GH-115869: Make jit_stencils.h reproducible (GH-127166)

Misc/NEWS.d/next/Build/2024-11-22-08-46-46.gh-issue-115869.UVLSKd.rst

@@ -0,0 +1 @@
+Make ``jit_stencils.h`` (which is produced during JIT builds) reproducible.

Tools/jit/_stencils.py

@@ -202,7 +202,8 @@ def pad(self, alignment: int) -> None:
         """Pad the stencil to the given alignment."""
         offset = len(self.body)
         padding = -offset % alignment
-        self.disassembly.append(f"{offset:x}: {' '.join(['00'] * padding)}")
+        if padding:
+            self.disassembly.append(f"{offset:x}: {' '.join(['00'] * padding)}")
         self.body.extend([0] * padding)
 
     def remove_jump(self, *, alignment: int = 1) -> None:

Tools/jit/_targets.py

@@ -61,10 +61,11 @@ async def _parse(self, path: pathlib.Path) -> _stencils.StencilGroup:
         args = ["--disassemble", "--reloc", f"{path}"]
         output = await _llvm.maybe_run("llvm-objdump", args, echo=self.verbose)
         if output is not None:
+            # Make sure that full paths don't leak out (for reproducibility):
+            long, short = str(path), str(path.name)
             group.code.disassembly.extend(
-                line.expandtabs().strip()
+                line.expandtabs().strip().replace(long, short)
                 for line in output.splitlines()
-                if not line.isspace()
             )
         args = [
             "--elf-output-style=JSON",
@@ -90,9 +91,6 @@ async def _parse(self, path: pathlib.Path) -> _stencils.StencilGroup:
         if group.data.body:
             line = f"0: {str(bytes(group.data.body)).removeprefix('b')}"
             group.data.disassembly.append(line)
-        group.process_relocations(
-            known_symbols=self.known_symbols, alignment=self.alignment
-        )
         return group
 
     def _handle_section(self, section: _S, group: _stencils.StencilGroup) -> None:
@@ -122,6 +120,10 @@ async def _compile(
             f"-I{CPYTHON / 'Tools' / 'jit'}",
             "-O3",
             "-c",
+            # Shorten full absolute file paths in the generated code (like the
+            # __FILE__ macro and assert failure messages) for reproducibility:
+            f"-ffile-prefix-map={CPYTHON}=.",
+            f"-ffile-prefix-map={tempdir}=.",
             # This debug info isn't necessary, and bloats out the JIT'ed code.
             # We *may* be able to re-enable this, process it, and JIT it for a
             # nicer debugging experience... but that needs a lot more research:
@@ -167,7 +169,12 @@ async def _build_stencils(self) -> dict[str, _stencils.StencilGroup]:
                     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}
+        stencil_groups = {task.get_name(): task.result() for task in tasks}
+        for stencil_group in stencil_groups.values():
+            stencil_group.process_relocations(
+                known_symbols=self.known_symbols, alignment=self.alignment
+            )
+        return stencil_groups
 
     def build(
         self, out: pathlib.Path, *, comment: str = "", force: bool = False

Tools/jit/_writer.py

@@ -77,6 +77,6 @@ def dump(
     groups: dict[str, _stencils.StencilGroup], symbols: dict[str, int]
 ) -> typing.Iterator[str]:
     """Yield a JIT compiler line-by-line as a C header file."""
-    for opname, group in sorted(groups.items()):
+    for opname, group in groups.items():
         yield from _dump_stencil(opname, group)
     yield from _dump_footer(groups, symbols)

Tools/jit/build.py

@@ -8,7 +8,7 @@
 import _targets
 
 if __name__ == "__main__":
-    comment = f"$ {shlex.join([sys.executable] + sys.argv)}"
+    comment = f"$ {shlex.join([pathlib.Path(sys.executable).name] + sys.argv)}"
     parser = argparse.ArgumentParser(description=__doc__)
     parser.add_argument(
         "target", type=_targets.get_target, help="a PEP 11 target triple to compile for"