Description

markshannon

opened on Oct 8, 2025

Issue body actions

For AArch64 emit__GUARD_IS_TRUE_POP produces the following assembly code:

...
14: 54000041 b.ne 0x1c <_JIT_ENTRY+0x1c>
18: 14000002 b 0x20 <_JIT_ENTRY+0x20>
1c: 14000000 b 0x1c <_JIT_ENTRY+0x1c>
000000000000001c: R_AARCH64_JUMP26 _JIT_JUMP_TARGET

but it should, ideally, emit this:

14: 54000041 b.ne 0x14 <_JIT_ENTRY+0x14>
000000000000001c: R_AARCH64_CONDBR19 _JIT_JUMP_TARGET

This doesn't seem to be an issue for x86, possibly due to the way Clang is setup for AArch64, possibly just an artefact of llvm's code gen.
Either way, rather than rely on llvm eliminating these jumps we can do it in the JIT builder's assembly optimizer.
Currently we don't perform jump fusion, but it would be easy enough to do so.

Linked PRs

• GH-139757: JIT: Remove redundant branches to jumps in the assembly optimizer #140800
• gh-139757: Treat call specially in JIT assembly backend optimizer on x86-64 and AArch64 #142907
• gh-139757: Add _GUARD_NOS_COMPACT_ASCII for compact ascii optimization #143332
• gh-139757: Fix unintended bytecode specialization for non-ascii string #143352
• gh-139757: Add BINARY_OP_SUBSCR_USTR_INT #143389

Activity

1. picnixz

   added

   type-featureA feature request or enhancementA feature request or enhancement

   performancePerformance or resource usagePerformance or resource usage

   interpreter-core(Objects, Python, Grammar, and Parser dirs)(Objects, Python, Grammar, and Parser dirs)

   topic-JIT

   and removed

   performancePerformance or resource usagePerformance or resource usage

   on Oct 8, 2025
2. markshannon

   added a commit that references this issue on Dec 4, 2025

   GH-139757: Fix reference leaks introduced in GH-140800 (GH-142257)

   6825d5c
3. StanFromIreland

   added a commit that references this issue on Dec 6, 2025

   pythonGH-139757: Fix reference leaks introduced in pythonGH-140800 (p…

   f3ecc76
4. markshannon

   added a commit that references this issue on Dec 8, 2025

   GH-139757: JIT: Remove redundant branches to jumps in the assembly op…

   e0451ce
5. Fidget-Spinner commented on Dec 17, 2025

   

   Fidget-Spinner

   on Dec 17, 2025

   Member

   More actions

   The introduction of _remove_unreachable broke the JIT + FT builds on x86-64.

   I get this error:

   0.	Program arguments: clang-21 --target=x86_64-unknown-linux-gnu -c -o /tmp/tmpbw0scowv/_LOAD_ATTR_WITH_HINT_r11.o /tmp/tmpbw0scowv/_LOAD_ATTR_WITH_HINT_r11.s
   Stack dump without symbol names (ensure you have llvm-symbolizer in your PATH or set the environment var `LLVM_SYMBOLIZER_PATH` to point to it):
   
   

   Commenting it out fixes it.
6. Fidget-Spinner commented on Dec 17, 2025

   

   Fidget-Spinner

   on Dec 17, 2025

   Member

   More actions

   Alright, I narrowed it down: for some stencils, there are more than one entry points, e.g. if PyStackRef_CLOSE is outlined, then a stencil may have two entry points (one _JIT_ENTRY, one PyStackRef_CLOSE).

   The problem with that is that we don't treat call PyStackRef_CLOSE as a jump to a live block. So this optimization breaks.

   A possible fix is to just treat all function entries as live.
7. Fidget-Spinner

   added a commit that references this issue on Dec 17, 2025

   gh-139757: Treat call specially in JIT assembly backend optimizer on …

   8b64dd8
8. corona10

   added 5 commits that reference this issue on Jan 1, 2026

   pythongh-139757: Add _GUARD_NOS_COMPACT_ASCII for compact ascii optim…

   54af89d

   Merge remote-tracking branch 'upstream/main' into pythongh-139757-gua…

   a86857d

   gh-139757: Add _GUARD_NOS_COMPACT_ASCII for compact ascii optimization (

   315f474

   pythongh-139757: Fix unintended BINARY_OP_SUBSCR_STR_INT for non-ASCII

   1b6fa3c

   gh-139757: Fix unintended bytecode specialization for non-ascii string (

   61f2ad9
9. Fidget-Spinner

   added a commit that references this issue on Jan 4, 2026

   gh-139757: Add BINARY_OP_SUBSCR_USTR_INT (GH-143389)

   e6bfe4d
10. thunder-coding

    added 3 commits that reference this issue on Feb 15, 2026

    pythongh-139757: Add _GUARD_NOS_COMPACT_ASCII for compact ascii optim…

    9438b06

    pythongh-139757: Fix unintended bytecode specialization for non-ascii…

    5c6da5c

    pythongh-139757: Add BINARY_OP_SUBSCR_USTR_INT (pythonGH-143389)

    b773aae
11. markshannon commented on Mar 23, 2026

    

    markshannon

    on Mar 23, 2026

    MemberAuthor

    More actions

    I think this is done
12. markshannon

    closed this as completedon Mar 23, 2026

Issue actions