From ramyamenonc at gmail.com Thu Jul 7 21:43:53 2011 From: ramyamenonc at gmail.com (ramya menon) Date: Fri, 8 Jul 2011 10:13:53 +0530 Subject: [Hotspot] ptrace values Message-ID: Respected sir, I am pretty new to the simulator hotspot. Can you please provide me the ptrace input values for different blocks like memories and processors at 3GHz frequency for 45nm technology. It is really urgent. I am eagerly waiting for your reply. Thank you; Ramya Menon .C -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20110708/2cc080aa/attachment.html From skadron at cs.virginia.edu Fri Jul 8 11:51:08 2011 From: skadron at cs.virginia.edu (Kevin Skadron) Date: Fri, 08 Jul 2011 14:51:08 -0400 Subject: [Hotspot] ptrace values In-Reply-To: References: Message-ID: <4E17519C.5010506@cs.virginia.edu> I'm sorry we can't help you directly on this. HotSpot doesn't contain a power model; power values for the ptrace need to come from some external tool, such as McPat or Wattch. We don't have anything in releasable form at the moment. --Kevin On 7/8/11 12:43 AM, ramya menon wrote: > Respected sir, > > I am pretty new to the simulator hotspot. Can you please provide me the > ptrace input values for different blocks like memories and processors at > 3GHz frequency for 45nm technology. It is really urgent. I am eagerly > waiting for your reply. > > Thank you; > Ramya Menon .C > > > _______________________________________________ > HotSpot mailing list > HotSpot at mail.cs.virginia.edu > http://www.cs.virginia.edu/mailman/listinfo/hotspot From ostrowski.d at husky.neu.edu Mon Jul 11 19:46:06 2011 From: ostrowski.d at husky.neu.edu (Devon Ostrowski) Date: Mon, 11 Jul 2011 22:46:06 -0400 Subject: [Hotspot] ptrace values In-Reply-To: <4E17519C.5010506@cs.virginia.edu> References: <4E17519C.5010506@cs.virginia.edu> Message-ID: <4E1BB56E.4080203@husky.neu.edu> Hi Professor, I have obtained the Wattch model and ran some simulations with it. While the simulation statistics do show the power statistics for the functional units, there was no power trace file generated. Is there some option that needs to be set in order for one to be generated? Or does the file need to be manually created from the simulation statistics? Thank you very much for your time. Devon Ostrowski On 7/8/2011 2:51 PM, Kevin Skadron wrote: > I'm sorry we can't help you directly on this. HotSpot doesn't contain a > power model; power values for the ptrace need to come from some external > tool, such as McPat or Wattch. We don't have anything in releasable > form at the moment. > > --Kevin > > On 7/8/11 12:43 AM, ramya menon wrote: >> Respected sir, >> >> I am pretty new to the simulator hotspot. Can you please provide me the >> ptrace input values for different blocks like memories and processors at >> 3GHz frequency for 45nm technology. It is really urgent. I am eagerly >> waiting for your reply. >> >> Thank you; >> Ramya Menon .C >> >> >> _______________________________________________ >> HotSpot mailing list >> HotSpot at mail.cs.virginia.edu >> http://www.cs.virginia.edu/mailman/listinfo/hotspot > _______________________________________________ > HotSpot mailing list > HotSpot at mail.cs.virginia.edu > http://www.cs.virginia.edu/mailman/listinfo/hotspot From skadron at cs.virginia.edu Mon Jul 11 20:00:35 2011 From: skadron at cs.virginia.edu (Kevin Skadron) Date: Mon, 11 Jul 2011 23:00:35 -0400 Subject: [Hotspot] ptrace values In-Reply-To: <4E1BB56E.4080203@husky.neu.edu> References: <4E17519C.5010506@cs.virginia.edu> <4E1BB56E.4080203@husky.neu.edu> Message-ID: <4E1BB8D3.2030806@cs.virginia.edu> The ptrace file needs to be manually generated. It's best to modify the Wattch code that currently collects power statistics to write the ptrace file at the desired interval. /K On 7/11/2011 10:46 PM, Devon Ostrowski wrote: > Hi Professor, > > I have obtained the Wattch model and ran some simulations with it. > While the simulation statistics do show the power statistics for the > functional units, there was no power trace file generated. Is there > some option that needs to be set in order for one to be generated? Or > does the file need to be manually created from the simulation statistics? > > Thank you very much for your time. > Devon Ostrowski > > > On 7/8/2011 2:51 PM, Kevin Skadron wrote: >> I'm sorry we can't help you directly on this. HotSpot doesn't contain a >> power model; power values for the ptrace need to come from some external >> tool, such as McPat or Wattch. We don't have anything in releasable >> form at the moment. >> >> --Kevin >> >> On 7/8/11 12:43 AM, ramya menon wrote: >>> Respected sir, >>> >>> I am pretty new to the simulator hotspot. Can you please provide me the >>> ptrace input values for different blocks like memories and processors at >>> 3GHz frequency for 45nm technology. It is really urgent. I am eagerly >>> waiting for your reply. >>> >>> Thank you; >>> Ramya Menon .C >>> >>> >>> _______________________________________________ >>> HotSpot mailing list >>> HotSpot at mail.cs.virginia.edu >>> http://www.cs.virginia.edu/mailman/listinfo/hotspot >> _______________________________________________ >> HotSpot mailing list >> HotSpot at mail.cs.virginia.edu >> http://www.cs.virginia.edu/mailman/listinfo/hotspot > > _______________________________________________ > HotSpot mailing list > HotSpot at mail.cs.virginia.edu > http://www.cs.virginia.edu/mailman/listinfo/hotspot From fsamie at gmail.com Tue Jul 12 12:12:59 2011 From: fsamie at gmail.com (Farzad Samie) Date: Tue, 12 Jul 2011 23:42:59 +0430 Subject: [Hotspot] ptrace values Message-ID: For generating power trace file, you need to use a power model simulator (e.g. Wattch). But Wattch does not have the same functional units as HotSpot. You should use a power simulator like PTscalar which is based on Wattch and has the corresponding units of HotSpot. It has the same floorplann. You need to make a small change in its code i.e. print the power values into a file. That's it. It's simple. On Tue, Jul 12, 2011 at 11:30 PM, wrote: > Send HotSpot mailing list submissions to > hotspot at mail.cs.virginia.edu > > To subscribe or unsubscribe via the World Wide Web, visit > http://www.cs.virginia.edu/mailman/listinfo/hotspot > or, via email, send a message with subject or body 'help' to > hotspot-request at mail.cs.virginia.edu > > You can reach the person managing the list at > hotspot-owner at mail.cs.virginia.edu > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of HotSpot digest..." > > > Today's Topics: > > 1. Re: ptrace values (Devon Ostrowski) > 2. Re: ptrace values (Kevin Skadron) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Mon, 11 Jul 2011 22:46:06 -0400 > From: Devon Ostrowski > Subject: Re: [Hotspot] ptrace values > To: skadron at cs.virginia.edu > Cc: hotspot at mail.cs.virginia.edu > Message-ID: <4E1BB56E.4080203 at husky.neu.edu> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > Hi Professor, > > I have obtained the Wattch model and ran some simulations with it. > While the simulation statistics do show the power statistics for the > functional units, there was no power trace file generated. Is there > some option that needs to be set in order for one to be generated? Or > does the file need to be manually created from the simulation statistics? > > Thank you very much for your time. > Devon Ostrowski > > > On 7/8/2011 2:51 PM, Kevin Skadron wrote: > > I'm sorry we can't help you directly on this. HotSpot doesn't contain a > > power model; power values for the ptrace need to come from some external > > tool, such as McPat or Wattch. We don't have anything in releasable > > form at the moment. > > > > --Kevin > > > > On 7/8/11 12:43 AM, ramya menon wrote: > >> Respected sir, > >> > >> I am pretty new to the simulator hotspot. Can you please provide me the > >> ptrace input values for different blocks like memories and processors at > >> 3GHz frequency for 45nm technology. It is really urgent. I am eagerly > >> waiting for your reply. > >> > >> Thank you; > >> Ramya Menon .C > >> > >> > >> _______________________________________________ > >> HotSpot mailing list > >> HotSpot at mail.cs.virginia.edu > >> http://www.cs.virginia.edu/mailman/listinfo/hotspot > > _______________________________________________ > > HotSpot mailing list > > HotSpot at mail.cs.virginia.edu > > http://www.cs.virginia.edu/mailman/listinfo/hotspot > > > > ------------------------------ > > Message: 2 > Date: Mon, 11 Jul 2011 23:00:35 -0400 > From: Kevin Skadron > Subject: Re: [Hotspot] ptrace values > To: hotspot at mail.cs.virginia.edu > Message-ID: <4E1BB8D3.2030806 at cs.virginia.edu> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > The ptrace file needs to be manually generated. It's best to modify the > Wattch code that currently collects power statistics to write the ptrace > file at the desired interval. > > /K > > On 7/11/2011 10:46 PM, Devon Ostrowski wrote: > > Hi Professor, > > > > I have obtained the Wattch model and ran some simulations with it. > > While the simulation statistics do show the power statistics for the > > functional units, there was no power trace file generated. Is there > > some option that needs to be set in order for one to be generated? Or > > does the file need to be manually created from the simulation statistics? > > > > Thank you very much for your time. > > Devon Ostrowski > > > > > > On 7/8/2011 2:51 PM, Kevin Skadron wrote: > >> I'm sorry we can't help you directly on this. HotSpot doesn't contain a > >> power model; power values for the ptrace need to come from some external > >> tool, such as McPat or Wattch. We don't have anything in releasable > >> form at the moment. > >> > >> --Kevin > >> > >> On 7/8/11 12:43 AM, ramya menon wrote: > >>> Respected sir, > >>> > >>> I am pretty new to the simulator hotspot. Can you please provide me the > >>> ptrace input values for different blocks like memories and processors > at > >>> 3GHz frequency for 45nm technology. It is really urgent. I am eagerly > >>> waiting for your reply. > >>> > >>> Thank you; > >>> Ramya Menon .C > >>> > >>> > >>> _______________________________________________ > >>> HotSpot mailing list > >>> HotSpot at mail.cs.virginia.edu > >>> http://www.cs.virginia.edu/mailman/listinfo/hotspot > >> _______________________________________________ > >> HotSpot mailing list > >> HotSpot at mail.cs.virginia.edu > >> http://www.cs.virginia.edu/mailman/listinfo/hotspot > > > > _______________________________________________ > > HotSpot mailing list > > HotSpot at mail.cs.virginia.edu > > http://www.cs.virginia.edu/mailman/listinfo/hotspot > > > ------------------------------ > > _______________________________________________ > HotSpot mailing list > HotSpot at mail.cs.virginia.edu > http://www.cs.virginia.edu/mailman/listinfo/hotspot > > > End of HotSpot Digest, Vol 48, Issue 2 > ************************************** > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20110712/78a89ef1/attachment.html From bhanu at iith.ac.in Wed Jul 13 12:06:57 2011 From: bhanu at iith.ac.in (Bhanu Prathap T) Date: Thu, 14 Jul 2011 00:36:57 +0530 Subject: [Hotspot] flp generation In-Reply-To: References: Message-ID: Hello Sir, I am trying to get a floorplan file for one of ev6 processors using hotfloorplan , given the area and power of the each module. In the .desc file , i have given the areas each block and 1 ,3 ,1 for min , max aspect ratios , rotables respectively . I haven't given the connectivity information as i am not aware of how they are connected . alpha.pdf provided along with this mail is the output after simulation. I have also provided other input files for reference. I want to know why are the empty spaces being filled by _some random number. How is the area of total area of the floorplan fixed , where can i modify it. ( may be empty spaces were created as the total area is somehow fixed .) Hoping patiently for faster reply. Thanking You Sir. regards, Bhanu Prathap.T -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20110714/22edef84/attachment.html -------------- next part -------------- A non-text attachment was scrubbed... Name: alpha.desc Type: application/octet-stream Size: 1046 bytes Desc: not available Url : http://www.cs.virginia.edu/pipermail/hotspot/attachments/20110714/22edef84/attachment.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: alpha.pdf Type: application/pdf Size: 3901 bytes Desc: not available Url : http://www.cs.virginia.edu/pipermail/hotspot/attachments/20110714/22edef84/attachment.pdf -------------- next part -------------- A non-text attachment was scrubbed... Name: alpha.p Type: text/x-pascal Size: 408 bytes Desc: not available Url : http://www.cs.virginia.edu/pipermail/hotspot/attachments/20110714/22edef84/attachment.bin From ks4kk at virginia.edu Thu Jul 14 10:06:02 2011 From: ks4kk at virginia.edu (Karthik Sankaranarayanan) Date: Thu, 14 Jul 2011 10:06:02 -0700 Subject: [Hotspot] flp generation In-Reply-To: References: Message-ID: Hi Bhanu Pratap, The total area of the generated floorplan can be controlled by the weights "lambdaA, lambdaT and lambdaW" in the hotspot.config file. The default weights that come with HotFloorplan were tuned for the EV6-like flooplan that is released with HotSpot. You might want to tune them to suit your case. Increasing lambdaA will give rise to tighter floorplans while increasing lambdaT will give rise to "cooler" floorplans. Similarly (when the wire densities are specified), increasing lambdaW will focus the optimization more on lowering the wire length metric. Hope this helps. Thanks, -karthik 2011/7/13 Bhanu Prathap T > Hello Sir, > I am trying to get a floorplan file for one of ev6 > processors using hotfloorplan , given the area and power of the each module. > In the .desc file , i have given the areas each block and 1 ,3 ,1 for min , > max aspect ratios , rotables respectively . I haven't given the connectivity > information as i am not aware of how they are connected . > > alpha.pdf provided along with this mail is the output after simulation. I > have also provided other input files for reference. > > I want to know why are the empty spaces being filled by _some random > number. > How is the area of total area of the floorplan fixed , where can i modify > it. ( may be empty spaces were created as the total area is somehow fixed .) > > Hoping patiently for faster reply. Thanking You Sir. > > regards, > Bhanu Prathap.T > > > _______________________________________________ > 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/20110714/472858c6/attachment.html From yishuihan.zc at gmail.com Mon Jul 18 21:40:59 2011 From: yishuihan.zc at gmail.com (Chao Zhang) Date: Tue, 19 Jul 2011 12:40:59 +0800 Subject: [Hotspot] [problem report] out put file format of HotSpot Message-ID: Hi~ I use the tool with the command as follow in order, but got the output .svg file wrong. ./hotspot -c hotspot.config -f ev6.flp -p gcc.ptrace \ -steady_file gcc.steady -model_type grid cp gcc.steady gcc.init ./hotspot -c hotspot.config -init_file gcc.init -f ev6.flp \ -p gcc.ptrace -o gcc.ttrace -model_type grid ./hotspot -c hotspot.config -f ev6.flp -p gcc.ptrace \ -steady_file gcc.steady -model_type grid \ -grid_steady_file gcc.grid.steady ./grid_thermal_map.pl ev6.flp gcc.grid.steady > gcc_1.svg The output gcc.grid.steady is of format as which is not 2 number a line as demo. So do you know the reason, or just mine wrong action? Thanks very much! -- Chao Zhang Electronic Engineering of EECS Department Yuanpei Program of Peking University Tel?+86 151-200-78471 Email?yishuihan.zc at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20110719/8299bee0/attachment.html From Runjie at virginia.edu Mon Jul 18 22:31:02 2011 From: Runjie at virginia.edu (Runjie Zhang) Date: Mon, 18 Jul 2011 22:31:02 -0700 Subject: [Hotspot] [problem report] out put file format of HotSpot In-Reply-To: References: Message-ID: Hi, Chao Good catch. The three number format was actually used for our other work and it was accidentally released. I've fixed it and you can download it again to get the original grid.steady format. Sorry for the inconvenience and thanks for pointing this out. Runjie On Mon, Jul 18, 2011 at 9:40 PM, Chao Zhang wrote: > Hi~ > I use the tool with the command as follow in order, but got the output .svg > file wrong. > > ./hotspot -c hotspot.config -f ev6.flp -p gcc.ptrace \ > -steady_file gcc.steady -model_type grid > cp gcc.steady gcc.init > ./hotspot -c hotspot.config -init_file gcc.init -f ev6.flp \ > -p gcc.ptrace -o gcc.ttrace -model_type grid > ./hotspot -c hotspot.config -f ev6.flp -p gcc.ptrace \ > -steady_file gcc.steady -model_type grid \ > -grid_steady_file gcc.grid.steady > ./grid_thermal_map.pl ev6.flp gcc.grid.steady > gcc_1.svg > > > The output gcc.grid.steady is of format as which > is not 2 number a line as demo. > > So do you know the reason, or just mine wrong action? > > Thanks very much! > -- > Chao Zhang > Electronic Engineering of EECS Department > Yuanpei Program of Peking University > Tel?+86 151-200-78471 > Email?yishuihan.zc at gmail.com > > > _______________________________________________ > HotSpot mailing list > HotSpot at mail.cs.virginia.edu > http://www.cs.virginia.edu/mailman/listinfo/hotspot > > -- Runjie Zhang Computer Engineering University of Virginia -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20110718/31205e83/attachment.html