Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
P
psrdada_cpp
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Custom issue tracker
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Build
Pipelines
Jobs
Pipeline schedules
Artifacts
Deploy
Releases
Model registry
Operate
Environments
Monitor
Incidents
Service Desk
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
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
MPIfR-BDG
psrdada_cpp
Commits
db6e597f
Commit
db6e597f
authored
6 years ago
by
root
Browse files
Options
Downloads
Patches
Plain Diff
Fixed bug where transpose only worked on square arrays
parent
de359cec
Branches
Branches containing commit
Tags
Tags containing commit
No related merge requests found
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
psrdada_cpp/effelsberg/edd/src/ScaledTransposeTFtoTFT.cu
+4
-4
4 additions, 4 deletions
psrdada_cpp/effelsberg/edd/src/ScaledTransposeTFtoTFT.cu
psrdada_cpp/effelsberg/edd/test/src/ScaledTransposeTFtoTFTTester.cu
+1
-1
1 addition, 1 deletion
...p/effelsberg/edd/test/src/ScaledTransposeTFtoTFTTester.cu
with
5 additions
and
5 deletions
psrdada_cpp/effelsberg/edd/src/ScaledTransposeTFtoTFT.cu
+
4
−
4
View file @
db6e597f
...
...
@@ -37,7 +37,7 @@ void tf_to_tft_transpose(
char2
store_val
;
store_val
.
x
=
(
char
)((
val
.
x
-
offset
)
/
scale
);
store_val
.
y
=
(
char
)((
val
.
y
-
offset
)
/
scale
);
temp
[
samp_load_idx
*
n
samps_per_load
+
chan_load_idx
]
=
store_val
;
temp
[
samp_load_idx
*
n
chans
+
chan_load_idx
]
=
store_val
;
}
}
__syncthreads
();
...
...
@@ -47,11 +47,11 @@ void tf_to_tft_transpose(
{
for
(
int
samp_store_idx
=
threadIdx
.
x
;
samp_store_idx
<
nsamps_per_load
;
samp_store_idx
+=
block
Idx
.
x
)
samp_store_idx
+=
block
Dim
.
x
)
{
int
store_idx
=
(
load_offset
+
chan_store_idx
*
nsamps_per_packet
+
nsamps_per_load
*
sub_samp_load_idx
+
samp_store_idx
);
output
[
store_idx
]
=
temp
[
samp_store_idx
*
n
samps_per_load
+
chan_store_idx
];
output
[
store_idx
]
=
temp
[
samp_store_idx
*
n
chans
+
chan_store_idx
];
}
}
__syncthreads
();
...
...
@@ -86,7 +86,7 @@ void ScaledTransposeTFtoTFT::transpose(InputType const& input, OutputType& outpu
//assert sizes
assert
(
input
.
size
()
%
(
_nchans
*
_nsamps_per_packet
)
==
0
/* Input is not a multiple of _nchans * _nsamps_per_packet*/
);
output
.
resize
(
input
.
size
());
const
int
nsamps_per_load
=
1
6
;
const
int
nsamps_per_load
=
1
28
;
assert
((
_nsamps_per_packet
%
nsamps_per_load
)
==
0
);
const
int
nsamps
=
input
.
size
()
/
_nchans
;
int
shared_mem_bytes
=
sizeof
(
OutputType
::
value_type
)
*
_nchans
*
nsamps_per_load
;
...
...
This diff is collapsed.
Click to expand it.
psrdada_cpp/effelsberg/edd/test/src/ScaledTransposeTFtoTFTTester.cu
+
1
−
1
View file @
db6e597f
...
...
@@ -76,7 +76,7 @@ void ScaledTransposeTFtoTFTTester::compare_against_host(
TEST_F
(
ScaledTransposeTFtoTFTTester
,
counter_test
)
{
int
nchans
=
16
;
int
nsamps_per_packet
=
8192
/
nchans
;
int
nsamps_per_packet
=
8192
;
float
stdev
=
64.0
f
;
float
scale
=
4.0
f
;
int
nsamps
=
nsamps_per_packet
*
1024
;
...
...
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