Skip to content

L
LGPU FPP
Commit 2eb07322 authored by Fernando Pérez Panadero's avatar Fernando Pérez Panadero
Browse files
Options

Typos in Dirac.jl

parent 6a23c916
Hide whitespace changes
Inline Side-by-side
Showing with 5 additions and 7 deletions
src/Dirac/Dirac.jl
View file @ 2eb07322
...@@ -90,10 +90,10 @@ function Dw!(so, U, si, dpar::DiracParam, lp::SpaceParm{4,6,B,D}, ymws = "ymws", ...@@ -90,10 +90,10 @@ function Dw!(so, U, si, dpar::DiracParam, lp::SpaceParm{4,6,B,D}, ymws = "ymws",
else else
for i in 1:Int(lp.npls/2) for i in 1:Int(lp.npls/2)
z1 = exp(2im * pi * lp.ntw[i]/gp.ng) z1 = exp(2im * pi * lp.ntw[i]/gp.ng)
z2 = exp(2im * pi * lp.ntw[lp.npls-i]/gp.ng) z2 = exp(2im * pi * lp.ntw[lp.npls-i+1]/gp.ng)
CUDA.@sync begin CUDA.@sync begin
CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_field_tensor!(ymws.frc1, ymws.frc2, U, gp.Ubnd, i, lp.npls-i+1, z1, z2, lp) CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_field_tensor!(ymws.frc1, ymws.frc2, U, gp.Ubnd, i, lp.npls-i+1, z1, z2, lp)
CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_Dw_impr!(so, ymws.frc1, ymws.frc2, Gamma{i+9}, Gamma{lp.npls-i+8}, si, dpar.csw, lp) CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_Dw_impr!(so, ymws.frc1, ymws.frc2, Gamma{i+9}, Gamma{i+12}, si, dpar.csw, lp)
end end
end end
end end
...@@ -133,17 +133,15 @@ function DwdagDw!(so, U, si, dpar::DiracParam, st, lp::SpaceParm{4,6,B,D}, ymws ...@@ -133,17 +133,15 @@ function DwdagDw!(so, U, si, dpar::DiracParam, st, lp::SpaceParm{4,6,B,D}, ymws
if ymws == "ymws" || gp == "gp" if ymws == "ymws" || gp == "gp"
error("YM workspace and Gauge parameters must be included for the improved dirac operator.") error("YM workspace and Gauge parameters must be included for the improved dirac operator.")
else else
@timeit "Dw_improvement" begin
for i in 1:Int(lp.npls/2) for i in 1:Int(lp.npls/2)
z1 = exp(2im * pi * lp.ntw[i]/gp.ng) z1 = exp(2im * pi * lp.ntw[i]/gp.ng)
z2 = exp(2im * pi * lp.ntw[lp.npls-i]/gp.ng) z2 = exp(2im * pi * lp.ntw[lp.npls-i+1]/gp.ng)
CUDA.@sync begin CUDA.@sync begin
CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_field_tensor!(ymws.frc1, ymws.frc2, U, gp.Ubnd, i, lp.npls-i+1, z1, z2, lp) CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_field_tensor!(ymws.frc1, ymws.frc2, U, gp.Ubnd, i, lp.npls-i+1, z1, z2, lp)
CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_g5Dw_impr!(st, ymws.frc1, ymws.frc2, Gamma{i+9}, Gamma{i+12}, si, dpar.csw, lp) CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_g5Dw_impr!(st, ymws.frc1, ymws.frc2, Gamma{i+9}, Gamma{i+12}, si, dpar.csw, lp)
end end
end end
end end
end
end end
end end
...@@ -168,7 +166,7 @@ function DwdagDw!(so, U, si, dpar::DiracParam, st, lp::SpaceParm{4,6,B,D}, ymws ...@@ -168,7 +166,7 @@ function DwdagDw!(so, U, si, dpar::DiracParam, st, lp::SpaceParm{4,6,B,D}, ymws
else else
for i in 1:Int(lp.npls/2) for i in 1:Int(lp.npls/2)
z1 = exp(2im * pi * lp.ntw[i]/gp.ng) z1 = exp(2im * pi * lp.ntw[i]/gp.ng)
z2 = exp(2im * pi * lp.ntw[lp.npls-i]/gp.ng) z2 = exp(2im * pi * lp.ntw[lp.npls-i+1]/gp.ng)
CUDA.@sync begin CUDA.@sync begin
CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_field_tensor!(ymws.frc1, ymws.frc2, U, gp.Ubnd, i, lp.npls-i+1, z1, z2, lp) CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_field_tensor!(ymws.frc1, ymws.frc2, U, gp.Ubnd, i, lp.npls-i+1, z1, z2, lp)
CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_g5Dw_impr!(so, ymws.frc1, ymws.frc2, Gamma{i+9}, Gamma{i+12}, st, dpar.csw, lp) CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_g5Dw_impr!(so, ymws.frc1, ymws.frc2, Gamma{i+9}, Gamma{i+12}, st, dpar.csw, lp)
...@@ -230,7 +228,7 @@ function g5Dw!(so, U, si, dpar, lp::SpaceParm{4,6,B,D}, ymws = "ymws", gp = "gp" ...@@ -230,7 +228,7 @@ function g5Dw!(so, U, si, dpar, lp::SpaceParm{4,6,B,D}, ymws = "ymws", gp = "gp"
else else
for i in 1:Int(lp.npls/2) for i in 1:Int(lp.npls/2)
z1 = exp(2im * pi * lp.ntw[i]/gp.ng) z1 = exp(2im * pi * lp.ntw[i]/gp.ng)
z2 = exp(2im * pi * lp.ntw[lp.npls-i]/gp.ng) z2 = exp(2im * pi * lp.ntw[lp.npls-i+1]/gp.ng)
CUDA.@sync begin CUDA.@sync begin
CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_field_tensor!(ymws.frc1, ymws.frc2, U, gp.Ubnd, i, lp.npls-i+1, z1, z2, lp) CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_field_tensor!(ymws.frc1, ymws.frc2, U, gp.Ubnd, i, lp.npls-i+1, z1, z2, lp)
CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_g5Dw_impr!(so, ymws.frc1, ymws.frc2, Gamma{i+9}, Gamma{i+12}, si, dpar.csw, lp) CUDA.@cuda threads=lp.bsz blocks=lp.rsz krnl_g5Dw_impr!(so, ymws.frc1, ymws.frc2, Gamma{i+9}, Gamma{i+12}, si, dpar.csw, lp)
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment