Parse improvements #158
Open
Parse improvements #158
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
cbb36cb to
faa256c
Compare
faa256c to
0f5d4ca
Compare
0f5d4ca to
c83f74e
Compare
We now parse STEs on multiple cores, in 64k chunks. To keep output deterministic, we buffer the resulting output before reassembling it in order, but if the user values speed over determinism we expose the --no_ordered_output flag. This patch, together with writing output directly to the file, improves STE parse speed by 7.5x. The numbers below do not reflect that because the tool runtime also includes loading data from the disk, which is something we have not optimized. Measurements taken on a 4M STE dump, excluding FW dump time (so, with --skip_dump): Wall time Max RSS Before 3:24 12.0G This patch 1:00 8.3G This patch, no_ordered 0:57 8.5G Signed-off-by: Hamdan Igbaria <hamdani@nvidia.com> Signed-off-by: Vlad Dogaru <vdogaru@nvidia.com>
Instead of having one large string and printing it at once at the end of all data parsed, split this writes to be done on each level. This significantly reduces the memory footprint of the tool. This patch, together with STE parse parallelization, improves STE parse speed by 7.5x. The numbers below do not reflect that because the tool runtime also includes loading data from the disk, which is something we have not optimized. Measurements taken on a 4M STE dump, excluding FW dump time (so, with --skip_dump): Wall time Max RSS Before 0:57 8.5G Before, no_ordered 0:57 8.3G After 0:48 6.0G After, no_ordered 0:45 3.8G Signed-off-by: Hamdan Igbaria <hamdani@nvidia.com> Signed-off-by: Vlad Dogaru <vdogaru@nvidia.com>
c83f74e to
49cd1ae
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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.
No description provided.