Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
H
HPC_Imaging
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Build
Pipelines
Jobs
Pipeline schedules
Artifacts
Deploy
Releases
Package registry
Container registry
Model registry
Operate
Environments
Terraform modules
Monitor
Incidents
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
GitLab community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
Claudio Gheller
HPC_Imaging
Commits
f8d864ea
Commit
f8d864ea
authored
May 15, 2024
by
Claudio Gheller
Browse files
Options
Downloads
Patches
Plain Diff
added some comments
parent
4601caf6
No related branches found
No related tags found
No related merge requests found
Changes
6
Show whitespace changes
Inline
Side-by-side
Showing
6 changed files
Makefile
+3
-3
3 additions, 3 deletions
Makefile
cd
+0
-0
0 additions, 0 deletions
cd
gridding_cpu.c
+8
-7
8 additions, 7 deletions
gridding_cpu.c
gridding_nccl.cu
+6
-5
6 additions, 5 deletions
gridding_nccl.cu
main.c
+7
-0
7 additions, 0 deletions
main.c
timing.h
+0
-1
0 additions, 1 deletion
timing.h
with
24 additions
and
16 deletions
Makefile
+
3
−
3
View file @
f8d864ea
...
...
@@ -42,10 +42,10 @@ FFTWLIBS =
OPT
+=
-DUSE_FFTW
# use omp-ized version of fftw routines
#
OPT += -DHYBRID_FFTW
OPT
+=
-DHYBRID_FFTW
# switch on the OpenMP parallelization
#
OPT += -DUSE_OMP
OPT
+=
-DUSE_OMP
# ========================================================
...
...
@@ -92,7 +92,7 @@ OPT += -DGAUSS_HI_PRECISION
#OPT += -DCUFFTMP
# FULL NVIDIA GPU SUPPORT - Recommended for full NVIDIA GPU code execution
OPT
+=
-DFULL_NVIDIA
#
OPT += -DFULL_NVIDIA
ifeq
(FULL_NVIDIA,$(findstring FULL_NVIDIA,$(OPT)))
OPT
+=
-DCUDACC
-DNCCL_REDUCE
-DCUFFTMP
endif
...
...
This diff is collapsed.
Click to expand it.
cd
deleted
100644 → 0
+
0
−
0
View file @
4601caf6
This diff is collapsed.
Click to expand it.
gridding_cpu.c
+
8
−
7
View file @
f8d864ea
...
...
@@ -67,12 +67,12 @@ void gridding_data()
}
// closes reduce_method == REDUCE_RING
timing_wt
.
kernel
=
0
.
0
;
timing_wt
.
reduce
=
0
.
0
;
timing_wt
.
reduce_mpi
=
0
.
0
;
timing_wt
.
reduce_sh
=
0
.
0
;
timing_wt
.
compose
=
0
.
0
;
//CLAAAA
//
timing_wt.kernel = 0.0;
//
timing_wt.reduce = 0.0;
//
timing_wt.reduce_mpi = 0.0;
//
timing_wt.reduce_sh = 0.0;
//
timing_wt.compose = 0.0;
// calculate the resolution in radians
resolution
=
1
.
0
/
MAX
(
fabs
(
metaData
.
uvmin
),
fabs
(
metaData
.
uvmax
));
...
...
@@ -142,6 +142,7 @@ void gridding_data()
double
uumax
=
-
1e20
;
double
vvmax
=
-
1e20
;
/*
#pragma omp parallel reduction( min: uumin, vvmin) reduction( max: uumax, vvmax) num_threads(param.num_threads)
{
double my_uumin = 1e20;
...
...
@@ -165,7 +166,7 @@ void gridding_data()
}
//printf("UU, VV, min, max = %f %f %f %f\n", uumin, uumax, vvmin, vvmax);
*/
timing_wt
.
compose
+=
CPU_TIME_wt
-
start
;
...
...
This diff is collapsed.
Click to expand it.
gridding_nccl.cu
+
6
−
5
View file @
f8d864ea
...
...
@@ -52,11 +52,12 @@ void gridding_data(){
double
shift
=
(
double
)(
dx
*
yaxis
);
timing_wt
.
kernel
=
0.0
;
timing_wt
.
reduce
=
0.0
;
timing_wt
.
reduce_mpi
=
0.0
;
timing_wt
.
reduce_sh
=
0.0
;
timing_wt
.
compose
=
0.0
;
// CLAAAA
//timing_wt.kernel = 0.0;
//timing_wt.reduce = 0.0;
//timing_wt.reduce_mpi = 0.0;
//timing_wt.reduce_sh = 0.0;
//timing_wt.compose = 0.0;
// calculate the resolution in radians
resolution
=
1.0
/
MAX
(
fabs
(
metaData
.
uvmin
),
fabs
(
metaData
.
uvmax
));
...
...
This diff is collapsed.
Click to expand it.
main.c
+
7
−
0
View file @
f8d864ea
...
...
@@ -33,6 +33,13 @@ int main(int argc, char * argv[])
{
//CLAAAA
timing_wt
.
kernel
=
0
.
0
;
timing_wt
.
reduce
=
0
.
0
;
timing_wt
.
reduce_mpi
=
0
.
0
;
timing_wt
.
reduce_sh
=
0
.
0
;
timing_wt
.
compose
=
0
.
0
;
if
(
argc
>
1
)
{
strcpy
(
in
.
paramfile
,
argv
[
1
]);
...
...
This diff is collapsed.
Click to expand it.
timing.h
+
0
−
1
View file @
f8d864ea
...
...
@@ -77,7 +77,6 @@ typedef struct {
extern
timing_t
timing_wt
;
// wall-clock process timing, at Task 0
extern
double
start_tot
;
extern
double
reduce_shmem_time
;
extern
double
reduce_mpi_time
;
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment