From lhuang at cse.cuhk.edu.hk Wed Apr 8 05:41:14 2009 From: lhuang at cse.cuhk.edu.hk (lhuang at cse.cuhk.edu.hk) Date: Wed, 8 Apr 2009 20:41:14 +0800 Subject: [Hotspot] About power trace file for Alpha 21264 Message-ID: <1239194474.49dc9b6ac1f90@webmail.cse.cuhk.edu.hk> Hi, All I'm working on temperature esitmation of Alpha 21264. The blocks shown in die photo include FBOX, FP IBOX, BIU, EBOX, MBOX AND WIRING, INSTR FETCH, INSTR CACHE, and DATA CACHE. But the output of Wattch shows rename_power, bpred_power, window_power, lsq_power, regfile_power, icache_power, dcache_power, dcache2_power, alu_power, falu_power, resultbut_power, clock_power, fetch_power, dispatch_power, issue_power. The names are not one-to-one correspond. Could any one tell me how to map these power consumption values into blocks? Another question: EV6.flp in the HotSpot 4.1 package is a floorplan of Alpha 21264, right? Thanks in advance. Best regards, Huang Lin From ks4kk at cs.virginia.edu Wed Apr 8 06:16:03 2009 From: ks4kk at cs.virginia.edu (Karthik Sankaranarayanan) Date: Wed, 8 Apr 2009 09:16:03 -0400 (EDT) Subject: [Hotspot] About power trace file for Alpha 21264 In-Reply-To: <1239194474.49dc9b6ac1f90@webmail.cse.cuhk.edu.hk> References: <1239194474.49dc9b6ac1f90@webmail.cse.cuhk.edu.hk> Message-ID: The floorplan modeled in HotSpot is ``similar'' to the Alpha. You would have to modify simplescalar to model an Alpha-like architecture. Please refer to Figure 4 of the following paper for more details: http://www.cs.virginia.edu/~skadron/Papers/hotspot_isca03.pdf Thanks, -karthik On Wed, 8 Apr 2009, lhuang at cse.cuhk.edu.hk wrote: > Hi, All > > I'm working on temperature esitmation of Alpha 21264. The blocks shown in die > photo include FBOX, FP IBOX, BIU, EBOX, MBOX AND WIRING, INSTR FETCH, INSTR > CACHE, and DATA CACHE. But the output of Wattch shows rename_power, > bpred_power, window_power, lsq_power, regfile_power, icache_power, > dcache_power, dcache2_power, alu_power, falu_power, resultbut_power, > clock_power, fetch_power, dispatch_power, issue_power. The names are not > one-to-one correspond. Could any one tell me how to map these power consumption > values into blocks? > > Another question: EV6.flp in the HotSpot 4.1 package is a floorplan of Alpha > 21264, right? Thanks in advance. > > Best regards, > Huang Lin > _______________________________________________ > HotSpot mailing list > HotSpot at mail.cs.virginia.edu > http://www.cs.virginia.edu/mailman/listinfo/hotspot > From skadron at cs.virginia.edu Sat Apr 25 13:55:48 2009 From: skadron at cs.virginia.edu (Kevin Skadron) Date: Sat, 25 Apr 2009 16:55:48 -0400 Subject: [Hotspot] My Deep Thanks to Hotspot Team In-Reply-To: <858616.14650.qm@web53712.mail.re2.yahoo.com> References: <858616.14650.qm@web53712.mail.re2.yahoo.com> Message-ID: <49F378D4.8000504@cs.virginia.edu> Glad to hear it worked out so well! /K Mohamed Elsawaf wrote: > Dear Drs > > On behalf of our research group; I would like to express our deep thanks > for your help; we wouldn't be able to complete our paper without your > advices. > All simulations were done using Hotspot 4.1 simulator. > > I would like to share our latest paper "CPU dynamic thermal management > via thermal spare cores" published at IEEE. > > http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?isnumber=4810723&arnumber=4810755&count=65&index=31 > > also seminar presentation at http://elsawaf.8m.com/favorite_links_1.html > I hope you find it useful > > Thank you > M. A Elsawaf > Faculty of Engineering, Cairo University, Egypt. > > From mohamed_elsawaf at yahoo.com Sat Apr 25 12:24:16 2009 From: mohamed_elsawaf at yahoo.com (Mohamed Elsawaf) Date: Sat, 25 Apr 2009 12:24:16 -0700 (PDT) Subject: [Hotspot] My Deep Thanks to Hotspot Team Message-ID: <858616.14650.qm@web53712.mail.re2.yahoo.com> Dear Drs On behalf of our research group; I would like to express our deep thanks for your help; we wouldn't be able to complete our paper without your advices. All simulations were done using Hotspot 4.1 simulator.? I would like to share our latest paper "CPU dynamic thermal management via thermal spare cores" published at IEEE.? http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?isnumber=4810723&arnumber=4810755&count=65&index=31 also seminar presentation at http://elsawaf.8m.com/favorite_links_1.html I hope you find it useful Thank you M. A Elsawaf Faculty of Engineering, Cairo University, Egypt. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.cs.virginia.edu/pipermail/hotspot/attachments/20090425/746bb99f/attachment-0001.html