-
Notifications
You must be signed in to change notification settings - Fork 4.8k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[LRS-GL] Support for Align-SSE #12637
[LRS-GL] Support for Align-SSE #12637
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Please fix CI issues
4e0cc66
to
1b04502
Compare
1b04502
to
7bd1b7d
Compare
src/proc/align.cpp
Outdated
@@ -11,10 +11,30 @@ | |||
#include "align.h" | |||
#include "stream.h" | |||
|
|||
#ifdef RS2_USE_CUDA |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is it possible a platform will have both CUDA & SSE3 enabled?
Please check because if yes, we will have to choose as this may not compile
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
sure @Nir-Az. Will check. But it is the same case for pointcloud implementation as well.
librealsense/src/proc/pointcloud.cpp
Line 394 in d06f21b
std::shared_ptr<pointcloud> pointcloud::create() |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@Nir-Az, the user cannot choose to run with or without "SSE3". If CPU processing is preferred, LRS should automatically detect and use "SSE3" wherever possible.
If "BUILD_WITH_CUDA" flag is enabled, CUDA implementations will be preferred. User cannot dynamically choose between CUDA or SSE3.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes I know, just wanted to make sure that if the user build with Cuda, and gis CPU support SSSE,
He will include both headers.
We need to make sure the headers does not conflict.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Headers are not conflicting. Anyway, updated them to avoid such confusions.
Tracked on LRS-1007
With LRS-GL library, in case if user chooses CPU acceleration ('SSE3' is enabled during build), create librealsense::align_sse class object instead of librealsense::align.