linux/poky.git/meta/recipes-devtools/llvm, branch yocto-5.0.8

llvm: reduce size of -dbg package

2024-11-26T14:11:30+00:00

Unless DEBUG_BUILD is enabled, pass -g1 to massively reduce the size of the debug symbols Level 1 produces minimal information, enough for making backtraces in parts of the program that you don't plan to debug. This includes descriptions of functions and external variables, and line number tables, but no information about local variables. This makes the sstate objects a lot more manageable, and packaging faster. (From OE-Core rev: dac630ab5ee7aa6c5c7c294093adbd11b116c765) Signed-off-by: Steve Sakoman

llvm: Enable libllvm for native build

2024-08-26T12:18:44+00:00

The bpftrace recipe under meta-clang[1] needs llvm-objcopy [2] during do_confgure phase otherwise there comes below error: | CMake Error at tests/data/CMakeLists.txt:6 (find_program): | Could not find LLVM_OBJCOPY using the following names: llvm-objcopy, | llvm-objcopy-18, llvm18-objcopy The commit ec22bfa67f llvm: allow building libllvm in native builds, subject to PACKAGECONFIG [3] introduces libllvm to manage the llvm function for native build maybe because there is only mesa-native to use llvm-native. Considering there are other recipes such as bpftrace needs llvm-native, so enable libllvm for native build. [1] https://github.com/kraj/meta-clang [2] https://github.com/bpftrace/bpftrace/blob/master/tests/data/CMakeLists.txt [3] https://git.openembedded.org/openembedded-core/commit/?id=ec22bfa67f6f1766102501d4593ce29aafe8c166 (From OE-Core rev: bd451b9cd951778cc6cdb0fe13f87c723c153d27) Signed-off-by: Mingli Yu Signed-off-by: Richard Purdie Signed-off-by: Steve Sakoman

llvm: upgrade 18.1.5 -> 18.1.6

2024-07-17T12:36:13+00:00

Changelog: ========== - Fixes issues where LLVM is either generating the incorrect thunk for a function with aligned parameters or didn't correctly pass through the return value when StructRet was used. - -Xclang -target-feature -Xclang +unaligned-scalar-mem can be used to enable unaligned scalar memory accesses for CPUs that do not support unaligned vector accesses. -mno-strict-align will enable unaligned scalar and vector memory accesses. - Don't replace an aliasee with an alias that has weak linkage. This avoids incorrect linkage that can lead to using the wrong symbols during linking time. - This patch fixes build failures when compiling AVX512 code using -march=native on machines without AVX512. - Fixes crash in AArch64 backend when having true or false as operand for a fcmp instruction on IR level. - Fixes compiler crash when user specifies -mno-evex512 with AVX512 features but no AVX512VL. - Fixes a bug that tries to do VBROADCAST_LOAD for f16 without AVX2. (From OE-Core rev: 941474ed77f6f5397ff4f83a4e4dae1c3b9103d3) Signed-off-by: Wang Mingyu Signed-off-by: Alexandre Belloni Signed-off-by: Richard Purdie (cherry picked from commit 3cd5c40f5736506b2cfc23b180fa915b01d8220c) Signed-off-by: Steve Sakoman

llvm: Fix CVE-2024-0151

2024-07-09T13:02:55+00:00

(From OE-Core rev: d89c0df4316a02d9b1857bf8844adfa7dbb315d5) Signed-off-by: Deepthi Hemraj Signed-off-by: Steve Sakoman

llvm: Switch to using release tarballs

2024-06-05T12:57:12+00:00

git checkouts are in excess of 3G, which is not ideal for everyone to download/clone, instead switch to fetching release tarball which is ~126M as of 18.1.5 release (From OE-Core rev: 800e6576e4f3af10846af13c2f217f986c1afdb4) (From OE-Core rev: 251d626a7d3cf6dfb3c0d8617f6cace666c7ea74) Signed-off-by: Khem Raj Signed-off-by: Alexandre Belloni Signed-off-by: Richard Purdie Signed-off-by: Peter Marko Signed-off-by: Steve Sakoman

llvm: Upgrade to 18.1.5

2024-06-05T12:57:12+00:00

Brings 617a15a9eac9 [clang codegen] Fix MS ABI detection of user-provided constructors. (#90151) 20b9ed64ea07 [RISCV][ISel] Fix types in tryFoldSelectIntoOp (#90659) ece9d35f1a70 [GlobalISel] Fix store merging incorrectly classifying an unknown index expr as 0. (#90375) a7b8b890600a [X86] Enable EVEX512 when host CPU has AVX512 (#90479) 4da5b1417493 [GlobalISel] Don't form anyextending atomic loads. a96b04442c9f [AArch64] Remove invalid uabdl patterns. (#89272) aea091b70eda [clang][CoverageMapping] do not emit a gap region when either end doesn't have valid source locations (#89564) 58648f334d62 [X86][EVEX512] Check hasEVEX512 for canExtendTo512DQ (#90390) 6350acdb134d [CGP] Drop poison-generating flags after hoisting (#90382) f341c76b9461 [Clang] Handle structs with inner structs and no fields (#89126) abf6b13085fb [IRCE] Skip icmp ptr in InductiveRangeCheck::parseRangeCheckICmp (#89967) ee5bb0c95667 Fix Objective-C++ Sret of non-trivial data types on Windows ARM64 (#88671) 6dbaa89433f7 [clang-format] Fix a regression in ContinuationIndenter (#88414) 51ff7f38b633 [clang-format] Fix a regression in annotating TrailingReturnArrow (#86624) b544217fb31f [AMDGPU] Fix setting nontemporal in memory legalizer (#83815) 78b99c73ee4b [DAGCombiner] Fix miscompile bug in combineShiftOfShiftedLogic (#89616) 1aa91720cc4f [DAGCombiner] Pre-commit test case for miscompile bug in combineShiftOfShiftedLogic 35fea1032741 release/18.x: [clang-format] Correctly annotate braces in macros (#87953) b9b73814ad8a [libcxx] [modules] Add _LIBCPP_USING_IF_EXISTS on aligned_alloc (#89827) c0b48372d82a release/18.x: [clang-format] Revert breaking stream operators to previous default (#89016) 3b4ba7277bd7 [analyzer] Fix performance of getTaintedSymbolsImpl() (#89606) 7699b341b763 release/18.x: [clang-format] Fix a regression in annotating BK_BracedInit (#87450) fb865928c8e0 [GlobalISel] Fix fewerElementsVectorPhi to insert after G_PHIs (#87927) 111ae4509c96 [X86] Fix miscompile in combineShiftRightArithmetic 76cbd417af50 [X86] Pre-commit tests (NFC) e7c816b3cd3e [InstCombine] Fix unexpected overwriting in foldSelectWithSRem (#89539) 3685a599c866 ReleaseNote: Mention SpecialCaseList change (#89141) a981a4f7653c [X86] Always use 64-bit relocations in no-PIC large code model (#89101) 4ddac856c55f [analyzer] Fix a security.cert.env.InvalidPtr crash c6d63d4fc555 Bump version to 18.1.5 (#89291) (From OE-Core rev: 02df2fc6241ac8fb0e78f2fdff97a04e5c561d54) (From OE-Core rev: b01e94690c3c29cd1e666e671c98e500c13e1942) Signed-off-by: Khem Raj Signed-off-by: Richard Purdie Signed-off-by: Peter Marko Signed-off-by: Steve Sakoman

llvm: Upgrade to 18.1.4

2024-06-05T12:57:12+00:00

Brings following fixes * e6c3289804a6 [CMake][Release] Disable PGO (#88465) (#89000) * 028e425f86cc [MIPS] Fix the opcode of max.fmt and mina.fmt (#85609) * e3c832b37b0a Fix override keyword being print to the left side * 1deeee3f5da4 Revert "[Mips] Fix missing sign extension in expansion of sub-word atomic max (#77072)" * 995539ce05ba [LLD] [COFF] Don't add pseudo relocs for dangling references (#88487) * db67e6fb9ad1 [libc++] Fix -Wgnu-include-next in stddef.h (#88214) * 647fbc710840 [SelectionDAG] Prevent combination on inconsistent type in `combineCarryDiamond` (#84888) * eaae766a20fd [RISCV] Support rv{32, 64}e in the compiler builtins (#88252) * c24b41d71f2e github-upload-release.py: Fix bug preventing release creation (#84571) * c837970dd7e9 [Codegen][X86] Fix /HOTPATCH with clang-cl and inline asm (#87639) * d0ddcce21d91 [InstSimplify] Make sure the simplified value doesn't generate poison in threadBinOpOverSelect (#87075) * 4056cc29dfd3 Prepend all library intrinsics with `#` when building for Arm64EC (#87542) * 6e071cf30599 [SLP]Fix a crash if the argument of call was affected by minbitwidth analysis. * d89da2ac8839 [libcxx] coerce formatter precision to int (#87738) * b6ebea7972cd [SPARC] Implement L and H inline asm argument modifiers (#87259) * bffecba7ce4c [libc++] Simplify the implementation of (#86843) * 9899a2d76c8f [lit][ci] Publish lit wheels (#88072) * 3ceccbdb1995 [clang-format] Correctly annotate braces of empty ctors/dtors (#82097) * 429d62872525 [Headers] Don't declare unreachable() from stddef.h in C++ (#86748) * feba8727f805 [ConstantRange] Fix off by 1 bugs in UIToFP and SIToFP handling. (#86041) * e4259b583c92 [Float2Int] Pre-commit test for SIToFP/UIToFP ConstantRange bug. NFC * daca56d8e162 Bump version to 18.1.4 (#87715) (From OE-Core rev: adc2651a8e902af24fee6ff30a72f4b7c63bef6f) (From OE-Core rev: 56ef42d454fb13a0eff0a7210a37ec4b54a2afc3) Signed-off-by: Khem Raj Signed-off-by: Alexandre Belloni Signed-off-by: Richard Purdie Signed-off-by: Peter Marko Signed-off-by: Steve Sakoman

llvm: upgrade 18.1.2 -> 18.1.3

2024-05-23T15:55:01+00:00

0001-AsmMatcherEmitter-sort-ClassInfo-lists-by-name-as-we.patch refreshed for 18.1.3 Changelog: ============ -DFixes tsan failures for glibc's LoongArch and certain RISC-V ports when fstat is used. -transform.structured.convert_to_loops now properly deletes its target op. -Fix a llvm.usub.with.overflow.i128 wrong code generation regression that was introduced with LLVM 18.1.0. -MemorySanitizer on Linux can now run even when maximum-entropy address-space layout randomization is configured globally -Fixed a Clang 18.x regression which increased binary size and stack usage with -ftrivial-auto-var-init. (From OE-Core rev: c2666c09e9ef1f523cbbcc7c9b91fc36c4872823) Signed-off-by: Wang Mingyu Signed-off-by: Alexandre Belloni (cherry picked from commit d2159f92ddbb6b999c1d14ac62647b4a35360377) Signed-off-by: Steve Sakoman

llvm: Upgrade to 18.1.3 bugfix release

2024-04-11T07:47:35+00:00

Brings following changes * c13b7485b879 [clang] Avoid -Wshadow warning when init-capture named same as class field (#74512) * f249092ef26d [tsan] Refine fstat{,64} interceptors (#86625) * 76c721994a11 [clang][Sema] Fix a CTAD regression after 42239d2e9 (#86914) * e0f0c463b539 [mlir][transform] replace original op to loop ops (#83537) * 47aac495c895 [OpenMP][AIX] Affinity implementation for AIX (#84984) * a385a917da93 [Support] Fix color handling in formatted_raw_ostream (#86700) * 0cd4babe0342 Reapply [libcxx] [modules] Fix relative paths with absolute LIBCXX_INSTALL_MODULES_DIR (#86020) * cfaeee6a8994 release/18.x: [SystemZ] Fix overflow flag for i128 USUBO (#86491) * 767b61c159b3 [lld/ELF][X86] Respect outSecOff when checking if GOTPCREL can be relaxed (#86334) * 7c7c3d4a3a27 [ODS][NFC] Cast range.size() to int32_t in accumulation (#85629) * 0e501dbd932e [Mips] Restore wrong deletion of instruction 'and' in unsigned min/max processing. (#85902) * fbb27d16fa12 [Mips] Fix missing sign extension in expansion of sub-word atomic max (#77072) * e74c1678231a [PowerPC] provide CFI for ELF32 to unwind cr2, cr3, cr4 (#83098) * 1c7c16e58240 [NFC][PowerPC] use script to regenerate the CHECK lines * c2a57034eff0 [msan] Re-exec with no ASLR if memory layout is incompatible on Linux (#85142) * b74f6158460c [msan] Add 'MappingDesc::ALLOCATOR' type and check it is available (#85153) * 98a15dd2d728 [clang][CodeGen] Allow `memcpy` replace with trivial auto var init * 2498e3a07f3d [LoongArch] Assume no-op addrspacecasts by default (#82332) * 368dc8553c5a [SROA]: Only defer trying partial sized ptr or ptr vector types * 0e16af8e4cf3 [ARM] Update IsRestored for LR based on all returns (#82745) * 6f8016fda7d5 Pre-commit test showing bug #80287 * ef6d1ec07c69 Bump version to 18.1.3 (#86168) (From OE-Core rev: f6a561db603a6583969137fdd9fb4ec0138bff46) Signed-off-by: Khem Raj Signed-off-by: Richard Purdie

llvm: Upgrade to 18.1.2 bugfix release

2024-03-22T16:25:08+00:00

(From OE-Core rev: ba6435d13f9d565a92d3e6f15a04398c8a7133f6) Signed-off-by: Khem Raj Signed-off-by: Richard Purdie