summaryrefslogtreecommitdiffstats
path: root/Python/jit.c
Commit message (Collapse)AuthorAgeFilesLines
* JIT: Rename trampoline.c to shim.c (#142974)Diego Russo2025-12-191-9/+9
|
* gh-142419: Add mmap.set_name method for user custom annotation (gh-142480)Donghee Na2025-12-181-1/+1
|
* GH-142621: JIT: Avoid memory load for symbols within 4GB on AArch64 (GH-142820)Mark Shannon2025-12-171-0/+9
|
* GH-135379: Top of stack caching for the JIT. (GH-135465)Mark Shannon2025-12-111-3/+3
| | | | Uses three registers to cache values at the top of the evaluation stack This significantly reduces memory traffic for smaller, more common uops.
* GH-142305: JIT: Deduplicating GOT symbols in the trace (#142316)Diego Russo2025-12-101-26/+54
|
* GH-139757: JIT: Remove redundant branches to jumps in the assembly optimizer ↵Mark Shannon2025-12-081-1/+2
| | | | | | | | | (GH-140800) JIT: Remove redundant branches to jump in the assembly optimizer * Refactor JIT assembly optimizer making instructions instances not just strings * Remove redundant jumps and branches where legal to do so * Modifies _BINARY_OP_SUBSCR_STR_INT to avoid excessive inlining depth
* gh-141770: Annotate anonymous mmap usage if "-X dev" is used (gh-142079)Donghee Na2025-12-081-0/+4
|
* GH-141794: Limit size of generated machine code. (GH-142228)Mark Shannon2025-12-031-0/+4
| | | | | | * Factor out bodies of the largest uops, to reduce jit code size. * Factor out common assert, also reducing jit code size. * Limit size of jitted code for a single executor to 1MB.
* gh-139109: A new tracing JIT compiler frontend for CPython (GH-140310)Ken Jin2025-11-131-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This PR changes the current JIT model from trace projection to trace recording. Benchmarking: better pyperformance (about 1.7% overall) geomean versus current https://raw.githubusercontent.com/facebookexperimental/free-threading-benchmarking/refs/heads/main/results/bm-20251108-3.15.0a1%2B-7e2bc1d-JIT/bm-20251108-vultr-x86_64-Fidget%252dSpinner-tracing_jit-3.15.0a1%2B-7e2bc1d-vs-base.svg, 100% faster Richards on the most improved benchmark versus the current JIT. Slowdown of about 10-15% on the worst benchmark versus the current JIT. **Note: the fastest version isn't the one merged, as it relies on fixing bugs in the specializing interpreter, which is left to another PR**. The speedup in the merged version is about 1.1%. https://raw.githubusercontent.com/facebookexperimental/free-threading-benchmarking/refs/heads/main/results/bm-20251112-3.15.0a1%2B-f8a764a-JIT/bm-20251112-vultr-x86_64-Fidget%252dSpinner-tracing_jit-3.15.0a1%2B-f8a764a-vs-base.svg Stats: 50% more uops executed, 30% more traces entered the last time we ran them. It also suggests our trace lengths for a real trace recording JIT are too short, as a lot of trace too long aborts https://github.com/facebookexperimental/free-threading-benchmarking/blob/main/results/bm-20251023-3.15.0a1%2B-eb73378-CLANG%2CJIT/bm-20251023-vultr-x86_64-Fidget%252dSpinner-tracing_jit-3.15.0a1%2B-eb73378-pystats-vs-base.md . This new JIT frontend is already able to record/execute significantly more instructions than the previous JIT frontend. In this PR, we are now able to record through custom dunders, simple object creation, generators, etc. None of these were done by the old JIT frontend. Some custom dunders uops were discovered to be broken as part of this work gh-140277 The optimizer stack space check is disabled, as it's no longer valid to deal with underflow. Pros: * Ignoring the generated tracer code as it's automatically created, this is only additional 1k lines of code. The maintenance burden is handled by the DSL and code generator. * `optimizer.c` is now significantly simpler, as we don't have to do strange things to recover the bytecode from a trace. * The new JIT frontend is able to handle a lot more control-flow than the old one. * Tracing is very low overhead. We use the tail calling interpreter/computed goto interpreter to switch between tracing mode and non-tracing mode. I call this mechanism dual dispatch, as we have two dispatch tables dispatching to each other. Specialization is still enabled while tracing. * Better handling of polymorphism. We leverage the specializing interpreter for this. Cons: * (For now) requires tail calling interpreter or computed gotos. This means no Windows JIT for now :(. Not to fret, tail calling is coming soon to Windows though https://github.com/python/cpython/pull/139962 Design: * After each instruction, the `record_previous_inst` function/label is executed. This does as the name suggests. * The tracing interpreter lowers bytecode to uops directly so that it can obtain "fresh" values at the point of lowering. * The tracing version behaves nearly identical to the normal interpreter, in fact it even has specialization! This allows it to run without much of a slowdown when tracing. The actual cost of tracing is only a function call and writes to memory. * The tracing interpreter uses the specializing interpreter's deopt to naturally form the side exit chains. This allows it to side exit chain effectively, without repeating much code. We force a re-specializing when tracing a deopt. * The tracing interpreter can even handle goto errors/exceptions, but I chose to disable them for now as it's not tested. * Because we do not share interpreter dispatch, there is should be no significant slowdown to the original specializing interpreter on tailcall and computed got with JIT disabled. With JIT enabled, there might be a slowdown in the form of the JIT trying to trace. * Things that could have dynamic instruction pointer effects are guarded on. The guard deopts to a new instruction --- `_DYNAMIC_EXIT`.
* GH-136895: Update JIT builds to use LLVM 20 (#140329)Savannah Ostrowski2025-11-031-14/+58
| | | Co-authored-by: Emma Harper Smith <emma@emmatyping.dev>
* gh-139269: Fix unaligned memory access in JIT code patching functions ↵Shamil2025-10-181-11/+19
| | | | | | | | (GH-139271) * Use memcpy for patching values instead of direct assignment Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
* GH-135904: Implement assembler optimization for AArch64. (GH-139855)Mark Shannon2025-10-171-5/+22
|
* GH-137959: Replace shim code in jitted code with a single trampoline ↵Mark Shannon2025-08-211-13/+65
| | | | function. (GH-137961)
* GH-136410: Faster side exits by using a cold exit stub (GH-136411)Mark Shannon2025-08-011-1/+1
|
* GH-134291: Support older macOS deployment targets for JIT builds (GH-137211)Brandt Bucher2025-07-301-14/+1
|
* GH-135904: Improve the JIT's performance on macOS (GH-136528)Brandt Bucher2025-07-141-4/+7
|
* gh-132661: Implement PEP 750 (#132662)Lysandros Nikolaou2025-04-301-0/+2
| | | | | | | | | | | | | Co-authored-by: Lysandros Nikolaou <lisandrosnik@gmail.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> Co-authored-by: Adam Turner <9087854+AA-Turner@users.noreply.github.com> Co-authored-by: Hugo van Kemenade <1324225+hugovk@users.noreply.github.com> Co-authored-by: Wingy <git@wingysam.xyz> Co-authored-by: Koudai Aono <koxudaxi@gmail.com> Co-authored-by: Dave Peck <davepeck@gmail.com> Co-authored-by: Terry Jan Reedy <tjreedy@udel.edu> Co-authored-by: Paul Everitt <pauleveritt@me.com> Co-authored-by: sobolevn <mail@sobolevn.me>
* gh-131238: Add missing pycore_function.h includes for JIT compiler (#131571)Victor Stinner2025-03-211-0/+1
|
* GH-131498: Remove conditional stack effects (GH-131499)Mark Shannon2025-03-201-0/+7
| | | * Adds some missing #includes
* GH-130956: Only emit AArch64 trampolines for long jumps (GH-131041)Diego Russo2025-03-191-1/+11
|
* GH-129231: Group executable JIT code in memory (GH-129232)Diego Russo2025-02-021-5/+5
|
* GH-128842: Collect JIT memory stats (GH-128941)Diego Russo2025-02-021-0/+8
|
* gh-129354: Fix grammar in PyErr_FormatUnraisable() (#129475)Victor Stinner2025-01-311-1/+2
| | | Replace "on verb+ing" with "while verb+ing".
* gh-129354: Use PyErr_FormatUnraisable() function (#129435)Victor Stinner2025-01-301-1/+1
| | | Replace PyErr_WriteUnraisable() with PyErr_FormatUnraisable().
* GH-126195: Use M1 JIT memory protection APIs (GH-126196)Diego Russo2024-11-121-2/+16
|
* GH-125911: Rename big trampoline to "shim" (GH-126339)Savannah Ostrowski2024-11-051-7/+5
|
* GH-125911: Call combine_symbol_mask on the initial trampoline of a trace ↵Nadeshiko Manju2024-10-251-0/+1
| | | | (GH-125973)
* gh-125207: Fix MSVC 1935 build with JIT (#125209)Michael Droettboom2024-10-181-1/+1
| | | | | | | * gh-125207: Use {0} array initializers * Simplify, as suggested in PR * Revert change to explicitly specify length
* GH-119726: Deduplicate AArch64 trampolines within a trace (GH-123872)Diego Russo2024-10-021-8/+84
|
* GH-116017: Get rid of _COLD_EXITs (GH-120960)Brandt Bucher2024-07-011-1/+1
|
* GH-118844: Fix build failures when combining --disable-gil with ↵Savannah Ostrowski2024-05-111-0/+1
| | | | --enable-experimental-jit (GH-118935)
* GH-113464: Generate a more efficient JIT (GH-118512)Brandt Bucher2024-05-031-311/+276
|
* GH-115802: Use the GHC calling convention in JIT code (GH-118287)Brandt Bucher2024-05-011-15/+37
|
* GH-118306: Update JIT to use LLVM 18 (GH-118307)Savannah Ostrowski2024-04-291-0/+3
|
* GH-117512: Allow 64-bit JIT operands on 32-bit platforms (GH-117527)Brandt Bucher2024-04-061-28/+34
|
* GH-116422: Tier2 hot/cold splitting (GH-116813)Mark Shannon2024-03-261-9/+28
| | | | | Splits the "cold" path, deopts and exits, from the "hot" path, reducing the size of most jitted instructions, at the cost of slower exits.
* GH-116017: Put JIT code and data on the same page (GH-116845)Brandt Bucher2024-03-191-29/+23
|
* GH-116134: JIT aarch64-pc-windows-msvc (GH-116130)Brandt Bucher2024-03-041-0/+7
|
* GH-115802: JIT "small" code for Windows (GH-115964)Brandt Bucher2024-02-291-11/+15
|
* GH-115802: JIT "small" code for macOS and Linux (GH-115826)Brandt Bucher2024-02-261-16/+103
|
* GH-112354: Initial implementation of warm up on exits and trace-stitching ↵Mark Shannon2024-02-201-4/+9
| | | | (GH-114142)
* GH-113464: Add a JIT backend for tier 2 (GH-113465)Brandt Bucher2024-01-291-0/+369
Add an option (--enable-experimental-jit for configure-based builds or --experimental-jit for PCbuild-based ones) to build an *experimental* just-in-time compiler, based on copy-and-patch (https://fredrikbk.com/publications/copy-and-patch.pdf). See Tools/jit/README.md for more information on how to install the required build-time tooling.
generated by cgit v0.12 at 2026-01-15 20:52:45 (GMT)