fix(profiling): remove slow getpid call from memalloc path #11848
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
|
|
sanchda
approved these changes
Jan 2, 2025
Datadog ReportBranch report: ✅ 0 Failed, 289 Passed, 1179 Skipped, 8m 2.93s Total duration (28m 21.92s time saved) |
BenchmarksBenchmark execution time: 2025-01-02 19:16:11 Comparing candidate commit a69241a in PR branch Found 1 performance improvements and 0 performance regressions! Performance is the same for 393 metrics, 2 unstable metrics. scenario:flasksimple-profiler
|
memalloc uses getpid to detect whether the process has forked, so that we can unlock the memalloc lock in the child process (if it isn't already locked). Unfortunately the getpid call is quite slow. From the man page: "calls to getpid() always invoke the actual system call, rather than returning a cached value." Furthermore, we _always_ attempt to take the lock for allocations, even if we aren't going to sample them. So this is basically adding a syscall to every allocation. Move this logic out of the allocation path. Switch to using pthread_atfork handlers to ensure that the lock is held prior to forking, and unlock it in the parent and child after forking. This (maybe) has the added benefit of making sure the data structures are in a consistent state in the child process after forking. Unclear if that's an issue prior to this change, though. I may be missing some code that resets the profiler on fork anyway?
nsrip-dd
force-pushed
the
nick.ripley/remove-memalloc-getpid
branch
from
3277223 to
a69241a
Compare
nsrip-dd
changed the title
|
The backport to To backport manually, run these commands in your terminal: # Fetch latest updates from GitHub
git fetch
# Create a new working tree
git worktree add .worktrees/backport-2.17 2.17
# Navigate to the new working tree
cd .worktrees/backport-2.17
# Create a new branch
git switch --create backport-11848-to-2.17
# Cherry-pick the merged commit of this pull request and resolve the conflicts
git cherry-pick -x --mainline 1 6bfe77ede64278fadbd64131fa14ad123417c7ec
# Push it to GitHub
git push --set-upstream origin backport-11848-to-2.17
# Go back to the original working tree
cd ../..
# Delete the working tree
git worktree remove .worktrees/backport-2.17Then, create a pull request where the |
|
The backport to To backport manually, run these commands in your terminal: # Fetch latest updates from GitHub
git fetch
# Create a new working tree
git worktree add .worktrees/backport-2.19 2.19
# Navigate to the new working tree
cd .worktrees/backport-2.19
# Create a new branch
git switch --create backport-11848-to-2.19
# Cherry-pick the merged commit of this pull request and resolve the conflicts
git cherry-pick -x --mainline 1 6bfe77ede64278fadbd64131fa14ad123417c7ec
# Push it to GitHub
git push --set-upstream origin backport-11848-to-2.19
# Go back to the original working tree
cd ../..
# Delete the working tree
git worktree remove .worktrees/backport-2.19Then, create a pull request where the |
github-actions bot
pushed a commit
that referenced
this pull request
Jan 2, 2025
memalloc uses getpid to detect whether the process has forked, so that we can unlock the memalloc lock in the child process (if it isn't already locked). Unfortunately the getpid call is quite slow. From the man page: "calls to getpid() always invoke the actual system call, rather than returning a cached value." Furthermore, we _always_ attempt to take the lock for allocations, even if we aren't going to sample them. So this is basically adding a syscall to every allocation. Move this logic out of the allocation path. Switch to using pthread_atfork handlers to ensure that the lock is held prior to forking, and unlock it in the parent and child after forking. This (maybe) has the added benefit of making sure the data structures are in a consistent state in the child process after forking. Unclear if that's an issue prior to this change, though. I may be missing some code that resets the profiler on fork anyway? (cherry picked from commit 6bfe77e)
2 tasks
nsrip-dd
pushed a commit
that referenced
this pull request
Jan 8, 2025
…2.18] (#11849) Backport 6bfe77e from #11848 to 2.18. memalloc uses getpid to detect whether the process has forked, so that we can unlock the memalloc lock in the child process (if it isn't already locked). Unfortunately the getpid call is quite slow. From the man page: "calls to getpid() always invoke the actual system call, rather than returning a cached value." Furthermore, we _always_ attempt to take the lock for allocations, even if we aren't going to sample them. So this is basically adding a syscall to every allocation. Move this logic out of the allocation path. Switch to using pthread_atfork handlers to ensure that the lock is held prior to forking, and unlock it in the parent and child after forking. This (maybe) has the added benefit of making sure the data structures are in a consistent state in the child process after forking. Unclear if that's an issue prior to this change, though. I may be missing some code that resets the profiler on fork anyway?
taegyunkim
added
[DEPRECATED] backport 2.19
and removed
[DEPRECATED] backport 2.19
labels
github-actions bot
pushed a commit
that referenced
this pull request
Jan 15, 2025
memalloc uses getpid to detect whether the process has forked, so that we can unlock the memalloc lock in the child process (if it isn't already locked). Unfortunately the getpid call is quite slow. From the man page: "calls to getpid() always invoke the actual system call, rather than returning a cached value." Furthermore, we _always_ attempt to take the lock for allocations, even if we aren't going to sample them. So this is basically adding a syscall to every allocation. Move this logic out of the allocation path. Switch to using pthread_atfork handlers to ensure that the lock is held prior to forking, and unlock it in the parent and child after forking. This (maybe) has the added benefit of making sure the data structures are in a consistent state in the child process after forking. Unclear if that's an issue prior to this change, though. I may be missing some code that resets the profiler on fork anyway? (cherry picked from commit 6bfe77e)
2 tasks
taegyunkim
pushed a commit
that referenced
this pull request
Jan 15, 2025
…2.19] (#11964) Backport 6bfe77e from #11848 to 2.19. memalloc uses getpid to detect whether the process has forked, so that we can unlock the memalloc lock in the child process (if it isn't already locked). Unfortunately the getpid call is quite slow. From the man page: "calls to getpid() always invoke the actual system call, rather than returning a cached value." Furthermore, we _always_ attempt to take the lock for allocations, even if we aren't going to sample them. So this is basically adding a syscall to every allocation. Move this logic out of the allocation path. Switch to using pthread_atfork handlers to ensure that the lock is held prior to forking, and unlock it in the parent and child after forking. This (maybe) has the added benefit of making sure the data structures are in a consistent state in the child process after forking. Unclear if that's an issue prior to this change, though. I may be missing some code that resets the profiler on fork anyway? ## Checklist - [x] PR author has checked that all the criteria below are met - The PR description includes an overview of the change - The PR description articulates the motivation for the change - The change includes tests OR the PR description describes a testing strategy - The PR description notes risks associated with the change, if any - Newly-added code is easy to change - The change follows the [library release note guidelines](https://ddtrace.readthedocs.io/en/stable/releasenotes.html) - The change includes or references documentation updates if necessary - Backport labels are set (if [applicable](https://ddtrace.readthedocs.io/en/latest/contributing.html#backporting)) ## Reviewer Checklist - [x] Reviewer has checked that all the criteria below are met - Title is accurate - All changes are related to the pull request's stated goal - Avoids breaking [API](https://ddtrace.readthedocs.io/en/stable/versioning.html#interfaces) changes - Testing strategy adequately addresses listed risks - Newly-added code is easy to change - Release note makes sense to a user of the library - If necessary, author has acknowledged and discussed the performance implications of this PR as reported in the benchmarks PR comment - Backport labels are set in a manner that is consistent with the [release branch maintenance policy](https://ddtrace.readthedocs.io/en/latest/contributing.html#backporting) Co-authored-by: Nick Ripley <[email protected]>
nsrip-dd
added a commit
that referenced
this pull request
May 14, 2025
We added locking to memalloc, the memory profiler, in #11460 in order to address crashes. These locks made the crashes go away, but significantly increased the baseline overhead of the profiler and introduced subtle bugs. The locks we added turned out to be fundamentally incompatible with the global interpreter lock (GIL), at least with the implementation from #11460. This PR refactors the profiler to use the GIL exclusively for locking. First, we should acknowledge no-GIL and subinterpreters. As of right now, our module does not support either. A module has to explicitly opt-in to support either, so there is no risk of those modes being enabled under our feet. Supporting either mode is likely a repo-wide project. For now, we can assume the GIL exists. This work was motivated by overhead. We currently acquire and release locks in every memory allocation and free. Even when the locks aren't contended, allocations and frees are very frequent, and the extra works adds up. We add about ~8x overhead to the baselien cost of allocation just with our locking, not including the cost of actually sampling an allocation. We can't get rid of this overhead just by reducing sampling frequency. There are a few rules to follow in order to use the GIL correctly for locking: 1) The GIL is held when a C extension function is called, _except_ possibly in the raw allocator, which we do not profile 2) The GIL may be released during C Python API calls. Even if it is released, though, it will be held again after the call 3) Thus, the GIL creates critical sections only between C Python API calls, and the beginning and end of C extension functions. Modifications to shared state across those points are not atomic. 4) If we take a lock of our own in a C extension code (i.e. a pthread_mutex), and the extension code releases the GIL, then the program will deadlock due to lock order inversion. We can only safely take locks in C extension when the GIL is released. The crashes that #11460 addresed were due to breaking the first three rules. In particular, we could race on accessing the shared scratch buffer used when collecting tracebacks, which lead to double-frees. See #13185 for more details. Our mitigation involved using C locks around any access to the shared profiler state. We nearly broke rule 4 in the process. However, we used try-locks specifically out of a fear of introducing deadlocks. Try-locks mean that we attempt to acquire the lock, but return a failure if the lock is already held. This stopped deadlocks, but introduced bugs: For example: - If we failed to take the lock when trying to report allocation profile events, we'd raise an exception when it was in fact not reasonable for doing that to fail. See #12075. - memalloc_heap_untrack, which removes tracked allocations, was guarded with a try-lock. If we couldn't acquire the lock, we would fail to remove a record for an allocation and effectively leak memory. See #13317 - We attempted to make our locking fork-safe. The first attempt was inefficient; we made it less inefficient but the fix only "worked" because of try-locks. See #11848 Try-locks hide concurrency problems and we shouldn't use them. Using our own locks requires releasing the GIL before acquisition, and then re-acquiring the GIL. That adds unnecessary overhead. We don't inherently need to do any off-GIL work. So, we should try to just use the GIL as long as it is available. The basic refactor is actually pretty simple. In a nutshell, we rearrange the memalloc_add_event and memalloc_heap_track functions so that they make the sampling decision, then take a traceback, then insert the traceback into the appropriate data structure. Collecting a traceback can release the GIL, so we make sure that modifying the data structure happens completely after the traceback is collected. We also safeguard against the possibility that the profiler was stopped during sampling, if the GIL was released. This requires a small rearrangement of memalloc_stop to make sure that the sampling functions don't see partially-freed profiler data structures. For testing, I have mainly used the code from test_memealloc_data_race_regression. I also added a debug mode, enabled by compiling with MEMALLOC_TESTING_GIL_RELEASE, which releases the GIL at places where it would be expected. For performance I examined the overhead of profiling on a basic flask application.
2 tasks
nsrip-dd
added a commit
that referenced
this pull request
May 15, 2025
We added locking to memalloc, the memory profiler, in #11460 in order to address crashes. These locks made the crashes go away, but significantly increased the baseline overhead of the profiler and introduced subtle bugs. The locks we added turned out to be fundamentally incompatible with the global interpreter lock (GIL), at least with the implementation from #11460. This PR refactors the profiler to use the GIL exclusively for locking. First, we should acknowledge no-GIL and subinterpreters. As of right now, our module does not support either. A module has to explicitly opt-in to support either, so there is no risk of those modes being enabled under our feet. Supporting either mode is likely a repo-wide project. For now, we can assume the GIL exists. This work was motivated by overhead. We currently acquire and release locks in every memory allocation and free. Even when the locks aren't contended, allocations and frees are very frequent, and the extra works adds up. We add about ~8x overhead to the baselien cost of allocation just with our locking, not including the cost of actually sampling an allocation. We can't get rid of this overhead just by reducing sampling frequency. There are a few rules to follow in order to use the GIL correctly for locking: 1) The GIL is held when a C extension function is called, _except_ possibly in the raw allocator, which we do not profile 2) The GIL may be released during C Python API calls. Even if it is released, though, it will be held again after the call 3) Thus, the GIL creates critical sections only between C Python API calls, and the beginning and end of C extension functions. Modifications to shared state across those points are not atomic. 4) If we take a lock of our own in a C extension code (i.e. a pthread_mutex), and the extension code releases the GIL, then the program will deadlock due to lock order inversion. We can only safely take locks in C extension when the GIL is released. The crashes that #11460 addresed were due to breaking the first three rules. In particular, we could race on accessing the shared scratch buffer used when collecting tracebacks, which lead to double-frees. See #13185 for more details. Our mitigation involved using C locks around any access to the shared profiler state. We nearly broke rule 4 in the process. However, we used try-locks specifically out of a fear of introducing deadlocks. Try-locks mean that we attempt to acquire the lock, but return a failure if the lock is already held. This stopped deadlocks, but introduced bugs: For example: - If we failed to take the lock when trying to report allocation profile events, we'd raise an exception when it was in fact not reasonable for doing that to fail. See #12075. - memalloc_heap_untrack, which removes tracked allocations, was guarded with a try-lock. If we couldn't acquire the lock, we would fail to remove a record for an allocation and effectively leak memory. See #13317 - We attempted to make our locking fork-safe. The first attempt was inefficient; we made it less inefficient but the fix only "worked" because of try-locks. See #11848 Try-locks hide concurrency problems and we shouldn't use them. Using our own locks requires releasing the GIL before acquisition, and then re-acquiring the GIL. That adds unnecessary overhead. We don't inherently need to do any off-GIL work. So, we should try to just use the GIL as long as it is available. The basic refactor is actually pretty simple. In a nutshell, we rearrange the memalloc_add_event and memalloc_heap_track functions so that they make the sampling decision, then take a traceback, then insert the traceback into the appropriate data structure. Collecting a traceback can release the GIL, so we make sure that modifying the data structure happens completely after the traceback is collected. We also safeguard against the possibility that the profiler was stopped during sampling, if the GIL was released. This requires a small rearrangement of memalloc_stop to make sure that the sampling functions don't see partially-freed profiler data structures. For testing, I have mainly used the code from test_memealloc_data_race_regression. I also added a debug mode, enabled by compiling with MEMALLOC_TESTING_GIL_RELEASE, which releases the GIL at places where it would be expected. For performance I examined the overhead of profiling on a basic flask application.
nsrip-dd
added a commit
that referenced
this pull request
May 20, 2025
We added locking to memalloc, the memory profiler, in #11460 in order to address crashes. These locks made the crashes go away, but significantly increased the baseline overhead of the profiler and introduced subtle bugs. The locks we added turned out to be fundamentally incompatible with the global interpreter lock (GIL), at least with the implementation from #11460. This PR refactors the profiler to use the GIL exclusively for locking. First, we should acknowledge no-GIL and subinterpreters. As of right now, our module does not support either. A module has to explicitly opt-in to support either, so there is no risk of those modes being enabled under our feet. Supporting either mode is likely a repo-wide project. For now, we can assume the GIL exists. This work was motivated by overhead. We currently acquire and release locks in every memory allocation and free. Even when the locks aren't contended, allocations and frees are very frequent, and the extra works adds up. We add about ~8x overhead to the baselien cost of allocation just with our locking, not including the cost of actually sampling an allocation. We can't get rid of this overhead just by reducing sampling frequency. There are a few rules to follow in order to use the GIL correctly for locking: 1) The GIL is held when a C extension function is called, _except_ possibly in the raw allocator, which we do not profile 2) The GIL may be released during C Python API calls. Even if it is released, though, it will be held again after the call 3) Thus, the GIL creates critical sections only between C Python API calls, and the beginning and end of C extension functions. Modifications to shared state across those points are not atomic. 4) If we take a lock of our own in a C extension code (i.e. a pthread_mutex), and the extension code releases the GIL, then the program will deadlock due to lock order inversion. We can only safely take locks in C extension when the GIL is released. The crashes that #11460 addresed were due to breaking the first three rules. In particular, we could race on accessing the shared scratch buffer used when collecting tracebacks, which lead to double-frees. See #13185 for more details. Our mitigation involved using C locks around any access to the shared profiler state. We nearly broke rule 4 in the process. However, we used try-locks specifically out of a fear of introducing deadlocks. Try-locks mean that we attempt to acquire the lock, but return a failure if the lock is already held. This stopped deadlocks, but introduced bugs: For example: - If we failed to take the lock when trying to report allocation profile events, we'd raise an exception when it was in fact not reasonable for doing that to fail. See #12075. - memalloc_heap_untrack, which removes tracked allocations, was guarded with a try-lock. If we couldn't acquire the lock, we would fail to remove a record for an allocation and effectively leak memory. See #13317 - We attempted to make our locking fork-safe. The first attempt was inefficient; we made it less inefficient but the fix only "worked" because of try-locks. See #11848 Try-locks hide concurrency problems and we shouldn't use them. Using our own locks requires releasing the GIL before acquisition, and then re-acquiring the GIL. That adds unnecessary overhead. We don't inherently need to do any off-GIL work. So, we should try to just use the GIL as long as it is available. The basic refactor is actually pretty simple. In a nutshell, we rearrange the memalloc_add_event and memalloc_heap_track functions so that they make the sampling decision, then take a traceback, then insert the traceback into the appropriate data structure. Collecting a traceback can release the GIL, so we make sure that modifying the data structure happens completely after the traceback is collected. We also safeguard against the possibility that the profiler was stopped during sampling, if the GIL was released. This requires a small rearrangement of memalloc_stop to make sure that the sampling functions don't see partially-freed profiler data structures. For testing, I have mainly used the code from test_memealloc_data_race_regression. I also added a debug mode, enabled by compiling with MEMALLOC_TESTING_GIL_RELEASE, which releases the GIL at places where it would be expected. For performance I examined the overhead of profiling on a basic flask application.
nsrip-dd
added a commit
that referenced
this pull request
May 20, 2025
We added locking to memalloc, the memory profiler, in #11460 in order to address crashes. These locks made the crashes go away, but significantly increased the baseline overhead of the profiler and introduced subtle bugs. The locks we added turned out to be fundamentally incompatible with the global interpreter lock (GIL), at least with the implementation from #11460. This PR refactors the profiler to use the GIL exclusively for locking. First, we should acknowledge no-GIL and subinterpreters. As of right now, our module does not support either. A module has to explicitly opt-in to support either, so there is no risk of those modes being enabled under our feet. Supporting either mode is likely a repo-wide project. For now, we can assume the GIL exists. This work was motivated by overhead. We currently acquire and release locks in every memory allocation and free. Even when the locks aren't contended, allocations and frees are very frequent, and the extra works adds up. We add about ~8x overhead to the baselien cost of allocation just with our locking, not including the cost of actually sampling an allocation. We can't get rid of this overhead just by reducing sampling frequency. There are a few rules to follow in order to use the GIL correctly for locking: 1) The GIL is held when a C extension function is called, _except_ possibly in the raw allocator, which we do not profile 2) The GIL may be released during C Python API calls. Even if it is released, though, it will be held again after the call 3) Thus, the GIL creates critical sections only between C Python API calls, and the beginning and end of C extension functions. Modifications to shared state across those points are not atomic. 4) If we take a lock of our own in a C extension code (i.e. a pthread_mutex), and the extension code releases the GIL, then the program will deadlock due to lock order inversion. We can only safely take locks in C extension when the GIL is released. The crashes that #11460 addresed were due to breaking the first three rules. In particular, we could race on accessing the shared scratch buffer used when collecting tracebacks, which lead to double-frees. See #13185 for more details. Our mitigation involved using C locks around any access to the shared profiler state. We nearly broke rule 4 in the process. However, we used try-locks specifically out of a fear of introducing deadlocks. Try-locks mean that we attempt to acquire the lock, but return a failure if the lock is already held. This stopped deadlocks, but introduced bugs: For example: - If we failed to take the lock when trying to report allocation profile events, we'd raise an exception when it was in fact not reasonable for doing that to fail. See #12075. - memalloc_heap_untrack, which removes tracked allocations, was guarded with a try-lock. If we couldn't acquire the lock, we would fail to remove a record for an allocation and effectively leak memory. See #13317 - We attempted to make our locking fork-safe. The first attempt was inefficient; we made it less inefficient but the fix only "worked" because of try-locks. See #11848 Try-locks hide concurrency problems and we shouldn't use them. Using our own locks requires releasing the GIL before acquisition, and then re-acquiring the GIL. That adds unnecessary overhead. We don't inherently need to do any off-GIL work. So, we should try to just use the GIL as long as it is available. The basic refactor is actually pretty simple. In a nutshell, we rearrange the memalloc_add_event and memalloc_heap_track functions so that they make the sampling decision, then take a traceback, then insert the traceback into the appropriate data structure. Collecting a traceback can release the GIL, so we make sure that modifying the data structure happens completely after the traceback is collected. We also safeguard against the possibility that the profiler was stopped during sampling, if the GIL was released. This requires a small rearrangement of memalloc_stop to make sure that the sampling functions don't see partially-freed profiler data structures. For testing, I have mainly used the code from test_memealloc_data_race_regression. I also added a debug mode, enabled by compiling with MEMALLOC_TESTING_GIL_RELEASE, which releases the GIL at places where it would be expected. For performance I examined the overhead of profiling on a basic flask application.
nsrip-dd
added a commit
that referenced
this pull request
May 20, 2025
We added locking to memalloc, the memory profiler, in #11460 in order to address crashes. These locks made the crashes go away, but significantly increased the baseline overhead of the profiler and introduced subtle bugs. The locks we added turned out to be fundamentally incompatible with the global interpreter lock (GIL), at least with the implementation from #11460. This PR refactors the profiler to use the GIL exclusively for locking. First, we should acknowledge no-GIL and subinterpreters. As of right now, our module does not support either. A module has to explicitly opt-in to support either, so there is no risk of those modes being enabled under our feet. Supporting either mode is likely a repo-wide project. For now, we can assume the GIL exists. This work was motivated by overhead. We currently acquire and release locks in every memory allocation and free. Even when the locks aren't contended, allocations and frees are very frequent, and the extra works adds up. We add about ~8x overhead to the baselien cost of allocation just with our locking, not including the cost of actually sampling an allocation. We can't get rid of this overhead just by reducing sampling frequency. There are a few rules to follow in order to use the GIL correctly for locking: 1) The GIL is held when a C extension function is called, _except_ possibly in the raw allocator, which we do not profile 2) The GIL may be released during C Python API calls. Even if it is released, though, it will be held again after the call 3) Thus, the GIL creates critical sections only between C Python API calls, and the beginning and end of C extension functions. Modifications to shared state across those points are not atomic. 4) If we take a lock of our own in a C extension code (i.e. a pthread_mutex), and the extension code releases the GIL, then the program will deadlock due to lock order inversion. We can only safely take locks in C extension when the GIL is released. The crashes that #11460 addresed were due to breaking the first three rules. In particular, we could race on accessing the shared scratch buffer used when collecting tracebacks, which lead to double-frees. See #13185 for more details. Our mitigation involved using C locks around any access to the shared profiler state. We nearly broke rule 4 in the process. However, we used try-locks specifically out of a fear of introducing deadlocks. Try-locks mean that we attempt to acquire the lock, but return a failure if the lock is already held. This stopped deadlocks, but introduced bugs: For example: - If we failed to take the lock when trying to report allocation profile events, we'd raise an exception when it was in fact not reasonable for doing that to fail. See #12075. - memalloc_heap_untrack, which removes tracked allocations, was guarded with a try-lock. If we couldn't acquire the lock, we would fail to remove a record for an allocation and effectively leak memory. See #13317 - We attempted to make our locking fork-safe. The first attempt was inefficient; we made it less inefficient but the fix only "worked" because of try-locks. See #11848 Try-locks hide concurrency problems and we shouldn't use them. Using our own locks requires releasing the GIL before acquisition, and then re-acquiring the GIL. That adds unnecessary overhead. We don't inherently need to do any off-GIL work. So, we should try to just use the GIL as long as it is available. The basic refactor is actually pretty simple. In a nutshell, we rearrange the memalloc_add_event and memalloc_heap_track functions so that they make the sampling decision, then take a traceback, then insert the traceback into the appropriate data structure. Collecting a traceback can release the GIL, so we make sure that modifying the data structure happens completely after the traceback is collected. We also safeguard against the possibility that the profiler was stopped during sampling, if the GIL was released. This requires a small rearrangement of memalloc_stop to make sure that the sampling functions don't see partially-freed profiler data structures. For testing, I have mainly used the code from test_memealloc_data_race_regression. I also added a debug mode, enabled by compiling with MEMALLOC_TESTING_GIL_RELEASE, which releases the GIL at places where it would be expected. For performance I examined the overhead of profiling on a basic flask application.
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.
memalloc uses getpid to detect whether the process has forked, so that
we can unlock the memalloc lock in the child process (if it isn't
already locked). Unfortunately the getpid call is quite slow. From the
man page: "calls to getpid() always invoke the actual system call,
rather than returning a cached value." Furthermore, we always attempt
to take the lock for allocations, even if we aren't going to sample
them. So this is basically adding a syscall to every allocation.
Move this logic out of the allocation path. Switch to using
pthread_atfork handlers to ensure that the lock is held prior to
forking, and unlock it in the parent and child after forking. This
(maybe) has the added benefit of making sure the data structures are in
a consistent state in the child process after forking. Unclear if that's
an issue prior to this change, though. I may be missing some code that
resets the profiler on fork anyway?
Checklist
Reviewer Checklist