Working version

.gitignore

@@ -0,0 +1,7 @@
+output/*
+cnfg/*
+input/*
+
+!output/Template_run.log
+!cnfg/randfield
+!input/sfcf.in

output/Template_run.log

@@ -0,0 +1,184 @@
+Running sfcf.jl by Lattice_user. Name of the run: Template_run
+
+Calling: sfcf.jl -i ./input/sfcf.in -c ./cnfg/randfield 
+
+Version info:
+Julia Version 1.8.5
+Commit 17cfb8e65ea (2023-01-08 06:45 UTC)
+Platform Info:
+  OS: Linux (x86_64-linux-gnu)
+  CPU: 128 × AMD EPYC 7452 32-Core Processor
+  WORD_SIZE: 64
+  LIBM: libopenlibm
+  LLVM: libLLVM-13.0.1 (ORCJIT, znver2)
+  Threads: 1 on 128 virtual cores
+
+Reading input file from:./input/sfcf.in...
+
+Parameters:
+Lattice size:  (8, 8, 8, 8)
+Phi0         = [0.0, 0.0]
+PhiT         = [0.0, 0.0]
+cG           = 0.965530530296619
+kappa        = 0.1324767
+theta        = 0.5
+csw          = 1.239426196162344
+ct           = 0.98718528389891
+tolerance    = 1.0e-13
+maxiter      = 1000
+
+Reading gauge field with LGPU native convention from: ./cnfg/randfield...
+
+Gauge meassurements:
+
+Plaquette: 0.3306405864818687
+Topological charge: -0.057705150807682376
+SF coupling (dS/deta): 6.346811365600632
+
+
+Computing propagators from t=0 to the bulk
+CG converged for c=1 and s=1 after 50 iterations with absolute residue 0.0009654241012493543
+CG converged for c=2 and s=1 after 50 iterations with absolute residue 0.0009792505060407602
+CG converged for c=3 and s=1 after 50 iterations with absolute residue 0.0009410537206344275
+CG converged for c=1 and s=2 after 50 iterations with absolute residue 0.0009706380660725444
+CG converged for c=2 and s=2 after 50 iterations with absolute residue 0.0009348723257035598
+CG converged for c=3 and s=2 after 50 iterations with absolute residue 0.0009646157878217441
+Computing propagators from t=T to the bulk
+CG converged for c=1 and s=1 after 50 iterations with absolute residue 0.0009409155699679223
+CG converged for c=2 and s=1 after 50 iterations with absolute residue 0.0009591325218333536
+CG converged for c=3 and s=1 after 50 iterations with absolute residue 0.000993180324656404
+CG converged for c=1 and s=2 after 50 iterations with absolute residue 0.0009696740933237723
+CG converged for c=2 and s=2 after 50 iterations with absolute residue 0.0009753894671580818
+CG converged for c=3 and s=2 after 50 iterations with absolute residue 0.0009677391305951339
+
+Computing correlators
+
+Computing f1...
+
+f1 =
+7.968153189691683e-10
+
+Computing k1...
+
+k1 =
+5.773208341735963e-10 + 0.0im
+
+Computing fP...
+
+fP =
+0.0 + 0.0im
+0.39164297606334497 + 0.0im
+0.03838253084495811 + 0.0im
+0.0036829603591832434 + 0.0im
+0.00034990928932360324 + 0.0im
+3.1848709670960246e-5 + 0.0im
+2.83087866678673e-6 + 0.0im
+2.552038846719931e-7 + 0.0im
+
+Computing fA...
+
+fA =
+0.0 + 0.0im
+-0.28022427481948287 + 3.9466441386130056e-20im
+-0.028432536079057023 - 9.446402595355576e-22im
+-0.002717462099554097 + 4.301339185275744e-23im
+-0.0002553997594898297 - 2.8822074468644558e-24im
+-2.3190475107607993e-5 - 1.34416849539867e-24im
+-2.078551613006522e-6 - 3.0342745858421e-25im
+-1.8801834874904572e-7 - 1.7443292336247013e-26im
+
+Computing gP...
+
+gP =
+0.0 + 0.0im
+2.538597063821986e-7 + 0.0im
+2.8229409631152233e-6 + 0.0im
+3.096526353494133e-5 + 0.0im
+0.0003331843094846162 + 0.0im
+0.0036254847521787863 + 0.0im
+0.03837127818992878 + 0.0im
+0.39410791485294716 + 0.0im
+
+Computing gA...
+
+gA =
+0.0 + 0.0im
+-1.8696507552685477e-7 - 6.096317075547979e-27im
+-2.0658936442177055e-6 - 3.842068152788416e-26im
+-2.2674085372631826e-5 - 8.56261180963095e-25im
+-0.00024433351353064797 + 9.060212646869881e-24im
+-0.0026482638745495414 + 2.161009350294785e-22im
+-0.028194731180271545 + 5.375019620369574e-21im
+-0.28340818434504744 + 2.511320339710992e-20im
+
+Computing kV...
+
+kV =
+0.0 + 0.0im
+0.33632172225753315 - 1.0340860564396276e-20im
+0.03186802210515694 - 9.926167350636303e-23im
+0.0029308316253718663 - 1.1084220208210557e-22im
+0.0002663612299706971 + 3.171720661258017e-23im
+2.3221979105382962e-5 - 1.3495537858449775e-24im
+1.9759435694009697e-6 + 3.989827059409007e-25im
+1.7038427603748213e-7 + 2.5655860266867883e-26im
+
+Computing lV...
+
+lV =
+0.0 + 0.0im
+1.692538762882869e-7 - 1.2470313189733742e-26im
+1.972297868357158e-6 + 1.9703431420432253e-25im
+2.2655589268307515e-5 + 4.094974855373194e-24im
+0.0002539424459369091 - 2.540133797714921e-23im
+0.0028788471359388916 - 4.1193594505140788e-22im
+0.031803280548268274 + 2.5322755818956662e-21im
+0.3390357554359771 - 7.90564084104014e-21im
+
+Computing kT...
+
+kT =
+0.0 + 0.0im
+-0.30173882074287517 + 5.293955920339405e-22im
+-0.028918463611070797 - 1.548482106699267e-21im
+-0.002655774263798736 + 6.81045371001993e-23im
+-0.00024049035597286907 + 3.794691060087014e-23im
+-2.0961481048454104e-5 - 1.4734154661100807e-24im
+-1.7864824922599887e-6 + 1.9037001727701538e-25im
+-1.543158528170733e-7 + 2.862618452866008e-26im
+
+Computing lT...
+
+lT =
+0.0 + 0.0im
+-1.5314862576847136e-7 + 3.022494262990794e-26im
+-1.781979951825098e-6 + 3.204247815553729e-26im
+-2.0486213262577787e-5 - 4.26515003347655e-25im
+-0.00022976131918157333 - 4.718376077437893e-23im
+-0.002600005011614236 + 1.9521462456251453e-22im
+-0.028781733302634345 - 3.211666591672554e-21im
+-0.3047801284340767 + 6.988021814847987e-21im
+ ────────────────────────────────────────────────────────────────────────────────
+                                        Time                    Allocations      
+                               ───────────────────────   ────────────────────────
+       Tot / % measured:             108s /  89.1%           11.3GiB /  90.7%    
+
+ Section               ncalls     time    %tot     avg     alloc    %tot      avg
+ ────────────────────────────────────────────────────────────────────────────────
+ Propagator computa...      1    52.8s   54.7%   52.8s   5.99GiB   58.3%  5.99GiB
+   g5Dw                    12    2.80s    2.9%   233ms    220MiB    2.1%  18.3MiB
+   DwdagDw                600    732ms    0.8%  1.22ms   14.7MiB    0.1%  25.1KiB
+     g5Dw               1.20k    694ms    0.7%   578μs   11.1MiB    0.1%  9.46KiB
+     SF boundary fix    1.20k   37.2ms    0.0%  31.0μs   3.41MiB    0.0%  2.91KiB
+   SF boundary fix         12    702ms    0.7%  58.5ms   29.9MiB    0.3%  2.49MiB
+ Gauge field reading        1    19.7s   20.4%   19.7s   1.52GiB   14.8%  1.52GiB
+   Csw computation          1    16.5s   17.1%   16.5s   1.14GiB   11.1%  1.14GiB
+ Gauge measurements         1    18.2s   18.8%   18.2s   1.83GiB   17.8%  1.83GiB
+   Qtop measurement         1    8.15s    8.5%   8.15s    691MiB    6.6%   691MiB
+   Plaquette measur...      1    1.96s    2.0%   1.96s    157MiB    1.5%   157MiB
+   SF coupling meas...      1    1.82s    1.9%   1.82s    137MiB    1.3%   137MiB
+ Imput reading and ...      1    4.46s    4.6%   4.46s    529MiB    5.0%   529MiB
+   Allocating Dirac...      1    157ms    0.2%   157ms   1.15MiB    0.0%  1.15MiB
+   Allocating YMWor...      1   46.5μs    0.0%  46.5μs   1.03KiB    0.0%  1.03KiB
+ Correlators contra...      1    1.39s    1.4%   1.39s    428MiB    4.1%   428MiB
+ ────────────────────────────────────────────────────────────────────────────────

sfcf.jl

@@ -3,17 +3,18 @@ using TOML
 using TimerOutputs
 using ArgParse
 using CUDA
+using InteractiveUtils
 
 include("./src/io.jl")
 include("./src/meas.jl")
 include("./src/corr.jl")
 
-log_file = open("./output/"*params["Run"]["name"]*".log","w+")
-write_log()
 
 @timeit "Imput reading and space allocation" begin
     read_input()
     load_structs()
+    global log_file = open("./output/"*params["Run"]["name"]*".log","w+")
+    write_log()
     U = vector_field(SU3{Float64}, lp)
 end
 
@@ -31,3 +32,4 @@ print_timer(log_file)
 
 flush(log_file)
 close(log_file)
+

src/corr.jl

@@ -141,77 +141,77 @@ function compute_correlators()
 
     println(log_file,"\nComputing correlators")
 
-    println(log_file,"\nComputing f1...")
+    println(log_file,"\nComputing f1...\n")
     global f1 = f1_fun()
-    println(log_file,"f1 = \n")
+    println(log_file,"f1 =")
     println(log_file,f1)
     flush(log_file)
 
-    println(log_file,"\nComputing k1...")
+    println(log_file,"\nComputing k1...\n")
     global k1 = k1_fun()
-    println(log_file,"k1 = \n")
+    println(log_file,"k1 =")
     println(log_file,k1)
     flush(log_file)
 
-    println(log_file,"\nComputing fP...")
+    println(log_file,"\nComputing fP...\n")
     global fP = fP_fun()
-    println(log_file,"fP = \n")
+    println(log_file,"fP =")
     for i in 1:lp.iL[4]
         println(log_file,fP[i])
     end
     flush(log_file)
 
-    println(log_file,"\nComputing fA...")
+    println(log_file,"\nComputing fA...\n")
     global fA = fA_fun()
-    println(log_file,"fA = \n")
+    println(log_file,"fA =")
     for i in 1:lp.iL[4]
         println(log_file,fA[i])
     end
     flush(log_file)
 
-    println(log_file,"\nComputing gP...")
+    println(log_file,"\nComputing gP...\n")
     global gP = gP_fun()
-    println(log_file,"gP = \n")
+    println(log_file,"gP =")
     for i in 1:lp.iL[4]
         println(log_file,gP[i])
     end
     flush(log_file)
 
-    println(log_file,"\nComputing gA...")
+    println(log_file,"\nComputing gA...\n")
     global gA = gA_fun()
-    println(log_file,"gA = \n")
+    println(log_file,"gA =")
     for i in 1:lp.iL[4]
         println(log_file,gA[i])
     end
     flush(log_file)
 
-    println(log_file,"\nComputing kV...")
+    println(log_file,"\nComputing kV...\n")
     global kV = kV_fun()
-    println(log_file,"kV = \n")
+    println(log_file,"kV =")
     for i in 1:lp.iL[4]
         println(log_file,kV[i])
     end
     flush(log_file)
 
-    println(log_file,"\nComputing lV...")
+    println(log_file,"\nComputing lV...\n")
     global lV = lV_fun()
-    println(log_file,"lV = \n")
+    println(log_file,"lV =")
     for i in 1:lp.iL[4]
         println(log_file,lV[i])
     end
     flush(log_file)
 
-    println(log_file,"\nComputing kT...")
+    println(log_file,"\nComputing kT...\n")
     global kT = kT_fun()
-    println(log_file,"kT = \n")
+    println(log_file,"kT =")
     for i in 1:lp.iL[4]
         println(log_file,kT[i])
     end
     flush(log_file)
 
-    println(log_file,"\nComputing lT...")
+    println(log_file,"\nComputing lT...\n")
     global lT = lT_fun()
-    println(log_file,"lT = \n")
+    println(log_file,"lT =")
     for i in 1:lp.iL[4]
         println(log_file,lT[i])
     end

src/io.jl

@@ -12,7 +12,6 @@ Stores as global variables 'parsed_args' (info from the command line) and 'param
 """
 function read_input()
     global parsed_args = parse_commandline()
-    println(log_file,"Reading input file from:", parsed_args["i"], "...")
     global params = TOML.parsefile(parsed_args["i"])
     return nothing
 end
@@ -49,7 +48,7 @@ Returns the gauge field and computes the Csw term
 """
 function load_gauge_field()
     if !parsed_args["cern"]
-    println(log_file,"\nReading gauge field with LGPU native convention from: ", parsed_args["c"], "...\n")
+    println(log_file,"\nReading gauge field with LGPU native convention from: ", parsed_args["c"], "...")
         U,_ = read_cnfg(parsed_args["c"])
     else
     println(log_file,"\nReading gauge field with CERN convention from: ", parsed_args["c"], "...\n")
@@ -153,12 +152,14 @@ function write_log()
 
     print(log_file,"Calling: ")
     print(log_file,PROGRAM_FILE*" "); for x in ARGS; print(log_file,x*" "); end
-    println(log_file,"")
+    println(log_file,"\n")
 
     println(log_file,"Version info:")
-    println(log_file,versioninfo())
+    versioninfo(log_file)
     println(log_file,"")
 
+    println(log_file,"Reading input file from:", parsed_args["i"], "...\n")
+
     println(log_file,"Parameters:")
     println(log_file,"Lattice size:  ", lp.iL)
     println(log_file,"Phi0         = ", params["Space"]["phi0"])
@@ -170,8 +171,8 @@ function write_log()
     println(log_file,"ct           = ", dpar.ct)
     println(log_file,"tolerance    = ", params["Solver"]["tolerance"])
     println(log_file,"maxiter      = ", params["Solver"]["maxiter"])
-    println(log_file,"")
 
     flush(log_file)
     return nothing
 end
+

src/meas.jl

@@ -15,7 +15,7 @@ function compute_propagators()
 
     psi = scalar_field(Spinor{4,SU3fund{Float64}},lp)
 
-    println(log_file,"Computing propagators from t=0 to the bulk")
+    println(log_file,"\n\nComputing propagators from t=0 to the bulk")
     flush(log_file)
     niter = bndpropagator!(psi, U, dpar, dws, lp, params["Solver"]["maxiter"], params["Solver"]["tolerance"], 1, 1)
     println(log_file,"CG converged for c=1 and s=1 after ",niter," iterations with absolute residue ", CUDA.mapreduce(x -> norm(x), +, dws.sr))
@@ -82,7 +82,6 @@ function compute_propagators()
     niter = Tbndpropagator!(psi, U, dpar, dws, lp, params["Solver"]["maxiter"], params["Solver"]["tolerance"], 3, 2)
     println(log_file,"CG converged for c=3 and s=2 after ",niter," iterations with absolute residue ", CUDA.mapreduce(x -> norm(x), +, dws.sr))
     global B32 = Array(psi)
-    println(log_file,"\n")
     flush(log_file)
 
     return nothing
@@ -90,7 +89,7 @@ end
 
 function gauge_meas()
 
-    println(log_file,"\nGauge meassurements:")
+    println(log_file,"\nGauge meassurements:\n")
     println(log_file,"Plaquette: ", plaquette(U, lp, gp, ymws))
     println(log_file,"Topological charge: ", Qtop(U, gp, lp, ymws))
     println(log_file,"SF coupling (dS/deta): ",sfcoupling(U, lp, gp, ymws)[1])