You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I found a serious perf regression on nbody. The problem is that stop_tracing adds a _DEOPT when it sees ENTER_EXECUTOR, thus causing the traces to stop growing. I put _EXIT_TRACE previously, but forgot about it when the JIT code was undergoing refactoring.
The fix is just to pass the opcode of the exit op in. Will do that later.
I'd rather keep this open, otherwise it gives a false impression of being finished when there's some clean up to do.
Or, make new issues for the list above and then close this one.
Restore the optimization for stack size checks and its tests
I think we should abandon this and just clean up the code implementing this optimisation. It's generally very finnicky and I don't think it nets us much perf wins.
I think we should abandon this and just clean up the code implementing this optimisation. It's generally very finnicky and I don't think it nets us much perf wins.
Hi, @Fidget-Spinner
Refer: #144388. Should we convert _CHECK_STACK_SPACE to _CHECK_STACK_SPACE_OPERAND one by one? If we need this, i'll implement it.
Activity
Fidget-Spinner commentedon Nov 14, 2025
I found a serious perf regression on nbody. The problem is that
stop_tracingadds a_DEOPTwhen it seesENTER_EXECUTOR, thus causing the traces to stop growing. I put _EXIT_TRACE previously, but forgot about it when the JIT code was undergoing refactoring.The fix is just to pass the opcode of the exit op in. Will do that later.
gh-139109: JIT _EXIT_TRACE to ENTER_EXECUTOR rather than _DEOPT (GH-1…
GH-139109: Support switch/case dispatch with the tracing interpreter. (…
pythongh-139109: A new tracing JIT compiler frontend for CPython (pyt…
pythongh-139109: JIT _EXIT_TRACE to ENTER_EXECUTOR rather than _DEOPT (…
pythonGH-139109: Support switch/case dispatch with the tracing interp…
savannahostrowski commentedon Dec 10, 2025
@Fidget-Spinner Is this closable now? Can we track bugs etc. in new issues?
For example: #139109 (comment)?
Fidget-Spinner commentedon Dec 10, 2025
Yeah lets do follow ups separately
markshannon commentedon Dec 16, 2025
I'd rather keep this open, otherwise it gives a false impression of being finished when there's some clean up to do.
Or, make new issues for the list above and then close this one.
Fidget-Spinner commentedon Dec 16, 2025
I think we should abandon this and just clean up the code implementing this optimisation. It's generally very finnicky and I don't think it nets us much perf wins.
gh-139109: Add terminator to JIT code when halting due to invalid dep…
GH-139109: Partial reworking of JIT data structures (GH-144105)
cocolato commentedon Feb 2, 2026
Hi, @Fidget-Spinner
Refer: #144388. Should we convert
_CHECK_STACK_SPACEto_CHECK_STACK_SPACE_OPERANDone by one? If we need this, i'll implement it.gh-139109: Replace `_CHECK_STACK_SPACE` with `_CHECK_STACK_SPACE_OPER…
pythongh-139109: Replace `_CHECK_STACK_SPACE` with `_CHECK_STACK_SPAC…
pythonGH-139109: Partial reworking of JIT data structures (pythonGH-1…
pythongh-139109: Replace `_CHECK_STACK_SPACE` with `_CHECK_STACK_SPAC…
gh-139109: Document the trace recording interpreter in internaldocs (G…
pythongh-139109: Document the trace recording interpreter in internal…
pythongh-139109: Document the trace recording interpreter in internal…