LLVM 20.0.0git Release Notes

Warning

These are in-progress notes for the upcoming LLVM 20.0.0git release. Release notes for previous releases can be found on the Download Page.

Introduction

This document contains the release notes for the LLVM Compiler Infrastructure, release 20.0.0git. 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

  • Added a new IRNormalizer pass which aims to transform LLVM modules into a normal form by reordering and renaming instructions while preserving the same semantics. The normalizer makes it easier to spot semantic differences when diffing two modules which have undergone different passes.

  • …

Changes to the LLVM IR

  • Types are no longer allowed to be recursive.

  • The x86_mmx IR type has been removed. It will be translated to the standard vector type <1 x i64> in bitcode upgrade.

  • Renamed llvm.experimental.stepvector intrinsic to llvm.stepvector.

  • Added usub_cond and usub_sat operations to atomicrmw.

  • Introduced noalias.addrspace metadata.

  • Remove the following intrinsics which can be replaced with a bitcast:

    • llvm.nvvm.bitcast.f2i

    • llvm.nvvm.bitcast.i2f

    • llvm.nvvm.bitcast.d2ll

    • llvm.nvvm.bitcast.ll2d

  • Remove the following intrinsics which can be replaced with a funnel-shift:

    • llvm.nvvm.rotate.b32

    • llvm.nvvm.rotate.right.b64

    • llvm.nvvm.rotate.b64

  • Remove the following intrinsics which can be replaced with an addrspacecast:

    • llvm.nvvm.ptr.gen.to.global

    • llvm.nvvm.ptr.gen.to.shared

    • llvm.nvvm.ptr.gen.to.constant

    • llvm.nvvm.ptr.gen.to.local

    • llvm.nvvm.ptr.global.to.gen

    • llvm.nvvm.ptr.shared.to.gen

    • llvm.nvvm.ptr.constant.to.gen

    • llvm.nvvm.ptr.local.to.gen

  • Remove the following intrinsics which can be relaced with a load from addrspace(1) with an !invariant.load metadata

    • llvm.nvvm.ldg.global.i

    • llvm.nvvm.ldg.global.f

    • llvm.nvvm.ldg.global.p

  • Operand bundle values can now be metadata strings.

  • Fast math flags are now permitted on fptrunc and fpext.

Changes to LLVM infrastructure

Changes to building LLVM

  • Raised the minimum MSVC version to Visual Studio 2019 16.8.

Changes to TableGen

Changes to Interprocedural Optimizations

Changes to the AArch64 Backend

  • .balign N, 0, .p2align N, 0, .align N, 0 in code sections will now fill the required alignment space with a sequence of 0x0 bytes (the requested fill value) rather than NOPs.

  • Assembler/disassembler support has been added for Armv9.6-A (2024) architecture extensions.

  • Added support for the FUJITSU-MONAKA CPU.

Changes to the AMDGPU Backend

  • Removed llvm.amdgcn.flat.atomic.fadd and llvm.amdgcn.global.atomic.fadd intrinsics. Users should use the atomicrmw instruction with fadd and addrspace(0) or addrspace(1) instead.

Changes to the ARM Backend

  • .balign N, 0, .p2align N, 0, .align N, 0 in code sections will now fill the required alignment space with a sequence of 0x0 bytes (the requested fill value) rather than NOPs.

  • The default behavior for frame pointers in leaf functions has been updated. When the -fno-omit-frame-pointer option is specified, FPKeepKindStr is set to -mframe-pointer=all, meaning the frame pointer (FP) is now retained in leaf functions by default. To eliminate the frame pointer in leaf functions, you must explicitly use the -momit-leaf-frame-pointer option.

  • When using the MOVT or MOVW instructions, the Assembler will now check to ensure that any addend that is used is within a 16-bit signed value range. If the addend falls outside of this range, the LLVM backend will emit an error like so Relocation Not In Range.

Changes to the AVR Backend

Changes to the DirectX Backend

Changes to the Hexagon Backend

Changes to the LoongArch Backend

Changes to the MIPS Backend

Changes to the PowerPC Backend

  • The Linux ppc64 LLC default cpu is updated from ppc to ppc64.

  • The AIX LLC default cpu is updated from generic to pwr7.

Changes to the RISC-V Backend

  • .balign N, 0, .p2align N, 0, .align N, 0 in code sections will now fill the required alignment space with a sequence of 0x0 bytes (the requested fill value) rather than NOPs.

  • Added Syntacore SCR4 and SCR5 CPUs: -mcpu=syntacore-scr4/5-rv32/64

  • -mcpu=sifive-p470 was added.

  • Added Hazard3 CPU as taped out for RP2350: -mcpu=rp2350-hazard3 (32-bit only).

  • Fixed length vector support using RVV instructions now requires VLEN>=64. This means Zve32x and Zve32f will also require Zvl64b. The prior support was largely untested.

  • The Zvbc32e and Zvkgs extensions are now supported experimentally.

  • Added Smctr, Ssctr and Svvptc extensions.

  • -mcpu=syntacore-scr7 was added.

  • -mcpu=tt-ascalon-d8 was added.

  • -mcpu=mips-p8700 was added.

  • -mcpu=sifive-p550 was added.

  • The Zacas extension is no longer marked as experimental.

  • Added Smdbltrp, Ssdbltrp extensions to -march.

  • The Smmpm, Smnpm, Ssnpm, Supm, and Sspm pointer masking extensions are no longer marked as experimental.

  • The Sha extension is now supported.

  • The RVA23U64, RVA23S64, RVB23U64, and RVB23S64 profiles are no longer marked as experimental.

  • .insn <length>, <raw encoding> can be used to assemble 48- and 64-bit instructions from raw integer values.

  • .insn [<length>,] <raw encoding> now accepts absolute expressions for both expressions, so that they can be computed from constants and absolute symbols.

  • The following new inline assembly constraints and modifiers are accepted:

    • cr constraint meaning an RVC-encoding compatible GPR (x8-x15)

    • cf constraint meaning an RVC-encoding compatible FPR (f8-f15)

    • R constraint meaning an even-odd GPR pair (prints as the even register, but both registers in the pair are considered live).

    • N modifer meaning print the register encoding (0-31) rather than the name.

  • f and cf inline assembly constraints, when using F-/D-/H-in-X extensions, will use the relevant GPR rather than FPR. This makes inline assembly portable between e.g. F and Zfinx code.

  • Adds experimental assembler support for the Qualcomm uC ‘Xqcicsr` (CSR) extension.

  • Adds experimental assembler support for the Qualcomm uC ‘Xqcisls` (Scaled Load Store) extension.

  • Adds experimental assembler support for the Qualcomm uC ‘Xqcia` (Arithmetic) extension.

  • Adds experimental assembler support for the Qualcomm uC ‘Xqciac` (Load-Store Address Calculation) extension.

  • Adds experimental assembler support for the Qualcomm uC ‘Xqcics` (Conditonal Select) extension.

  • Adds experimental assembler support for the Qualcomm uC ‘Xqcilsm` (Load Store Multiple) extension.

  • Adds experimental assembler support for the Qualcomm uC ‘Xqcicli` (Conditional Load Immediate) extension.

  • Adds experimental assembler support for the Qualcomm uC ‘Xqcicm` (Conditonal Move) extension.

  • Adds experimental assembler support for the Qualcomm uC ‘Xqciint` (Interrupts) extension.

  • Added Sdext and Sdtrig extensions.

Changes to the WebAssembly Backend

  • The default target CPU, “generic”, now enables the -mnontrapping-fptoint and -mbulk-memory flags, which correspond to the Bulk Memory Operations and Non-trapping float-to-int Conversions language features, which are widely implemented in engines.

  • A new Lime1 target CPU is added, -mcpu=lime1. This CPU follows the definition of the Lime1 CPU here, and enables -mmultivalue, -mmutable-globals, -mcall-indirect-overlong, -msign-ext, -mbulk-memory-opt, -mnontrapping-fptoint, and -mextended-const.

  • Support for the new standardized Exception Handling proposal is added. The legacy Exception Handling proposal is still supported, and turned on by the newly added -wasm-use-legacy-eh option. Given that major web browsers still default to the legacy EH proposal, this option is turned on by default for the moment.

Changes to the Windows Target

Changes to the X86 Backend

  • .balign N, 0x90, .p2align N, 0x90, and .align N, 0x90 in code sections now fill the required alignment space with repeating 0x90 bytes, rather than using optimised NOP filling. Optimised NOP filling fills the space with NOP instructions of various widths, not just those that use the 0x90 byte encoding. To use optimised NOP filling in a code section, leave off the “fillval” argument, i.e. .balign N, .p2align N or .align N respectively.

  • Due to the removal of the x86_mmx IR type, functions with x86_mmx arguments or return values will use a different, incompatible, calling convention ABI. Such functions are not generally seen in the wild (Clang never generates them!), so this is not expected to result in real-world compatibility problems.

  • Support ISA of AVX10.2-256 and AVX10.2-512.

  • Supported instructions of MOVRS AND AVX10.2

  • Supported ISA of SM4(EVEX).

  • Supported ISA of MSR_IMM.

  • Supported -mcpu=diamondrapids

  • Supported emitting relocation types for x86-64 target:

    • R_X86_64_CODE_4_GOTPCRELX

    • R_X86_64_CODE_4_GOTTPOFF

    • R_X86_64_CODE_4_GOTPC32_TLSDESC

    • R_X86_64_CODE_6_GOTTPOFF

Changes to the OCaml bindings

Changes to the Python bindings

Changes to the C API

  • The following symbols are deleted due to the removal of the x86_mmx IR type:

    • LLVMX86_MMXTypeKind

    • LLVMX86MMXTypeInContext

    • LLVMX86MMXType

  • The following functions are added to further support non-null-terminated strings:

  • LLVMGetNamedFunctionWithLength

  • LLVMGetNamedGlobalWithLength

  • The following functions are added to access the LLVMContextRef associated with LLVMValueRef and LLVMBuilderRef objects:

    • LLVMGetValueContext

    • LLVMGetBuilderContext

  • The new pass manager can now be invoked with a custom alias analysis pipeline, using the LLVMPassBuilderOptionsSetAAPipeline function.

  • It is now also possible to run the new pass manager on a single function, by calling LLVMRunPassesOnFunction instead of LLVMRunPasses.

  • Support for creating instructions with custom synchronization scopes has been added:

    • LLVMGetSyncScopeID to map a synchronization scope name to an ID.

    • LLVMBuildFenceSyncScope, LLVMBuildAtomicRMWSyncScope and LLVMBuildAtomicCmpXchgSyncScope versions of the existing builder functions with an additional synchronization scope ID parameter.

    • LLVMGetAtomicSyncScopeID and LLVMSetAtomicSyncScopeID to get and set the synchronization scope of any atomic instruction.

    • LLVMIsAtomic to check if an instruction is atomic, for use with the above functions. Because of backwards compatibility, LLVMIsAtomicSingleThread and LLVMSetAtomicSingleThread continue to work with any instruction type.

  • The LLVMSetPersonalityFn and LLVMSetInitializer APIs now support clearing the personality function and initializer respectively by passing a null pointer.

  • The following functions are added to allow iterating over debug records attached to instructions:

    • LLVMGetFirstDbgRecord

    • LLVMGetLastDbgRecord

    • LLVMGetNextDbgRecord

    • LLVMGetPreviousDbgRecord

  • Added LLVMAtomicRMWBinOpUSubCond and LLVMAtomicRMWBinOpUSubSat to LLVMAtomicRMWBinOp enum for AtomicRMW instructions.

Changes to the CodeGen infrastructure

Changes to the Metadata Info

Changes to the Debug Info

Changes to the LLVM tools

  • llvm-objcopy now supports the following options for Mach-O: --globalize-symbol, --globalize-symbols, --keep-global-symbol, --keep-global-symbols, --localize-symbol, --localize-symbols, --skip-symbol, --skip-symbols.

Changes to LLDB

  • It is now recommended that LLDB be built with Python >= 3.8, but no changes have been made to the supported Python versions. The next release, LLDB 21, will require Python >= 3.8.

  • LLDB now supports inline diagnostics for the expression evaluator and command line parser.

    Old:

    (lldb) p a+b
error: <user expression 0>:1:1: use of undeclared identifier 'a'
    1 | a+b
      | ^
error: <user expression 0>:1:3: use of undeclared identifier 'b'
    1 | a+b
      |   ^

    New:

    (lldb) p a+b
         ˄ ˄
         │ ╰─ error: use of undeclared identifier 'b'
         ╰─ error: use of undeclared identifier 'a'

  • LLDB can now read the fpmr register from AArch64 Linux processes and core files.

  • Program stdout/stderr redirection will now open the file with O_TRUNC flag, make sure to truncate the file if path already exists.

    • eg. settings set target.output-path/target.error-path <path/to/file>

  • A new setting target.launch-working-dir can be used to set a persistent cwd that is used by default by process launch and run.

  • LLDB now parses shared libraries in parallel, resulting in an average 2x speedup when attaching (only available on Darwin platforms) and launching (available on all platforms).

  • It is now possible to implement lldb commands in Python that use lldb’s native command-line parser. In particular, that allows per-option/argument completion, with all the basic completers automatically supported and auto-generated help. The command template file in lldb/examples/python/cmdtemplate.py has been updated to show how to use this.

  • Breakpoints on “inlined call sites” are now supported. Previous to this fix, breakpoints on source lines that only contained inlined call sites would be moved to the next source line, causing you to miss the inlined executions.

  • On the command line, LLDB now limits tab completions to your terminal width to avoid wrapping.

    Old:

    Available completions:
        _regexp-attach    -- Attach to process by ID or name.
        _regexp-break     -- Set a breakpoint using one of several shorthand
formats.
        _regexp-bt        -- Show backtrace of the current thread's call sta
ck. Any numeric argument displays at most that many frames. The argument 'al
l' displays all threads. Use 'settings set frame-format' to customize the pr
inting of individual frames and 'settings set thread-format' to customize th
e thread header. Frame recognizers may filter thelist. Use 'thread backtrace
-u (--unfiltered)' to see them all.
        _regexp-display   -- Evaluate an expression at every stop (see 'help
target stop-hook'.)

    New:

    Available completions:
        _regexp-attach    -- Attach to process by ID or name.
        _regexp-break     -- Set a breakpoint using one of several shorth...
        _regexp-bt        -- Show backtrace of the current thread's call ...
        _regexp-display   -- Evaluate an expression at every stop (see 'h...

  • DWARF indexing speed (for binaries not using the debug_names index) increased by 30-60%.

  • Minidumps generated by LLDB now support:

    • 64 bit memory (due to 64b support, Minidumps are now paged to disk while being written).

    • Capturing of TLS variables.

    • Multiple signals or exceptions, including breakpoints.

  • New Core File API. This gives greater control on the data captured into the core file, relative to the existing process save-core styles.

  • When opening ELF core files, LLDB will print additional information about the signal that killed the process and the disassembly view will display actual (relocated) targets of the jump instructions instead of raw offsets encoded in the instruction. This matches existing behavior for live processes.

    Old:

    * thread #1: tid = 329384, 0x0000000000401262, name = 'a.out', stop reason = signal SIGSEGV

0x7f1e3193e0a7 <+23>:  ja     0xfe100        ; <+112>

    New:

    * thread #1: tid = 329384, 0x0000000000401262, name = 'a.out', stop reason = SIGSEGV: address not mapped to object (fault address: 0x0)

0x7f1e3193e0a7 <+23>:  ja     0x7f1e3193e100 ; <+112>

  • lldb-server now listens to a single port for gdbserver connections and provides that port to the connection handler processes. This means that only 2 ports need to be opened in the firewall (one for the lldb-server platform, one for gdbserver connections). In addition, due to this work, lldb-server now works on Windows in the server mode.

Changes to BOLT

Changes to Sanitizers

Other Changes

External Open Source Projects Using LLVM 20.0.0git

  • A project…

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.