Movatterモバイル変換


[0]ホーム

URL:


Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
Appearance settings

GH-98831: Add DECREF_INPUTS(), expanding to DECREF() each stack input#100205

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to ourterms of service andprivacy statement. We’ll occasionally send you account related emails.

Already on GitHub?Sign in to your account

Merged
gvanrossum merged 1 commit intopython:mainfromgvanrossum:decref-inputs
Dec 17, 2022

Conversation

gvanrossum
Copy link
Member

@gvanrossumgvanrossum commentedDec 13, 2022
edited by bedevere-bot
Loading

The presence of this macro indicates that a particular instruction may be considered for conversion to a register-based format (seefaster-cpython/ideas#485).

An invariant (currently unchecked) is thatDEOPT_IF() may only occurbeforeDECREF_INPUTS(), andERROR_IF() may only occurafter it. One reason not to check this is that there are a few places where we inserttwoDECREF_INPUTS() calls, in different branches of the code. The invariant checking would have to be able to do some flow control analysis to understand this.

Note that many instructions, especially specialized ones, can't be converted to use this macro straightforwardly. This is because the generator currently only generates plainPy_DECREF(variable) statements, and cannot generate things like_Py_DECREF_SPECIALIZED() let alone deal with_PyList_AppendTakeRef().

The presence of this macro indicates that a particular instructionmay be considered for conversion to a register-based format(seefaster-cpython/ideas#485).An invariant (currently unchecked) is that `DEOPT_IF()` may onlyoccur *before* `DECREF_INPUTS()`, and `ERROR_IF()` may only occur*after* it. One reason not to check this is that there are a fewplaces where we insert *two* `DECREF_INPUTS()` calls, in differentbranches of the code. The invariant checking would have to be ableto do some flow control analysis to understand this.Note that many instructions, especially specialized ones,can't be converted to use this macro straightforwardly.This is because the generator currently only generates plain`Py_DECREF(variable)` statements, and cannot generatethings like `_Py_DECREF_SPECIALIZED()` let alone deal with`_PyList_AppendTakeRef()`.
@gvanrossum
Copy link
MemberAuthor

Some thoughts about how the code would change for the register VM (to be done in a future PR):

  • We need a separate way to mark an instruction as register-based in the DSL input, mayberegister inst(NAME, ...)
  • A register-based instruction may have at most three stack effects (inputs and outputs) in total, assigned tooparg,oparg2, andoparg3
  • For those instructions,DECREF_INPUTS() expands to nothing
  • The preamble would assign each input an oparg and then generate code to initialize the variable usingfast_locals[opargN] instead of usingPEEK(N)
  • The postamble assigns the output variable the next oparg and writes it usingPy_SETREF(fast_locals[opargN], variable)

@iritkatriel
Copy link
Member

  • A register-based instruction may have at most three stack effects (inputs and outputs) in total, assigned tooparg,oparg2, andoparg3

Should we make it oparg1, oparg2, oparg3? (oparg can stay for now, for "legacy" instructions that were not converted yet).

gvanrossum reacted with thumbs up emoji

@pythonpython deleted a comment fromnetlifybotDec 13, 2022
@gvanrossum
Copy link
MemberAuthor

I was hoping you'd review this PR,@iritkatriel. We can work on the 'register inst' syntax next.

@gvanrossumgvanrossum merged commit9cdd2fa intopython:mainDec 17, 2022
@gvanrossumgvanrossum deleted the decref-inputs branchDecember 17, 2022 04:46
gvanrossum added a commit to gvanrossum/cpython that referenced this pull requestDec 17, 2022
… input (python#100205)The presence of this macro indicates that a particular instructionmay be considered for conversion to a register-based format(seefaster-cpython/ideas#485).An invariant (currently unchecked) is that `DEOPT_IF()` may onlyoccur *before* `DECREF_INPUTS()`, and `ERROR_IF()` may only occur*after* it. One reason not to check this is that there are a fewplaces where we insert *two* `DECREF_INPUTS()` calls, in differentbranches of the code. The invariant checking would have to be ableto do some flow control analysis to understand this.Note that many instructions, especially specialized ones,can't be converted to use this macro straightforwardly.This is because the generator currently only generates plain`Py_DECREF(variable)` statements, and cannot generatethings like `_Py_DECREF_SPECIALIZED()` let alone deal with`_PyList_AppendTakeRef()`.
shihai1991 added a commit to shihai1991/cpython that referenced this pull requestDec 18, 2022
* origin/main: (1306 commits)  CorrectCVE-2020-10735 documentation (python#100306)pythongh-100272: Fix JSON serialization of OrderedDict (pythonGH-100273)pythongh-93649: Split tracemalloc tests from _testcapimodule.c (python#99551)  Docs: Use `PY_VERSION_HEX` for version comparison (python#100179)pythongh-97909: Fix markup for `PyMethodDef` members (python#100089)pythongh-99240: Reset pointer to NULL when the pointed memory is freed in argument parsing (python#99890)pythongh-99240: Reset pointer to NULL when the pointed memory is freed in argument parsing (python#99890)pythonGH-98831: Add DECREF_INPUTS(), expanding to DECREF() each stack input (python#100205)pythongh-78707: deprecate passing >1 argument to `PurePath.[is_]relative_to()` (pythonGH-94469)pythongh-99540: Constant hash for _PyNone_Type to aid reproducibility (pythonGH-99541)pythongh-100039: enhance __signature__ to work with str and callables (pythonGH-100168)pythongh-99830: asyncio: Document returns of remove_{reader,writer} (python#100302)  "Compound statement" docs: Fix with-statement step indexing (python#100286)pythonGH-90043: Handle NaNs in COMPARE_OP_FLOAT_JUMP (pythonGH-100278)  Improve stats presentation for calls. (pythonGH-100274)  Better stats for `LOAD_ATTR` and `STORE_ATTR` (pythonGH-100295)pythongh-81057: Move the Cached Parser Dummy Name to _PyRuntimeState (python#100277)  Document that zipfile's pwd parameter is a `bytes` object (python#100209)pythongh-99767: mark `PyTypeObject.tp_watched` as internal use only in table (python#100271)  Fix typo in introduction.rst (python#100266)  ...
carljm added a commit to carljm/cpython that referenced this pull requestDec 19, 2022
* main:pythongh-89727: Fix os.walk RecursionError on deep trees (python#99803)  Docs: Don't upload CI artifacts (python#100330)pythongh-94912: Added marker for non-standard coroutine function detection (python#99247)  CorrectCVE-2020-10735 documentation (python#100306)pythongh-100272: Fix JSON serialization of OrderedDict (pythonGH-100273)pythongh-93649: Split tracemalloc tests from _testcapimodule.c (python#99551)  Docs: Use `PY_VERSION_HEX` for version comparison (python#100179)pythongh-97909: Fix markup for `PyMethodDef` members (python#100089)pythongh-99240: Reset pointer to NULL when the pointed memory is freed in argument parsing (python#99890)pythongh-99240: Reset pointer to NULL when the pointed memory is freed in argument parsing (python#99890)pythonGH-98831: Add DECREF_INPUTS(), expanding to DECREF() each stack input (python#100205)pythongh-78707: deprecate passing >1 argument to `PurePath.[is_]relative_to()` (pythonGH-94469)
Sign up for freeto join this conversation on GitHub. Already have an account?Sign in to comment
Reviewers

@iritkatrieliritkatrieliritkatriel approved these changes

Assignees
No one assigned
Labels
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

3 participants
@gvanrossum@iritkatriel@bedevere-bot

[8]ページ先頭

©2009-2025 Movatter.jp