Issue 884022: dynamic execution profiling vs opcode prediction

➜

This issue tracker has been migrated to GitHub , and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

This issue has been migrated to GitHub: https://github.com/python/cpython/issues/39865

classification

Type:	Stage:
Title:	dynamic execution profiling vs opcode prediction
Components:	None	Versions:

process

Created on 2004年01月25日 06:45 by aimacintyre, last changed 2022年04月11日 14:56 by admin. This issue is now closed.

Files
File name	Uploaded	Description	Edit
dyn_exec_prof_vs_opc_pred-ceval.c.patch	aimacintyre, 2004年01月25日 06:45	disable opcode prediction when doing dynamic execution profiling

Messages (2)
msg45319 - (view)	Author: Andrew I MacIntyre (aimacintyre) * (Python triager)	Date: 2004年01月25日 06:45
This patch is a proposed fix/work around for bug #884020. It disables opcode prediction when dynamic execution profiling is in effect, so the profiling counters at the top of the main interpreter loop in eval_frame() are updated for each opcode. It does give rise to warnings about unused labels (with gcc anyway) which could be resolved by #define'ing all the prediction machinery away.
msg45320 - (view)	Author: Raymond Hettinger (rhettinger) * (Python committer)	Date: 2004年02月08日 20:05
Logged In: YES user_id=80475 Accepted and applied: Python/ceval.c 2.375 FWIW, I think the significance of dynamic opcode counts is overrated as it tends to say more about the application being profiled and its coding style than it says about the interpreter.

homepage