fdurastante/psblas3
1
0
Fork 0
mirror of https://github.com/sfilippone/psblas3.git synced 2026-10-07 07:04:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'development' into cmake

Browse Source
This commit is contained in:
sfilippone
2025-04-10 17:36:41 +02:00
parent aeb3ee5fb6 922f7dd278
commit 09f40e653d
9 changed files with 57 additions and 56 deletions
Download Patch File Download Diff File Expand all files Collapse all files
base/serial/impl/psb_c_coo_impl.F90
+2 -2
View File
@@ -2879,9 +2879,9 @@ subroutine psb_c_coo_csput_a(nz,ia,ja,val,a,imin,imax,jmin,jmax,info)
if (info /= 0) write(0,*) name,' point 0:',info,isza,nza,nz
! Build phase. Must handle reallocations in a sensible way.
if (isza < (nza+nz)) then
write(0,*) ' before reallocate in csput ',psb_errstatus_fatal(),info
!write(0,*) ' before reallocate in csput ',psb_errstatus_fatal(),info
call a%reallocate(max(nza+nz,int(1.5*isza)))
write(0,*) ' after reallocate in csput ',psb_errstatus_fatal(),info
!write(0,*) ' after reallocate in csput ',psb_errstatus_fatal(),info
endif
isza = a%get_size()
if (isza < (nza+nz)) then
base/serial/impl/psb_c_csr_impl.F90
+12 -12
View File
@@ -152,7 +152,7 @@ contains
!$omp parallel do private(i,j, acc) schedule(static)
do i=1,m
acc = czero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -164,7 +164,7 @@ contains
!$omp parallel do private(i,j, acc)
do i=1,m
acc = czero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -176,7 +176,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = czero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -192,7 +192,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = czero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -204,7 +204,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = czero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -216,7 +216,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = czero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -231,7 +231,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = czero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -243,7 +243,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = czero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -255,7 +255,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = czero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -270,7 +270,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = czero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -282,7 +282,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = czero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -294,7 +294,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = czero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
base/serial/impl/psb_d_coo_impl.F90
+2 -2
View File
@@ -2879,9 +2879,9 @@ subroutine psb_d_coo_csput_a(nz,ia,ja,val,a,imin,imax,jmin,jmax,info)
if (info /= 0) write(0,*) name,' point 0:',info,isza,nza,nz
! Build phase. Must handle reallocations in a sensible way.
if (isza < (nza+nz)) then
write(0,*) ' before reallocate in csput ',psb_errstatus_fatal(),info
!write(0,*) ' before reallocate in csput ',psb_errstatus_fatal(),info
call a%reallocate(max(nza+nz,int(1.5*isza)))
write(0,*) ' after reallocate in csput ',psb_errstatus_fatal(),info
!write(0,*) ' after reallocate in csput ',psb_errstatus_fatal(),info
endif
isza = a%get_size()
if (isza < (nza+nz)) then
base/serial/impl/psb_d_csr_impl.F90
+12 -12
View File
@@ -152,7 +152,7 @@ contains
!$omp parallel do private(i,j, acc) schedule(static)
do i=1,m
acc = dzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -164,7 +164,7 @@ contains
!$omp parallel do private(i,j, acc)
do i=1,m
acc = dzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -176,7 +176,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = dzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -192,7 +192,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = dzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -204,7 +204,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = dzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -216,7 +216,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = dzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -231,7 +231,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = dzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -243,7 +243,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = dzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -255,7 +255,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = dzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -270,7 +270,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = dzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -282,7 +282,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = dzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -294,7 +294,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = dzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
base/serial/impl/psb_s_coo_impl.F90
+2 -2
View File
@@ -2879,9 +2879,9 @@ subroutine psb_s_coo_csput_a(nz,ia,ja,val,a,imin,imax,jmin,jmax,info)
if (info /= 0) write(0,*) name,' point 0:',info,isza,nza,nz
! Build phase. Must handle reallocations in a sensible way.
if (isza < (nza+nz)) then
write(0,*) ' before reallocate in csput ',psb_errstatus_fatal(),info
!write(0,*) ' before reallocate in csput ',psb_errstatus_fatal(),info
call a%reallocate(max(nza+nz,int(1.5*isza)))
write(0,*) ' after reallocate in csput ',psb_errstatus_fatal(),info
!write(0,*) ' after reallocate in csput ',psb_errstatus_fatal(),info
endif
isza = a%get_size()
if (isza < (nza+nz)) then
base/serial/impl/psb_s_csr_impl.F90
+12 -12
View File
@@ -152,7 +152,7 @@ contains
!$omp parallel do private(i,j, acc) schedule(static)
do i=1,m
acc = szero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -164,7 +164,7 @@ contains
!$omp parallel do private(i,j, acc)
do i=1,m
acc = szero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -176,7 +176,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = szero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -192,7 +192,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = szero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -204,7 +204,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = szero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -216,7 +216,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = szero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -231,7 +231,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = szero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -243,7 +243,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = szero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -255,7 +255,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = szero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -270,7 +270,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = szero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -282,7 +282,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = szero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -294,7 +294,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = szero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
base/serial/impl/psb_z_coo_impl.F90
+2 -2
View File
@@ -2879,9 +2879,9 @@ subroutine psb_z_coo_csput_a(nz,ia,ja,val,a,imin,imax,jmin,jmax,info)
if (info /= 0) write(0,*) name,' point 0:',info,isza,nza,nz
! Build phase. Must handle reallocations in a sensible way.
if (isza < (nza+nz)) then
write(0,*) ' before reallocate in csput ',psb_errstatus_fatal(),info
!write(0,*) ' before reallocate in csput ',psb_errstatus_fatal(),info
call a%reallocate(max(nza+nz,int(1.5*isza)))
write(0,*) ' after reallocate in csput ',psb_errstatus_fatal(),info
!write(0,*) ' after reallocate in csput ',psb_errstatus_fatal(),info
endif
isza = a%get_size()
if (isza < (nza+nz)) then
base/serial/impl/psb_z_csr_impl.F90
+12 -12
View File
@@ -152,7 +152,7 @@ contains
!$omp parallel do private(i,j, acc) schedule(static)
do i=1,m
acc = zzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -164,7 +164,7 @@ contains
!$omp parallel do private(i,j, acc)
do i=1,m
acc = zzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -176,7 +176,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = zzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -192,7 +192,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = zzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -204,7 +204,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = zzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -216,7 +216,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = zzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -231,7 +231,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = zzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -243,7 +243,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = zzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -255,7 +255,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = zzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -270,7 +270,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = zzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -282,7 +282,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = zzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
@@ -294,7 +294,7 @@ contains
!$omp parallel do private(i,j,acc)
do i=1,m
acc = zzero
!$omp simd
!$omp simd reduction(+:acc)
do j=irp(i), irp(i+1)-1
acc = acc + val(j) * x(ja(j))
enddo
cuda/Makefile
+1
View File
@@ -51,6 +51,7 @@ lib: objs ilib cudalib spgpulib
$(AR) $(LIBNAME) $(OBJS)
/bin/cp -p $(LIBNAME) $(LIBDIR)
$(COBJS): spgpuinc
objs: spgpuinc $(OBJS) iobjs cudaobjs spgpuobjs
/bin/cp -p *$(.mod) $(MODDIR)