Performance improvements in serialization#553
Merged
jmachowinski merged 2 commits intoros2:rollingfrom Feb 19, 2026
Merged
Conversation
The dynamic_cast here turned out to be quite costly. Replacing it by the variant like function, we see a ~10% cpu usage reduction in our use case. Signed-off-by: Janosch Machowinski <J.Machowinski@cellumation.com>
This gives another performance improvement of around 2% of cpu usage in our usage scenario. Signed-off-by: Janosch Machowinski <J.Machowinski@cellumation.com>
jmachowinski
force-pushed
the
rolling
branch
from
c56eee4 to
014c629
Compare
Contributor
Author
|
Pulls: #553
|
fujitatomoya
approved these changes
Feb 18, 2026
Contributor
fujitatomoya
left a comment
fujitatomoya
left a comment
There was a problem hiding this comment.
lgtm with green CI.
i would like to approval from @eboasson or someone else before merge.
biodranik
approved these changes
Feb 18, 2026
biodranik
left a comment
biodranik
left a comment
There was a problem hiding this comment.
Thanks, LGTM
The change eliminates dynamic_cast and virtual functions' overhead by using the CRTP pattern https://en.cppreference.com/w/cpp/language/crtp.html
eboasson
approved these changes
Feb 19, 2026
Collaborator
eboasson
left a comment
eboasson
left a comment
There was a problem hiding this comment.
LGTM (not that I am much of a C++ programmer)
Merging this is going to require some rework on #550. For that I only took the existing code and added a few variations and a few functions, so I am hopeful it is straightforward. The trouble is that I don't really know whether my hope is well-founded ("I am not much of a ..." &c.). Perhaps you could check @jmachowinski ?
Contributor
Author
Contributor
Author
|
@Mergifyio backport jazzy kilted |
✅ Backports have been createdDetails
|
mergify bot
pushed a commit
that referenced
this pull request
Feb 19, 2026
* perf: Fixed high CPU usage due to dynamic_cast The dynamic_cast here turned out to be quite costly. Replacing it by the variant like function, we see a ~10% cpu usage reduction in our use case. Signed-off-by: Janosch Machowinski <J.Machowinski@cellumation.com> * perf: Replaced virtual functions by templates This gives another performance improvement of around 2% of cpu usage in our usage scenario. Signed-off-by: Janosch Machowinski <J.Machowinski@cellumation.com> --------- Signed-off-by: Janosch Machowinski <J.Machowinski@cellumation.com> Co-authored-by: Janosch Machowinski <J.Machowinski@cellumation.com> (cherry picked from commit 5147e01)
mergify bot
pushed a commit
that referenced
this pull request
Feb 19, 2026
* perf: Fixed high CPU usage due to dynamic_cast The dynamic_cast here turned out to be quite costly. Replacing it by the variant like function, we see a ~10% cpu usage reduction in our use case. Signed-off-by: Janosch Machowinski <J.Machowinski@cellumation.com> * perf: Replaced virtual functions by templates This gives another performance improvement of around 2% of cpu usage in our usage scenario. Signed-off-by: Janosch Machowinski <J.Machowinski@cellumation.com> --------- Signed-off-by: Janosch Machowinski <J.Machowinski@cellumation.com> Co-authored-by: Janosch Machowinski <J.Machowinski@cellumation.com> (cherry picked from commit 5147e01)
jmachowinski
added a commit
that referenced
this pull request
Feb 20, 2026
* perf: Fixed high CPU usage due to dynamic_cast The dynamic_cast here turned out to be quite costly. Replacing it by the variant like function, we see a ~10% cpu usage reduction in our use case. * perf: Replaced virtual functions by templates This gives another performance improvement of around 2% of cpu usage in our usage scenario. --------- (cherry picked from commit 5147e01) Signed-off-by: Janosch Machowinski <J.Machowinski@cellumation.com> Co-authored-by: Janosch Machowinski <jmachowinski@users.noreply.github.com> Co-authored-by: Janosch Machowinski <J.Machowinski@cellumation.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Optimized functions that showed up in perf
Did you use Generative AI?
No