From 77b1e03b395f56dbf6cfae60f90210c8b34fc028 Mon Sep 17 00:00:00 2001 From: gabrielequatrana Date: Thu, 11 Apr 2024 22:33:44 +0200 Subject: [PATCH] Added test --- test/block_krylov/gpu/Makefile | 50 ++ test/block_krylov/gpu/getp.f90 | 169 +++++ test/block_krylov/gpu/psb_dbf_compare.F90 | 368 +++++++++++ test/block_krylov/gpu/psb_dbf_sample.F90 | 319 +++++++++ test/block_krylov/gpu/psb_dpde_gen.F90 | 762 ++++++++++++++++++++++ test/block_krylov/kernel/Makefile | 4 +- test/block_krylov/kernel/dpdegenmm.F90 | 45 +- test/block_krylov/psb_dbf_sample.f90 | 8 +- test/block_krylov/psb_perf_test.f90 | 296 +++++++++ 9 files changed, 1994 insertions(+), 27 deletions(-) create mode 100644 test/block_krylov/gpu/Makefile create mode 100644 test/block_krylov/gpu/getp.f90 create mode 100644 test/block_krylov/gpu/psb_dbf_compare.F90 create mode 100644 test/block_krylov/gpu/psb_dbf_sample.F90 create mode 100644 test/block_krylov/gpu/psb_dpde_gen.F90 create mode 100644 test/block_krylov/psb_perf_test.f90 diff --git a/test/block_krylov/gpu/Makefile b/test/block_krylov/gpu/Makefile new file mode 100644 index 00000000..e7235d77 --- /dev/null +++ b/test/block_krylov/gpu/Makefile @@ -0,0 +1,50 @@ +TOPDIR=../../.. +include $(TOPDIR)/Make.inc +# +# Libraries used +# +LIBDIR=$(TOPDIR)/lib/ +PSBLIBDIR=$(TOPDIR)/lib/ +OPTDIR=$(LIBDIR) +PSBINCDIR=$(TOPDIR)/include +PSBMODDIR=$(TOPDIR)/modules +PSBLAS_LIB= -L$(LIBDIR) -L$(PSBLIBDIR) $(LCUDA) -lpsb_ext -lpsb_util -lpsb_krylov -lpsb_prec -lpsb_base +INCDIR=$(TOPDIR)/include +MODDIR=$(TOPDIR)/modules + +LDLIBS=$(PSBLDLIBS) +LDLIBSG=$(PSBGPULDLIBS) + +FINCLUDES=$(FMFLAG)$(MODDIR) $(FMFLAG)$(INCDIR) $(FMFLAG). $(FMFLAG)$(PSBMODDIR) $(FMFLAG)$(PSBINCDIR) $(LIBRSB_DEFINES) + +DBFOBJS=getp.o psb_dbf_sample.o +DCOMP=psb_dbf_compare.o +DPDEGEN=psb_dpde_gen.o +EXEDIR=./runs + +all: runsd psb_dbf_sample psb_dbf_compare psb_dpde_gen + +runsd: + (if test ! -d runs ; then mkdir runs; fi) + +psb_dbf_sample.o: getp.o + +psb_dbf_sample: $(DBFOBJS) + $(FLINK) $(LOPT) $(DBFOBJS) -fopenmp -o psb_dbf_sample $(FINCLUDES) $(PSBLAS_LIB) $(LDLIBSG) + /bin/mv psb_dbf_sample $(EXEDIR) + +psb_dbf_compare: $(DCOMP) + $(FLINK) $(LOPT) $(DCOMP) -fopenmp -o psb_dbf_compare $(FINCLUDES) $(PSBLAS_LIB) $(LDLIBSG) + /bin/mv psb_dbf_compare $(EXEDIR) + +psb_dpde_gen: $(DPDEGEN) + $(FLINK) $(LOPT) $(DPDEGEN) -fopenmp -o psb_dpde_gen $(FINCLUDES) $(PSBLAS_LIB) $(LDLIBSG) + /bin/mv psb_dpde_gen $(EXEDIR) + +clean: + /bin/rm -f $(DBFOBJS) $(DCOMP) $(DPDEGEN) *$(.mod) $(EXEDIR)/psb_*f_sample + +lib: + (cd ../../; make library) +verycleanlib: + (cd ../../; make veryclean) diff --git a/test/block_krylov/gpu/getp.f90 b/test/block_krylov/gpu/getp.f90 new file mode 100644 index 00000000..197cb646 --- /dev/null +++ b/test/block_krylov/gpu/getp.f90 @@ -0,0 +1,169 @@ +! +! Parallel Sparse BLAS version 3.5 +! (C) Copyright 2006-2018 +! Salvatore Filippone +! Alfredo Buttari +! +! Redistribution and use in source and binary forms, with or without +! modification, are permitted provided that the following conditions +! are met: +! 1. Redistributions of source code must retain the above copyright +! notice, this list of conditions and the following disclaimer. +! 2. Redistributions in binary form must reproduce the above copyright +! notice, this list of conditions, and the following disclaimer in the +! documentation and/or other materials provided with the distribution. +! 3. The name of the PSBLAS group or the names of its contributors may +! not be used to endorse or promote products derived from this +! software without specific written permission. +! +! THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +! ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED +! TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR +! PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE PSBLAS GROUP OR ITS CONTRIBUTORS +! BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +! CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +! SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +! INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +! CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +! ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +! POSSIBILITY OF SUCH DAMAGE. +! +! +Module getp + interface get_parms + module procedure get_dparms + end interface + +contains + ! + ! Get iteration parameters from the command line + ! + subroutine get_dparms(ctxt,mtrx_file,rhs_file,filefmt,kmethd,ptype,part,& + & afmt,nrhs,istopc,itmax,itrace,itrs,eps) + use psb_base_mod + type(psb_ctxt_type) :: ctxt + character(len=2) :: filefmt + character(len=40) :: kmethd, mtrx_file, rhs_file, ptype + character(len=20) :: part + integer(psb_ipk_) :: nrhs,iret,istopc,itmax,itrace,itrs + character(len=40) :: charbuf + real(psb_dpk_) :: eps + character :: afmt*5 + integer(psb_ipk_) :: np, iam + integer(psb_ipk_) :: inparms(40), ip, inp_unit + character(len=1024) :: filename + + call psb_info(ctxt,iam,np) + if (iam == 0) then + if (command_argument_count()>0) then + call get_command_argument(1,filename) + inp_unit = 30 + open(inp_unit,file=filename,action='read',iostat=info) + if (info /= 0) then + write(psb_err_unit,*) 'Could not open file ',filename,' for input' + call psb_abort(ctxt) + stop + else + write(psb_err_unit,*) 'Opened file ',trim(filename),' for input' + end if + else + inp_unit=psb_inp_unit + end if + ! Read Input Parameters + read(inp_unit,*) ip + if (ip >= 5) then + read(inp_unit,*) mtrx_file + read(inp_unit,*) rhs_file + read(inp_unit,*) filefmt + read(inp_unit,*) kmethd + read(inp_unit,*) ptype + read(inp_unit,*) afmt + read(inp_unit,*) part + + call psb_bcast(ctxt,mtrx_file) + call psb_bcast(ctxt,rhs_file) + call psb_bcast(ctxt,filefmt) + call psb_bcast(ctxt,kmethd) + call psb_bcast(ctxt,ptype) + call psb_bcast(ctxt,afmt) + call psb_bcast(ctxt,part) + + if (ip >= 7) then + read(inp_unit,*) nrhs + else + nrhs=4 + endif + if (ip >= 8) then + read(inp_unit,*) istopc + else + istopc=1 + endif + if (ip >= 9) then + read(inp_unit,*) itmax + else + itmax=500 + endif + if (ip >= 10) then + read(inp_unit,*) itrace + else + itrace=-1 + endif + if (ip >= 11) then + read(inp_unit,*) itrs + else + itrs = 1 + endif + if (ip >= 12) then + read(inp_unit,*) eps + else + eps=1.d-6 + endif + inparms(1) = nrhs + inparms(2) = istopc + inparms(3) = itmax + inparms(4) = itrace + inparms(5) = itrs + call psb_bcast(ctxt,inparms(1:5)) + call psb_bcast(ctxt,eps) + + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Solving matrix : ",a)') mtrx_file + write(psb_out_unit,'("Number of processors : ",i3)') np + write(psb_out_unit,'("Data distribution : ",a)') part + write(psb_out_unit,'("Iterative method : ",a)') kmethd + write(psb_out_unit,'("Preconditioner : ",a)') ptype + write(psb_out_unit,'("Number of RHS : ",i3)') nrhs + write(psb_out_unit,'("Number of iterations : ",i3)') itrs + write(psb_out_unit,'("Storage format : ",a)') afmt(1:3) + write(psb_out_unit,'(" ")') + else + write(psb_err_unit,*) 'Wrong format for input file' + call psb_abort(ctxt) + stop 1 + end if + if (inp_unit /= psb_inp_unit) then + close(inp_unit) + end if + else + ! Receive Parameters + call psb_bcast(ctxt,mtrx_file) + call psb_bcast(ctxt,rhs_file) + call psb_bcast(ctxt,filefmt) + call psb_bcast(ctxt,kmethd) + call psb_bcast(ctxt,ptype) + call psb_bcast(ctxt,afmt) + call psb_bcast(ctxt,part) + + call psb_bcast(ctxt,inparms(1:5)) + nrhs = inparms(1) + istopc = inparms(2) + itmax = inparms(3) + itrace = inparms(4) + itrs = inparms(5) + call psb_bcast(ctxt,eps) + + end if + + end subroutine get_dparms + +end module getp diff --git a/test/block_krylov/gpu/psb_dbf_compare.F90 b/test/block_krylov/gpu/psb_dbf_compare.F90 new file mode 100644 index 00000000..56977850 --- /dev/null +++ b/test/block_krylov/gpu/psb_dbf_compare.F90 @@ -0,0 +1,368 @@ +program psb_dbf_sample + + use psb_base_mod + use psb_prec_mod + use psb_krylov_mod + use psb_util_mod + use psb_cuda_mod + use getp + + implicit none + + ! input parameters + character(len=40) :: mtrx_file = "pde80.mtx" + character(len=2) :: filefmt = "MM" + character(len=40) :: kmethd = "GMRES" + character(len=40) :: ptype = "NONE" + character(len=20) :: part = "GRAPH" + character(len=5) :: afmt = "CSR" + character(len=5) :: agfmt = "ELG" + integer(psb_ipk_) :: nrhs = 6 + integer(psb_ipk_) :: istopbg = 1 + integer(psb_ipk_) :: istoprg = 2 + integer(psb_ipk_) :: itmax = 500 + integer(psb_ipk_) :: itrace = -1 + integer(psb_ipk_) :: itrs = 30 + real(psb_dpk_) :: eps = 1.d-7 + + ! sparse matrices + type(psb_dspmat_type) :: a, agpu + type(psb_ldspmat_type) :: aux_a + + ! preconditioner data + type(psb_dprec_type) :: prec + + ! dense matrices + real(psb_dpk_), allocatable, target :: aux_b(:,:) + real(psb_dpk_), allocatable, save :: x_mv_glob(:,:), r_mv_glob(:,:) + real(psb_dpk_), allocatable, save :: x_col_glob(:), r_col_glob(:) + real(psb_dpk_), pointer :: b_mv_glob(:,:) + real(psb_dpk_), pointer :: b_col_glob(:) + type(psb_d_multivect_type) :: b_mv, x_mv, r_mv + type(psb_d_vect_type) :: b_col, x_col, r_col + type(psb_d_multivect_cuda) :: gpumold_mv + type(psb_d_vect_cuda) :: gpumold_col + type(psb_i_vect_cuda) :: imold + integer(psb_ipk_) :: m + real(psb_dpk_) :: random_value + + ! molds + type(psb_d_cuda_csrg_sparse_mat), target :: acsrg + type(psb_d_cuda_hlg_sparse_mat), target :: ahlg + type(psb_d_cuda_elg_sparse_mat), target :: aelg + class(psb_d_base_sparse_mat), pointer :: agmold + + ! communications data structure + type(psb_desc_type) :: desc_a + type(psb_ctxt_type) :: ctxt + integer(psb_ipk_) :: iam, np + integer(psb_lpk_) :: lnp + + ! solver paramters + integer(psb_ipk_) :: iter, ierr, ircode, methd, istopc + integer(psb_epk_) :: amatsize, precsize, descsize + real(psb_dpk_) :: err, cond + + ! other variables + character(len=20) :: name + integer(psb_ipk_) :: reps = 50 + integer(psb_ipk_) :: i, j, info, rep + real(psb_dpk_) :: tb1, tb2, tr1, tr2 + real(psb_dpk_), allocatable :: resmx(:), tb(:), tr(:) + real(psb_dpk_) :: resmxp + integer(psb_ipk_), allocatable :: ivg(:) + integer(psb_ipk_), parameter :: iunit=12 + logical :: print_matrix = .true. + + call psb_init(ctxt) + call psb_info(ctxt,iam,np) + call psb_cuda_init(ctxt) + if (iam < 0) then + ! This should not happen, but just in case + call psb_exit(ctxt) + stop + endif + + name='psb_perf_test' + if(psb_errstatus_fatal()) goto 9999 + info=psb_success_ + call psb_set_errverbosity(itwo) + ! + ! Hello world + ! + if (iam == psb_root_) then + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Welcome to PSBLAS version: ",a)') psb_version_string_ + write(psb_out_unit,'("This is the ",a," sample program")') trim(name) + write(psb_out_unit,'("Number of processors: ",i8)')np + write(psb_out_unit,'(" ")') + end if + write(*,*) 'Process ',iam,' running on device: ', psb_cuda_getDevice(),' out of', psb_cuda_getDeviceCount() + write(*,*) 'Process ',iam,' device ', psb_cuda_getDevice(),' is a: ', trim(psb_cuda_DeviceName()) + + select case(psb_toupper(agfmt)) + case('ELG') + agmold => aelg + case('HLG') + agmold => ahlg + case('CSRG') + agmold => acsrg + case default + write(*,*) 'Unknown format defaulting to CSRG' + agmold => acsrg + end select + + if (iam == psb_root_) then + select case(psb_toupper(filefmt)) + case('MM') + call mm_mat_read(aux_a,info,iunit=iunit,filename=mtrx_file) + case ('HB') + call hb_read(aux_a,info,iunit=iunit,b=aux_b,filename=mtrx_file) + case default + info = -1 + write(psb_err_unit,*) 'Wrong choice for fileformat ', filefmt + end select + if (info /= psb_success_) then + write(psb_err_unit,*) 'Error while reading input matrix ' + call psb_abort(ctxt) + end if + + m = aux_a%get_nrows() + call psb_bcast(ctxt,m) + + ! At this point aux_b may still be unallocated + if (size(aux_b) == m*nrhs) then + ! if any rhs were present, broadcast the first one + write(psb_err_unit,'("Ok, got an rhs ")') + b_mv_glob =>aux_b(:,:) + else + write(psb_out_unit,'("Generating an rhs...")') + write(psb_out_unit,'("Number of RHS: ",i3)') nrhs + write(psb_out_unit,'(" ")') + call psb_realloc(m,nrhs,aux_b,ircode) + if (ircode /= 0) then + call psb_errpush(psb_err_alloc_dealloc_,name) + goto 9999 + endif + + b_mv_glob => aux_b(:,:) + do i=1, m + do j=1, nrhs + b_mv_glob(i,j) = done + !call random_number(random_value) + !b_mv_glob(i,j) = random_value + enddo + enddo + endif + else + call psb_bcast(ctxt,m) + end if + + ! switch over different partition types + select case(psb_toupper(part)) + case('BLOCK') + if (iam == psb_root_) write(psb_out_unit,'("Partition type: block")') + call psb_matdist(aux_a,a,ctxt,desc_a,info,fmt=afmt,parts=part_block) + case('GRAPH') + if (iam == psb_root_) then + write(psb_out_unit,'("Partition type: graph vector")') + write(psb_out_unit,'(" ")') + call aux_a%cscnv(info,type='csr') + lnp = np + call build_mtpart(aux_a,lnp) + endif + call psb_barrier(ctxt) + call distr_mtpart(psb_root_,ctxt) + call getv_mtpart(ivg) + call psb_matdist(aux_a,a,ctxt,desc_a,info,fmt=afmt,vg=ivg) + case default + if (iam == psb_root_) write(psb_out_unit,'("Partition type: block")') + call psb_matdist(aux_a,a,ctxt,desc_a,info,fmt=afmt,parts=part_block) + end select + + ! building the preconditioner + call prec%init(ctxt,ptype,info) + call prec%build(a,desc_a,info) + if (info /= psb_success_) then + call psb_errpush(psb_err_from_subroutine_,name,a_err='psb_precbld') + goto 9999 + end if + + ! Set GPU matrix + call a%cscnv(agpu,info,mold=agmold) + call desc_a%cnv(mold=imold) + if ((info /= 0).or.(psb_get_errstatus()/=0)) then + write(0,*) 'From cscnv ',info + call psb_error() + stop + end if + + if (iam == psb_root_) then + allocate(tb(reps),tr(reps)) + tb = dzero + tr = dzero + end if + + if(iam == psb_root_) then + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Starting BGMRES")') + write(psb_out_unit,'(" ")') + end if + + do rep=1,reps + call psb_scatter(b_mv_glob,b_mv,desc_a,info,root=psb_root_,mold=gpumold_mv) + call psb_geall(x_mv,desc_a,info,nrhs) + call x_mv%zero() + call psb_geasb(x_mv,desc_a,info,mold=gpumold_mv) + call psb_geall(r_mv,desc_a,info,nrhs) + call r_mv%zero() + call psb_geasb(r_mv,desc_a,info) + + call psb_barrier(ctxt) + tb1 = psb_wtime() + + call psb_krylov(kmethd,a,prec,b_mv,x_mv,eps,desc_a,info,& + & itmax=itmax,iter=iter,err=err,itrace=itrace,& + & itrs=itrs,istop=istopbg) + + call psb_barrier(ctxt) + tb2 = psb_wtime() - tb1 + call psb_amx(ctxt,tb2) + + if (iam == psb_root_) then + tb(rep) = tb2 + write(*,*) 'Time', rep, tb(rep) + end if + call psb_barrier(ctxt) + + call psb_gefree(b_mv,desc_a,info) + call psb_gefree(x_mv,desc_a,info) + end do + + if(iam == psb_root_) then + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Finished BGMRES")') + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Starting sGMRES")') + write(psb_out_unit,'(" ")') + end if + + call psb_barrier(ctxt) + + do rep=1,reps + do i=1,nrhs + + b_col_glob => aux_b(:,i) + call psb_scatter(b_col_glob,b_col,desc_a,info,root=psb_root_,mold=gpumold_col) + call psb_geall(x_col,desc_a,info) + call x_col%zero() + call psb_geasb(x_col,desc_a,info,mold=gpumold_col) + call psb_geall(r_col,desc_a,info) + call r_col%zero() + call psb_geasb(r_col,desc_a,info) + + cond = dzero + + call psb_barrier(ctxt) + tr1 = psb_wtime() + + call psb_krylov(kmethd,a,prec,b_col,x_col,eps,desc_a,info,& + & itmax=itmax,iter=iter,err=err,itrace=itrace,& + & istop=istoprg,irst=itrs,cond=cond) + + call psb_barrier(ctxt) + tr2 = psb_wtime() - tr1 + call psb_amx(ctxt,tr2) + + if (iam == psb_root_) then + tr(rep) = tr(rep) + tr2 + end if + call psb_barrier(ctxt) + + call psb_gefree(b_col, desc_a,info) + call psb_gefree(x_col, desc_a,info) + end do + + if (iam == psb_root_) then + write(*,*) 'Time', rep, tr(rep) + end if + end do + + call psb_barrier(ctxt) + + if(iam == psb_root_) then + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Finished sGMRES")') + write(psb_out_unit,'(" ")') + end if + + call psb_spfree(a, desc_a,info) + call prec%free(info) + call psb_cdfree(desc_a,info) + call psb_cuda_exit() + call psb_exit(ctxt) + + return + + call psb_geaxpby(done,b_mv,dzero,r_mv,desc_a,info) + call psb_spmm(-done,a,x_mv,done,r_mv,desc_a,info) + + resmx = psb_genrm2(r_mv,desc_a,info) + resmxp = psb_geamax(r_mv,desc_a,info) + + amatsize = a%sizeof() + descsize = desc_a%sizeof() + precsize = prec%sizeof() + + call psb_sum(ctxt,amatsize) + call psb_sum(ctxt,descsize) + call psb_sum(ctxt,precsize) + + call psb_gather(x_mv_glob,x_mv,desc_a,info,root=psb_root_) + if (info == psb_success_) call psb_gather(r_mv_glob,r_mv,desc_a,info,root=psb_root_) + if (info /= psb_success_) goto 9999 + + if (iam == psb_root_) then + call prec%descr(info) + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Computed solution on: ",i8," processors")')np + write(psb_out_unit,'("Matrix: ",a)')mtrx_file + write(psb_out_unit,'("Storage format for A: ",a)')a%get_fmt() + write(psb_out_unit,'("Storage format for DESC_A: ",a)')desc_a%get_fmt() + write(psb_out_unit,'("Total memory occupation for A: ",i12)')amatsize + write(psb_out_unit,'("Total memory occupation for PREC: ",i12)')precsize + write(psb_out_unit,'("Total memory occupation for DESC_A: ",i12)')descsize + write(psb_out_unit,'("Iterations to convergence: ",i12)')iter + write(psb_out_unit,'("Error estimate on exit: ",es12.5)')err + !write(psb_out_unit,'("Time to buil prec.: ",es12.5)')tprec + ! write(psb_out_unit,'("Time to solve system: ",es12.5)')t2 + !write(psb_out_unit,'("Time per iteration: ",es12.5)')t2/(iter) + !write(psb_out_unit,'("Total time: ",es12.5)')t2+tprec + write(psb_out_unit,'("Residual norm 2: ",es12.5)')maxval(resmx) + write(psb_out_unit,'("Residual norm inf: ",es12.5)')resmxp + write(psb_out_unit,'(a8,4(2x,a20))') 'I','X(I)','R(I)','B(I)' + if (print_matrix) then + do i=1,m + write(psb_out_unit,993) i, x_mv_glob(i,:), r_mv_glob(i,:), b_mv_glob(i,:) + end do + end if + end if + +998 format(i8,4(2x,g20.14)) +993 format(i6,4(1x,e12.6)) + + call psb_gefree(b_mv,desc_a,info) + call psb_gefree(x_mv,desc_a,info) + call psb_spfree(a,desc_a,info) + call prec%free(info) + call psb_cdfree(desc_a,info) + call psb_cuda_exit() + call psb_exit(ctxt) + + return + +9999 call psb_error(ctxt) + + return + +end program psb_dbf_sample diff --git a/test/block_krylov/gpu/psb_dbf_sample.F90 b/test/block_krylov/gpu/psb_dbf_sample.F90 new file mode 100644 index 00000000..f2cbf99b --- /dev/null +++ b/test/block_krylov/gpu/psb_dbf_sample.F90 @@ -0,0 +1,319 @@ +program psb_dbf_sample + + use psb_base_mod + use psb_prec_mod + use psb_krylov_mod + use psb_util_mod +#ifdef HAVE_CUDA + use psb_cuda_mod +#endif + use getp + + implicit none + + ! input parameters + character(len=40) :: kmethd, ptype, mtrx_file, rhs_file + + ! sparse matrices + type(psb_dspmat_type) :: a, agpu + type(psb_ldspmat_type) :: aux_a + + ! preconditioner data + type(psb_dprec_type) :: prec + + ! dense matrices + real(psb_dpk_), allocatable, target :: aux_b(:,:) + real(psb_dpk_), allocatable, save :: x_mv_glob(:,:), r_mv_glob(:,:) + real(psb_dpk_), pointer :: b_mv_glob(:,:) + type(psb_d_multivect_type) :: b_mv, x_mv, r_mv + type(psb_d_multivect_cuda) :: gpumold + type(psb_i_vect_cuda) :: imold + integer(psb_ipk_) :: m, nrhs + real(psb_dpk_) :: random_value + + ! molds + type(psb_d_cuda_csrg_sparse_mat), target :: acsrg + type(psb_d_cuda_hlg_sparse_mat), target :: ahlg + type(psb_d_cuda_elg_sparse_mat), target :: aelg + class(psb_d_base_sparse_mat), pointer :: agmold, acmold + + ! communications data structure + type(psb_desc_type) :: desc_a + type(psb_ctxt_type) :: ctxt + integer(psb_ipk_) :: iam, np + integer(psb_lpk_) :: lnp + + ! solver paramters + integer(psb_ipk_) :: iter, itmax, ierr, itrace, ircode, methd, istopc, itrs + integer(psb_epk_) :: amatsize, precsize, descsize + real(psb_dpk_) :: err, eps + + ! input parameters + character(len=5) :: afmt, agfmt = "ELG" + character(len=20) :: name, part + character(len=2) :: filefmt + integer(psb_ipk_), parameter :: iunit=12 + + ! other variables + integer(psb_ipk_) :: i, j, info + real(psb_dpk_) :: t1, t2, tprec + real(psb_dpk_), allocatable :: resmx(:), res(:,:) + real(psb_dpk_) :: resmxp + integer(psb_ipk_), allocatable :: ivg(:) + logical :: print_matrix = .true. + + call psb_init(ctxt) + call psb_info(ctxt,iam,np) + call psb_cuda_init(ctxt) + + if (iam < 0) then + ! This should not happen, but just in case + call psb_exit(ctxt) + stop + endif + + name='psb_dbf_sample' + if(psb_errstatus_fatal()) goto 9999 + info=psb_success_ + call psb_set_errverbosity(itwo) + ! + ! Hello world + ! + if (iam == psb_root_) then + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Welcome to PSBLAS version: ",a)') psb_version_string_ + write(psb_out_unit,'("This is the ",a," sample program")') trim(name) + write(psb_out_unit,'(" ")') + end if +#ifdef HAVE_CUDA + write(*,*) 'Process ',iam,' running on device: ', psb_cuda_getDevice(),' out of', psb_cuda_getDeviceCount() + write(*,*) 'Process ',iam,' device ', psb_cuda_getDevice(),' is a: ', trim(psb_cuda_DeviceName()) +#endif + ! + ! get parameters + ! + call get_parms(ctxt,mtrx_file,rhs_file,filefmt,kmethd,ptype,& + & part,afmt,nrhs,istopc,itmax,itrace,itrs,eps) + + select case(psb_toupper(agfmt)) + case('ELG') + agmold => aelg + case('HLG') + agmold => ahlg + case('CSRG') + agmold => acsrg + case default + write(*,*) 'Unknown format defaulting to CSRG' + agmold => acsrg + end select + + call psb_barrier(ctxt) + t1 = psb_wtime() + + if (iam == psb_root_) then + select case(psb_toupper(filefmt)) + case('MM') + ! For Matrix Market we have an input file for the matrix + ! and an (optional) second file for the RHS. + call mm_mat_read(aux_a,info,iunit=iunit,filename=mtrx_file) + if (info == psb_success_) then + if (rhs_file /= 'NONE') then + call mm_array_read(aux_b,info,iunit=iunit,filename=rhs_file) + end if + end if + + case ('HB') + ! For Harwell-Boeing we have a single file which may or may not + ! contain an RHS. + call hb_read(aux_a,info,iunit=iunit,b=aux_b,filename=mtrx_file) + + case default + info = -1 + write(psb_err_unit,*) 'Wrong choice for fileformat ', filefmt + end select + + if (info /= psb_success_) then + write(psb_err_unit,*) 'Error while reading input matrix ' + call psb_abort(ctxt) + end if + + m = aux_a%get_nrows() + call psb_bcast(ctxt,m) + + ! At this point aux_b may still be unallocated + if (size(aux_b) == m*nrhs) then + ! if any rhs were present, broadcast the first one + write(psb_err_unit,'("Ok, got an rhs ")') + b_mv_glob =>aux_b(:,:) + else + write(psb_out_unit,'("Generating an rhs...")') + write(psb_out_unit,'("Number of RHS: ",i3)') nrhs + write(psb_out_unit,'(" ")') + call psb_realloc(m,nrhs,aux_b,ircode) + if (ircode /= 0) then + call psb_errpush(psb_err_alloc_dealloc_,name) + goto 9999 + endif + + b_mv_glob => aux_b(:,:) + do i=1, m + do j=1, nrhs + b_mv_glob(i,j) = done + !call random_number(random_value) + !b_mv_glob(i,j) = random_value + enddo + enddo + endif + + else + call psb_bcast(ctxt,m) + + end if + + ! switch over different partition types + select case(psb_toupper(part)) + case('BLOCK') + if (iam == psb_root_) write(psb_out_unit,'("Partition type: block")') + call psb_matdist(aux_a,a,ctxt,desc_a,info,fmt=afmt,parts=part_block) + + case('GRAPH') + if (iam == psb_root_) then + write(psb_out_unit,'("Partition type: graph vector")') + write(psb_out_unit,'(" ")') + call aux_a%cscnv(info,type='csr') + lnp = np + call build_mtpart(aux_a,lnp) + + endif + call psb_barrier(ctxt) + call distr_mtpart(psb_root_,ctxt) + call getv_mtpart(ivg) + call psb_matdist(aux_a,a,ctxt,desc_a,info,fmt=afmt,vg=ivg) + + case default + if (iam == psb_root_) write(psb_out_unit,'("Partition type: block")') + call psb_matdist(aux_a,a,ctxt,desc_a,info,fmt=afmt,parts=part_block) + end select + + call a%cscnv(agpu,info,mold=agmold) + if ((info /= 0).or.(psb_get_errstatus()/=0)) then + write(0,*) 'From cscnv ',info + call psb_error() + stop + end if + call desc_a%cnv(mold=imold) + + call psb_scatter(b_mv_glob,b_mv,desc_a,info,root=psb_root_,mold=gpumold) + call psb_geall(x_mv,desc_a,info,nrhs) + call x_mv%zero() + call psb_geasb(x_mv,desc_a,info,mold=gpumold) + call psb_geall(r_mv,desc_a,info,nrhs) + call r_mv%zero() + call psb_geasb(r_mv,desc_a,info) + + t2 = psb_wtime() - t1 + call psb_amx(ctxt, t2) + + if (iam == psb_root_) then + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Time to read and partition matrix : ",es12.5)')t2 + write(psb_out_unit,'(" ")') + end if + + ! building the preconditioner + call prec%init(ctxt,ptype,info) + t1 = psb_wtime() + call prec%build(a,desc_a,info) + tprec = psb_wtime()-t1 + if (info /= psb_success_) then + call psb_errpush(psb_err_from_subroutine_,name,a_err='psb_precbld') + goto 9999 + end if + + call psb_amx(ctxt,tprec) + + if(iam == psb_root_) then + write(psb_out_unit,'("Preconditioner time: ",es12.5)')tprec + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Starting algorithm")') + write(psb_out_unit,'(" ")') + end if + + call psb_barrier(ctxt) + t1 = psb_wtime() + + call psb_krylov(kmethd,agpu,prec,b_mv,x_mv,eps,desc_a,info,& + & itmax=itmax,iter=iter,err=err,itrace=itrace,& + & itrs=itrs,istop=istopc) + + call psb_barrier(ctxt) + t2 = psb_wtime() - t1 + call psb_amx(ctxt,t2) + + if(iam == psb_root_) then + write(psb_out_unit,'("Finished algorithm")') + write(psb_out_unit,'(" ")') + end if + + call psb_geaxpby(done,b_mv,dzero,r_mv,desc_a,info) + call psb_spmm(-done,a,x_mv,done,r_mv,desc_a,info) + + resmx = psb_genrm2(r_mv,desc_a,info) + resmxp = psb_geamax(r_mv,desc_a,info) + + amatsize = a%sizeof() + descsize = desc_a%sizeof() + precsize = prec%sizeof() + + call psb_sum(ctxt,amatsize) + call psb_sum(ctxt,descsize) + call psb_sum(ctxt,precsize) + + call psb_gather(x_mv_glob,x_mv,desc_a,info,root=psb_root_) + if (info == psb_success_) call psb_gather(r_mv_glob,r_mv,desc_a,info,root=psb_root_) + if (info /= psb_success_) goto 9999 + + if (iam == psb_root_) then + call prec%descr(info) + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Computed solution on: ",i8," processors")')np + write(psb_out_unit,'("Matrix: ",a)')mtrx_file + write(psb_out_unit,'("Storage format for A: ",a)')a%get_fmt() + write(psb_out_unit,'("Storage format for DESC_A: ",a)')desc_a%get_fmt() + write(psb_out_unit,'("Total memory occupation for A: ",i12)')amatsize + write(psb_out_unit,'("Total memory occupation for PREC: ",i12)')precsize + write(psb_out_unit,'("Total memory occupation for DESC_A: ",i12)')descsize + write(psb_out_unit,'("Iterations to convergence: ",i12)')iter + write(psb_out_unit,'("Error estimate on exit: ",es12.5)')err + write(psb_out_unit,'("Time to buil prec.: ",es12.5)')tprec + write(psb_out_unit,'("Time to solve system: ",es12.5)')t2 + write(psb_out_unit,'("Time per iteration: ",es12.5)')t2/(iter) + write(psb_out_unit,'("Total time: ",es12.5)')t2+tprec + write(psb_out_unit,'("Residual norm 2: ",es12.5)')maxval(resmx) + write(psb_out_unit,'("Residual norm inf: ",es12.5)')resmxp + write(psb_out_unit,'(a8,4(2x,a20))') 'I','X(I)','R(I)','B(I)' + if (print_matrix) then + do i=1,m + write(psb_out_unit,993) i, x_mv_glob(i,:), r_mv_glob(i,:), b_mv_glob(i,:) + end do + end if + end if + +998 format(i8,4(2x,g20.14)) +993 format(i6,4(1x,e12.6)) + + call psb_gefree(b_mv,desc_a,info) + call psb_gefree(x_mv,desc_a,info) + call psb_spfree(a,desc_a,info) + call prec%free(info) + call psb_cdfree(desc_a,info) + call psb_cuda_exit() + call psb_exit(ctxt) + + return + +9999 call psb_error(ctxt) + + return + +end program psb_dbf_sample diff --git a/test/block_krylov/gpu/psb_dpde_gen.F90 b/test/block_krylov/gpu/psb_dpde_gen.F90 new file mode 100644 index 00000000..a8c97d06 --- /dev/null +++ b/test/block_krylov/gpu/psb_dpde_gen.F90 @@ -0,0 +1,762 @@ +module psb_d_pde3d_mod + + use psb_base_mod, only : psb_dpk_, psb_ipk_, psb_lpk_, psb_desc_type,& + & psb_dspmat_type, psb_d_multivect_type, dzero,& + & psb_d_base_sparse_mat, psb_d_base_multivect_type, & + & psb_i_base_vect_type, psb_l_base_vect_type + + interface + function d_func_3d(x,y,z) result(val) + import :: psb_dpk_ + real(psb_dpk_), intent(in) :: x,y,z + real(psb_dpk_) :: val + end function d_func_3d + end interface + + interface psb_gen_pde3d + module procedure psb_d_gen_pde3d + end interface psb_gen_pde3d + +contains + + function d_null_func_3d(x,y,z) result(val) + + real(psb_dpk_), intent(in) :: x,y,z + real(psb_dpk_) :: val + + val = dzero + + end function d_null_func_3d + ! + ! functions parametrizing the differential equation + ! + + ! + ! Note: b1, b2 and b3 are the coefficients of the first + ! derivative of the unknown function. The default + ! we apply here is to have them zero, so that the resulting + ! matrix is symmetric/hermitian and suitable for + ! testing with CG and FCG. + ! When testing methods for non-hermitian matrices you can + ! change the B1/B2/B3 functions to e.g. done/sqrt((3*done)) + ! + function b1(x,y,z) + use psb_base_mod, only : psb_dpk_, done, dzero + implicit none + real(psb_dpk_) :: b1 + real(psb_dpk_), intent(in) :: x,y,z + b1=done/sqrt((3*done)) + end function b1 + function b2(x,y,z) + use psb_base_mod, only : psb_dpk_, done, dzero + implicit none + real(psb_dpk_) :: b2 + real(psb_dpk_), intent(in) :: x,y,z + b2=done/sqrt((3*done)) + end function b2 + function b3(x,y,z) + use psb_base_mod, only : psb_dpk_, done, dzero + implicit none + real(psb_dpk_) :: b3 + real(psb_dpk_), intent(in) :: x,y,z + b3=done/sqrt((3*done)) + end function b3 + function c(x,y,z) + use psb_base_mod, only : psb_dpk_, done, dzero + implicit none + real(psb_dpk_) :: c + real(psb_dpk_), intent(in) :: x,y,z + c=dzero + end function c + function a1(x,y,z) + use psb_base_mod, only : psb_dpk_, done, dzero + implicit none + real(psb_dpk_) :: a1 + real(psb_dpk_), intent(in) :: x,y,z + a1=done/80 + end function a1 + function a2(x,y,z) + use psb_base_mod, only : psb_dpk_, done, dzero + implicit none + real(psb_dpk_) :: a2 + real(psb_dpk_), intent(in) :: x,y,z + a2=done/80 + end function a2 + function a3(x,y,z) + use psb_base_mod, only : psb_dpk_, done, dzero + implicit none + real(psb_dpk_) :: a3 + real(psb_dpk_), intent(in) :: x,y,z + a3=done/80 + end function a3 + function g(x,y,z) + use psb_base_mod, only : psb_dpk_, done, dzero + implicit none + real(psb_dpk_) :: g + real(psb_dpk_), intent(in) :: x,y,z + g = dzero + if (x == done) then + g = done + else if (x == dzero) then + g = exp(y**2-z**2) + end if + end function g + + + ! + ! subroutine to allocate and fill in the coefficient matrix and + ! the rhs. + ! + subroutine psb_d_gen_pde3d(ctxt,idim,a,bmv,xmv,nrhs,desc_a,afmt,info,& + & f,amold,vmold,imold,partition,nrl,iv,tnd) + use psb_base_mod + use psb_util_mod + ! + ! Discretizes the partial differential equation + ! + ! a1 dd(u) a2 dd(u) a3 dd(u) b1 d(u) b2 d(u) b3 d(u) + ! - ------ - ------ - ------ + ----- + ------ + ------ + c u = f + ! dxdx dydy dzdz dx dy dz + ! + ! with Dirichlet boundary conditions + ! u = g + ! + ! on the unit cube 0<=x,y,z<=1. + ! + ! + ! Note that if b1=b2=b3=c=0., the PDE is the Laplace equation. + ! + implicit none + integer(psb_ipk_) :: idim + type(psb_dspmat_type) :: a + type(psb_d_multivect_type) :: xmv,bmv + integer(psb_ipk_) :: nrhs + type(psb_desc_type) :: desc_a + type(psb_ctxt_type) :: ctxt + integer(psb_ipk_) :: info + character(len=*) :: afmt + procedure(d_func_3d), optional :: f + class(psb_d_base_sparse_mat), optional :: amold + class(psb_d_base_multivect_type), optional :: vmold + class(psb_i_base_vect_type), optional :: imold + integer(psb_ipk_), optional :: partition, nrl,iv(:) + logical, optional :: tnd + ! Local variables. + + integer(psb_ipk_), parameter :: nb=20 + type(psb_d_csc_sparse_mat) :: acsc + type(psb_d_coo_sparse_mat) :: acoo + type(psb_d_csr_sparse_mat) :: acsr + real(psb_dpk_) :: zt(nb,nrhs),x,y,z + integer(psb_ipk_) :: nnz,nr,nlr,i,j,ii,ib,k, partition_ + integer(psb_lpk_) :: m,n,glob_row,nt + integer(psb_ipk_) :: ix,iy,iz,ia,indx_owner + ! For 3D partition + ! Note: integer control variables going directly into an MPI call + ! must be 4 bytes, i.e. psb_mpk_ + integer(psb_mpk_) :: npdims(3), npp, minfo + integer(psb_ipk_) :: npx,npy,npz, iamx,iamy,iamz,mynx,myny,mynz + integer(psb_ipk_), allocatable :: bndx(:),bndy(:),bndz(:) + ! Process grid + integer(psb_ipk_) :: np, iam + integer(psb_ipk_) :: icoeff + integer(psb_lpk_), allocatable :: irow(:),icol(:),myidx(:) + real(psb_dpk_), allocatable :: val(:) + ! deltah dimension of each grid cell + ! deltat discretization time + real(psb_dpk_) :: deltah, sqdeltah, deltah2 + real(psb_dpk_), parameter :: rhs=dzero,one=done,zero=dzero + real(psb_dpk_) :: t0, t1, t2, t3, tasb, talc, ttot, tgen, tcdasb + integer(psb_ipk_) :: err_act + procedure(d_func_3d), pointer :: f_ + logical :: tnd_ + character(len=20) :: name, ch_err,tmpfmt + + info = psb_success_ + name = 'create_matrix' + call psb_erractionsave(err_act) + + call psb_info(ctxt, iam, np) + + + if (present(f)) then + f_ => f + else + f_ => d_null_func_3d + end if + + deltah = done/(idim+2) + sqdeltah = deltah*deltah + deltah2 = (2*done)* deltah + + if (present(partition)) then + if ((1<= partition).and.(partition <= 3)) then + partition_ = partition + else + write(*,*) 'Invalid partition choice ',partition,' defaulting to 3' + partition_ = 3 + end if + else + partition_ = 3 + end if + + ! initialize array descriptor and sparse matrix storage. provide an + ! estimate of the number of non zeroes + + m = (1_psb_lpk_*idim)*idim*idim + n = m + nnz = ((n*7)/(np)) + if(iam == psb_root_) write(psb_out_unit,'("Generating Matrix (size=",i0,")...")')n + t0 = psb_wtime() + select case(partition_) + case(1) + ! A BLOCK partition + if (present(nrl)) then + nr = nrl + else + ! + ! Using a simple BLOCK distribution. + ! + nt = (m+np-1)/np + nr = max(0,min(nt,m-(iam*nt))) + end if + + nt = nr + call psb_sum(ctxt,nt) + if (nt /= m) then + write(psb_err_unit,*) iam, 'Initialization error ',nr,nt,m + info = -1 + call psb_barrier(ctxt) + call psb_abort(ctxt) + return + end if + + ! + ! First example of use of CDALL: specify for each process a number of + ! contiguous rows + ! + call psb_cdall(ctxt,desc_a,info,nl=nr) + myidx = desc_a%get_global_indices() + nlr = size(myidx) + + case(2) + ! A partition defined by the user through IV + + if (present(iv)) then + if (size(iv) /= m) then + write(psb_err_unit,*) iam, 'Initialization error: wrong IV size',size(iv),m + info = -1 + call psb_barrier(ctxt) + call psb_abort(ctxt) + return + end if + else + write(psb_err_unit,*) iam, 'Initialization error: IV not present' + info = -1 + call psb_barrier(ctxt) + call psb_abort(ctxt) + return + end if + + ! + ! Second example of use of CDALL: specify for each row the + ! process that owns it + ! + call psb_cdall(ctxt,desc_a,info,vg=iv) + myidx = desc_a%get_global_indices() + nlr = size(myidx) + + case(3) + ! A 3-dimensional partition + + ! A nifty MPI function will split the process list + npdims = 0 + call mpi_dims_create(np,3,npdims,info) + npx = npdims(1) + npy = npdims(2) + npz = npdims(3) + + allocate(bndx(0:npx),bndy(0:npy),bndz(0:npz)) + ! We can reuse idx2ijk for process indices as well. + call idx2ijk(iamx,iamy,iamz,iam,npx,npy,npz,base=0) + ! Now let's split the 3D cube in hexahedra + call dist1Didx(bndx,idim,npx) + mynx = bndx(iamx+1)-bndx(iamx) + call dist1Didx(bndy,idim,npy) + myny = bndy(iamy+1)-bndy(iamy) + call dist1Didx(bndz,idim,npz) + mynz = bndz(iamz+1)-bndz(iamz) + + ! How many indices do I own? + nlr = mynx*myny*mynz + allocate(myidx(nlr)) + ! Now, let's generate the list of indices I own + nr = 0 + do i=bndx(iamx),bndx(iamx+1)-1 + do j=bndy(iamy),bndy(iamy+1)-1 + do k=bndz(iamz),bndz(iamz+1)-1 + nr = nr + 1 + call ijk2idx(myidx(nr),i,j,k,idim,idim,idim) + end do + end do + end do + if (nr /= nlr) then + write(psb_err_unit,*) iam,iamx,iamy,iamz, 'Initialization error: NR vs NLR ',& + & nr,nlr,mynx,myny,mynz + info = -1 + call psb_barrier(ctxt) + call psb_abort(ctxt) + end if + + ! + ! Third example of use of CDALL: specify for each process + ! the set of global indices it owns. + ! + call psb_cdall(ctxt,desc_a,info,vl=myidx) + + case default + write(psb_err_unit,*) iam, 'Initialization error: should not get here' + info = -1 + call psb_barrier(ctxt) + call psb_abort(ctxt) + return + end select + + + if (info == psb_success_) call psb_spall(a,desc_a,info,nnz=nnz,& + & dupl=psb_dupl_err_) + ! define rhs from boundary conditions; also build initial guess + if (info == psb_success_) call psb_geall(xmv,desc_a,info,n=nrhs) + if (info == psb_success_) call psb_geall(bmv,desc_a,info,n=nrhs) + + call psb_barrier(ctxt) + talc = psb_wtime()-t0 + + if (info /= psb_success_) then + info=psb_err_from_subroutine_ + ch_err='allocation rout.' + call psb_errpush(info,name,a_err=ch_err) + goto 9999 + end if + + ! we build an auxiliary matrix consisting of one row at a + ! time; just a small matrix. might be extended to generate + ! a bunch of rows per call. + ! + allocate(val(20*nb),irow(20*nb),& + &icol(20*nb),stat=info) + if (info /= psb_success_ ) then + info=psb_err_alloc_dealloc_ + call psb_errpush(info,name) + goto 9999 + endif + + + ! loop over rows belonging to current process in a block + ! distribution. + + call psb_barrier(ctxt) + t1 = psb_wtime() + do ii=1, nlr,nb + ib = min(nb,nlr-ii+1) + icoeff = 1 + do k=1,ib + i=ii+k-1 + ! local matrix pointer + glob_row=myidx(i) + ! compute gridpoint coordinates + call idx2ijk(ix,iy,iz,glob_row,idim,idim,idim) + ! x, y, z coordinates + x = (ix-1)*deltah + y = (iy-1)*deltah + z = (iz-1)*deltah + zt(k,:) = f_(x,y,z) + ! internal point: build discretization + ! + ! term depending on (x-1,y,z) + ! + val(icoeff) = -a1(x,y,z)/sqdeltah-b1(x,y,z)/deltah2 + if (ix == 1) then + zt(k,:) = g(dzero,y,z)*(-val(icoeff)) + zt(k,:) + else + call ijk2idx(icol(icoeff),ix-1,iy,iz,idim,idim,idim) + irow(icoeff) = glob_row + icoeff = icoeff+1 + endif + ! term depending on (x,y-1,z) + val(icoeff) = -a2(x,y,z)/sqdeltah-b2(x,y,z)/deltah2 + if (iy == 1) then + zt(k,:) = g(x,dzero,z)*(-val(icoeff)) + zt(k,:) + else + call ijk2idx(icol(icoeff),ix,iy-1,iz,idim,idim,idim) + irow(icoeff) = glob_row + icoeff = icoeff+1 + endif + ! term depending on (x,y,z-1) + val(icoeff)=-a3(x,y,z)/sqdeltah-b3(x,y,z)/deltah2 + if (iz == 1) then + zt(k,:) = g(x,y,dzero)*(-val(icoeff)) + zt(k,:) + else + call ijk2idx(icol(icoeff),ix,iy,iz-1,idim,idim,idim) + irow(icoeff) = glob_row + icoeff = icoeff+1 + endif + + ! term depending on (x,y,z) + val(icoeff)=(2*done)*(a1(x,y,z)+a2(x,y,z)+a3(x,y,z))/sqdeltah & + & + c(x,y,z) + call ijk2idx(icol(icoeff),ix,iy,iz,idim,idim,idim) + irow(icoeff) = glob_row + icoeff = icoeff+1 + ! term depending on (x,y,z+1) + val(icoeff)=-a3(x,y,z)/sqdeltah+b3(x,y,z)/deltah2 + if (iz == idim) then + zt(k,:) = g(x,y,done)*(-val(icoeff)) + zt(k,:) + else + call ijk2idx(icol(icoeff),ix,iy,iz+1,idim,idim,idim) + irow(icoeff) = glob_row + icoeff = icoeff+1 + endif + ! term depending on (x,y+1,z) + val(icoeff)=-a2(x,y,z)/sqdeltah+b2(x,y,z)/deltah2 + if (iy == idim) then + zt(k,:) = g(x,done,z)*(-val(icoeff)) + zt(k,:) + else + call ijk2idx(icol(icoeff),ix,iy+1,iz,idim,idim,idim) + irow(icoeff) = glob_row + icoeff = icoeff+1 + endif + ! term depending on (x+1,y,z) + val(icoeff)=-a1(x,y,z)/sqdeltah+b1(x,y,z)/deltah2 + if (ix==idim) then + zt(k,:) = g(done,y,z)*(-val(icoeff)) + zt(k,:) + else + call ijk2idx(icol(icoeff),ix+1,iy,iz,idim,idim,idim) + irow(icoeff) = glob_row + icoeff = icoeff+1 + endif + + end do + call psb_spins(icoeff-1,irow,icol,val,a,desc_a,info) + if(info /= psb_success_) exit + call psb_geins(ib,myidx(ii:ii+ib-1),zt(1:ib,:),bmv,desc_a,info) + if(info /= psb_success_) exit + zt(:,:)=dzero + call psb_geins(ib,myidx(ii:ii+ib-1),zt(1:ib,:),xmv,desc_a,info) + if(info /= psb_success_) exit + end do + + tgen = psb_wtime()-t1 + if(info /= psb_success_) then + info=psb_err_from_subroutine_ + ch_err='insert rout.' + call psb_errpush(info,name,a_err=ch_err) + goto 9999 + end if + + deallocate(val,irow,icol) + + call psb_barrier(ctxt) + t1 = psb_wtime() + call psb_cdasb(desc_a,info,mold=imold) + tcdasb = psb_wtime()-t1 + call psb_barrier(ctxt) + t1 = psb_wtime() + if (info == psb_success_) then + if (present(amold)) then + call psb_spasb(a,desc_a,info,mold=amold,bld_and=tnd) + else + call psb_spasb(a,desc_a,info,afmt=afmt,bld_and=tnd) + end if + end if + call psb_barrier(ctxt) + if(info /= psb_success_) then + info=psb_err_from_subroutine_ + ch_err='asb rout.' + call psb_errpush(info,name,a_err=ch_err) + goto 9999 + end if + if (info == psb_success_) call psb_geasb(xmv,desc_a,info,mold=vmold) + if (info == psb_success_) call psb_geasb(bmv,desc_a,info,mold=vmold) + if(info /= psb_success_) then + info=psb_err_from_subroutine_ + ch_err='asb rout.' + call psb_errpush(info,name,a_err=ch_err) + goto 9999 + end if + tasb = psb_wtime()-t1 + call psb_barrier(ctxt) + ttot = psb_wtime() - t0 + + call psb_amx(ctxt,talc) + call psb_amx(ctxt,tgen) + call psb_amx(ctxt,tasb) + call psb_amx(ctxt,ttot) + if(iam == psb_root_) then + tmpfmt = a%get_fmt() + write(psb_out_unit,'("The matrix has been generated and assembled in ",a3," format.")')& + & tmpfmt + write(psb_out_unit,'("-allocation time : ",es12.5)') talc + write(psb_out_unit,'("-coeff. gen. time : ",es12.5)') tgen + write(psb_out_unit,'("-desc asbly time : ",es12.5)') tcdasb + write(psb_out_unit,'("- mat asbly time : ",es12.5)') tasb + write(psb_out_unit,'("-total time : ",es12.5)') ttot + + end if + call psb_erractionrestore(err_act) + return + +9999 call psb_error_handler(ctxt,err_act) + + return + end subroutine psb_d_gen_pde3d + +end module psb_d_pde3d_mod + +program dpdegen + use psb_base_mod + use psb_util_mod + use psb_prec_mod + use psb_krylov_mod + use psb_ext_mod + use psb_cuda_mod + use psb_d_pde3d_mod + + ! input parameters + character(len=40) :: kmethd = "GMRES" + character(len=40) :: ptype = "NONE" + character(len=5) :: agfmt = "CSRG" + integer(psb_ipk_) :: nrhs = 50 + integer(psb_ipk_) :: istopbg = 1 + integer(psb_ipk_) :: istoprg = 2 + integer(psb_ipk_) :: itmax = 500 + integer(psb_ipk_) :: itrace = -1 + integer(psb_ipk_) :: itrs = 100 + real(psb_dpk_) :: eps = 1.d-7 + integer(psb_ipk_) :: idim = 10 + logical :: tnd = .false. + + + ! sparse matrix + type(psb_dspmat_type) :: a, aux_a + + ! preconditioner data + type(psb_dprec_type) :: prec + + ! miscellaneous + real(psb_dpk_) :: tb1, tb2, tr1, tr2 + real(psb_dpk_), allocatable :: tb(:), tr(:) + + ! descriptor + type(psb_desc_type) :: desc_a + + ! dense matrices + type(psb_d_multivect_type), target :: x_mv, b_mv + type(psb_d_vect_type), target :: x_col, b_col + type(psb_d_multivect_cuda) :: gpumold_mv + type(psb_d_vect_cuda) :: gpumold_col + type(psb_i_vect_cuda) :: imold + real(psb_dpk_), allocatable :: b_mv_glob(:,:) + real(psb_dpk_), allocatable :: b_col_glob(:) + + ! blacs parameters + type(psb_ctxt_type) :: ctxt + integer :: iam, np + + ! solver parameters + real(psb_dpk_) :: err, cond + integer(psb_ipk_) :: reps = 2 + + ! molds + type(psb_d_cuda_elg_sparse_mat), target :: aelg + type(psb_d_cuda_csrg_sparse_mat), target :: acsrg + type(psb_d_cuda_hlg_sparse_mat), target :: ahlg + class(psb_d_base_sparse_mat), pointer :: agmold + + ! other variables + integer(psb_ipk_) :: info, i, j, m_problem, iter, rep + character(len=20) :: name, ch_err + real(psb_dpk_) :: random_value + + ! Init environment + info=psb_success_ + call psb_init(ctxt) + call psb_info(ctxt,iam,np) + call psb_cuda_init(ctxt) + if (iam < 0) then + ! This should not happen, but just in case + call psb_exit(ctxt) + stop + endif + if(psb_get_errstatus() /= 0) goto 9999 + name='pdegenmm-cuda' + ! + ! Hello world + ! + if (iam == psb_root_) then + write(*,*) 'Welcome to PSBLAS version: ',psb_version_string_ + write(*,*) 'This is the ',trim(name),' sample program' + write(psb_out_unit,'("Number of processors: ",i8)')np + end if + write(*,*) 'Process ',iam,' running on device: ', psb_cuda_getDevice(),' out of', psb_cuda_getDeviceCount() + write(*,*) 'Process ',iam,' device ', psb_cuda_getDevice(),' is a: ', trim(psb_cuda_DeviceName()) + + ! allocate and fill in the coefficient matrix and initial vectors + ! + call psb_barrier(ctxt) + t1 = psb_wtime() + call psb_gen_pde3d(ctxt,idim,a,b_mv,x_mv,nrhs,desc_a,'CSR ',info,partition=3,tnd=tnd) + call psb_barrier(ctxt) + t2 = psb_wtime() - t1 + if(info /= psb_success_) then + info=psb_err_from_subroutine_ + ch_err='create_matrix' + call psb_errpush(info,name,a_err=ch_err) + goto 9999 + end if + if (iam == psb_root_) write(psb_out_unit,'("Overall matrix creation time : ",es12.5)')t2 + if (iam == psb_root_) write(psb_out_unit,'(" ")') + + ! building the preconditioner + call prec%init(ctxt,ptype,info) + call prec%build(a,desc_a,info) + if (info /= psb_success_) then + call psb_errpush(psb_err_from_subroutine_,name,a_err='psb_precbld') + goto 9999 + end if + + select case(psb_toupper(agfmt)) + case('ELG') + agmold => aelg + case('HLG') + agmold => ahlg + case('CSRG') + agmold => acsrg + case default + write(*,*) 'Unknown format defaulting to HLG' + agmold => ahlg + end select + call a%cscnv(info,mold=agmold) + call desc_a%cnv(mold=imold) + if ((info /= 0).or.(psb_get_errstatus()/=0)) then + write(0,*) 'From cscnv ',info + call psb_error() + stop + end if + + ! Set RHS + call psb_geall(b_mv_glob,desc_a,info,n=nrhs) + do i=1,b_mv%get_nrows() + do j=1,b_mv%get_ncols() + call random_number(random_value) + b_mv_glob(i,j) = random_value + end do + end do + + if (iam == psb_root_) then + allocate(tb(reps),tr(reps)) + tb = dzero + tr = dzero + end if + + do rep=1,reps + call psb_scatter(b_mv_glob,b_mv,desc_a,info,root=psb_root_,mold=gpumold_mv) + call psb_geall(x_mv,desc_a,info,nrhs) + call x_mv%zero() + call psb_geasb(x_mv,desc_a,info,mold=gpumold_mv) + + call psb_barrier(ctxt) + tb1 = psb_wtime() + + call psb_krylov(kmethd,a,prec,b_mv,x_mv,eps,desc_a,info,& + & itmax=itmax,iter=iter,err=err,itrace=itrace,& + & itrs=itrs,istop=istopbg) + + call psb_barrier(ctxt) + tb2 = psb_wtime() - tb1 + call psb_amx(ctxt,tb2) + + if (iam == psb_root_) then + tb(rep) = tb2 + write(*,*) 'Time', rep, tb(rep), iter + end if + call psb_barrier(ctxt) + + call psb_gefree(b_mv,desc_a,info) + call psb_gefree(x_mv,desc_a,info) + end do + + if(iam == psb_root_) then + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Finished BGMRES")') + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Starting sGMRES")') + write(psb_out_unit,'(" ")') + end if + + call psb_barrier(ctxt) + + do rep=1,reps + do i=1,nrhs + b_col_glob = b_mv_glob(:,i) + call psb_scatter(b_col_glob,b_col,desc_a,info,root=psb_root_,mold=gpumold_col) + call psb_geall(x_col,desc_a,info) + call x_col%zero() + call psb_geasb(x_col,desc_a,info,mold=gpumold_col) + + cond = dzero + + call psb_barrier(ctxt) + tr1 = psb_wtime() + + call psb_krylov(kmethd,a,prec,b_col,x_col,eps,desc_a,info,& + & itmax=itmax,iter=iter,err=err,itrace=itrace,& + & istop=istoprg,irst=itrs,cond=cond) + + call psb_barrier(ctxt) + tr2 = psb_wtime() - tr1 + call psb_amx(ctxt,tr2) + + if (iam == psb_root_) then + tr(rep) = tr(rep) + tr2 + end if + call psb_barrier(ctxt) + + call psb_gefree(b_col, desc_a,info) + call psb_gefree(x_col, desc_a,info) + end do + + if (iam == psb_root_) then + write(*,*) 'Time', rep, tr(rep), iter + end if + end do + + call psb_barrier(ctxt) + + if(iam == psb_root_) then + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Finished sGMRES")') + write(psb_out_unit,'(" ")') + end if + + ! + ! cleanup storage and exit + ! + call psb_spfree(a,desc_a,info) + call psb_cdfree(desc_a,info) + if(info /= psb_success_) then + info=psb_err_from_subroutine_ + ch_err='free routine' + call psb_errpush(info,name,a_err=ch_err) + goto 9999 + end if + + call psb_cuda_exit() + call psb_exit(ctxt) + return + +9999 continue + + call psb_error(ctxt) + +end program dpdegen diff --git a/test/block_krylov/kernel/Makefile b/test/block_krylov/kernel/Makefile index 330babad..a5c07d4e 100644 --- a/test/block_krylov/kernel/Makefile +++ b/test/block_krylov/kernel/Makefile @@ -8,11 +8,11 @@ PSBLIBDIR=$(TOPDIR)/lib/ OPTDIR=$(LIBDIR) PSBINCDIR=$(TOPDIR)/include PSBMODDIR=$(TOPDIR)/modules -PSBLAS_LIB= -L$(LIBDIR) -L$(PSBLIBDIR) $(LCUDA) -lpsb_ext -lpsb_util -lpsb_base +PSBLAS_LIB= -L$(LIBDIR) -L$(PSBLIBDIR) $(LCUDA) -lpsb_ext -lpsb_util -lpsb_base -lpsb_krylov INCDIR=$(TOPDIR)/include MODDIR=$(TOPDIR)/modules -LDLIBS=$(PSBGPULDLIBS) +LDLIBS=$(PSBGPULDLIBS) FINCLUDES=$(FMFLAG)$(MODDIR) $(FMFLAG)$(INCDIR) $(FMFLAG). $(FMFLAG)$(PSBMODDIR) $(FMFLAG)$(PSBINCDIR) $(LIBRSB_DEFINES) diff --git a/test/block_krylov/kernel/dpdegenmm.F90 b/test/block_krylov/kernel/dpdegenmm.F90 index cece5417..070e8465 100644 --- a/test/block_krylov/kernel/dpdegenmm.F90 +++ b/test/block_krylov/kernel/dpdegenmm.F90 @@ -662,7 +662,7 @@ program pdegenmm !call get_parms(ctxt,nrhs,acfmt,agfmt,idim,tnd) nrhs=2 acfmt='CSR' - agfmt='HLG' + agfmt='CSRG' idim=2 tnd=.false. call psb_init_timers() @@ -804,27 +804,30 @@ program pdegenmm x1 = b_mv%get_vect() x2 = b_mv_g%get_vect() -! ! TODO test AXPBY -! call psb_geall(xg,desc_a,info) -! call psb_geasb(xg,desc_a,info,mold=tmold) -! call xg%set(done) -! call xg%sync() -! call psb_geall(bg,desc_a,info) -! call psb_geasb(bg,desc_a,info,mold=tmold) -! !call bg%set(done+done) - -! ! ! TODO: Non funziona spgpuDaxpby (axpbyMultiVecDeviceDouble) -! call psb_geaxpby(done,xg,dzero,bg,desc_a,info) -! call psb_cuda_DeviceSync() - -! write(*,*) 'BG ', bg%is_dev(), bg%is_host(), bg%is_sync() -! call bg%sync() -! write(*,*) 'BG ', bg%is_dev(), bg%is_host(), bg%is_sync() -! do i=1,8 -! write(*,*) bg%v%v(i) -! end do + ! TODO test AXPBY + call psb_geall(xg,desc_a,info) + call psb_geasb(xg,desc_a,info,mold=tmold) + call xg%set(done) + !call xg%sync() + call psb_geall(bg,desc_a,info) + call psb_geasb(bg,desc_a,info,mold=tmold) + !call bg%set(done+done) + do i=1,8 + write(*,*) xg%v%v(i) + end do -! return +! ! TODO: Non funziona spgpuDaxpby (axpbyMultiVecDeviceDouble) + call psb_geaxpby(done,xg,dzero,bg,desc_a,info) + call psb_cuda_DeviceSync() + + write(*,*) 'BG ', bg%is_dev(), bg%is_host(), bg%is_sync() + call bg%sync() + write(*,*) 'BG ', bg%is_dev(), bg%is_host(), bg%is_sync() + do i=1,8 + write(*,*) bg%v%v(i) + end do + + return ! call x_mv_g%set(done) ! call x_mv_g%sync() diff --git a/test/block_krylov/psb_dbf_sample.f90 b/test/block_krylov/psb_dbf_sample.f90 index a5f593b9..d6be4716 100644 --- a/test/block_krylov/psb_dbf_sample.f90 +++ b/test/block_krylov/psb_dbf_sample.f90 @@ -52,7 +52,7 @@ program psb_dbf_sample real(psb_dpk_), allocatable :: resmx(:) real(psb_dpk_) :: resmxp integer(psb_ipk_), allocatable :: ivg(:) - logical :: print_matrix = .true. + logical :: print_matrix = .false. call psb_init(ctxt) call psb_info(ctxt,iam,np) @@ -133,9 +133,9 @@ program psb_dbf_sample b_mv_glob => aux_b(:,:) do i=1, m do j=1, nrhs - b_mv_glob(i,j) = done - !call random_number(random_value) - !b_mv_glob(i,j) = random_value + !b_mv_glob(i,j) = done + call random_number(random_value) + b_mv_glob(i,j) = random_value enddo enddo endif diff --git a/test/block_krylov/psb_perf_test.f90 b/test/block_krylov/psb_perf_test.f90 new file mode 100644 index 00000000..ecc3b55a --- /dev/null +++ b/test/block_krylov/psb_perf_test.f90 @@ -0,0 +1,296 @@ +program psb_perf_test + + use psb_base_mod + use psb_prec_mod + use psb_krylov_mod + use psb_util_mod + + implicit none + + ! input parameters + character(len=40) :: mtrx_file = "pde80.mtx" + character(len=2) :: filefmt = "MM" + character(len=40) :: kmethd = "GMRES" + character(len=40) :: ptype = "NONE" + character(len=20) :: part = "GRAPH" + character(len=5) :: afmt = "CSR" + integer(psb_ipk_) :: nrhs = 4 + integer(psb_ipk_) :: istopbg = 1 + integer(psb_ipk_) :: istoprg = 2 + integer(psb_ipk_) :: itmax = 500 + integer(psb_ipk_) :: itrace = -1 + integer(psb_ipk_) :: itrs = 10 + real(psb_dpk_) :: eps = 1.d-7 + + integer(psb_ipk_) :: status + + ! sparse matrices + type(psb_dspmat_type) :: a + type(psb_ldspmat_type) :: aux_a + integer(psb_ipk_), parameter :: iunit=12 + + ! preconditioner data + type(psb_dprec_type) :: prec + + ! dense matrices + real(psb_dpk_), allocatable, target :: aux_b(:,:) + real(psb_dpk_), allocatable, save :: x_mv_glob(:,:), r_mv_glob(:,:) + real(psb_dpk_), allocatable, save :: x_col_glob(:), r_col_glob(:) + real(psb_dpk_), pointer :: b_mv_glob(:,:) + real(psb_dpk_), pointer :: b_col_glob(:) + type(psb_d_multivect_type) :: b_mv, x_mv, r_mv + type(psb_d_vect_type) :: b_col, x_col, r_col + integer(psb_ipk_) :: m + real(psb_dpk_) :: random_value + + ! communications data structure + type(psb_desc_type) :: desc_a + type(psb_ctxt_type) :: ctxt + integer(psb_ipk_) :: iam, np + integer(psb_lpk_) :: lnp + + ! solver paramters + integer(psb_ipk_) :: iter, ierr, ircode, methd + integer(psb_epk_) :: amatsize, precsize, descsize + real(psb_dpk_) :: err, cond + + ! other variables + character(len=20) :: name + integer(psb_ipk_) :: i, j, info, rep, reps = 10 + real(psb_dpk_) :: tb1, tb2, tr1, tr2 + real(psb_dpk_), allocatable :: resmx(:), tb(:), tr(:) + real(psb_dpk_) :: resmxp + integer(psb_ipk_), allocatable :: ivg(:) + + call psb_init(ctxt) + call psb_info(ctxt,iam,np) + if (iam < 0) then + ! This should not happen, but just in case + call psb_exit(ctxt) + stop + endif + + name='psb_perf_test' + if(psb_errstatus_fatal()) goto 9999 + info=psb_success_ + call psb_set_errverbosity(itwo) + + call psb_barrier(ctxt) + + if (iam == psb_root_) then + select case(psb_toupper(filefmt)) + case('MM') + call mm_mat_read(aux_a,info,iunit=iunit,filename=mtrx_file) + case ('HB') + call hb_read(aux_a,info,iunit=iunit,b=aux_b,filename=mtrx_file) + case default + info = -1 + write(psb_err_unit,*) 'Wrong choice for fileformat ', filefmt + end select + if (info /= psb_success_) then + write(psb_err_unit,*) 'Error while reading input matrix ' + call psb_abort(ctxt) + end if + + m = aux_a%get_nrows() + call psb_bcast(ctxt,m) + + ! At this point aux_b may still be unallocated + if (size(aux_b) == m*nrhs) then + ! if any rhs were present, broadcast the first one + write(psb_err_unit,'("Ok, got an rhs ")') + b_mv_glob =>aux_b(:,:) + else + write(psb_out_unit,'("Generating an rhs...")') + write(psb_out_unit,'("Number of RHS: ",i3)') nrhs + write(psb_out_unit,'(" ")') + call psb_realloc(m,nrhs,aux_b,ircode) + if (ircode /= 0) then + call psb_errpush(psb_err_alloc_dealloc_,name) + goto 9999 + endif + + b_mv_glob => aux_b(:,:) + do i=1, m + do j=1, nrhs + b_mv_glob(i,j) = done + !call random_number(random_value) + !b_mv_glob(i,j) = random_value + enddo + enddo + endif + else + call psb_bcast(ctxt,m) + end if + + ! switch over different partition types + select case(psb_toupper(part)) + case('BLOCK') + if (iam == psb_root_) write(psb_out_unit,'("Partition type: block")') + call psb_matdist(aux_a,a,ctxt,desc_a,info,fmt=afmt,parts=part_block) + case('GRAPH') + if (iam == psb_root_) then + write(psb_out_unit,'("Partition type: graph vector")') + write(psb_out_unit,'(" ")') + call aux_a%cscnv(info,type='csr') + lnp = np + call build_mtpart(aux_a,lnp) + endif + call psb_barrier(ctxt) + call distr_mtpart(psb_root_,ctxt) + call getv_mtpart(ivg) + call psb_matdist(aux_a,a,ctxt,desc_a,info,fmt=afmt,vg=ivg) + case default + if (iam == psb_root_) write(psb_out_unit,'("Partition type: block")') + call psb_matdist(aux_a,a,ctxt,desc_a,info,fmt=afmt,parts=part_block) + end select + + ! building the preconditioner + call prec%init(ctxt,ptype,info) + call prec%build(a,desc_a,info) + if (info /= psb_success_) then + call psb_errpush(psb_err_from_subroutine_,name,a_err='psb_precbld') + goto 9999 + end if + + if (iam == psb_root_) then + allocate(tb(reps),tr(reps)) + tb = dzero + tr = dzero + end if + + do rep=1,reps + + call psb_scatter(b_mv_glob,b_mv,desc_a,info,root=psb_root_) + call psb_geall(x_mv,desc_a,info,nrhs) + call x_mv%zero() + call psb_geasb(x_mv,desc_a,info) + call psb_geall(r_mv,desc_a,info,nrhs) + call r_mv%zero() + call psb_geasb(r_mv,desc_a,info) + + if(iam == psb_root_) then + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Starting BGMRES")') + write(psb_out_unit,'(" ")') + end if + + call psb_barrier(ctxt) + tb1 = psb_wtime() + + call psb_krylov(kmethd,a,prec,b_mv,x_mv,eps,desc_a,info,& + & itmax=itmax,iter=iter,err=err,itrace=itrace,& + & itrs=itrs,istop=istopbg) + + call psb_barrier(ctxt) + tb2 = psb_wtime() - tb1 + call psb_amx(ctxt,tb2) + + if (iam == psb_root_) then + tb(rep) = tb2 + end if + call psb_barrier(ctxt) + + call psb_geaxpby(done,b_mv,dzero,r_mv,desc_a,info) + call psb_spmm(-done,a,x_mv,done,r_mv,desc_a,info) + resmx = psb_genrm2(r_mv,desc_a,info) + resmxp = psb_geamax(r_mv,desc_a,info) + call psb_gather(x_mv_glob,x_mv,desc_a,info,root=psb_root_) + if (info == psb_success_) call psb_gather(r_mv_glob,r_mv,desc_a,info,root=psb_root_) + if (info /= psb_success_) goto 9999 + + if(iam == psb_root_) then + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Finished BGMRES")') + write(psb_out_unit,'(" ")') + end if + + call psb_gefree(b_mv, desc_a,info) + call psb_gefree(x_mv, desc_a,info) + + if(iam == psb_root_) then + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Starting sGMRES")') + write(psb_out_unit,'(" ")') + end if + + call psb_barrier(ctxt) + + do i=1,nrhs + if(iam == psb_root_) then + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Step ",i6)')i + write(psb_out_unit,'(" ")') + end if + + b_col_glob => aux_b(:,i) + call psb_scatter(b_col_glob,b_col,desc_a,info,root=psb_root_) + call psb_geall(x_col,desc_a,info) + call x_col%zero() + call psb_geasb(x_col,desc_a,info) + call psb_geall(r_col,desc_a,info) + call r_col%zero() + call psb_geasb(r_col,desc_a,info) + + cond = dzero + + call psb_barrier(ctxt) + tr1 = psb_wtime() + + call psb_krylov(kmethd,a,prec,b_col,x_col,eps,desc_a,info,& + & itmax=itmax,iter=iter,err=err,itrace=itrace,& + & istop=2,irst=itrs,cond=cond) + + call psb_barrier(ctxt) + tr2 = psb_wtime() - tr1 + call psb_amx(ctxt,tr2) + + if (iam == psb_root_) then + tr(rep) = tr(rep) + tr2 + end if + call psb_barrier(ctxt) + + call psb_geaxpby(done,b_col,dzero,r_col,desc_a,info) + call psb_spmm(-done,a,x_col,done,r_col,desc_a,info) + resmx = psb_genrm2(r_col,desc_a,info) + resmxp = psb_geamax(r_col,desc_a,info) + call psb_gather(x_col_glob,x_col,desc_a,info,root=psb_root_) + if (info == psb_success_) call psb_gather(r_col_glob,r_col,desc_a,info,root=psb_root_) + if (info /= psb_success_) goto 9999 + + call psb_gefree(b_col, desc_a,info) + call psb_gefree(x_col, desc_a,info) + end do + + end do + + call psb_barrier(ctxt) + + if(iam == psb_root_) then + write(psb_out_unit,'(" ")') + write(psb_out_unit,'("Finished sGMRES")') + write(psb_out_unit,'(" ")') + write(*,*) 'TB ', sum(tb)/size(tb), ' TR ', sum(tr)/size(tr) + end if + + call psb_spfree(a, desc_a,info) + call prec%free(info) + call psb_cdfree(desc_a,info) + call psb_exit(ctxt) + +! open(unit=10,file='test.csv',status='replace',action='write',iostat=status) + +! do i=1,size(x_mv_glob,1) +! write(10, 993) x_mv_glob(i,:) +! end do + +! 993 format(1x,*(g0, "; ")) + +! return + +9999 call psb_error(ctxt) + + return + +end program psb_perf_test +