactly!

> Thanks.
>
>
> _______________________________________________
> HotSpot mailing list
> HotSpot at mail.cs.virginia.edu
> http://www.cs.virginia.edu/mailman/listinfo/hotspot
>

From puskar.budhathoki at gmail.com  Tue May 21 07:22:19 2013
From: puskar.budhathoki at gmail.com (Puskar Budhathoki)
Date: Tue, 21 May 2013 15:22:19 +0400
Subject: [Hotspot] 3D Stacking in HotSpot
Message-ID: <CAP=ZmBYTsovhx4CYh47QsPU7=8DQZOXu0ON8tR8oagTfikuNCw@mail.gmail.com>

Hi,

I want to use HotSpot for thermal approximation of 3D floor planning. I
have some confusion, could you please clarify it?



1) Do we need to have same height and width for each layer of
floorplanning? and for each layer also, eg. layer0 independently need to
have same width and height?



2) As there are different benchmark available like ami33, ami49, xerox,
apte, hp, etc, so for this benchmark, how to tune the configuration file
for the thermal approximation and how about the input power file(ptrace)?



3)Using 3D stacking capability, we are able to approximate 3D layered
temperature of each block. Is it possible to graphically represent(as eg.
gcc.pdf) of each and every layer ?





Regards,

Puskar Budhathoki
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130521/ed621bfd/attachment.html 

From mostafa.saied at ejust.edu.eg  Tue May 21 09:15:53 2013
From: mostafa.saied at ejust.edu.eg (mossaied remeah)
Date: Tue, 21 May 2013 15:15:53 +0200
Subject: [Hotspot] request to join mailing list
Message-ID: <CACd347NbsHPw4=epkNYNFUzCT9X31Zz7GxtTjcojYWCvANa0Jw@mail.gmail.com>

Dear professor,

I have installed hotspot and I'd like to join the hotspot mailing list.

My email is

mostafa.saied at ejust.edu.eg

Thanks in adavnce

-- 
Egypt-Japan University of science and Technology (EJUST)
ECE Department
Mostafa Said Sayed Remeah
01003163829
01119216928
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130521/3dc95df0/attachment.html 

From wh6p at virginia.edu  Tue May 21 10:32:02 2013
From: wh6p at virginia.edu (Wei Huang)
Date: Tue, 21 May 2013 09:32:02 -0500
Subject: [Hotspot] 3D Stacking in HotSpot
In-Reply-To: <CAP=ZmBYTsovhx4CYh47QsPU7=8DQZOXu0ON8tR8oagTfikuNCw@mail.gmail.com>
References: <CAP=ZmBYTsovhx4CYh47QsPU7=8DQZOXu0ON8tR8oagTfikuNCw@mail.gmail.com>
Message-ID: <CANYeXBrd+FnOYS-qMBdnjby3=aWEYmWoi8uaBZuZXWWZGLPexQ@mail.gmail.com>

Hello, please see embedded answers below.


On Tue, May 21, 2013 at 6:22 AM, Puskar Budhathoki <
puskar.budhathoki at gmail.com> wrote:

> Hi,
>
> I want to use HotSpot for thermal approximation of 3D floor planning. I
> have some confusion, could you please clarify it?
>
1) Do we need to have same height and width for each layer of
> floorplanning? and for each layer also, eg. layer0 independently need to
> have same width and height?
>
Yes, all the layers have to have the same die size.

>
>
> 2) As there are different benchmark available like ami33, ami49, xerox,
> apte, hp, etc, so for this benchmark, how to tune the configuration file
> for the thermal approximation and how about the input power file(ptrace)?
>
>
>
HotSpot only provides thermal simulations. You'll have to obtain the power
file elsewhere (e.g. McPAT, Wattch, etc).

As for floorplan,  you also need to generate that yourself by observing a
die photograph. If you are exploring different hypothetical floorplan
possibilities, the HotFloorplanner tool inside HotSpot may be useful.

For other configs such as cooling solutions, if you are simulating typical
cooling packages, one of the default configs in HotSpot should provide
reasonable results.

> 3)Using 3D stacking capability, we are able to approximate 3D layered
> temperature of each block. Is it possible to graphically represent(as eg.
> gcc.pdf) of each and every layer ?
>

If you mean GUI for generating floorplan or power trace, HotSpot doesn't
support that. If you want to look at the thermal map of each layer, you can
always (easily) modify the source to dump temperature information for all
layers. There are example dump temperature subroutines you can use as a
template.

Hope this helps.

-Wei

>
>
>
>
> Regards,
>
> Puskar Budhathoki
>
> _______________________________________________
> HotSpot mailing list
> HotSpot at mail.cs.virginia.edu
> http://www.cs.virginia.edu/mailman/listinfo/hotspot
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130521/cd8cd562/attachment.html 

From mostafa.saied at ejust.edu.eg  Tue May 21 13:24:25 2013
From: mostafa.saied at ejust.edu.eg (mossaied remeah)
Date: Tue, 21 May 2013 19:24:25 +0200
Subject: [Hotspot] Problem in running hotspot from Excel file
Message-ID: <CACd347MusPLZm8EqzZVOs18oPmeipG0DVUgyfacAcSVR=A_fAg@mail.gmail.com>

Hi all,

I have installed HotSpot using MinGW on Windows 7, when I try to run it
using "Run Hotspot.exe" in Floorplan sheet I got the following error:

"Run-time error: '1004'
Application-defined or object-defined error"

Can anyone help me in this!?

Thanks in advance

-- 
Egypt-Japan University of science and Technology (EJUST)
ECE Department
Mostafa Said Sayed Remeah
01003163829
01119216928
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130521/c6b9878b/attachment.html 

From wh6p at virginia.edu  Tue May 21 14:46:46 2013
From: wh6p at virginia.edu (Wei Huang)
Date: Tue, 21 May 2013 13:46:46 -0500
Subject: [Hotspot] Problem in running hotspot from Excel file
In-Reply-To: <CACd347MusPLZm8EqzZVOs18oPmeipG0DVUgyfacAcSVR=A_fAg@mail.gmail.com>
References: <CACd347MusPLZm8EqzZVOs18oPmeipG0DVUgyfacAcSVR=A_fAg@mail.gmail.com>
Message-ID: <CANYeXBqVth+vtCr0OPtVrFhm1LH2EKKwQs186C2hHBtC9cHeiA@mail.gmail.com>

Have you confirmed that hotspot.exe can be run in windows command line
first?
If yes, then it might be an issue with Excel compatibility. What version of
Excel are you using? Have you enabled macros?

-Wei


On Tue, May 21, 2013 at 12:24 PM, mossaied remeah <
mostafa.saied at ejust.edu.eg> wrote:

> Hi all,
>
> I have installed HotSpot using MinGW on Windows 7, when I try to run it
> using "Run Hotspot.exe" in Floorplan sheet I got the following error:
>
> "Run-time error: '1004'
> Application-defined or object-defined error"
>
> Can anyone help me in this!?
>
> Thanks in advance
>
> --
> Egypt-Japan University of science and Technology (EJUST)
> ECE Department
> Mostafa Said Sayed Remeah
> 01003163829
> 01119216928
>
> _______________________________________________
> HotSpot mailing list
> HotSpot at mail.cs.virginia.edu
> http://www.cs.virginia.edu/mailman/listinfo/hotspot
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130521/d2fb9a45/attachment.html 

From ks4kk at virginia.edu  Wed May 22 16:59:14 2013
From: ks4kk at virginia.edu (Karthik Sankaranarayanan)
Date: Wed, 22 May 2013 16:59:14 -0400
Subject: [Hotspot] About using dump temperature subroutine in HotSpot
In-Reply-To: <CAP=ZmBayvFs8b1CY2ATJ=_1ssG6eM0e5eeyBNEY85Zb5JnPK5w@mail.gmail.com>
References: <CAP=ZmBayvFs8b1CY2ATJ=_1ssG6eM0e5eeyBNEY85Zb5JnPK5w@mail.gmail.com>
Message-ID: <CAPE+kwDujuD4JSYCnb8_zq+siGttXJARUy+FBp2SyTxak1XimA@mail.gmail.com>

Hi,

To print out the grid temperatures of the internal layers, you would have
to modify the source code. dump_top_layer_temp_grid just writes out the
layer0 temperatures. (as shown in its code below:)

for(i=0;  i < model->rows; i++)
        for(j=0;  j < model->cols; j++)
            fprintf(fp, "%d\t%.2f\n", i*model->cols+j,
temp->cuboid[0][i][j]);

For other layers, you would have to change temp->cuboid[0][i][j] to
temp->cuboid[n][i][j] where n is the layer number.

For example, the following code prints layer n of the most recently
computed steady state temperatures onto stdout

for(i=0;  i < model->rows; i++)
        for(j=0;  j < model->cols; j++)
            printf("%d\t%.2f\n", i*model->cols+j,
model->last_steady->cuboid[n][i][j]);


Hope this helps,
-karthik



On Wed, May 22, 2013 at 1:19 PM, Puskar Budhathoki <
puskar.budhathoki at gmail.com> wrote:

> Hi,
> I am doing 3D layered simulation using HotSpot 3D stacking capability and
> i want to generate  grid steady file of each layer so that i can plot them
> using grid thermal map. I tried to use  dump_top_layer_temp_grid(model,
> file, model->last_steady); so that it can write grid steady value in file
> but i have no idea how to use it and where to use it?
>
> could you please help me to write grid steady value of each layer ?
>
> Hoping for positive response.
>
> Reagrds,
> Puskar Budhathoki
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20130522/6f6181e1/attachment.html