Skip to content

gh-109039: Branch prediction for Tier 2 interpreter #109038

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.

Sign up for GitHub

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

Already on GitHub? Sign in to your account

Merged
gvanrossum merged 17 commits into from
Sep 11, 2023
Merged

gh-109039: Branch prediction for Tier 2 interpreter #109038

Changes from 1 commit
Commits
Show all changes
17 commits
Select commit Hold shift + click to select a range
44db701
inst() and macro() may need cache size metadata
gvanrossum Sep 6, 2023
ff29ab3
Add cache entry to *POP_JUMP_IF_* instructions
gvanrossum Sep 6, 2023
ebc91a2
Fix test_dis (also fixes test_peepholer)
gvanrossum Sep 7, 2023
072bb38
Fix test_monitoring
gvanrossum Sep 7, 2023
896ae53
Include pycore_bitutils.h in instrumentation.c
gvanrossum Sep 7, 2023
0eb5b90
Follow likely jumps in trace
gvanrossum Sep 7, 2023
25bfb3d
Merge remote-tracking branch 'upstream/main' into count-branches
gvanrossum Sep 7, 2023
a9c0805
Require 16 iterations before optimizing
gvanrossum Sep 7, 2023
7dfb94c
Fix existing uops tests
gvanrossum Sep 7, 2023
73eb60f
Add test for branch prediction
gvanrossum Sep 7, 2023
fbd322a
Initialize POP_JUMP_IF* counters to 0x5555
gvanrossum Sep 7, 2023
ed045d7
Update counter in INSTRUMENTED_POP_JUMP_IF_FALSE
gvanrossum Sep 8, 2023
1850988
Simplify writing of _PyOpcode_Caches
gvanrossum Sep 8, 2023
4f1684c
Fix test_huntrleaks under -Xuops
gvanrossum Sep 8, 2023
cb2cf12
Fix test_dis under -Xuops
gvanrossum Sep 8, 2023
d74670c
Merge branch 'main' into count-branches
gvanrossum Sep 11, 2023
41463a5
Update magic number
gvanrossum Sep 11, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Next Next commit
inst() and macro() may need cache size metadata
  • Loading branch information
@gvanrossum
gvanrossum committed Sep 6, 2023
commit 44db70146bdbebcb666f8c7365eb24399ac957e8
17 changes: 15 additions & 2 deletions Tools/cases_generator/generate_cases.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -549,10 +549,23 @@ def write_metadata(self, metadata_filename: str, pymetadata_filename: str) -> No
"=",
";",
):
for name, _ in self.families.items():
family_member_names: set[str] = set()
Comment thread
gvanrossum marked this conversation as resolved.
for name, family in self.families.items():
family_member_names.update(family.members)
instr = self.instrs[name]
if instr.cache_offset > 0:
self.out.emit(f'[{name}] = {instr.cache_offset},')
self.out.emit(f"[{name}] = {instr.cache_offset},")
for instr in self.instrs.values():
if (
not instr.family
and instr.cache_offset
and instr.kind == "inst"
and not instr.name.startswith("INSTRUMENTED_")
):
self.out.emit(f"[{instr.name}] = {instr.cache_offset},")
for mac in self.macro_instrs.values():
if mac.name not in family_member_names and mac.cache_offset > 0:
self.out.emit(f"[{mac.name}] = {mac.cache_offset},")
# Irregular case:
self.out.emit('[JUMP_BACKWARD] = 1,')

Expand Down