From eae030c9d1d607d1c14e4ade99cb5caea6403efd Mon Sep 17 00:00:00 2001
From: aamir <araihan@ece.ubc.ca>
Date: Sat, 3 Nov 2018 17:03:58 -0700
Subject: merged with memory subsytem. Regression is passing but tensorcore
 kernel is stuck in deadlock

---
 configs/QuadroFX5600/gpuwattch_quadrofx5600.xml | 538 ------------------------
 1 file changed, 538 deletions(-)
 delete mode 100644 configs/QuadroFX5600/gpuwattch_quadrofx5600.xml

(limited to 'configs/QuadroFX5600/gpuwattch_quadrofx5600.xml')

diff --git a/configs/QuadroFX5600/gpuwattch_quadrofx5600.xml b/configs/QuadroFX5600/gpuwattch_quadrofx5600.xml
deleted file mode 100644
index 2c5a6fc..0000000
--- a/configs/QuadroFX5600/gpuwattch_quadrofx5600.xml
+++ /dev/null
@@ -1,538 +0,0 @@
-<?xml version="1.0" ?>
-<component id="root" name="root">
-	<component id="system" name="system">
-		<!--McPAT will skip the components if number is set to 0 -->
-		<param name="GPU_Architecture" value="0"/><!-- 0-G80; 1-Fermi; others not supported -->
-		<param name="number_of_cores" value="16"/>
-		<param name="architecture" value="2"/> <!-- fermi:1 quadro:2 other: undefined-->
-		<param name="number_of_L1Directories" value="0"/>
-		<param name="number_of_L2Directories" value="0"/>
-		<param name="number_of_L2s" value="0"/> <!-- This number means how many L2 clusters in each cluster there can be multiple banks/ports -->
-		<param name="number_of_L3s" value="0"/> <!-- This number means how many L3 clusters -->
-		<param name="number_of_NoCs" value="1"/> 
-		<param name="homogeneous_cores" value="1"/><!--1 means homo -->
-		<param name="homogeneous_L2s" value="1"/>
-		<param name="homogeneous_L1Directorys" value="1"/>
-		<param name="homogeneous_L2Directorys" value="1"/>
-		<param name="homogeneous_L3s" value="1"/>
-		<param name="homogeneous_ccs" value="1"/><!--cache coherece hardware -->
-		<param name="homogeneous_NoCs" value="0"/>
-		<param name="core_tech_node" value="65"/><!-- nm -->
-		<param name="target_core_clockrate" value="650"/><!--MHz -->
-		<param name="temperature" value="400"/> <!-- Kelvin -->
-		<param name="number_cache_levels" value="2"/>
-		<param name="interconnect_projection_type" value="0"/><!--0: agressive wire technology; 1: conservative wire technology -->
-		<param name="device_type" value="0"/><!--0: HP(High Performance Type); 1: LSTP(Low standby power) 2: LOP (Low Operating Power)  -->
-		<param name="longer_channel_device" value="1"/><!-- 0 no use; 1 use when possible -->
-		<param name="machine_bits" value="32"/>
-		<param name="virtual_address_width" value="32"/>
-		<param name="physical_address_width" value="32"/>
-		<param name="virtual_memory_page_size" value="4096"/>
-		<param name="idle_core_power" value="1.54"/><!-- idle core power for GTX479 -->
-		<param name="TOT_INST" value="2" />
-		<param name="FP_INT" value="5" />
-		<param name="IC_H" value="10" />
-		<param name="IC_M" value="10" />
-		<param name="DC_RH" value="10" />
-		<param name="DC_RM" value="10" />
-		<param name="DC_WH" value="10" />
-		<param name="DC_WM" value="10" />
-		<param name="TC_H" value="20" />
-		<param name="TC_M" value="20" />
-		<param name="CC_H" value="10" />
-		<param name="CC_M" value="10" />
-		<param name="SHRD_ACC" value="20" />
-		<param name="REG_RD" value="0.25" />
-		<param name="REG_WR" value="0.4" />
-		<param name="NON_REG_OPs" value="0.4" />
-		<param name="SP_ACC" value="0.6" />
-		<param name="SFU_ACC" value="0.7" />
-		<param name="FPU_ACC" value="0.25" />
-		<param name="MEM_RD" value="0.3" />
-		<param name="MEM_WR" value="0.4" />
-		<param name="MEM_PRE" value="0.7" />
-		<param name="L2_RH" value="1" />
-		<param name="L2_RM" value="1" />
-		<param name="L2_WH" value="1" />
-		<param name="L2_WM" value="1" />
-		<param name="NOC_A" value="80" />
-		<param name="PIPE_A" value="0.8" />
-		<param name="IDLE_CORE_N" value="1"/>
-		<param name="CONST_DYNAMICN" value="9" />
-		<stat name="num_idle_cores" value="0"/><!-- Average Number of idle cores during this period -->
-		<stat name="total_cycles" value="total_cycles_match_mcpat"/>
-		<stat name="idle_cycles" value="idle_cycles_match_mcpat"/>
-		<stat name="busy_cycles"  value="busy_cycles_match_mcpat"/>
-			<!--This page size(B) is complete different from the page size in Main memo secction. this page size is the size of 
-			virtual memory from OS/Archi perspective; the page size in Main memo secction is the actuall physical line in a DRAM bank  -->
-		<!-- *********************** cores ******************* -->
-		<component id="system.core0" name="core0">
-			<!-- Core property -->
-			<param name="clock_rate" value="650"/>
-			<param name="instruction_length" value="32"/>
-			<param name="opcode_width" value="9"/>
-			<!-- address width determins the tag_width in Cache, LSQ and buffers in cache controller 
-			default value is machine_bits, if not set --> 
-			<param name="machine_type" value="1"/><!-- 1 inorder; 0 OOO-->
-			<!-- inorder/OoO -->
-			<param name="number_hardware_threads" value="32"/>
-			<!-- number_instruction_fetch_ports(icache ports) is always 1 in single-thread processor,
-			it only may be more than one in SMT processors. BTB ports always equals to fetch ports since 
-			branch information in consective branch instructions in the same fetch group can be read out from BTB once.--> 
-			<param name="fetch_width" value="1"/>
-			<!-- fetch_width determins the size of cachelines of L1 cache block -->
-			<param name="number_instruction_fetch_ports" value="1"/>
-			<param name="decode_width" value="1"/>
-			<!-- decode_width determins the number of ports of the 
-			renaming table (both RAM and CAM) scheme -->
-			<param name="issue_width" value="1"/>
-			<!-- issue_width determins the number of ports of Issue window and other logic 
-			as in the complexity effective proccessors paper; issue_width==dispatch_width -->
-			<param name="commit_width" value="1"/>
-			<!-- commit_width determins the number of ports of register files -->
-			<param name="fp_issue_width" value="1"/>
-			<param name="prediction_width" value="0"/> 
-			<!-- number of branch instructions can be predicted simultannouesl-->
-			<!-- Current version of McPAT does not distinguish int and floating point pipelines 
-			Theses parameters are reserved for future use.--> 
-			<param name="pipelines_per_core" value="1,1"/>
-			<!--integer_pipeline and floating_pipelines, if the floating_pipelines is 0, then the pipeline is shared-->
-			<param name="pipeline_depth" value="8,8"/>
-			<!-- pipeline depth of int and fp, if pipeline is shared, the second number is the average cycles of fp ops -->
-			<!-- issue and exe unit-->
-			<param name="ALU_per_core" value="1"/>
-			<!-- contains an adder, a shifter, and a logical unit -->
-			<param name="MUL_per_core" value="1"/>
-			<!-- For MUL and Div -->
-			<param name="FPU_per_core" value="8"/>		
-			<!-- buffer between IF and ID stage -->
-			<param name="instruction_buffer_size" value="1"/>
-			<!-- buffer between ID and sche/exe stage -->
-			<param name="decoded_stream_buffer_size" value="1"/>
-			<param name="instruction_window_scheme" value="0"/><!-- 0 PHYREG based, 1 RSBASED-->
-			<!-- McPAT support 2 types of OoO cores, RS based and physical reg based-->
-			<param name="instruction_window_size" value="1"/>
-			<param name="fp_instruction_window_size" value="1"/>
-			<!-- the instruction issue Q as in Alpha 21264; The RS as in Intel P6 -->
-			<param name="ROB_size" value="0"/>
-			<!-- each in-flight instruction has an entry in ROB -->
-			<!-- registers -->
-
-			<!-- SM parameters  Added by Syed Gilani -->
-			<param name="rf_banks" value="16"/>
-			<param name="simd_width" value="8"/>
-			<param name="collector_units" value="16"/>
-			<param name="core_clock_ratio" value="2"/>
-			<param name="warp_size" value="32"/>
-
-			<param name="archi_Regs_IRF_size" value="8192"/>			
-			<param name="archi_Regs_FRF_size" value="32"/>
-			<!--  if OoO processor, phy_reg number is needed for renaming logic, 
-			renaming logic is for both integer and floating point insts.  -->
-			<param name="phy_Regs_IRF_size" value="32"/>
-			<param name="phy_Regs_FRF_size" value="32"/>
-			<!-- rename logic -->
-			<param name="rename_scheme" value="0"/>
-			<!-- can be RAM based(0) or CAM based(1) rename scheme 
-			RAM-based scheme will have free list, status table;
-			CAM-based scheme have the valid bit in the data field of the CAM 
-			both RAM and CAM need RAM-based checkpoint table, checkpoint_depth=# of in_flight instructions;
-			Detailed RAT Implementation see TR -->
-			<param name="register_windows_size" value="0"/>
-			<!-- how many windows in the windowed register file, sun processors;
-			no register windowing is used when this number is 0 -->
-			<!-- In OoO cores, loads and stores can be issued whether inorder(Pentium Pro) or (OoO)out-of-order(Alpha),
-			They will always try to exeute out-of-order though. -->
-			<param name="LSU_order" value="inorder"/>
-			<param name="store_buffer_size" value="32"/>
-			<!-- By default, in-order cores do not have load buffers -->
-			<param name="load_buffer_size" value="32"/>	
-			<!-- number of ports refer to sustainable concurrent memory accesses --> 
-			<param name="memory_ports" value="1"/>	
-			<!-- max_allowed_in_flight_memo_instructions determins the # of ports of load and store buffer
-			as well as the ports of Dcache which is connected to LSU -->	
-			<!-- dual-pumped Dcache can be used to save the extra read/write ports -->
-			<param name="RAS_size" value="1"/>						
-			<!-- general stats, defines simulation periods;require total, idle, and busy cycles for senity check  -->
-			<!-- please note: if target architecture is X86, then all the instrucions refer to (fused) micro-ops -->
-			<stat name="total_instructions" value="total_instructions_match_mcpat"/>
-			<stat name="int_instructions" value="int_instruction_match_mcpat"/>
-			<stat name="fp_instructions" value="flt_instruction_match_mcpat"/>
-			<stat name="branch_instructions" value="branch_instruction_match_mcpat"/>
-			<stat name="branch_mispredictions" value="0"/>
-			<stat name="load_instructions" value="load_instruction_match_mcpat"/>
-			<stat name="store_instructions" value="store_instruction_match_mcpat"/>
-			<stat name="committed_instructions" value="total_instructions_match_mcpat"/>
-			<stat name="committed_int_instructions" value="int_instruction_match_mcpat"/>
-			<stat name="committed_fp_instructions" value="flt_instruction_match_mcpat"/>
-			<stat name="pipeline_duty_cycle" value="0.6"/><!--<=1, runtime_ipc/peak_ipc; averaged for all cores if homogenous -->
-			<!-- the following cycle stats are used for heterogeneouse cores only, 
-				please ignore them if homogeneouse cores -->
-			<stat name="total_cycles" value="total_cycles_match_mcpat"/>
-		    <stat name="idle_cycles" value="idle_cycles_match_mcpat"/>
-		    <stat name="busy_cycles"  value="busy_cycles_match_mcpat"/>
-			<!-- instruction buffer stats -->
-			<!-- ROB stats, both RS and Phy based OoOs have ROB
-			performance simulator should capture the difference on accesses,
-			otherwise, McPAT has to guess based on number of commited instructions. -->
-			<stat name="ROB_reads" value="263886"/>
-			<stat name="ROB_writes" value="263886"/>
-			<!-- RAT accesses -->
-			<stat name="rename_accesses" value="263886"/>
-			<stat name="fp_rename_accesses" value="263886"/>
-			<!-- decode and rename stage use this, should be total ic - nop -->
-			<!-- Inst window stats -->
-			<stat name="inst_window_reads" value="263886"/>
-			<stat name="inst_window_writes" value="263886"/>
-			<stat name="inst_window_wakeup_accesses" value="263886"/>
-			<stat name="fp_inst_window_reads" value="263886"/>
-			<stat name="fp_inst_window_writes" value="263886"/>
-			<stat name="fp_inst_window_wakeup_accesses" value="263886"/>
-			<!--  RF accesses -->
-			<stat name="int_regfile_reads" value="int_register_read_access_match_mcpat"/>
-			<stat name="float_regfile_reads" value="int_register_write_access_match_mcpat"/>
-			<stat name="int_regfile_writes" value="float_register_read_access_match_mcpat"/>
-			<stat name="float_regfile_writes" value="float_register_write_access_match_mcpat"/>
-			
-			<!-- The following stat is for operand collector power - Added by Syed -->
-			<stat name="non_rf_operands" value="0"/>
-
-			<!-- accesses to the working reg -->
-			<stat name="function_calls" value="0"/>
-			<stat name="context_switches" value="0"/> <!--not used in the McPAT -->
-			<!-- Number of Windowes switches (number of function calls and returns)-->
-			<!-- Alu stats by default, the processor has one FPU that includes the divider and 
-			 multiplier. The fpu accesses should include accesses to multiplier and divider  -->
-			<stat name="ialu_accesses" value="ialu_accesses_match_mcpat"/>			
-			<stat name="fpu_accesses" value="fpu_accesses_match_mcpat"/>
-			<stat name="mul_accesses" value="mul_accesses_match_mcpat"/>
-			<stat name="cdb_alu_accesses" value="0"/>
-			<stat name="cdb_mul_accesses" value="0"/>
-			<stat name="cdb_fpu_accesses" value="0"/>
-			<!-- multiple cycle accesses should be counted multiple times, 
-			otherwise, McPAT can use internal counter for different floating point instructions 
-			to get final accesses. But that needs detailed info for floating point inst mix -->
-			<!--  currently the performance simulator should 
-			make sure all the numbers are final numbers, 
-			including the explicit read/write accesses, 
-			and the implicite accesses such as replacements and etc.
-			Future versions of McPAT may be able to reason the implicite access
-			based on param and stats of last level cache
-			The same rule applies to all cache access stats too!  -->
-			<!-- following is AF for max power computation. 
-				Do not change them, unless you understand them-->
-			<stat name="IFU_duty_cycle" value="0.25"/>			
-			<stat name="LSU_duty_cycle" value="0.25"/>
-			<stat name="MemManU_I_duty_cycle" value="1"/>
-			<stat name="MemManU_D_duty_cycle" value="0.25"/>
-			<stat name="ALU_duty_cycle" value="0.9"/>
-			<stat name="MUL_duty_cycle" value="0.5"/>
-			<stat name="FPU_duty_cycle" value="1"/><!-- FPU numbers are already average -->
-			<stat name="ALU_cdb_duty_cycle" value="0.9"/>
-			<stat name="MUL_cdb_duty_cycle" value="0.5"/>
-			<stat name="FPU_cdb_duty_cycle" value="0.4"/>
-			<component id="system.core0.predictor" name="PBT">
-				<!-- branch predictor; tournament predictor see Alpha implementation -->
-				<param name="local_predictor_size" value="10,3"/>
-				<param name="local_predictor_entries" value="1024"/>
-				<param name="global_predictor_entries" value="4096"/>
-				<param name="global_predictor_bits" value="2"/>
-				<param name="chooser_predictor_entries" value="4096"/>
-				<param name="chooser_predictor_bits" value="2"/>
-				<!-- These parameters can be combined like below in next version
-				<param name="load_predictor" value="10,3,1024"/>
-				<param name="global_predictor" value="4096,2"/>
-				<param name="predictor_chooser" value="4096,2"/>
-				-->
-			</component>
-			<component id="system.core0.itlb" name="itlb">
-				<param name="number_entries" value="1"/>
-				<stat name="total_accesses" value="1"/>
-				<stat name="total_misses" value="0"/>
-				<stat name="conflicts" value="0"/>	
-				<!-- there is no write requests to itlb although writes happen to itlb after miss, 
-				which is actually a replacement -->
-			</component>
-			<component id="system.core0.icache" name="icache">
-				<!-- there is no write requests to itlb although writes happen to it after miss, 
-				which is actually a replacement -->
-				<param name="icache_config" value="16384,32,4,1,1,3,8,0"/>
-				<!-- the parameters are capacity,block_width, associativity, bank, throughput w.r.t. core clock, latency w.r.t. core clock,output_width, cache policy -->
-				<!-- cache_policy;//0 no write or write-though with non-write allocate;1 write-back with write-allocate -->
-				<param name="buffer_sizes" value="16, 16, 16,0"/>
-				<!-- cache controller buffer sizes: miss_buffer_size(MSHR),fill_buffer_size,prefetch_buffer_size,wb_buffer_size--> 
-				<stat name="read_accesses" value="total_instructions_match_mcpat"/>
-				<stat name="read_misses" value="0"/>
-				<stat name="conflicts" value="0"/>				
-			</component>
-			<component id="system.core0.dtlb" name="dtlb">
-				<param name="number_entries" value="1"/>
-				<stat name="total_accesses" value="1"/>
-				<stat name="total_misses" value="0"/>
-				<stat name="conflicts" value="0"/>	
-			</component>
-			<component id="system.core0.ccache" name="ccache">
-			        <!-- all the buffer related are optional -->
-				<param name="ccache_config" value="65536,32,4,1,1,3,8,0"/>
-				<param name="buffer_sizes" value="16, 16, 16, 0"/>
-				<!-- cache controller buffer sizes: miss_buffer_size(MSHR),fill_buffer_size,prefetch_buffer_size,wb_buffer_size-->	
-				<stat name="read_accesses" value="ccache_read_accesses_match_mcpat"/>
-				<stat name="write_accesses" value="0"/>
-				<stat name="read_misses" value="ccache_read_misses_match_mcpat"/>
-				<stat name="write_misses" value="0"/>
-				<stat name="conflicts" value="0"/>	
-			</component>
-			<component id="system.core0.tcache" name="tcache">
-			        <!-- all the buffer related are optional -->
-				<param name="tcache_config" value="16384,32,4,1,1,3,8,0"/>
-				<param name="buffer_sizes" value="16, 16, 16, 0"/>
-				<!-- cache controller buffer sizes: miss_buffer_size(MSHR),fill_buffer_size,prefetch_buffer_size,wb_buffer_size-->	
-				<stat name="read_accesses" value="tcache_read_accesses_match_mcpat"/>
-				<stat name="write_accesses" value="0"/>
-				<stat name="read_misses" value="tcache_read_misses_match_mcpat"/>
-				<stat name="write_misses" value="0"/>
-				<stat name="conflicts" value="0"/>	
-			</component>
-      <!--model the shared memory by mimicing dcache-->
-			<component id="system.core0.sharedmemory" name="sharedmemory">
-			        <!-- all the buffer related are optional -->
-				<param name="sharedmemory_config" value="16384,16,1,16,1,3,16,0"/>
-				<!-- the parameters are capacity,block_width, associativity, bank, throughput w.r.t. core clock, latency w.r.t. core clock,output_width, cache policy -->
-				<param name="buffer_sizes" value="16, 16, 16, 16"/>
-				<!-- cache controller buffer sizes: miss_buffer_size(MSHR),fill_buffer_size,prefetch_buffer_size,wb_buffer_size-->	
-				<stat name="read_accesses" value="sharedmemory_read_access_match_mcpat"/>
-				<stat name="write_accesses" value="sharedmemory_write_access_match_mcpat"/>
-				<stat name="read_misses" value="0"/>
-				<stat name="write_misses" value="0"/>
-				<stat name="conflicts" value="0"/>	
-			</component>
-			<component id="system.core0.dcache" name="dcache">
-			        <!-- all the buffer related are optional -->
-				<param name="dcache_config" value="16384,32,4,1,1,3,8,0"/>
-				<param name="buffer_sizes" value="16, 16, 16, 0"/>
-				<!-- cache controller buffer sizes: miss_buffer_size(MSHR),fill_buffer_size,prefetch_buffer_size,wb_buffer_size-->	
-				<stat name="read_accesses" value="dcache_read_access_match_mcpat"/>
-				<stat name="write_accesses" value="dcache_write_access_match_mcpat"/>
-				<stat name="read_misses" value="dcache_read_miss_match_mcpat"/>
-				<stat name="write_misses" value="dcache_write_miss_match_mcpat"/>
-				<stat name="conflicts" value="0"/>	
-			</component>
-			<component id="system.core0.BTB" name="BTB">
-			        <!-- all the buffer related are optional -->
-				<param name="BTB_config" value="8192,4,2,1, 1,3"/>
-				<!-- the parameters are capacity,block_width,associativity,bank, throughput w.r.t. core clock, latency w.r.t. core clock,-->
-			</component>
-	</component>
-		<component id="system.L1Directory0" name="L1Directory0">
-				<param name="Directory_type" value="0"/>
-			    <!--0 cam based shadowed tag. 1 directory cache -->	
-				<param name="Dir_config" value="2048,1,0,1, 4, 4,8"/>
-				<!-- the parameters are capacity,block_width, associativity,bank, throughput w.r.t. core clock, latency w.r.t. core clock,-->
-			    <param name="buffer_sizes" value="8, 8, 8, 8"/>	
-				<!-- all the buffer related are optional -->
-			    <param name="clockrate" value="1300"/>
-				<param name="ports" value="1,1,1"/>
-				<!-- number of r, w, and rw search ports -->
-				<param name="device_type" value="0"/>
-				<!-- altough there are multiple access types, 
-				Performance simulator needs to cast them into reads or writes
-				e.g. the invalidates can be considered as writes -->
-				<stat name="read_accesses" value="800000"/>
-				<stat name="write_accesses" value="27276"/>
-				<stat name="read_misses" value="1632"/>
-				<stat name="write_misses" value="183"/>
-				<stat name="conflicts" value="20"/>	
-				<stat name="duty_cycle" value="0.45"/>	
-		</component>
-		<component id="system.L2Directory0" name="L2Directory0">
-				<param name="Directory_type" value="1"/>
-			    <!--0 cam based shadowed tag. 1 directory cache -->	
-				<param name="Dir_config" value="1048576,16,16,1,2, 100"/>
-				<!-- the parameters are capacity,block_width, associativity,bank, throughput w.r.t. core clock, latency w.r.t. core clock,-->
-			    <param name="buffer_sizes" value="8, 8, 8, 8"/>	
-				<!-- all the buffer related are optional -->
-			    <param name="clockrate" value="1300"/>
-				<param name="ports" value="1,1,1"/>
-				<!-- number of r, w, and rw search ports -->
-				<param name="device_type" value="0"/>
-				<!-- altough there are multiple access types, 
-				Performance simulator needs to cast them into reads or writes
-				e.g. the invalidates can be considered as writes -->
-				<stat name="read_accesses" value="58824"/>
-				<stat name="write_accesses" value="27276"/>
-				<stat name="read_misses" value="1632"/>
-				<stat name="write_misses" value="183"/>
-				<stat name="conflicts" value="100"/>
-			    <stat name="duty_cycle" value="0.45"/>		
-		</component>
-		<component id="system.L20" name="L20">
-			<!-- all the buffer related are optional -->
-				<param name="L2_config" value="786432,64,16,1, 4,23, 64, 1"/>
-			    <!-- consider 4-way bank interleaving for Niagara 1 -->
-				<!-- the parameters are capacity,block_width, associativity, bank, throughput w.r.t. core clock, latency w.r.t. core clock,output_width, cache policy -->
-				<param name="buffer_sizes" value="16, 16, 16, 16"/>
-				<!-- cache controller buffer sizes: miss_buffer_size(MSHR),fill_buffer_size,prefetch_buffer_size,wb_buffer_size-->	
-				<param name="clockrate" value="1300"/>
-				<param name="ports" value="1,1,1"/>
-				<!-- number of r, w, and rw ports -->
-				<param name="device_type" value="0"/>
-				<stat name="read_accesses" value="200000"/>
-				<stat name="write_accesses" value="0"/>
-				<stat name="read_misses" value="0"/>
-				<stat name="write_misses" value="0"/>
-				<stat name="conflicts" value="0"/>	
-			    <stat name="duty_cycle" value="0.5"/>	
-		</component>
-		
-<!--**********************************************************************-->
-<component id="system.L30" name="L30">
-				<param name="L3_config" value="1048576,64,16,1, 2,100, 64,1"/>
-				<!-- the parameters are capacity,block_width, associativity, bank, throughput w.r.t. core clock, latency w.r.t. core clock,output_width, cache policy -->
-				<param name="clockrate" value="3500"/>
-				<param name="ports" value="1,1,1"/>
-				<!-- number of r, w, and rw ports -->
-				<param name="device_type" value="0"/>
-				<param name="buffer_sizes" value="16, 16, 16, 16"/>
-				<!-- cache controller buffer sizes: miss_buffer_size(MSHR),fill_buffer_size,prefetch_buffer_size,wb_buffer_size-->	
-				<stat name="read_accesses" value="58824"/>
-				<stat name="write_accesses" value="27276"/>
-				<stat name="read_misses" value="1632"/>
-				<stat name="write_misses" value="183"/>
-				<stat name="conflicts" value="0"/>	
-	            <stat name="duty_cycle" value="0.35"/>	
-		</component>
-
-
-<!--**********************************************************************-->
-		<component id="system.NoC0" name="noc0">
-			<param name="clockrate" value="650"/>
-			<param name="type" value="1"/>
-			<!-- 1 NoC, O bus -->
-			<param name="horizontal_nodes" value="4"/>
-			<param name="vertical_nodes" value="1"/>
-			<param name="has_global_link" value="0"/>
-			<!-- 1 has global link, 0 does not have global link -->
-			<param name="link_throughput" value="1"/><!--w.r.t clock -->
-			<param name="link_latency" value="1"/><!--w.r.t clock -->
-			<!-- througput >= latency -->
-			<!-- Router architecture -->
-			<param name="input_ports" value="7"/>
-			<param name="output_ports" value="7"/>
-			<param name="virtual_channel_per_port" value="1"/>
-			<!-- input buffer; in classic routers only input ports need buffers -->
-			<param name="flit_bits" value="32"/>
-			<param name="input_buffer_entries_per_vc" value="1"/><!--VCs within the same ports share input buffers whose size is propotional to the number of VCs-->
-			<param name="chip_coverage" value="1"/>
-			<!-- When multiple NOC present, one NOC will cover part of the whole chip. chip_coverage <=1 -->
-			<stat name="total_accesses" value="360000"/>
-			<!-- This is the number of total accesses within the whole network not for each router -->
-			<stat name="duty_cycle" value="0.6"/>
-		</component>	
-<!--**********************************************************************-->
-<!--**********************************************************************-->
-
-		<component id="system.mem" name="mem">
-			<!-- Main memory property -->
-			<param name="mem_tech_node" value="65"/>
-			<param name="device_clock" value="200"/><!--MHz, this is clock rate of the actual memory device, not the FSB -->
-			<param name="peak_transfer_rate" value="3200"/><!--MB/S-->
-			<param name="internal_prefetch_of_DRAM_chip" value="4"/>
-			<!-- 2 for DDR, 4 for DDR2, 8 for DDR3...-->
-			<!-- the device clock, peak_transfer_rate, and the internal prefetch decide the DIMM property -->
-			<!-- above numbers can be easily found from Wikipedia -->
-			<param name="capacity_per_channel" value="4096"/> <!-- MB -->
-			<!-- capacity_per_Dram_chip=capacity_per_channel/number_of_dimms/number_ranks/Dram_chips_per_rank
-			Current McPAT assumes single DIMMs are used.--> 		
-			<param name="number_ranks" value="2"/>
-			<param name="num_banks_of_DRAM_chip" value="8"/>			
-			<param name="Block_width_of_DRAM_chip" value="64"/> <!-- B -->
-			<param name="output_width_of_DRAM_chip" value="8"/>
-			<!--number of Dram_chips_per_rank=" 72/output_width_of_DRAM_chip-->
-			<!--number of Dram_chips_per_rank=" 72/output_width_of_DRAM_chip-->
-			<param name="page_size_of_DRAM_chip" value="8"/> <!-- 8 or 16 -->
-			<param name="burstlength_of_DRAM_chip" value="8"/>
-			<stat name="memory_accesses" value="1052"/>
-			<stat name="memory_reads" value="1052"/>
-			<stat name="memory_writes" value="1052"/>									
-		</component>
-		<component id="system.mc" name="mc">
-			<!-- Memeory controllers are for DDR(2,3...) DIMMs -->
-			<!-- current version of McPAT uses published values for base parameters of memory controller
-			improvments on MC will be added in later versions. -->
-			<param name="type" value="0"/> <!-- 1: low power; 0 high performance -->
-			<param name="mc_clock" value="800"/><!--DIMM IO bus clock rate MHz DDR2-400 for Niagara 1--> 
-			<param name="peak_transfer_rate" value="9600"/><!--MB/S  Syed: Quadro FX 5800 has 76.8GB/s mem transfer rate with 6 MCs -->
-			<param name="block_size" value="64"/><!--B-->
-			<param name="number_mcs" value="8"/><!-- 6 GDDR5 memory controllers  -->
-			<!-- current McPAT only supports homogeneous memory controllers -->
-			<param name="memory_channels_per_mc" value="2"/>
-			<param name="number_ranks" value="1"/>
-			<param name="withPHY" value="0"/>
-			<!-- # of ranks of each channel-->
-			<param name="req_window_size_per_channel" value="16"/>
-			<param name="IO_buffer_size_per_channel" value="16"/>
-			<param name="databus_width" value="32"/>
-			<param name="addressbus_width" value="32"/>
-			<param name="PRT_entries" value="32"/>
-			<!-- # of empirical DRAM model parameter -->
-			<param name="dram_cmd_coeff" value="0"/>
-			<param name="dram_act_coeff" value="0"/>
-			<param name="dram_nop_coeff" value="0"/>
-			<param name="dram_activity_coeff" value="0"/>
-			<param name="dram_pre_coeff" value="1.53846154e-8f"/>
-			<param name="dram_rd_coeff" value="3.07692308e-8f"/>
-			<param name="dram_wr_coeff" value="1.30769231e-8f"/>
-			<param name="dram_req_coeff" value="0"/>
-			<param name="dram_const_coeff" value="0"/>
-
-			<!-- McPAT will add the control bus width to the addressbus width automatically -->
-			<stat name="memory_accesses" value="memory_accesses_match_mcpat"/>
-			<stat name="memory_reads" value="memory_reads_match_mcpat"/>
-			<stat name="memory_writes" value="memory_writes_match_mcpat"/>
-			<!-- McPAT does not track individual mc, instead, it takes the total accesses and calculate 
-			the average power per MC or per channel. This is sufficent for most application. 
-			Further trackdown can be easily added in later versions. -->  			
-		</component>
-<!--**********************************************************************-->
-		<component id="system.niu" name="niu">
-			<!-- On chip 10Gb Ethernet NIC, including XAUI Phy and MAC controller  -->
-			<!-- For a minimum IP packet size of 84B at 10Gb/s, a new packet arrives every 67.2ns. 
-				 the low bound of clock rate of a 10Gb MAC is 150Mhz -->
-			<param name="type" value="0"/> <!-- 1: low power; 0 high performance -->
-			<param name="clockrate" value="350"/>
-			<param name="number_units" value="0"/> <!-- unlike PCIe and memory controllers, each Ethernet controller only have one port -->
-			<stat name="duty_cycle" value="1.0"/> <!-- achievable max load <= 1.0 -->
-			<stat name="total_load_perc" value="0.7"/> <!-- ratio of total achived load to total achivable bandwidth  -->
-			<!-- McPAT does not track individual nic, instead, it takes the total accesses and calculate 
-			the average power per nic or per channel. This is sufficent for most application. -->  			
-		</component>
-<!--**********************************************************************-->
-		<component id="system.pcie" name="pcie">
-			<!-- On chip PCIe controller, including Phy-->
-			<!-- For a minimum PCIe packet size of 84B at 8Gb/s per lane (PCIe 3.0), a new packet arrives every 84ns. 
-				 the low bound of clock rate of a PCIe per lane logic is 120Mhz -->
-			<param name="type" value="0"/> <!-- 1: low power; 0 high performance -->
-			<param name="withPHY" value="1"/>
-			<param name="clockrate" value="350"/>
-			<param name="number_units" value="0"/>
-			<param name="num_channels" value="8"/> <!-- 2 ,4 ,8 ,16 ,32 -->
-			<stat name="duty_cycle" value="1.0"/> <!-- achievable max load <= 1.0 -->
-			<stat name="total_load_perc" value="0.7"/> <!-- Percentage of total achived load to total achivable bandwidth  -->
-			<!-- McPAT does not track individual pcie controllers, instead, it takes the total accesses and calculate 
-			the average power per pcie controller or per channel. This is sufficent for most application. -->  			
-		</component>
-<!--**********************************************************************-->
-		<component id="system.flashc" name="flashc">
-		    <param name="number_flashcs" value="0"/>
-			<param name="type" value="1"/> <!-- 1: low power; 0 high performance -->
-            <param name="withPHY" value="1"/>
-			<param name="peak_transfer_rate" value="200"/><!--Per controller sustainable reak rate MB/S -->
-			<stat name="duty_cycle" value="1.0"/> <!-- achievable max load <= 1.0 -->
-			<stat name="total_load_perc" value="0.7"/> <!-- Percentage of total achived load to total achivable bandwidth  -->
-			<!-- McPAT does not track individual flash controller, instead, it takes the total accesses and calculate 
-			the average power per fc or per channel. This is sufficent for most application -->  			
-		</component>
-<!--**********************************************************************-->
-
-		</component>
-</component>
-- 
cgit v1.3