add benchmark_from_scratch flag #1881
Open
Conversation
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
ginkgo-bot
added
reg:benchmarking
pratikvn
requested changes
Jul 22, 2025
benchmark/solver/solver_common.hpp
Outdated
| FLAGS_max_iters) | ||
| ->generate(state.system_matrix)); | ||
| generate_timer->toc(); | ||
| // if we benchmark from scrach, we use the new generated solver. |
There was a problem hiding this comment.
nit:
Suggested change
| // if we benchmark from scrach, we use the new generated solver. | |
| // if we benchmark from scratch, we use the new generated solver. |
| generate_timer->toc(); | ||
| // if we benchmark from scrach, we use the new generated solver. | ||
| if (FLAGS_benchmark_from_scratch) { | ||
| solver = generated_solver; |
There was a problem hiding this comment.
If benchmark from scratch is off, then the we use the generated solver from the warmup (line 477), right ? I think then we dont use FLAGS_max_iters and use FLAGS_warmup_max_iters instead.
There was a problem hiding this comment.
yes. we still use max_iters for apply but the solver is used once from warmup.
I just realize this will be broken when no warmup iteration
There was a problem hiding this comment.
the issue when no warmup iteration is fixed now
Co-authored-by: Pratik Nayak <pratikvn@protonmail.com>
yhmtsai
force-pushed
the
check_solver_without_reinit
branch
from
7ab9521 to
156a081
Compare
yhmtsai
commented
Jul 30, 2025
| @@ -509,19 +515,27 @@ struct SolverBenchmark : Benchmark<solver_benchmark_state<Generator>> { | |||
| exec->get_master()->add_logger(gen_logger); | |||
| } | |||
|
|
|||
| auto precond = precond_factory.at(precond_name)(exec); | |||
| solver = generate_solver(exec, give(precond), solver_name, | |||
There was a problem hiding this comment.
free entry in the previous output is deleting the workspace from previous warmup apply not the generation. It introduces the changes for test output
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
This PR allows user to select whether to benchmark the solver from scratch or not.
The previous behavior is to benchmark the solver from scratch every time, which will benchmark the workspace initialization every time.
The proposed behavior is to benchamark the same solver object such that we only need to initialize the workspace in the first run (or warmup).
User can set it
--benchmark_from_scratch=trueagain to benchmark the solver with initialization.