diff options
author | Miss Islington (bot) <31488909+miss-islington@users.noreply.github.com> | 2025-07-15 15:10:37 (GMT) |
---|---|---|
committer | GitHub <noreply@github.com> | 2025-07-15 15:10:37 (GMT) |
commit | 9ae74e94a06865aee96827838ce16c8924cbe86f (patch) | |
tree | 71932366c8f66e7d486570a471fd147d48cbf09b /Objects/codeobject.c | |
parent | c939963b53600062f0978c99559e22ab39135402 (diff) | |
download | cpython-9ae74e94a06865aee96827838ce16c8924cbe86f.zip cpython-9ae74e94a06865aee96827838ce16c8924cbe86f.tar.gz cpython-9ae74e94a06865aee96827838ce16c8924cbe86f.tar.bz2 |
[3.14] gh-136396: Include instrumentation when creating new copies of the bytecode (GH-136525) (GH-136657)
Previously, we assumed that instrumentation would happen for all copies of
the bytecode if the instrumentation version on the code object didn't match
the per-interpreter instrumentation version. That assumption was incorrect:
instrumentation will exit early if there are no new "events," even if there
is an instrumentation version mismatch.
To fix this, include the instrumented opcodes when creating new copies of
the bytecode, rather than replacing them with their uninstrumented variants.
I don't think we have to worry about races between instrumentation and creating
new copies of the bytecode: instrumentation and new bytecode creation cannot happen
concurrently. Instrumentation requires that either the world is stopped or the
code object's per-object lock is held and new bytecode creation requires holding
the code object's per-object lock.
(cherry picked from commit d995922198304a6de19ac1bec3e36d1e886d8468)
Co-authored-by: mpage <mpage@meta.com>
Co-authored-by: Kumar Aditya <kumaraditya@python.org>
Diffstat (limited to 'Objects/codeobject.c')
-rw-r--r-- | Objects/codeobject.c | 19 |
1 files changed, 18 insertions, 1 deletions
diff --git a/Objects/codeobject.c b/Objects/codeobject.c index ba178abc..42e0216 100644 --- a/Objects/codeobject.c +++ b/Objects/codeobject.c @@ -3330,12 +3330,29 @@ _PyCodeArray_New(Py_ssize_t size) return arr; } +// Get the underlying code unit, leaving instrumentation +static _Py_CODEUNIT +deopt_code_unit(PyCodeObject *code, int i) +{ + _Py_CODEUNIT *src_instr = _PyCode_CODE(code) + i; + _Py_CODEUNIT inst = { + .cache = FT_ATOMIC_LOAD_UINT16_RELAXED(*(uint16_t *)src_instr)}; + int opcode = inst.op.code; + if (opcode < MIN_INSTRUMENTED_OPCODE) { + inst.op.code = _PyOpcode_Deopt[opcode]; + assert(inst.op.code < MIN_SPECIALIZED_OPCODE); + } + // JIT should not be enabled with free-threading + assert(inst.op.code != ENTER_EXECUTOR); + return inst; +} + static void copy_code(_Py_CODEUNIT *dst, PyCodeObject *co) { int code_len = (int) Py_SIZE(co); for (int i = 0; i < code_len; i += _PyInstruction_GetLength(co, i)) { - dst[i] = _Py_GetBaseCodeUnit(co, i); + dst[i] = deopt_code_unit(co, i); } _PyCode_Quicken(dst, code_len, 1); } |