195 lines
6.4 KiB
INI
195 lines
6.4 KiB
INI
|
# Cache size
|
||
|
//-size (bytes) 528
|
||
|
//-size (bytes) 4096
|
||
|
//-size (bytes) 262144
|
||
|
//-size (bytes) 1048576
|
||
|
//-size (bytes) 2097152
|
||
|
//-size (bytes) 4194304
|
||
|
//-size (bytes) 8388608
|
||
|
//-size (bytes) 16777216
|
||
|
//-size (bytes) 33554432
|
||
|
//-size (bytes) 134217728
|
||
|
//-size (bytes) 268435456
|
||
|
//-size (bytes) 536870912
|
||
|
//-size (bytes) 67108864
|
||
|
//-size (bytes) 536870912
|
||
|
//-size (bytes) 1073741824
|
||
|
# For 3D DRAM memory please use Gb as units
|
||
|
-size (Gb) 1
|
||
|
|
||
|
# Line size
|
||
|
//-block size (bytes) 8
|
||
|
-block size (bytes) 128
|
||
|
|
||
|
# To model Fully Associative cache, set associativity to zero
|
||
|
//-associativity 0
|
||
|
//-associativity 2
|
||
|
//-associativity 4
|
||
|
-associativity 1
|
||
|
//-associativity 16
|
||
|
|
||
|
-read-write port 1
|
||
|
-exclusive read port 0
|
||
|
-exclusive write port 0
|
||
|
-single ended read ports 0
|
||
|
|
||
|
# Multiple banks connected using a bus
|
||
|
-UCA bank count 8
|
||
|
//-technology (u) 0.032
|
||
|
//-technology (u) 0.040
|
||
|
//-technology (u) 0.065
|
||
|
-technology (u) 0.078
|
||
|
//-technology (u) 0.080
|
||
|
|
||
|
# following three parameters are meaningful only for main memories
|
||
|
|
||
|
//-page size (bits) 8192
|
||
|
-burst length 4
|
||
|
-internal prefetch width 1
|
||
|
|
||
|
# following parameter can have one of five values -- (itrs-hp, itrs-lstp, itrs-lop, lp-dram, comm-dram)
|
||
|
//-Data array cell type - "itrs-hp"
|
||
|
//-Data array cell type - "itrs-lstp"
|
||
|
//-Data array cell type - "itrs-lop"
|
||
|
-Data array cell type - "comm-dram"
|
||
|
|
||
|
# following parameter can have one of three values -- (itrs-hp, itrs-lstp, itrs-lop)
|
||
|
//-Data array peripheral type - "itrs-hp"
|
||
|
-Data array peripheral type - "itrs-lstp"
|
||
|
//-Data array peripheral type - "itrs-lop"
|
||
|
|
||
|
# following parameter can have one of five values -- (itrs-hp, itrs-lstp, itrs-lop, lp-dram, comm-dram)
|
||
|
-Tag array cell type - "itrs-hp"
|
||
|
//-Tag array cell type - "itrs-lstp"
|
||
|
|
||
|
# following parameter can have one of three values -- (itrs-hp, itrs-lstp, itrs-lop)
|
||
|
-Tag array peripheral type - "itrs-hp"
|
||
|
//-Tag array peripheral type - "itrs-lstp"
|
||
|
|
||
|
# Bus width include data bits and address bits required by the decoder
|
||
|
//-output/input bus width 16
|
||
|
//-output/input bus width 64
|
||
|
-output/input bus width 64
|
||
|
|
||
|
// 300-400 in steps of 10
|
||
|
-operating temperature (K) 350
|
||
|
|
||
|
# Type of memory - cache (with a tag array) or ram (scratch ram similar to a register file)
|
||
|
# or main memory (no tag array and every access will happen at a page granularity Ref: CACTI 5.3 report)
|
||
|
//-cache type "cache"
|
||
|
//-cache type "ram"
|
||
|
//-cache type "main memory"
|
||
|
-cache type "3D memory or 2D main memory"
|
||
|
|
||
|
## Parameters for 3D DRAM
|
||
|
-page size (bits) 16384
|
||
|
//-page size (bits) 8192
|
||
|
-burst depth 8
|
||
|
-IO width 4
|
||
|
-system frequency (MHz) 533
|
||
|
|
||
|
-stacked die count 1
|
||
|
-partitioning granularity 0 // 0: coarse-grained rank-level; 1: fine-grained rank-level
|
||
|
//-TSV projection 1 // 0: ITRS aggressive; 1: industrial conservative
|
||
|
|
||
|
## End of parameters for 3D DRAM
|
||
|
|
||
|
|
||
|
# to model special structure like branch target buffers, directory, etc.
|
||
|
# change the tag size parameter
|
||
|
# if you want cacti to calculate the tagbits, set the tag size to "default"
|
||
|
-tag size (b) "default"
|
||
|
//-tag size (b) 45
|
||
|
|
||
|
# fast - data and tag access happen in parallel
|
||
|
# sequential - data array is accessed after accessing the tag array
|
||
|
# normal - data array lookup and tag access happen in parallel
|
||
|
# final data block is broadcasted in data array h-tree
|
||
|
# after getting the signal from the tag array
|
||
|
-access mode (normal, sequential, fast) - "fast"
|
||
|
//-access mode (normal, sequential, fast) - "normal"
|
||
|
//-access mode (normal, sequential, fast) - "sequential"
|
||
|
|
||
|
|
||
|
# DESIGN OBJECTIVE for UCA (or banks in NUCA)
|
||
|
-design objective (weight delay, dynamic power, leakage power, cycle time, area) 0:0:0:0:10
|
||
|
|
||
|
# Percentage deviation from the minimum value
|
||
|
# Ex: A deviation value of 10:1000:1000:1000:1000 will try to find an organization
|
||
|
# that compromises at most 10% delay.
|
||
|
# NOTE: Try reasonable values for % deviation. Inconsistent deviation
|
||
|
# percentage values will not produce any valid organizations. For example,
|
||
|
# 0:0:100:100:100 will try to identify an organization that has both
|
||
|
# least delay and dynamic power. Since such an organization is not possible, CACTI will
|
||
|
# throw an error. Refer CACTI-6 Technical report for more details
|
||
|
-deviate (delay, dynamic power, leakage power, cycle time, area) 50:100000:100000:100000:1000000
|
||
|
|
||
|
# Objective for NUCA
|
||
|
-NUCAdesign objective (weight delay, dynamic power, leakage power, cycle time, area) 100:100:0:0:100
|
||
|
-NUCAdeviate (delay, dynamic power, leakage power, cycle time, area) 10:10000:10000:10000:10000
|
||
|
|
||
|
# Set optimize tag to ED or ED^2 to obtain a cache configuration optimized for
|
||
|
# energy-delay or energy-delay sq. product
|
||
|
# Note: Optimize tag will disable weight or deviate values mentioned above
|
||
|
# Set it to NONE to let weight and deviate values determine the
|
||
|
# appropriate cache configuration
|
||
|
//-Optimize ED or ED^2 (ED, ED^2, NONE): "ED"
|
||
|
//-Optimize ED or ED^2 (ED, ED^2, NONE): "ED^2"
|
||
|
-Optimize ED or ED^2 (ED, ED^2, NONE): "NONE"
|
||
|
|
||
|
-Cache model (NUCA, UCA) - "UCA"
|
||
|
//-Cache model (NUCA, UCA) - "NUCA"
|
||
|
|
||
|
# In order for CACTI to find the optimal NUCA bank value the following
|
||
|
# variable should be assigned 0.
|
||
|
-NUCA bank count 0
|
||
|
|
||
|
# NOTE: for nuca network frequency is set to a default value of
|
||
|
# 5GHz in time.c. CACTI automatically
|
||
|
# calculates the maximum possible frequency and downgrades this value if necessary
|
||
|
|
||
|
# By default CACTI considers both full-swing and low-swing
|
||
|
# wires to find an optimal configuration. However, it is possible to
|
||
|
# restrict the search space by changing the signaling from "default" to
|
||
|
# "fullswing" or "lowswing" type.
|
||
|
-Wire signaling (fullswing, lowswing, default) - "Global_30"
|
||
|
//-Wire signaling (fullswing, lowswing, default) - "default"
|
||
|
//-Wire signaling (fullswing, lowswing, default) - "lowswing"
|
||
|
|
||
|
//-Wire inside mat - "global"
|
||
|
-Wire inside mat - "semi-global"
|
||
|
-Wire outside mat - "global"
|
||
|
//-Wire outside mat - "semi-global"
|
||
|
|
||
|
-Interconnect projection - "conservative"
|
||
|
//-Interconnect projection - "aggressive"
|
||
|
|
||
|
# Contention in network (which is a function of core count and cache level) is one of
|
||
|
# the critical factor used for deciding the optimal bank count value
|
||
|
# core count can be 4, 8, or 16
|
||
|
//-Core count 4
|
||
|
-Core count 8
|
||
|
//-Core count 16
|
||
|
-Cache level (L2/L3) - "L3"
|
||
|
|
||
|
-Add ECC - "true"
|
||
|
|
||
|
//-Print level (DETAILED, CONCISE) - "CONCISE"
|
||
|
-Print level (DETAILED, CONCISE) - "DETAILED"
|
||
|
|
||
|
# for debugging
|
||
|
//-Print input parameters - "true"
|
||
|
-Print input parameters - "false"
|
||
|
# force CACTI to model the cache with the
|
||
|
# following Ndbl, Ndwl, Nspd, Ndsam,
|
||
|
# and Ndcm values
|
||
|
-Force cache config - "true"
|
||
|
//-Force cache config - "false"
|
||
|
-Ndwl 16
|
||
|
-Ndbl 16
|
||
|
-Nspd 1
|
||
|
-Ndcm 1
|
||
|
-Ndsam1 1
|
||
|
-Ndsam2 1
|
||
|
|