Skip to content
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

Jump to bottom

GH-113464: Speed up JIT builds #122839

Merged
merged 2 commits into from
Aug 14, 2024
+19 −5
Merged

GH-113464: Speed up JIT builds #122839

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
60b7e71
Only include one case when compiling a stencil
brandtbucher Jul 27, 2024
a32111d
Add a comment about faster compilation
brandtbucher Aug 14, 2024
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
18 changes: 15 additions & 3 deletions Tools/jit/_targets.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -182,15 +182,27 @@ async def _compile(

async def _build_stencils(self) -> dict[str, _stencils.StencilGroup]:
generated_cases = PYTHON_EXECUTOR_CASES_C_H.read_text()
opnames = sorted(re.findall(r"\n {8}case (\w+): \{\n", generated_cases))
cases_and_opnames = sorted(
re.findall(
r"\n {8}(case (\w+): \{\n.*?\n {8}\})", generated_cases, flags=re.DOTALL
)
)
tasks = []
with tempfile.TemporaryDirectory() as tempdir:
work = pathlib.Path(tempdir).resolve()
async with asyncio.TaskGroup() as group:
coro = self._compile("trampoline", TOOLS_JIT / "trampoline.c", work)
tasks.append(group.create_task(coro, name="trampoline"))
for opname in opnames:
coro = self._compile(opname, TOOLS_JIT_TEMPLATE_C, work)
template = TOOLS_JIT_TEMPLATE_C.read_text()
for case, opname in cases_and_opnames:
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Would it be good to add a comment here about why each of these is compiled independently (e.g., because of the performance benefit)? I'm not sure that would be abundantly clear if I just stumbled upon this code.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good idea, I'll add a comment.

# Write out a copy of the template with *only* this case
# inserted. This is about twice as fast as #include'ing all
# of executor_cases.c.h each time we compile (since the C
# compiler wastes a bunch of time parsing the dead code for
# all of the other cases):
c = work / f"{opname}.c"
c.write_text(template.replace("CASE", case))
coro = self._compile(opname, c, work)
tasks.append(group.create_task(coro, name=opname))
return {task.get_name(): task.result() for task in tasks}

Expand Down
6 changes: 4 additions & 2 deletions Tools/jit/template.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -84,6 +84,8 @@ do { \
#undef WITHIN_STACK_BOUNDS
#define WITHIN_STACK_BOUNDS() 1

#define TIER_TWO 2

_Py_CODEUNIT *
_JIT_ENTRY(_PyInterpreterFrame *frame, _PyStackRef *stack_pointer, PyThreadState *tstate)
{
Expand All @@ -107,9 +109,9 @@ _JIT_ENTRY(_PyInterpreterFrame *frame, _PyStackRef *stack_pointer, PyThreadState
OPT_STAT_INC(uops_executed);
UOP_STAT_INC(uopcode, execution_count);

// The actual instruction definitions (only one will be used):
switch (uopcode) {
#include "executor_cases.c.h"
// The actual instruction definition gets inserted here:
CASE
default:
Py_UNREACHABLE();
}
Expand Down
Loading