Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
T
Tensor-Core Correlator
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Iterations
Wiki
Requirements
Jira
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Locked files
Build
Pipelines
Jobs
Pipeline schedules
Test cases
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
Code review analytics
Issue analytics
Insights
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
ResearchAndDevelopment
Tensor-Core Correlator
Commits
2fdfa0b4
Commit
2fdfa0b4
authored
2 years ago
by
John Romein
Browse files
Options
Downloads
Patches
Plain Diff
Added support for Jetson AGX Orin.
parent
296cc7f5
No related branches found
No related tags found
No related merge requests found
Changes
1
Show whitespace changes
Inline
Side-by-side
Showing
1 changed file
libtcc/TCCorrelator.cu
+3
-3
3 additions, 3 deletions
libtcc/TCCorrelator.cu
with
3 additions
and
3 deletions
libtcc/TCCorrelator.cu
+
3
−
3
View file @
2fdfa0b4
...
@@ -22,7 +22,7 @@
...
@@ -22,7 +22,7 @@
#error this architecture has no suitable tensor cores
#error this architecture has no suitable tensor cores
#endif
#endif
#if __CUDA_ARCH__ != 700 && __CUDA_ARCH__ != 720 && __CUDA_ARCH__ != 750 && __CUDA_ARCH__ != 800 && __CUDA_ARCH__ != 860
#if __CUDA_ARCH__ != 700 && __CUDA_ARCH__ != 720 && __CUDA_ARCH__ != 750 && __CUDA_ARCH__ != 800 && __CUDA_ARCH__ != 860
&& __CUDA_ARCH__ != 870
#define PORTABLE // unknown architecture -> write visibilities in portable way (via shared memory)
#define PORTABLE // unknown architecture -> write visibilities in portable way (via shared memory)
#endif
#endif
...
@@ -317,7 +317,7 @@ __device__ inline void storeVisibilities(Visibilities visibilities, unsigned cha
...
@@ -317,7 +317,7 @@ __device__ inline void storeVisibilities(Visibilities visibilities, unsigned cha
unsigned
recvX
=
firstReceiverX
+
NR_RECEIVERS_PER_TCM_X
*
x
+
((
threadIdx
.
x
>>
2
)
&
2
);
unsigned
recvX
=
firstReceiverX
+
NR_RECEIVERS_PER_TCM_X
*
x
+
((
threadIdx
.
x
>>
2
)
&
2
);
unsigned
polY
=
threadIdx
.
x
&
1
;
unsigned
polY
=
threadIdx
.
x
&
1
;
unsigned
polX
=
(
threadIdx
.
x
>>
1
)
&
1
;
unsigned
polX
=
(
threadIdx
.
x
>>
1
)
&
1
;
#elif (__CUDA_ARCH__ == 720 && NR_BITS == 8) || __CUDA_ARCH__ == 750 || __CUDA_ARCH__ == 800 || __CUDA_ARCH__ == 860
#elif (__CUDA_ARCH__ == 720 && NR_BITS == 8) || __CUDA_ARCH__ == 750 || __CUDA_ARCH__ == 800 || __CUDA_ARCH__ == 860
|| __CUDA_ARCH__ == 870
unsigned
recvY
=
firstReceiverY
+
NR_RECEIVERS_PER_TCM_Y
*
y
+
((
threadIdx
.
x
>>
3
)
&
3
);
unsigned
recvY
=
firstReceiverY
+
NR_RECEIVERS_PER_TCM_Y
*
y
+
((
threadIdx
.
x
>>
3
)
&
3
);
unsigned
recvX
=
firstReceiverX
+
NR_RECEIVERS_PER_TCM_X
*
x
+
((
threadIdx
.
x
>>
1
)
&
1
);
unsigned
recvX
=
firstReceiverX
+
NR_RECEIVERS_PER_TCM_X
*
x
+
((
threadIdx
.
x
>>
1
)
&
1
);
unsigned
polY
=
(
threadIdx
.
x
>>
2
)
&
1
;
unsigned
polY
=
(
threadIdx
.
x
>>
2
)
&
1
;
...
@@ -331,7 +331,7 @@ __device__ inline void storeVisibilities(Visibilities visibilities, unsigned cha
...
@@ -331,7 +331,7 @@ __device__ inline void storeVisibilities(Visibilities visibilities, unsigned cha
storeVisibility
(
visibilities
,
channel
,
baseline
,
recvY
,
recvX
,
0
,
1
,
polY
,
polX
,
skipCheckY
,
skipCheckX
,
sum
.
x
[
4
],
sum
.
x
[
5
]);
storeVisibility
(
visibilities
,
channel
,
baseline
,
recvY
,
recvX
,
0
,
1
,
polY
,
polX
,
skipCheckY
,
skipCheckX
,
sum
.
x
[
4
],
sum
.
x
[
5
]);
storeVisibility
(
visibilities
,
channel
,
baseline
,
recvY
,
recvX
,
1
,
0
,
polY
,
polX
,
skipCheckY
,
skipCheckX
,
sum
.
x
[
2
],
sum
.
x
[
3
]);
storeVisibility
(
visibilities
,
channel
,
baseline
,
recvY
,
recvX
,
1
,
0
,
polY
,
polX
,
skipCheckY
,
skipCheckX
,
sum
.
x
[
2
],
sum
.
x
[
3
]);
storeVisibility
(
visibilities
,
channel
,
baseline
,
recvY
,
recvX
,
1
,
1
,
polY
,
polX
,
skipCheckY
,
skipCheckX
,
sum
.
x
[
6
],
sum
.
x
[
7
]);
storeVisibility
(
visibilities
,
channel
,
baseline
,
recvY
,
recvX
,
1
,
1
,
polY
,
polX
,
skipCheckY
,
skipCheckX
,
sum
.
x
[
6
],
sum
.
x
[
7
]);
#elif (__CUDA_ARCH__ == 720 && NR_BITS == 8) || __CUDA_ARCH__ == 750 || __CUDA_ARCH__ == 800 || __CUDA_ARCH__ == 860
#elif (__CUDA_ARCH__ == 720 && NR_BITS == 8) || __CUDA_ARCH__ == 750 || __CUDA_ARCH__ == 800 || __CUDA_ARCH__ == 860
|| __CUDA_ARCH__ == 870
storeVisibility
(
visibilities
,
channel
,
baseline
,
recvY
,
recvX
,
0
,
0
,
polY
,
polX
,
skipCheckY
,
skipCheckX
,
sum
.
x
[
0
],
sum
.
x
[
1
]);
storeVisibility
(
visibilities
,
channel
,
baseline
,
recvY
,
recvX
,
0
,
0
,
polY
,
polX
,
skipCheckY
,
skipCheckX
,
sum
.
x
[
0
],
sum
.
x
[
1
]);
#if NR_BITS == 8 || NR_BITS == 16
#if NR_BITS == 8 || NR_BITS == 16
storeVisibility
(
visibilities
,
channel
,
baseline
,
recvY
,
recvX
,
0
,
2
,
polY
,
polX
,
skipCheckY
,
skipCheckX
,
sum
.
x
[
4
],
sum
.
x
[
5
]);
storeVisibility
(
visibilities
,
channel
,
baseline
,
recvY
,
recvX
,
0
,
2
,
polY
,
polX
,
skipCheckY
,
skipCheckX
,
sum
.
x
[
4
],
sum
.
x
[
5
]);
...
...
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