aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* gcc: Add 15.1.0Hans-Christian Noren Egtvedt2025-05-0515-1/+879
| | | | | | | | | | | https://gcc.gnu.org/pipermail/gcc-announce/2025/000185.html See upstream changes at https://gcc.gnu.org/gcc-15/changes.html Add the new version clean up the patches. Remove GCC 14 patches applied to GCC 15. Signed-off-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no>
* binutils: add version 2.44Hans-Christian Noren Egtvedt2025-04-3010-2/+569
| | | | | | | | | | | | https://lists.gnu.org/archive/html/info-gnu/2025-02/msg00001.html Add the new version rebasing the patches we carry on top. Drop patches to gold, since this linker has been removed from upstream. Add a new Kconfig entry that removes visibility of enabling gold linker when selecting binutils version >= 2.44. Signed-off-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no>
* packages:mold: add version 2.37.1Hans-Christian Noren Egtvedt2025-04-262-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add newly released mold version 2.37.1 from upstream https://github.com/rui314/mold 2.37.1 - Fixed a bug where mold incorrectly reported a spurious "duplicate symbol" error when LTO was enabled. (c95476d) 2.37.0 New Features - If an undefined weak symbol is not resolved to a defined symbol at link time, the linker can choose whether to promote the symbol to a dynamic symbol or not. If promoted, the weak symbol has another chance to be resolved to a defined symbol at load time. Otherwise, it is resolved to address 0 at link time. Previously, mold always resolved remaining undefined weak symbols in an executable to address 0 at link time. Now, you can instruct the linker to promote them to dynamic symbols using -z dynamic-undefined-weak. (1822e47) Bug Fixes and Compatibility Improvements - [x86-64] The relocation types R_X86_64_CODE_4_{GOTPCRELX,GOTTPOFF,GOTPC32_TLSDESC} and R_X86_64_CODE_6_GOTTPOFF are now supported. These relocations are for Intel APX (Advanced Performance Extensions), which extends the number of general-purpose registers from 16 to 32. (83152ac, a17202d) - [ARM32] The R_ARM_THM_JUMP8 relocation type is now supported. (1fbbcec) - [ARM32] Previously, the .ARM.exidx section (which contains exception-handling records) was not subject to garbage collection, even when --gc-sections was specified. This prevented all functions from being garbage-collected, as they were referenced by exception-handling records. Now, mold correctly garbage-collects unused .ARM.exidx records and functions. (16f7599) - Previously, --compress-debug-sections was ignored if --separate-debug-file was specified. Now, mold compresses debug information sections even when they are in a separate debug file. (bab7dd1) Signed-off-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no>
* newlib: Add patch for powerpc errorsChris Packham2025-04-261-0/+109
| | | | | | | | | Add a patch to resolve some errors for powerpc-unknown-elf configurations. Submitted upstream as https://sourceware.org/pipermail/newlib/2025/021708.html Signed-off-by: Chris Packham <judge.packham@gmail.com>
* Add ncurses patch to handle debian mingw32 toolchainKeith Packard2025-04-261-0/+30
| | | | Signed-off-by: Keith Packard <keithp@keithp.com>
* Update to picolibc 1.8.10Keith Packard2025-04-263-4/+4
| | | | Signed-off-by: Keith Packard <keithp@keithp.com>
* companion_libs: Make picolibc build compatible with 1.8.10Keith Packard2025-04-262-11/+42
| | | | | | | | | | | | | | | | | | | | | Picolibc 1.8.10 reworked the configuration option for single/multi thread support and removed some unused options for exit handler support. Map LIBC_PICOLIBC_MULTITHREAD to the new single-thread option or the old newlib-retargetable-locking+newlib-multithread options depending on whether the picolibc version knows about the single-thread option. Remove support for newlib-register-fini, newlib-atexit-dynamic-alloc, newlib-global-atexit and lite-exit options as all of those relate to the legacy exit handler code. There hasn't ever been any way to enable this using crosstool-ng, so none of these options have ever done anything. The help strings for each of of the unused configuration variables have been updated. Signed-off-by: Keith Packard <keithp@keithp.com>
* picolibc: Add LIBC_PICOLIBC_GCC_LIBSTDCXX_TARGET_CXXFLAGSKeith Packard2025-04-262-0/+11
| | | | | | | This provides additional flags used when building libstdc++ for picolibc. Signed-off-by: Keith Packard <keithp@keithp.com>
* gcc: Add CC_GCC_LIBSTDCXX_TARGET_CXXFLAGSKeith Packard2025-04-262-0/+9
| | | | | | | | | This parallels the newlib-nano setting, NEWLIB_NANO_GCC_LIBSTDCXX_TARGET_CXXFLAGS, and provides a place to specify options that control how libstdc++ is build, such as -fno-exceptions. Signed-off-by: Keith Packard <keithp@keithp.com>
* gdb: Add gdb-16.3Chris Packham2025-04-243-8/+8
| | | | | | https://sourceware.org/pipermail/gdb-announce/2025/000145.html Signed-off-by: Chris Packham <judge.packham@gmail.com>
* CI: Use arm-none-eabi to get newlib tarballChris Packham2025-04-241-1/+1
| | | | | | | | Use arm-none-eabi instead of arm-picolibc-eabi. The none-eabi configuration includes picolibc and newlib so it will ensure both are downloaded. Signed-off-by: Chris Packham <judge.packham@gmail.com>
* Patch to allow for building on ol8Bruce McCulloch2025-04-221-2/+2
| | | | Signed-off-by: Bruce McCulloch <bruce.mcculloch@oracle.com>
* CI: add riscv64-unknown-linux-gnuXeonacid2025-04-221-0/+1
| | | | | Helps finding regressions like #2315 and #2316 Signed-off-by: Xeonacid <h.dwwwwww@gmail.com>
* gcc/riscv: Fix incorrect jal with HIDDEN_JUMPTARGETXeonacid2025-04-227-0/+1082
| | | | | | | | Similar to #2315, backport commit 45116f3 for gcc 7,8,9. Gcc 7 is the first version to support RISC-V. Signed-off-by: Xeonacid <h.dwwwwww@gmail.com>
* glibc/riscv: Fix incorrect jal with HIDDEN_JUMPTARGETLevi Zim2025-04-226-0/+372
| | | | | | | | | | | | | This fixes BFD assertion fail in elfnn-riscv.c when building glibc 2.29~2.34 with binutils(e.g. version 2.40 used in #2270). Versions older than 2.29 have merge conflicts with this patch so they are not taken care of. (And I don't think anyone will be interested in them as riscv64 support is introduced in 2.27) Close #2270 Signed-off-by: Levi Zim <rsworktech@outlook.com>
* Fix missing $ for variableQuentin Boswank2025-04-161-1/+1
| | | | | | | Fixes https://github.com/crosstool-ng/crosstool-ng/issues/2321 Credit to @jimbomorrison for spotting it. Signed-off-by: Quentin Boswank <qubos@outlook.de>
* add samples for mips64-ol9u5Bruce McCulloch2025-04-042-0/+32
| | | | Signed-off-by: Bruce McCulloch <bruce.mcculloch@oracle.com>
* arch: Add support for RX gcc buildDuy Nguyen2025-04-044-0/+23
| | | | Signed-off-by: Duy Nguyen <duy.nguyen.xa@renesas.com>
* zstd: Bring in upstream fix for MSYSChris Packham2025-03-101-0/+197
| | | | Signed-off-by: Chris Packham <judge.packham@gmail.com>
* Fix disabled libstdc++-v3Quentin Boswank2025-03-101-2/+2
| | | | | the config var is unset and not just `n`. Signed-off-by: Quentin Boswank <qubos@outlook.de>
* contrib/gcc-test-suite: Add configuration options including qemu setupKeith Packard2025-03-015-9/+96
| | | | | | | | | | This makes the options necessary to run the gcc test suite configurable in the crosstool-ng config file. That includes the ability to run the test suite using qemu instead of on a remote host. Signed-off-by: Keith Packard <keithp@keithp.com>
* ncurses: Pass --disable-widec to configureChris Packham2025-02-241-0/+2
| | | | | | | | | The default for wide char support has changed in ncurses-6.5. This causes build failures for gdb. Pass --disable-widec to prevent the build errors. Fixes #2292, #2294 Signed-off-by: Chris Packham <judge.packham@gmail.com>
* Make GCC 12.4.0 compatible with MinGW >=11Oliver Old2025-02-241-0/+379
| | | | | | | | | This patch defines `WIN32_LEAN_AND_MEAN` before the `windows.h` include statements to avoid name clashes with the `abort()` member function in `IXMLDOMDocument`, which would now be pulled in via `msxml.h` without the `WIN32_LEAN_AND_MEAN` definition. Signed-off-by: Oliver Old <mail@oold.dev>
* binutils: Don't enable plugins with static toolchainChris Packham2025-02-171-0/+1
| | | | | | | | | A static toolchain by definition can't load plugins and GCC will reject the --enable-plugin configure option if we're trying to build a static toolchain. Fixes #2288 Signed-off-by: Chris Packham <judge.packham@gmail.com>
* packages/picolibc: Update to version 1.8.9Keith Packard2025-02-173-4/+4
| | | | | | Support latest upstream release Signed-off-by: Keith Packard <keithp@keithp.com>
* Use no-name flag for gzipandrin-geiger2025-02-171-1/+1
| | | | | | Don't save the timestamp in the gzip header to ensure reproducibility. Signed-off-by: andrin-geiger <me@darkdampsquib.com>
* gmp: Add upstream patch for C23 compatibilityChris Packham2025-02-081-0/+31
| | | | | | | | Add a patch from upstream GMP that updates some custom configure checks for C23 compatibility (gnu32 will be the default for GCC 15). Reported-by: Marc Poulhiès <dkm@kataplop.net> Signed-off-by: Chris Packham <judge.packham@gmail.com>
* mold: Mark versions 2.31.0-2.33.0 as obsoleteChris Packham2025-02-083-0/+3
| | | | | | | 2.36.0 is the latest version. There's no reason to keep these older ones around. Signed-off-by: Chris Packham <judge.packham@gmail.com>
* mpc, mpfr: Remove unused milestonesChris Packham2025-02-083-4/+0
| | | | | | | | An older version of MPC had issues with some versions of MPFR. We don't actually have either version of these packages anymore so we no longer need these milestones. Signed-off-by: Chris Packham <judge.packham@gmail.com>
* packages: Drop obsolete versionsChris Packham2025-02-0813-167/+0
| | | | | | | Post release drop versions of some packages that were marked as obsolete. Signed-off-by: Chris Packham <judge.packham@gmail.com>
* Bump make 4.4 to 4.4.1Michał Zagórski2025-02-073-8/+8
| | | | | | https://lists.gnu.org/archive/html/info-gnu/2023-02/msg00011.html Signed-off-by: Michał Zagórski <zagura6+github@gmail.com>
* Add ncurses 6.5Michał Zagórski2025-02-072-0/+4
| | | | | | https://invisible-island.net/ncurses/announce.html Signed-off-by: Michał Zagórski <zagura6+github@gmail.com>
* Add automake 1.17Michał Zagórski2025-02-072-0/+8
| | | | | | https://lists.gnu.org/archive/html/autotools-announce/2024-07/msg00000.html Signed-off-by: Michał Zagórski <zagura6+github@gmail.com>
* Add autoconf 2.72Michał Zagórski2025-02-072-0/+8
| | | | | | https://lists.gnu.org/archive/html/autotools-announce/2023-12/msg00003.html Signed-off-by: Michał Zagórski <zagura6+github@gmail.com>
* Add bison 3.8.2Michał Zagórski2025-02-072-0/+8
| | | | | | https://lists.gnu.org/archive/html/bison-announce/2021-09/msg00002.html Signed-off-by: Michał Zagórski <zagura6+github@gmail.com>
* Add libiconv 1.18Michał Zagórski2025-02-072-0/+4
| | | | | | https://savannah.gnu.org/news/?id=10703 Signed-off-by: Michał Zagórski <zagura6+github@gmail.com>
* Add libtool 2.5.4Michał Zagórski2025-02-072-0/+8
| | | | | | https://lists.gnu.org/archive/html/libtool/2024-12/msg00004.html Signed-off-by: Michał Zagórski <zagura6+github@gmail.com>
* Add libexpat 2.6.4Michał Zagórski2025-02-072-0/+16
| | | | | | https://github.com/libexpat/libexpat/releases/tag/R_2_6_4 Signed-off-by: Michał Zagórski <zagura6+github@gmail.com>
* arch: nios2: Qualify selectsChris Packham2025-02-051-2/+2
| | | | | | | | | | | | | | | | | | Previous commits added selects so that the nios2 architecture only used GDB/GLIBC versions that still support it however these weren't qualified leading to kconfig complaints such as WARNING: unmet direct dependencies detected for GLIBC_REQUIRE_older_than_2_41 Depends on [n]: LIBC_GLIBC_SHOW [=n] Selected by [y]: - ARCH_NIOS2 [=y] && GEN_CHOICE_ARCH [=y] Qualify these selects with if DEBUG_GDB and if LIBC_GLIBC to stop the complaints. Fixes: 9d91cebe ("gdb: Add 16.1") Fixes: 87605d65 ("glibc: Add 2.41") Signed-off-by: Chris Packham <judge.packham@gmail.com>
* Add GMP 6.3.0Marc Poulhiès2025-02-042-0/+8
| | | | | | GMP 6.2.1 may fail some autoconf test when using GCC master (to-be-15) branch. Signed-off-by: Marc Poulhiès <dkm@kataplop.net>
* samples: x86_64-w64-mingw32,arm-cortexa9_neon-linux-gnueabihf pin gettextChris Packham2025-02-031-1/+1
| | | | | | | The latest gettext has problems building this configuration. Pin the version to gettext-0.22.5 which works fine. Signed-off-by: Chris Packham <judge.packham@gmail.com>
* gdb: Add 16.2Chris Packham2025-02-033-8/+8
| | | | | | https://sourceware.org/pipermail/gdb-announce/2025/000144.html Signed-off-by: Chris Packham <judge.packham@gmail.com>
* libelf: Update configure tests for GCC14 part2Chris Packham2025-02-021-0/+51
| | | | | | More configure checks that need updating. Signed-off-by: Chris Packham <judge.packham@gmail.com>
* samples: x86_64-unknown-linux-gnu,arm-picolibc-eabi: UpdateChris Packham2025-02-021-2/+2
| | | | | | Update the x86_64-unknown-linux-gnu,arm-picolibc-eabi sample config. Signed-off-by: Chris Packham <judge.packham@gmail.com>
* gettext: Add 0.23.1Chris Packham2025-02-023-1/+9
| | | | | | Add the latest gettext version. Signed-off-by: Chris Packham <judge.packham@gmail.com>
* gettext: Add fix for tparm check in gettext-0.20.1Chris Packham2025-02-021-0/+49
| | | | | | | | Some canadian configs that use uclibc are pinned to gettext-0.20.1. We need to update some of the configure checks to avoid issues with newer versions of GCC. Signed-off-by: Chris Packham <judge.packham@gmail.com>
* glibc: Add 2.41Chris Packham2025-02-015-1/+106
| | | | | | | | | https://sourceware.org/pipermail/libc-announce/2025/000045.html Support for nios2*-*-linux-gnu has been removed in this version so the nios2 architecture now requires and older glibc. Signed-off-by: Chris Packham <judge.packham@gmail.com>
* packages:mold: add version 2.36.0Hans-Christian Noren Egtvedt2025-02-012-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add newly released mold version 2.36.0 from upstream https://github.com/rui314/mold New Features - The --package-metadata=<string> option has been added to embed a given string into the .note.package section. This option is designed for build scripts that generate binary packages, such as .rpm or .deb, to include package metadata in each executable. It simplifies the process of identifying the corresponding package for a given executable or core file. (7ddc8f4) - [ARM][PowerPC] We've improved the algorithm for creating range extension thunks to reduce memory usage and improve speed. For example, linking clang-19 for ARM64 is now ~7% faster than before. (9fc0ace) - [RISC-V][LoongArch] We've improved the algorithm for code-shrinking linker relaxation to reduce memory usage and improve speed. For example, linking clang-19 for RISC-V is now ~4% faster than before. (3234d88) Bug Fixes and Compatibility Improvements - mold created a bad relocation for an IFUNC if the linker's output file type was a shared library and the symbol was exported. This bug could cause a segmentation fault of a linked program. The problem has now been fixed. (a297859) - [RISC-V] mold could produce incorrect code as a result of code-shrinking relaxation for the R_RISCV_HI20 relocation. That type of relocation was used rarely because it is not PC-relative. That being said, if your program used the relocation, and the relocation targets were at a low address (from 0x1f800 to 0x20000), your program would crash at runtime due to the linker's bug. The issue has now been resolved. (eec3f6b) - [RISC-V][LoongArch] When the linker removed instructions from a function as a result of code-shrinking relaxation, the function symbol's size in the output file should be updated to reflect the result of relaxation, even though doing it is mostly cosmetic. mold did not do that. Now, mold sets correct sizes to output function symbols. (e6345d5) - [LoongArch] Binaries linked with mold now work on 64 KiB page systems. Previously, only up to 16 KiB pages were supported. (2d7b6b2) - [s390x] The s390x processor-specific ABI requires the linker to reserve the first three slots of the .got section for the runtime. mold, however, reserved only two slots and used the third for itself. Even though we did not observe issues in the wild, it was a violation of the psABI. The problem has now been fixed. (dfce2fc) Signed-off-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no>
* crosstool-ng update to ol9-{u3,u4,u5} and ol8-{u9, u10} for aarch64 and x86_64Bruce McCulloch2025-01-2820-0/+290
| | | | Signed-off-by: Bruce McCulloch <bruce.mcculloch@oracle.com>
* samples: Disable gdb in or1k-unknown-linux-muslChris Packham2025-01-271-2/+0
| | | | | | | | | | | | | | MUSL's or1k support lacks definitions for elf_gregset_t which leads to failures like [ERROR] sysroot/usr/include/sys/procfs.h:29:9: error: 'elf_gregset_t' does not name a type; did you mean 'gregset_t'? [ERROR] sysroot/usr/include/sys/procfs.h:54:9: error: 'elf_gregset_t' does not name a type; did you mean 'gregset_t'? [ERROR] sysroot/usr/include/sys/procfs.h:55:9: error: 'elf_fpregset_t' does not name a type Disable GDB support in the or1k-unknown-linux-musl config. Fixes #2266 Signed-off-by: Chris Packham <judge.packham@gmail.com>
generated by cgit v1.2.3 (git 2.39.1) at 2025-07-21 01:27:56 +0000