Change how cache manager handles child process trace cache for rocpd#1033
Merged
dgaliffiAMD merged 36 commits intodevelopfrom Oct 24, 2025
Merged
Change how cache manager handles child process trace cache for rocpd#1033dgaliffiAMD merged 36 commits intodevelopfrom
dgaliffiAMD merged 36 commits intodevelopfrom
Conversation
mradosav-amd
force-pushed
the
users/mradosav-amd/add-vllm-v1-support
branch
from
691f804 to
23d87d8
Compare
mradosav-amd
force-pushed
the
users/mradosav-amd/add-vllm-v1-support
branch
2 times, most recently
from
be37764 to
c438aaf
Compare
github-actions
bot
added
the
github actions
mradosav-amd
force-pushed
the
users/mradosav-amd/add-vllm-v1-support
branch
6 times, most recently
from
6f4b2bb to
f1b6e47
Compare
mradosav-amd
force-pushed
the
users/mradosav-amd/add-vllm-v1-support
branch
2 times, most recently
from
8b435a7 to
6e28312
Compare
dgaliffiAMD
requested changes
Sep 30, 2025
Contributor
dgaliffiAMD
left a comment
dgaliffiAMD
left a comment
There was a problem hiding this comment.
Thanks @mradosav-amd, I'm still reviewing but here are some comments on my first pass.
projects/rocprofiler-systems/source/lib/rocprof-sys/library.cpp
Outdated
Show resolved
Hide resolved
dgaliffiAMD
reviewed
Sep 30, 2025
mradosav-amd
force-pushed
the
users/mradosav-amd/add-vllm-v1-support
branch
2 times, most recently
from
a063b8e to
7ae9221
Compare
mradosav-amd
force-pushed
the
users/mradosav-amd/add-vllm-v1-support
branch
4 times, most recently
from
a69b671 to
b67b0ee
Compare
mradosav-amd
removed
the
github actions
Co-authored-by: Aleksandar Djordjevic <[email protected]>
This reverts commit ddabf6d.
This reverts commit 973da3a1ba99d99d529af5269d30e177092f9bfa.
mradosav-amd
force-pushed
the
users/mradosav-amd/add-vllm-v1-support
branch
4 times, most recently
from
52c0ac8 to
6e96292
Compare
mradosav-amd
force-pushed
the
users/mradosav-amd/add-vllm-v1-support
branch
3 times, most recently
from
2a58ec7 to
2590eef
Compare
mradosav-amd
force-pushed
the
users/mradosav-amd/add-vllm-v1-support
branch
from
2590eef to
1d046b3
Compare
dgaliffiAMD
force-pushed
the
users/mradosav-amd/add-vllm-v1-support
branch
from
9300c69 to
c7110bc
Compare
dgaliffiAMD
force-pushed
the
users/mradosav-amd/add-vllm-v1-support
branch
from
72c2865 to
74411a9
Compare
systems-assistant bot
pushed a commit
to ROCm/rocprofiler-systems
that referenced
this pull request
Oct 24, 2025
for rocpd (#1033) * Change how cache manager handles child process trace cache * Sampling and backtrace metrics to cache * Apply cmake formatting * Fix parsing of metadata json * Code clean up * Fix build nlohmann json from source * Fix storage parsed finished callback * Revert sampling for child process * Change cache file name generating * Fix thread start stop * Fix process start end timestamp * Applied suggestions from code review * Try with late start of flushing task thread * Change dockerfiles for ci * Revert changes on github workflows * Remove json_fwd.hpp include * fix dump * Build nlohmann/json by default Signed-off-by: David Galiffi <[email protected]> * Update location of build artifacts for nlohmann/json Signed-off-by: David Galiffi <[email protected]> * Revert use_output_suffix * Remove unused logs * Fix cache store inside counter due to structure change * Remove decode tests from debian ci * Fix issue where all databases have the same UUID (#1499) Co-authored-by: Aleksandar Djordjevic <[email protected]> * Removing the cpack and install steps to save space * Revert "Remove decode tests from debian ci" This reverts commit ddabf6dd142dcf438e6b8997b8abe86f2c868468. * Revert "Removing the cpack and install steps to save space" This reverts commit 973da3a1ba99d99d529af5269d30e177092f9bfa. * Add prepare-runner job as dependency to clean up the space * Fix formatting * Free up even more space * Remove verbose for workflows * remove hw_counters from ext_data * move space clean up inside container * try to remove external folder to free up space * Check space * Refactor Cleanup to it's own step [rocm-systems] ROCm/rocm-systems#1033 (commit 8806be1)
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.
Motivation
This PR has goal to add support to rocprofiler-systems to be able to handle AI workloads for vLLM v1.
vLLM v1 has different approach to process handling than vLLM v0, this PR will cover both approaches.
Resolves Tickets:
Technical Details
Child processes will create trace cache file and metadata file in
tmpdirectory. Root process will gather all files under it, and create rocpd database.Test Plan
Profile AI workload with vLLM v0 and v1 engine.
Test Result
rocprofiler-systems should generate rocpd databases for each process, both for v0 and v1 engine.
Submission Checklist