Repository navigation
gh-75876: Correct the memory use declared by some bigmem tests - #155307
Merged
serhiy-storchaka merged 1 commit intoAug 7, 2026
Merged
serhiy-storchaka merged 1 commit into
serhiy-storchaka merged 1 commit into
Conversation
Copy link
Copy Markdown
Member
Sorry, something went wrong.
Measured with a limited address space and the peak RSS of the process: * TupleTest.test_repeat_large_2 keeps 3 tuples of pointers alive, as its siblings do, but declared the count for a 32-bit build. * The md5 tests hash the data and a io.BytesIO() copy of it. * testDecompress4G keeps the data, the compressed and the decompressed bytes alive, and the decompressor grows its output buffer. * test_huge_string_builtins uses 3 times the string size, not 2. * test_large_subn uses a bit more than the 18 times it declared. Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Copy link
Copy Markdown
|
Thanks @serhiy-storchaka for the PR 🌮🎉.. I'm working now to backport this PR to: 3.13, 3.14, 3.15. |
Sorry, something went wrong.
Copy link
Copy Markdown
|
GH-155312 is a backport of this pull request to the 3.15 branch. |
Sorry, something went wrong.
Copy link
Copy Markdown
|
GH-155313 is a backport of this pull request to the 3.14 branch. |
Sorry, something went wrong.
Copy link
Copy Markdown
|
GH-155314 is a backport of this pull request to the 3.13 branch. |
Sorry, something went wrong.
serhiy-storchaka
added a commit
that referenced
this pull request
Aug 7, 2026
…GH-155307) (GH-155314) Measured with a limited address space and the peak RSS of the process: * TupleTest.test_repeat_large_2 keeps 3 tuples of pointers alive, as its siblings do, but declared the count for a 32-bit build. * The md5 tests hash the data and a io.BytesIO() copy of it. * testDecompress4G keeps the data, the compressed and the decompressed bytes alive, and the decompressor grows its output buffer. * test_huge_string_builtins uses 3 times the string size, not 2. * test_large_subn uses a bit more than the 18 times it declared. (cherry picked from commit e469fa9) Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
serhiy-storchaka
added a commit
that referenced
this pull request
Aug 7, 2026
…GH-155307) (GH-155313) Measured with a limited address space and the peak RSS of the process: * TupleTest.test_repeat_large_2 keeps 3 tuples of pointers alive, as its siblings do, but declared the count for a 32-bit build. * The md5 tests hash the data and a io.BytesIO() copy of it. * testDecompress4G keeps the data, the compressed and the decompressed bytes alive, and the decompressor grows its output buffer. * test_huge_string_builtins uses 3 times the string size, not 2. * test_large_subn uses a bit more than the 18 times it declared. (cherry picked from commit e469fa9) Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
hugovk
pushed a commit
that referenced
this pull request
Aug 8, 2026
…GH-155307) (#155312) gh-75876: Correct the memory use declared by some bigmem tests (GH-155307) Measured with a limited address space and the peak RSS of the process: * TupleTest.test_repeat_large_2 keeps 3 tuples of pointers alive, as its siblings do, but declared the count for a 32-bit build. * The md5 tests hash the data and a io.BytesIO() copy of it. * testDecompress4G keeps the data, the compressed and the decompressed bytes alive, and the decompressor grows its output buffer. * test_huge_string_builtins uses 3 times the string size, not 2. * test_large_subn uses a bit more than the 18 times it declared. (cherry picked from commit e469fa9) Co-authored-by: Serhiy Storchaka <storchaka@gmail.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
mbeijen
pushed a commit
to mbeijen/cpython
that referenced
this pull request
Aug 14, 2026
…ythonGH-155307) Measured with a limited address space and the peak RSS of the process: * TupleTest.test_repeat_large_2 keeps 3 tuples of pointers alive, as its siblings do, but declared the count for a 32-bit build. * The md5 tests hash the data and a io.BytesIO() copy of it. * testDecompress4G keeps the data, the compressed and the decompressed bytes alive, and the decompressor grows its output buffer. * test_huge_string_builtins uses 3 times the string size, not 2. * test_large_subn uses a bit more than the 18 times it declared. Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.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
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.
Some bigmem tests declare less memory than they use. Measured as the peak RSS of a process running a single test with -M:
test_repeat_large_2 keeps 3 tuples of pointers alive, as test_repeat_small and test_repeat_large do, but declared 12 instead of pointer_size * 3, which is the count for a 32-bit build.
The md5 tests pass the data to check_file_digest(), which copies all of it into an io.BytesIO().
testDecompress4G keeps the data, the compressed and the decompressed bytes alive at the same time, and the decompressor grows its output buffer.
These are the only tests which use more memory than they declare. Every other bigmem test was measured too, except the two test_marshal tests which declare 256 GiB and cannot be run anywhere: the largest excess is 1.04x, and it is 20 MiB in the median.