shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity
llvm-project/llvm/docs/ReleaseNotes.md
Chris Cotter a591a44653
[tsan] Introduce Adaptive Delay Scheduling to TSAN (#178836) 
This commit introduces an "adaptive delay" feature to the
ThreadSanitizer runtime to improve race detection by perturbing thread
schedules. At various synchronization points (atomic operations,
mutexes, and thread lifecycle events), the runtime may inject small
delays (spin loops, yields, or sleeps) to explore different thread
interleavings and expose data races that would otherwise occur only in
rare execution orders.

This change is inspired by prior work, which is discussed in more detail
on

https://discourse.llvm.org/t/rfc-tsan-implementing-a-fuzz-scheduler-for-tsan/80969.
In short, https://reviews.llvm.org/D65383 was an earlier unmerged
attempt at adding a random delays. Feedback on the RFC led to the
version in this commit, aiming to limit the amount of delay.

The adaptive delay feature uses a configurable time budget and tiered
sampling strategy to balance race exposure against performance impact.
It prioritizes high-value synchronization points with clear
happens-before relationships: relaxed atomics receive lightweight spin
delays with low sampling, synchronizing atomics (acquire / release /
seq_cst) receive moderate delays with higher sampling, and mutex and
thread lifecycle operations receive the longest delays with highest
sampling.

The feature is disabled by default and incurs minimal overhead when not
enabled. Nearly all checks are guarded by an inline check on a global
variable that is only set when enable_adaptive_delay=1. Microbenchmarks
with tight loops of atomic operations showed no meaningful performance
difference between an unmodified TSAN runtime and this version when
running with empty TSAN_OPTIONS.

An LLM assisted in writing portions of the adaptive delay logic,
including the TimeBudget class, tiering concept, address sampler, and
per-thread quota system. I reviewed the output and made amendments to
reduce duplication and simplify the behavior. I also replaced the LLM's
original double-based calculation logic with the integer-based Percent
class. The LLM also helped write unit test cases for Percent.

cc @dvyukov 

## Examples

I used the delay scheduler to find novel bugs that rarely or never
occurred with the unmodified TSAN runtime. Some of the bugs below were
found with earlier versions of the delay scheduler that I iterated on,
but with this most recent implementation in this PR, I can still find
the bugs far more reliably than with the standard TSAN runtime.

- A use-after-free in the
[BlazingMQ](https://github.com/bloomberg/blazingmq) broker during
ungraceful producer disconnect.
 - Race in stdexec: https://github.com/NVIDIA/stdexec/pull/1395
- Race in stdexec's MPSC queue:
https://github.com/NVIDIA/stdexec/pull/1812
- A few races in [BDE](https://github.com/bloomberg/bde) thread enabled
data structures/algorithms.
- The "Data race on variable a" test from
https://ceur-ws.org/Vol-2344/paper9.pdf is more reliably reproduced with
more aggressive adaptive scheduler options

# Outstanding work

- The
[RFC](https://discourse.llvm.org/t/rfc-tsan-implementing-a-fuzz-scheduler-for-tsan/80969)
suggests moving the scheduler to sanitizer_common, so that ASAN can
leverage this. This should be done (should it be done in this PR?).
 - Missing interceptors for libdispatch
2026-02-16 12:20:40 +01:00

7.3 KiB
Raw Blame History

LLVM {{env.config.release}} Release Notes

```{warning} These are in-progress notes for the upcoming LLVM {{env.config.release}}
             release. Release notes for previous releases can be found on
             [the Download Page](https://releases.llvm.org/download.html).
```

Introduction

This document contains the release notes for the LLVM Compiler Infrastructure, release {{env.config.release}}. Here we describe the status of LLVM, including major improvements from the previous release, improvements in various subprojects of LLVM, and some of the current users of the code. All LLVM releases may be downloaded from the LLVM releases web site.

For more information about LLVM, including information about the latest release, please check out the main LLVM web site. If you have questions or comments, the Discourse forums is a good place to ask them.

Note that if you are reading this file from a Git checkout or the main LLVM web page, this document applies to the next release, not the current one. To see the release notes for a specific release, please see the releases page.

Non-comprehensive list of changes in this release

  • ...

Changes to the LLVM IR

  • Removed llvm.convert.to.fp16 and llvm.convert.from.fp16 intrinsics. These are equivalent to fptrunc and fpext with half with a bitcast.

  • "denormal-fp-math" and "denormal-fp-math-f32" string attributes were migrated to first-class denormal_fpenv attribute.

  • The "nooutline" attribute is now writen as nooutline. Existing IR and bitcode will be automatically updated.

Changes to LLVM infrastructure

  • Removed Constant::isZeroValue. It was functionally identical to Constant::isNullValue for all types except floating-point negative zero. All callers should use isNullValue instead. isZeroValue will be reintroduced in the future with bitwise-all-zeros semantics to support non-zero null pointers.

  • Removed TypePromoteFloat legalization from SelectionDAG

Changes to building LLVM

Changes to TableGen

Changes to Interprocedural Optimizations

Changes to Vectorizers

Changes to the AArch64 Backend

Changes to the AMDGPU Backend

  • Initial support for gfx1310

Changes to the ARM Backend

Changes to the AVR Backend

Changes to the DirectX Backend

Changes to the Hexagon Backend

Changes to the LoongArch Backend

  • DWARF fission is now compatible with linker relaxations, allowing -gsplit-dwarf and -mrelax to be used together when building for the LoongArch platform.

Changes to the MIPS Backend

Changes to the NVPTX Backend

  • The default SM version has been changed from sm_30 to sm_75. sm_75 is the oldest GPU variant compatible with the widest range of recent major CUDA Toolkit versions (11/12/13).

Changes to the PowerPC Backend

Changes to the RISC-V Backend

  • llvm-objdump now has support for --symbolize-operands with RISC-V.
  • -mcpu=spacemit-x100 was added.
  • Change P extension version to match the 019 draft specification. Encoded in -march as 0p19.
  • Mnemonics for MOP/HINT-based instructions (lpad, pause, ntl.*, c.ntl.*, sspush, sspopchk, ssrdp, c.sspush, c.sspopchk) are now always available in the assembler and disassembler without requiring their respective extensions.
  • Adds experimental assembler support for the 'Zvabd` (RISC-V Integer Vector Absolute Difference) extension.
  • -mcpu=spacemit-a100 was added.

Changes to the WebAssembly Backend

Changes to the Windows Target

  • The .seh_startchained and .seh_endchained assembly instructions have been removed and replaced with a new .seh_splitchained instruction.

Changes to the X86 Backend

  • .att_syntax directive is now emitted for assembly files when AT&T syntax is in use. This matches the behaviour of Intel syntax and aids with compatibility when changing the default Clang syntax to the Intel syntax.

Changes to the OCaml bindings

Changes to the Python bindings

Changes to the C API

Changes to the CodeGen infrastructure

Changes to the Metadata Info

Changes to the Debug Info

Changes to the LLVM tools

  • llvm-objcopy no longer corrupts the symbol table when --update-section is called for ELF files.
  • FileCheck option -check-prefix now accepts a comma-separated list of prefixes, making it an alias of the existing -check-prefixes option.

Changes to LLDB

FreeBSD

Userspace Debugging

  • Support for MIPS64 has been removed.
  • The minimum assumed FreeBSD version is now 14. The effect of which is that watchpoints are assumed to be supported.

Kernel Debugging

  • The crashed thread is now automatically selected on start.
  • Threads are listed in incrmental order by pid then by tid.

Changes to BOLT

Changes to Sanitizers

  • Add a random delay into ThreadSanitizer to help find rare thread interleavings.

Other Changes

External Open Source Projects Using LLVM {{env.config.release}}

Additional Information

A wide variety of additional information is available on the LLVM web page, in particular in the documentation section. The web page also contains versions of the API documentation which is up-to-date with the Git version of the source code. You can access versions of these documents specific to this release by going into the llvm/docs/ directory in the LLVM tree.

If you have any questions or comments about LLVM, please feel free to contact us via the Discourse forums.