[prof] regenerate CODEGEN patch from gg_tt.mad, finally including the…

… CUDACPP_RUNTIME_SKIPXBINCHECKS patch madgraph5#968 (on top of madgraph5#969)

This includes
- optionally skip xbin checks if CUDACPP_RUNTIME_SKIPXBINCHECKS is set

The only files that still need to be patched are
- 4 in patch.common: Source/makefile, Source/genps.inc, Source/dsample.f, SubProcesses/makefile
- 4 in patch.P1: auto_dsig1.f, auto_dsig.f, driver.f, matrix1.f

./CODEGEN/generateAndCompare.sh gg_tt --mad --nopatch
git diff --no-ext-diff -R gg_tt.mad/Source/makefile gg_tt.mad/Source/genps.inc gg_tt.mad/SubProcesses/makefile gg_tt.mad/Source/dsample.f > CODEGEN/PLUGIN/CUDACPP_SA_OUTPUT/MG5aMC_patches/PROD/patch.common
git diff --no-ext-diff -R gg_tt.mad/SubProcesses/P1_gg_ttx/auto_dsig1.f gg_tt.mad/SubProcesses/P1_gg_ttx/auto_dsig.f gg_tt.mad/SubProcesses/P1_gg_ttx/driver.f gg_tt.mad/SubProcesses/P1_gg_ttx/matrix1.f > CODEGEN/PLUGIN/CUDACPP_SA_OUTPUT/MG5aMC_patches/PROD/patch.P1
git checkout gg_tt.mad

(Later checked that regenerating gg_tt.mad is ok)

epochX/cudacpp/CODEGEN/PLUGIN/CUDACPP_SA_OUTPUT/MG5aMC_patches/PROD/patch.P1

@@ -1,5 +1,5 @@
 diff --git b/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/auto_dsig.f a/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/auto_dsig.f
-index 7bff4b945..0c5869973 100644
+index 7bff4b9455..0c58699731 100644
 --- b/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/auto_dsig.f
 +++ a/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/auto_dsig.f
 @@ -312,8 +312,10 @@ C      entries to the grid for the MC over helicity configuration
@@ -14,7 +14,7 @@ index 7bff4b945..0c5869973 100644
        IF(GROUPED_MC_GRID_STATUS.EQ.0) THEN
  C       If we were in the initialization phase of the grid for MC over
 diff --git b/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/auto_dsig1.f a/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/auto_dsig1.f
-index 4fbb8e6ba..77aff307b 100644
+index 4fbb8e6ba7..77aff307b8 100644
 --- b/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/auto_dsig1.f
 +++ a/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/auto_dsig1.f
 @@ -125,6 +125,7 @@ C     Continue only if IMODE is 0, 4 or 5
@@ -241,7 +241,7 @@ index 4fbb8e6ba..77aff307b 100644
        END
 
 diff --git b/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/driver.f a/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/driver.f
-index 1124a9164..f205ce6fd 100644
+index 1124a9164a..f205ce6fd9 100644
 --- b/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/driver.f
 +++ a/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/driver.f
 @@ -74,16 +74,95 @@ c      common/to_colstats/ncols,ncolflow,ncolalt,ic
@@ -396,7 +396,7 @@ index 1124a9164..f205ce6fd 100644
        open(unit=lun,file=tempname,status='old',ERR=20)
        fopened=.true.
 diff --git b/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/matrix1.f a/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/matrix1.f
-index e73e654d4..3072054f2 100644
+index e73e654d49..3072054f2d 100644
 --- b/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/matrix1.f
 +++ a/epochX/cudacpp/gg_tt.mad/SubProcesses/P1_gg_ttx/matrix1.f
 @@ -72,7 +72,10 @@ C

epochX/cudacpp/CODEGEN/PLUGIN/CUDACPP_SA_OUTPUT/MG5aMC_patches/PROD/patch.common

@@ -1,5 +1,5 @@
 diff --git b/epochX/cudacpp/gg_tt.mad/Source/dsample.f a/epochX/cudacpp/gg_tt.mad/Source/dsample.f
-index e18ba7c03..35e52cf44 100644
+index e18ba7c03d..899021c378 100644
 --- b/epochX/cudacpp/gg_tt.mad/Source/dsample.f
 +++ a/epochX/cudacpp/gg_tt.mad/Source/dsample.f
 @@ -169,7 +169,9 @@ c
@@ -67,7 +67,7 @@ index e18ba7c03..35e52cf44 100644
        data Minvar(1,1)/-1/       !No special variable mapping
 
  c-----
-@@ -1240,7 +1250,20 @@ c
+@@ -1240,7 +1250,26 @@ c
  c     Local
  c
        integer  im, ip,ij,icount,it_warned
@@ -86,10 +86,16 @@ index e18ba7c03..35e52cf44 100644
 +      save xbin_min0_saved, xbin_max1_saved
 +      data xbin_min0_saved/xbinarraydim*.false./
 +      data xbin_max1_saved/xbinarraydim*.false./
++
++      character*255 env_name, env_value
++      integer env_length, env_status
++      logical first, skipxbinchecks
++      data first, skipxbinchecks/.true., .false./
++      save first, skipxbinchecks
  c
  c     External
  c
-@@ -1291,15 +1314,29 @@ c         write(*,'(7f11.5)')(ddum(j)*real(ng),j=1,dim)
+@@ -1291,15 +1320,29 @@ c         write(*,'(7f11.5)')(ddum(j)*real(ng),j=1,dim)
        endif
        if (ituple .eq. 1) then
  c         write(*,*) 'Getting variable',ipole,j,minvar(j,ipole)
@@ -128,7 +134,7 @@ index e18ba7c03..35e52cf44 100644
  c
  c     Line which allows us to keep choosing same x
  c
-@@ -1312,10 +1349,10 @@ c            write(*,*) 'Reusing num',j,nzoom,tx(2,j)
+@@ -1312,10 +1355,10 @@ c            write(*,*) 'Reusing num',j,nzoom,tx(2,j)
              call ntuple(ddum(j),max(xbin_min,dble(int(tx(2,j)))),
       $           min(xbin_max,dble(int(tx(2,j))+1)),j,ipole)
 
@@ -142,11 +148,33 @@ index e18ba7c03..35e52cf44 100644
 
  c            write(*,'(2i6,4e15.5)') nzoom,j,ddum(j),tx(2,j),
  c     $           max(xbin_min,dble(int(tx(2,j)))),
-@@ -1389,10 +1426,10 @@ c
+@@ -1378,7 +1421,19 @@ c     to the fact that the grids are required to be separated by 1e-14. Since
+ c     double precision is about 18 digits, we expect things to agree to
+ c     3 digit accuracy.
+ c
+-      if (abs(ddum(j)-xbin(x,ij))/(ddum(j)+1d-22) .gt. 1e-3) then
++      if (first) then
++        env_name = 'CUDACPP_RUNTIME_SKIPXBINCHECKS'
++        call get_environment_variable(env_name, env_value, env_length, env_status)
++        if( env_status.eq.0 ) then
++          skipxbinchecks = .true.
++        endif
++      endif
++
++      if (skipxbinchecks) then
++        if (first) then
++          write(6,*) 'WARNING: skipping xbin checks (CUDACPP_RUNTIME_SKIPXBINCHECKS is set)'
++        endif
++      else if (abs(ddum(j)-xbin(x,ij))/(ddum(j)+1d-22) .gt. 1e-3) then
+          if (icount .lt. 5) then
+             write(*,'(a,i4,2e14.6,1e12.4)')
+      &           'Warning xbin not returning correct x', ij,
+@@ -1389,10 +1444,11 @@ c
           endif
           icount=icount+1
        endif
 -      if (x .lt. xmin .or. x .gt. xmax) then
++      first = .false.
 +c     if (x .lt. xmin .or. x .gt. xmax) then
  c         write(*,'(a,4i4,2f24.16,1e10.2)') 'Bad x',ij,int(xbin_min),ip,
  c     &        int(xbin_max),xmin,x,xmax-xmin
@@ -156,7 +184,7 @@ index e18ba7c03..35e52cf44 100644
        wgt = wgt * xo * dble(xbin_max-xbin_min)
  c      print*,'Returning x',ij,ipole,j,x
 diff --git b/epochX/cudacpp/gg_tt.mad/Source/genps.inc a/epochX/cudacpp/gg_tt.mad/Source/genps.inc
-index a59181c70..af7e0efbc 100644
+index a59181c708..af7e0efbce 100644
 --- b/epochX/cudacpp/gg_tt.mad/Source/genps.inc
 +++ a/epochX/cudacpp/gg_tt.mad/Source/genps.inc
 @@ -30,7 +30,8 @@ c*************************************************************************
@@ -170,7 +198,7 @@ index a59181c70..af7e0efbc 100644
        PARAMETER(MAXTRIES=25)
  C     To pass the helicity configuration chosen by the DiscreteSampler to 
 diff --git b/epochX/cudacpp/gg_tt.mad/Source/makefile a/epochX/cudacpp/gg_tt.mad/Source/makefile
-index 617f10b93..00c73099a 100644
+index 617f10b93e..00c73099a0 100644
 --- b/epochX/cudacpp/gg_tt.mad/Source/makefile
 +++ a/epochX/cudacpp/gg_tt.mad/Source/makefile
 @@ -120,7 +120,7 @@ $(LIBDIR)libiregi.a: $(IREGIDIR)
@@ -195,7 +223,7 @@ index 617f10b93..00c73099a 100644
 +cleanall: cleanSource # THIS IS THE ONE
 +	for i in `ls -d ../SubProcesses/P*`; do cd $$i; make cleanavxs; cd -; done;
 diff --git b/epochX/cudacpp/gg_tt.mad/SubProcesses/makefile a/epochX/cudacpp/gg_tt.mad/SubProcesses/makefile
-index 348c283be..504e41ea5 100644
+index 348c283be7..504e41ea52 100644
 --- b/epochX/cudacpp/gg_tt.mad/SubProcesses/makefile
 +++ a/epochX/cudacpp/gg_tt.mad/SubProcesses/makefile
 @@ -1,6 +1,40 @@