| commit | c8121b99a99fe1785add732aa062039b7c5fdd32 | [log] [tgz] |
|---|---|---|
| author | Fangrui Song <i@maskray.me> | Tue Apr 15 09:10:04 2025 -0700 |
| committer | GitHub <noreply@github.com> | Tue Apr 15 09:10:04 2025 -0700 |
| tree | 71a56cf9f3cee9b093688703b68c4939cd404698 | |
| parent | d41e517748e2dbb51e27bed217f3dd7a4c4fb86f [diff] |
[RISCV] Xqcilb: remove RISCVMCExpr::VK_QC_E_JUMP_PLT and drop `@plt` parsing Follow-up to the just landed #135044 . Remove `@plt` parsing (only needed by legacy `call foo@plt`). MCParser's `@` parsing is problematic. Supporting target variations like (`foo+2@plt foo@plt+2 (foo+2)@plt`) involves messy hacks. We should refrain from adding new `@` uses. Remove unneeded `RISCVMCExpr::VK_QC_E_JUMP_PLT` (should only be used when an instruction might have multiple reasonable relocations https://maskray.me/blog/2025-03-16-relocation-generation-in-assemblers). --- GCC's initial initial RISC-V port made a mistake by having both `call foo` (non-PIC) and `call foo@plt` (PIC), likely misled by x86/SystemZ. It was determined that the `@plt` was not needed. Since R_RISCV_CALL had questionable undefined weak semantics in GNU ld (which has been removed then), we kept R_RISCV_CALL_PLT and deprecated R_RISCV_CALL. For RISC-V instructions, we only keep `@` in call/jump for backward compatibility and discourage it for all other instructions. ( There is disagreement about whether `PLT` in `JUMP_PLT` is useful or misleading. MaskRay's opnion: For new branch relocations with procedure call semantics, use `_CALL` and avoid `_PLT` in the relocation name. `_PLT` should only be used in data directives (e.g. R_RISCV_PLT32) to indicate that the address of a function is not significant. ) Pull Request: https://github.com/llvm/llvm-project/pull/135507
Welcome to the LLVM project!
This repository contains the source code for LLVM, a toolkit for the construction of highly optimized compilers, optimizers, and run-time environments.
The LLVM project has multiple components. The core of the project is itself called “LLVM”. This contains all of the tools, libraries, and header files needed to process intermediate representations and convert them into object files. Tools include an assembler, disassembler, bitcode analyzer, and bitcode optimizer.
C-like languages use the Clang frontend. This component compiles C, C++, Objective-C, and Objective-C++ code into LLVM bitcode -- and from there into object files, using LLVM.
Other components include: the libc++ C++ standard library, the LLD linker, and more.
Consult the Getting Started with LLVM page for information on building and running LLVM.
For information on how to contribute to the LLVM project, please take a look at the Contributing to LLVM guide.
Join the LLVM Discourse forums, Discord chat, LLVM Office Hours or Regular sync-ups.
The LLVM project has adopted a code of conduct for participants to all modes of communication within the project.