This PR implements frame caching in the RemoteUnwinder class to significantly reduce memory reads when profiling remote processes with deep call stacks.
When cache_frames=True, the unwinder stores the frame chain from each sample and reuses unchanged portions in subsequent samples. Since most profiling samples capture similar call stacks (especially the parent frames), this optimization avoids repeatedly reading the same frame data from the target process.
The implementation adds a last_profiled_frame field to the thread state that tracks where the previous sample stopped. On the next sample, if the current frame chain reaches this marker, the cached frames from that point onward are reused instead of being re-read from remote memory.
The sampling profiler now enables frame caching by default.
Loading a small data which does not even involve arbitrary code execution
could consume arbitrary large amount of memory. There were three issues:
* PUT and LONG_BINPUT with large argument (the C implementation only).
Since the memo is implemented in C as a continuous dynamic array, a single
opcode can cause its resizing to arbitrary size. Now the sparsity of
memo indices is limited.
* BINBYTES, BINBYTES8 and BYTEARRAY8 with large argument. They allocated
the bytes or bytearray object of the specified size before reading into
it. Now they read very large data by chunks.
* BINSTRING, BINUNICODE, LONG4, BINUNICODE8 and FRAME with large
argument. They read the whole data by calling the read() method of
the underlying file object, which usually allocates the bytes object of
the specified size before reading into it. Now they read very large data
by chunks.
Also add comprehensive benchmark suite to measure performance and memory
impact of chunked reading optimization in PR #119204.
Features:
- Normal mode: benchmarks legitimate pickles (time/memory metrics)
- Antagonistic mode: tests malicious pickles (DoS protection)
- Baseline comparison: side-by-side comparison of two Python builds
- Support for truncated data and sparse memo attack vectors
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Co-authored-by: Gregory P. Smith <greg@krypto.org>
Add readline.get_pre_input_hook() to retrieve the current pre-input
hook. This allows applications to save and restore the hook without
overwriting user settings.
Added atomic operations to `scanner_begin()` and `scanner_end()` to prevent
race conditions on the `executing` flag in free-threaded builds. Also added
tests for concurrent usage of the `re` module.
Without the atomic operations, `test_scanner_concurrent_access()` triggers
`assert(self->executing)` failures, or a thread sanitizer run emits errors.
* gh-141801: Use accessors for ASN1_STRING fields
While ASN1_STRING is currently exposed, it is better to use the
accessors. See https://github.com/openssl/openssl/issues/29117 where, if
the type were opaque, OpenSSL's X509 objects could be much more
memory-efficient.
* Update Modules/_ssl.c
Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
* Update Modules/_ssl.c
Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
---------
Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
Added a critical section to protect the states of `ReaderObj` and `WriterObj` in the free-threading build. Without the critical sections, both new free-threading tests were crashing.
* Promote _PyObject_Dump() as a public function.
* Keep _PyObject_Dump() alias to PyUnstable_Object_Dump()
for backward compatibility.
* Replace _PyObject_Dump() with PyUnstable_Object_Dump().
Co-authored-by: Peter Bierma <zintensitydev@gmail.com>
Co-authored-by: Kumar Aditya <kumaraditya@python.org>
Co-authored-by: Petr Viktorin <encukou@gmail.com>
Adapted from a patch for Python 3.14 submitted to the Debian BTS by John
https://bugs.debian.org/1105111#20
Co-authored-by: John David Anglin <dave.anglin@bell.net>
- Introduce a new field in the GC state to store the frame that initiated garbage collection.
- Update RemoteUnwinder to include options for including "<native>" and "<GC>" frames in the stack trace.
- Modify the sampling profiler to accept parameters for controlling the inclusion of native and GC frames.
- Enhance the stack collector to properly format and append these frames during profiling.
- Add tests to verify the correct behavior of the profiler with respect to native and GC frames, including options to exclude them.
Co-authored-by: Pablo Galindo Salgado <pablogsal@gmail.com>
This PR changes the current JIT model from trace projection to trace recording. Benchmarking: better pyperformance (about 1.7% overall) geomean versus current https://raw.githubusercontent.com/facebookexperimental/free-threading-benchmarking/refs/heads/main/results/bm-20251108-3.15.0a1%2B-7e2bc1d-JIT/bm-20251108-vultr-x86_64-Fidget%252dSpinner-tracing_jit-3.15.0a1%2B-7e2bc1d-vs-base.svg, 100% faster Richards on the most improved benchmark versus the current JIT. Slowdown of about 10-15% on the worst benchmark versus the current JIT. **Note: the fastest version isn't the one merged, as it relies on fixing bugs in the specializing interpreter, which is left to another PR**. The speedup in the merged version is about 1.1%. https://raw.githubusercontent.com/facebookexperimental/free-threading-benchmarking/refs/heads/main/results/bm-20251112-3.15.0a1%2B-f8a764a-JIT/bm-20251112-vultr-x86_64-Fidget%252dSpinner-tracing_jit-3.15.0a1%2B-f8a764a-vs-base.svg
Stats: 50% more uops executed, 30% more traces entered the last time we ran them. It also suggests our trace lengths for a real trace recording JIT are too short, as a lot of trace too long aborts https://github.com/facebookexperimental/free-threading-benchmarking/blob/main/results/bm-20251023-3.15.0a1%2B-eb73378-CLANG%2CJIT/bm-20251023-vultr-x86_64-Fidget%252dSpinner-tracing_jit-3.15.0a1%2B-eb73378-pystats-vs-base.md .
This new JIT frontend is already able to record/execute significantly more instructions than the previous JIT frontend. In this PR, we are now able to record through custom dunders, simple object creation, generators, etc. None of these were done by the old JIT frontend. Some custom dunders uops were discovered to be broken as part of this work gh-140277
The optimizer stack space check is disabled, as it's no longer valid to deal with underflow.
Pros:
* Ignoring the generated tracer code as it's automatically created, this is only additional 1k lines of code. The maintenance burden is handled by the DSL and code generator.
* `optimizer.c` is now significantly simpler, as we don't have to do strange things to recover the bytecode from a trace.
* The new JIT frontend is able to handle a lot more control-flow than the old one.
* Tracing is very low overhead. We use the tail calling interpreter/computed goto interpreter to switch between tracing mode and non-tracing mode. I call this mechanism dual dispatch, as we have two dispatch tables dispatching to each other. Specialization is still enabled while tracing.
* Better handling of polymorphism. We leverage the specializing interpreter for this.
Cons:
* (For now) requires tail calling interpreter or computed gotos. This means no Windows JIT for now :(. Not to fret, tail calling is coming soon to Windows though https://github.com/python/cpython/pull/139962
Design:
* After each instruction, the `record_previous_inst` function/label is executed. This does as the name suggests.
* The tracing interpreter lowers bytecode to uops directly so that it can obtain "fresh" values at the point of lowering.
* The tracing version behaves nearly identical to the normal interpreter, in fact it even has specialization! This allows it to run without much of a slowdown when tracing. The actual cost of tracing is only a function call and writes to memory.
* The tracing interpreter uses the specializing interpreter's deopt to naturally form the side exit chains. This allows it to side exit chain effectively, without repeating much code. We force a re-specializing when tracing a deopt.
* The tracing interpreter can even handle goto errors/exceptions, but I chose to disable them for now as it's not tested.
* Because we do not share interpreter dispatch, there is should be no significant slowdown to the original specializing interpreter on tailcall and computed got with JIT disabled. With JIT enabled, there might be a slowdown in the form of the JIT trying to trace.
* Things that could have dynamic instruction pointer effects are guarded on. The guard deopts to a new instruction --- `_DYNAMIC_EXIT`.
Add PyUnstable_ThreadState_SetStackProtection() and
PyUnstable_ThreadState_ResetStackProtection() functions
to set the stack base address and stack size of a Python
thread state.
Co-authored-by: Petr Viktorin <encukou@gmail.com>
Many functions related to compiling or parsing Python code, such as
compile(), ast.parse(), symtable.symtable(),
and importlib.abc.InspectLoader.source_to_code() now allow to pass
the module name used when filtering syntax warnings.
* gh-137109: refactor warning about threads when forking
This splits the OS API specific functionality to get the number of threads out
from the fallback Python method and warning raising code itself. This way the
OS APIs can be queried before we've run
`os.register_at_fork(after_in_parent=...)` registered functions which
themselves may (re)start threads that would otherwise be detected.
This is best effort. If the OS APIs are either unavailable or fail, the
warning generating code still falls back to looking at the Python threading
state after the CPython interpreter world has been restarted and the
after_in_parent calls have been made. The common case for most Linux and macOS
environments should work today.
This also lines up with the existing TODO refactoring, we may choose to expose
this API to get the number of OS threads in the `os` module in the future.
* NEWS entry
* avoid "function-prototype" compiler warning?
Fix error in assertion which causes failure if pos is equal to PY_SSIZE_T_MAX.
Fix undefined behavior in read() and readinto() if pos is larger that the size
of the underlying buffer.
* Refer to bytes objects or bytes-like objects instead of strings.
* Remove backticks -- they do not have effect on formatting.
* Re-wrap lines to ensure the pydoc output fits in 80 coluimns.
* Remove references to the 1024 bytes limit.
Example of fixed warnings on 32-bit Windows:
Python\remote_debugging.c(24,53): warning C4244: 'function':
conversion from 'uint64_t' to 'uintptr_t', possible loss of data
Modules\_remote_debugging_module.c(789,44): warning C4244:
'function': conversion from 'uint64_t' to 'size_t', possible loss
of data
The test just before the cast ensures that the cast cannot overflow.
Fix the warning on 32-bit Windows:
Modules\_randommodule.c(525,28): warning C4244: '=': conversion
from 'uint64_t' to 'Py_ssize_t', possible loss of data
