LLVM 20.1.7 Release Notes¶
Introduction¶
This document contains the release notes for the LLVM Compiler Infrastructure, release 20.1.7. 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.
The SPIR-V backend is now an official LLVM target, providing OpenCL and SYCL conformance and establishing a foundation for broader applicability to other APIs, including Vulkan, GLSL, and HLSL. This backend aims to offer a unified approach for diverse compute and graphics workloads, providing a robust alternative to the Khronos SPIR-V LLVM Translator.
…
Changes to the LLVM IR¶
Types are no longer allowed to be recursive.
The
x86_mmxIR type has been removed. It will be translated to the standard vector type<1 x i64>in bitcode upgrade.Renamed
llvm.experimental.stepvectorintrinsic tollvm.stepvector.Added
usub_condandusub_satoperations toatomicrmw.Introduced
noalias.addrspacemetadata.Remove the following intrinsics which can be replaced with a
bitcast:llvm.nvvm.bitcast.f2illvm.nvvm.bitcast.i2fllvm.nvvm.bitcast.d2llllvm.nvvm.bitcast.ll2d
Remove the following intrinsics which can be replaced with a funnel-shift:
llvm.nvvm.rotate.b32llvm.nvvm.rotate.right.b64llvm.nvvm.rotate.b64
Remove the following intrinsics which can be replaced with an
addrspacecast:llvm.nvvm.ptr.gen.to.globalllvm.nvvm.ptr.gen.to.sharedllvm.nvvm.ptr.gen.to.constantllvm.nvvm.ptr.gen.to.localllvm.nvvm.ptr.global.to.genllvm.nvvm.ptr.shared.to.genllvm.nvvm.ptr.constant.to.genllvm.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.illvm.nvvm.ldg.global.fllvm.nvvm.ldg.global.p
Operand bundle values can now be metadata strings.
Fast math flags are now permitted on
fptruncandfpext.
Changes to LLVM infrastructure¶
Two methods that use Instruction pointers as code positions (moveBefore, getFirstNonPHI) have been deprecated in favour of overloads and variants that use
BasicBlock::iterators instead. The pointer-flavoured methods will be removed in a future release. This work is part of the RemoveDIs project, the documentation for which contains instructions for updating call-sites using the deprecated methods.
Changes to building LLVM¶
Raised the minimum MSVC version to Visual Studio 2019 16.8.
Deprecated support for building compiler-rt with
LLVM_ENABLE_PROJECTS. Users should instead useLLVM_ENABLE_RUNTIMES, either through the runtimes or the bootstrapping build.Deprecated support for building libc with
LLVM_ENABLE_PROJECTS. Users should instead useLLVM_ENABLE_RUNTIMES, either through the runtimes or the bootstrapping build.
Changes to TableGen¶
The ARMTargetDefEmitter now binds Funtion Multi Versioning features to the corresponding AArch64 Architecture Extensions such that their dependencies can be autogenerated using TableGen.
Changes to Interprocedural Optimizations¶
Added RelLookupTableConverterPass to LTO post-link pass pipeline.
Changes to the AArch64 Backend¶
.balign N, 0,.p2align N, 0,.align N, 0in code sections will now fill the required alignment space with a sequence of0x0bytes (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.
Updated feature dependency in Armv9.6 for FEAT_FAMINMAX, FEAT_LUT and FEAT_FP8, now they depend only on FEAT_NEON.
Changes to the AMDGPU Backend¶
Initial support for gfx950
Improved
llvm.memcpy,llvm.memmoveandllvm.memsetloweringFixed expansion of 64-bit flat address space
atomicrmwandcmpxchgoperations which may access private memory.noalias.addrspacemetadat may be used to avoid the expansion if the target address is known to not be on the stack.Fix compile failures when emitting unreachable functions.
Removed
llvm.amdgcn.flat.atomic.faddandllvm.amdgcn.global.atomic.faddintrinsics. Users should use the atomicrmw instruction withfaddand addrspace(0) or addrspace(1) instead.
Changes to the ARM Backend¶
.balign N, 0,.p2align N, 0,.align N, 0in code sections will now fill the required alignment space with a sequence of0x0bytes (the requested fill value) rather than NOPs.The default behavior for frame pointers in leaf functions has been updated. When the
-fno-omit-frame-pointeroption is specified,FPKeepKindStris 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-pointeroption.When using the
MOVTorMOVWinstructions, 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 soRelocation Not In Range.
Changes to the AVR Backend¶
Changes to the DirectX Backend¶
Changes to the Hexagon Backend¶
The default Hexagon architecture version in ELF object files produced by the tools such as llvm-mc is changed to v68. This version will be set if the user does not provide the CPU version in the command line.
Changes to the LoongArch Backend¶
Incorrect GOT usage for
non-dso_localfunction calls in large code model is fixed.A gprof support issue is fixed.
A SDAG hang issue caused by
ISD::CONCAT_VECTORSis fixed.A compiler crash issue when converting
halftoi32is fixed.Almost all of
la64v1.1instructions can now be generated. The full list isfrecipe.s,frecipe.d,frsqrte.s,frsqrte.d,vfrecipe.s,vfrecipe.d,vfrsqrte.s,vfrsqrte.d,xvfrecipe.s,xvfrecipe.d,xvfrsqrte.s,xvfrsqrte.d,sc.q,amcas.b,amcas.h,amcas.w,amcas.d,amcas_db.b,amcas_db.h,amcas_db.w,amcas_db.d,amswap.b,amswap.h,amswap_db.b,amswap_db.h,amadd.b,amadd.h,amadd_db.b,amadd_db.h. Optionally generate instructionsdbar 0x700,div.w,div.wu,mod.wandmod.wuwhen related target features are enabled.llacq.w,screl.w,llacq.dandscrel.dcannot be generated yet.An llc option called
-loongarch-annotate-tablejumpis added to annotate table jump instruction in the.discard.tablejump_annotatesection. A typical user of these annotations is theobjtoolin Linux kernel.The default cpu in
MCSubtargetInfois changed fromla464togeneric-la64. In addition, thelsxfeature is added togeneric-la64.CFI instructions now allow register names and aliases, previously only numbers were allowed.
RuntimeDyldnow supports LoongArch, which means that programs relying onMCJITcan now work..balign N, 0,.p2align N, 0,.align N, 0in code sections will now fill the required alignment space with a sequence of0x0bytes (the requested fill value) rather than NOPs.%ld_pcrel_20,%gd_pcrel_20and%desc_pcrel_20operand modifiers are supported by assembler.A machine function pass called
LoongArch Merge Base Offsetis added to merge the offset of address calculation into the offset field of instructions in a global address lowering sequence.The
LoopDataPrefetchpass can now work on LoongArch, but it is disabled by default due to the bad effect on Fortran benchmarks.Enable alias analysis by default.
Avoid indirect branch jumps using the
$raregister.Other optimizations.
Changes to the MIPS Backend¶
Changes to the PowerPC Backend¶
The Linux
ppc64LLC default cpu is updated fromppctoppc64.Replaced PPCMergeStringPool with GlobalMerge.
Disabled vsx and altivec when -msoft-float is used.
Added support for -mcpu=pwr11 -mtune=pwr11.
Implemented BCD assist builtins.
Expanded global named register support.
Updated to use tablegen’s MatchRegisterName().
Fixed saving of Link Register when using ROP Protect.
Fixed SUBREG_TO_REG handling in the RegisterCoalescer.
Fixed data layout alignment of i128 to 16.
Fixed codegen for transparent_union function parameters.
Added an error for incorrect use of memory operands.
Other various bug fixes and codegen improvements.
AIX Specific:
LLC default cpu is updated from
generictopwr7.Fixed handling in emitGlobalConstantImpl to emit aliases to subobjects at proper offsets.
Enabled aggressive merging of constants to reduce TOC entries.
Changes to the RISC-V Backend¶
.balign N, 0,.p2align N, 0,.align N, 0in code sections will now fill the required alignment space with a sequence of0x0bytes (the requested fill value) rather than NOPs.Added Syntacore SCR4 and SCR5 CPUs:
-mcpu=syntacore-scr4/5-rv32/64-mcpu=sifive-p470was 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
Zvbc32eandZvkgsextensions are now supported experimentally.Added
Smctr,SsctrandSvvptcextensions.-mcpu=syntacore-scr7was added.-mcpu=tt-ascalon-d8was added.-mcpu=mips-p8700was added.-mcpu=sifive-p550was added.The
Zacasextension is no longer marked as experimental.Added Smdbltrp, Ssdbltrp extensions to -march.
The
Smmpm,Smnpm,Ssnpm,Supm, andSspmpointer masking extensions are no longer marked as experimental.The
Shaextension 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:
crconstraint meaning an RVC-encoding compatible GPR (x8-x15)cfconstraint meaning an RVC-encoding compatible FPR (f8-f15)Rconstraint meaning an even-odd GPR pair (prints as the even register, but both registers in the pair are considered live).cRconstraint meaning an RVC-encoding compatible even-odd GPR Pair (prints as an even register betweenx8andx14, but both registers in the pair are considered live).Nmodifer meaning print the register encoding (0-31) rather than the name.
fandcfinline 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.
Adds experimental assembler support for the Qualcomm uC ‘Xqcilo` (Large Offset Load Store) extension.
Added
SdextandSdtrigextensions.
Changes to the SystemZ Backend¶
Added support for the IBM z17 processor and the arch15 cpu architecture.
Added support for
__builtin_setjumpand__builtin_longjmp.Improve inlining heuristics to fix compile time explosion in certain cases.
Improve various cost functions.
Improve compatibility of the assembler parser with the GNU assembler.
Changes to the WebAssembly Backend¶
The default target CPU, “generic”, now enables the
-mnontrapping-fptointand-mbulk-memoryflags, 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-ehoption. 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, 0x90in code sections now fill the required alignment space with repeating0x90bytes, rather than using optimised NOP filling. Optimised NOP filling fills the space with NOP instructions of various widths, not just those that use the0x90byte encoding. To use optimised NOP filling in a code section, leave off the “fillval” argument, i.e..balign N,.p2align Nor.align Nrespectively.Due to the removal of the
x86_mmxIR type, functions withx86_mmxarguments 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
AVX10.2-256andAVX10.2-512.Supported instructions of
MOVRS AND AVX10.2Supported ISA of
SM4(EVEX).Supported ISA of
MSR_IMM.Supported
-mcpu=diamondrapidsSupported emitting relocation types for x86-64 target:
R_X86_64_CODE_4_GOTPCRELXR_X86_64_CODE_4_GOTTPOFFR_X86_64_CODE_4_GOTPC32_TLSDESCR_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_mmxIR type:LLVMX86_MMXTypeKindLLVMX86MMXTypeInContextLLVMX86MMXType
The following functions are added to further support non-null-terminated strings:
LLVMGetNamedFunctionWithLengthLLVMGetNamedGlobalWithLengthThe following functions are added to access the
LLVMContextRefassociated withLLVMValueRefandLLVMBuilderRefobjects:LLVMGetValueContextLLVMGetBuilderContext
The new pass manager can now be invoked with a custom alias analysis pipeline, using the
LLVMPassBuilderOptionsSetAAPipelinefunction.It is now also possible to run the new pass manager on a single function, by calling
LLVMRunPassesOnFunctioninstead ofLLVMRunPasses.Support for creating instructions with custom synchronization scopes has been added:
LLVMGetSyncScopeIDto map a synchronization scope name to an ID.LLVMBuildFenceSyncScope,LLVMBuildAtomicRMWSyncScopeandLLVMBuildAtomicCmpXchgSyncScopeversions of the existing builder functions with an additional synchronization scope ID parameter.LLVMGetAtomicSyncScopeIDandLLVMSetAtomicSyncScopeIDto get and set the synchronization scope of any atomic instruction.LLVMIsAtomicto check if an instruction is atomic, for use with the above functions. Because of backwards compatibility,LLVMIsAtomicSingleThreadandLLVMSetAtomicSingleThreadcontinue to work with any instruction type.
The
LLVMSetPersonalityFnandLLVMSetInitializerAPIs 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:
LLVMGetFirstDbgRecordLLVMGetLastDbgRecordLLVMGetNextDbgRecordLLVMGetPreviousDbgRecord
Added
LLVMAtomicRMWBinOpUSubCondandLLVMAtomicRMWBinOpUSubSattoLLVMAtomicRMWBinOpenum for AtomicRMW instructions.
Changes to the CodeGen infrastructure¶
GlobalOpt can now statically resolve calls to multi-versioned functions when targeting AArch64. These calls would otherwise be routed through an IFunc resolver function. This optimization can be applied when the caller is either a multi-versioned function itself, or it is compiled with a sufficiently high set of architecture features (including the
targetattribute, and command line options).
Changes to the Metadata Info¶
Multi-versioned functions targeting AArch64 are annotated with new metadata named
fmv-features. The metadata string value consists of a comma-separated list of Function Multi Versioning feature names as defined in the Arm C Language Extensions (ACLE).
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.llvm-objcopy now prints the correct file path in the error message when the output file specified by
--dump-sectioncannot be opened.llvm-cxxfilt now supports demangling call expressions encoded using
cpinstead ofcl.llvm-objdump now supports printing the file header, load section header and auxiliary header for XCOFF object files under the
--private-headersoption.
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'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-dircan be used to set a persistent cwd that is used by default byprocess launchandrun.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.pyhas 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_namesindex) increased by 30-60%.The
frame diagnosenow works on ELF-based systems. After a crash, LLDB will try to determine the likely cause of the signal, matching Darwin behavior. This feature requires using a newlldb-serverversion and (like Darwin) only works on x86 binaries.* thread #1, name = 'a.out', stop reason = signal SIGSEGV: address not mapped to object (fault address=0x4) frame #0: 0x00005555555551aa a.out`GetSum(f=0x0000555555558018) at main.c:21:37 18 } 19 20 int GetSum(struct Foo *f) { -> 21 return SumTwoIntegers(f->a, f->b->d ? 0 : 1); 22 } 23 24 int main() { Likely cause: f->b->d accessed 0x4Minidumps 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-corestyles.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-servernow 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 thelldb-serverplatform, one for gdbserver connections). In addition, due to this work,lldb-servernow works on Windows in the server mode.LLDB can now read the
fpmrregister from AArch64 Linux processes and core files.Support was added for debugging AArch64 Linux programs that use the Guarded Control Stack extension (GCS). This includes live processes and core files.
LLDB now supports execution of user expressions for non-trivial cases for LoongArch and RISC-V targets, like function calls, when some code needs to be executed on the target.
LLDB now supports optionally enabled/disabled register sets (particularly floating point registers) for RISC-V 64. This happens for targets like
RV64IMACorRV64IMACV, that have no floating point registers. The change is applied to native debugging and core-file usage.LLDB now supports core-file for LoongArch.
LLDB now supports hardware breakpoint and watchpoint for LoongArch.
LLDB now supports vector registers for LoongArch when debugging a live process.
Incorrect floating-point register DWARF numbers for LoongArch were fixed.
Support was added for handling the GDB Remote Protocol
xpacket in the format introduced by GDB 16.2. LLDB currently uses a different format forxand LLDB is now able to handle both formats. At some point in the future support for LLDB’s format ofxwill be removed.
Changes to BOLT¶
Changes to Sanitizers¶
Changes to the Profile Runtime¶
On platforms where
atexit-registered functions are not called when a DSO isdlclose’d, a mechanism is added that implements this missing functionality for calls toatexitin the profile runtime. This is currently only enabled on AIX.
Other Changes¶
External Open Source Projects Using LLVM 20.1.7¶
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.
