aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2024-10-10Modify the enabled languages in stage2 builds to the macro @WITH_LANGUAGES@.Ariel Xiong4-36/+60
Use ./configure --with-languages=c,c++,fortran to adjust the enabled languages. This change only affects the GNU toolchain.
2024-10-10Update GCC branch tag - see ↵Tommy Murphy1-1/+1
https://github.com/riscv-collab/riscv-gnu-toolchain/issues/1575
2024-10-10Enable RV32 build of Linux/musl toolchain - effectively reverse the ↵Tommy Murphy6-18/+3
restrictions imposed by this commit: https://github.com/riscv-collab/riscv-gnu-toolchain/commit/629c67e0a93ec03edd3dfab60a2b8ad9c1768a2a
2024-10-10Merge pull request #1570 from cmuellner/ubuntu-24.04Kito Cheng2-5/+5
CI: Drop Ubuntu 20.04 and add Ubuntu 24.04
2024-10-09Remove link to configure script - see ↵Tommy Murphy1-2/+0
https://github.com/riscv-collab/riscv-gnu-toolchain/pull/1562#discussion_r1793167825
2024-10-09Small updates to the out-of-tree sources sectionTommy Murphy1-5/+9
2024-10-09Bump uClibc from v1.0.48-7-gad21b95c9 to v1.0.50. Addresses ↵Tommy Murphy1-0/+0
https://github.com/riscv-collab/riscv-gnu-toolchain/issues/1563
2024-10-09Bump musl from v1.2.2 to v1.2.5Tommy Murphy1-0/+0
2024-09-30CI/CD: Drop Ubuntu 20.04Christoph Müllner2-2/+2
We have build issues in Ubuntu 20.04 with QEMU, because of an outdated host dependency (glib2 is 2.64.6 but >= 2.66 is required). Ubuntu LTS releases have a standard support lifecycle of 5 years, so Ubuntu 20.04 support will be ending in about 6 months. Let's drop Ubuntu 20.04 support now, so we don't need a workaround for the QEMU issue. Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
2024-09-30CI/CD: Add ubuntu-24.04Christoph Müllner2-5/+5
We are currently using Ubuntu 20.04, Ubuntu 22.04, and Ubuntu latest as base images for our CI/CD builds. This patch add Ubuntu 24.04 and makes it the default for jobs where we only build on one OS. Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
2024-09-26Update fail testcases with multi-lib.Jiawei2-0/+7
2024-09-26Update gfortan and python fail cases.Jiawei2-0/+9
2024-09-26Update fail case condition.Jiawei2-4/+4
2024-09-26Categorize fail testcases.Jiawei6-17/+17
2024-09-26Bump gcc into 14.2 release version.Jiawei4-65/+82
2024-09-19Bump qemu from 9.0.2 to 9.1.0Christoph Müllner1-0/+0
Test results (with --enable-multilib): ``` ========= Summary of gcc testsuite ========= | # of unexpected case / # of unique unexpected case | gcc | g++ | gfortran | rv32imac/ ilp32/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv32imafdc/ ilp32d/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv64imac/ lp64/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv64imafdc/ lp64d/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | ``` Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
2024-08-27Merge pull request #1475 from CARV-ICS-FORTH/uclibc2024.09.032024.08.28Kito Cheng5-858/+1730
Add uClibc-ng support for NOMMU Linux builds
2024-08-27Bump qemu from 9.0.1 to 9.0.2Christoph Müllner1-0/+0
Test results (with --enable-multilib): ``` ========= Summary of gcc testsuite ========= | # of unexpected case / # of unique unexpected case | gcc | g++ | gfortran | rv32imac/ ilp32/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv32imafdc/ ilp32d/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv64imac/ lp64/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv64imafdc/ lp64d/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | ``` Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
2024-08-27Bump glibc from 2.39 to 2.40Christoph Müllner1-0/+0
Test results (with --enable-multilib): ``` ========= Summary of gcc testsuite ========= | # of unexpected case / # of unique unexpected case | gcc | g++ | gfortran | rv32imac/ ilp32/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv32imafdc/ ilp32d/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv64imac/ lp64/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv64imafdc/ lp64d/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | ``` Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
2024-08-27Bump binutils from 2.43 to 2.43.1Christoph Müllner2-1/+1
Test results (with --enable-multilib): ``` ========= Summary of gcc testsuite ========= | # of unexpected case / # of unique unexpected case | gcc | g++ | gfortran | rv32imac/ ilp32/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv32imafdc/ ilp32d/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv64imac/ lp64/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv64imafdc/ lp64d/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | ``` Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
2024-08-26Bump gdb into 15.1 release version.Jiawei2-1/+1
2024-08-26Install OpenMP to sysroot (#1541)Pengcheng Wang1-2/+2
Or we can't find headers and libraries.
2024-08-15show progress while clonning submodulesOsman Karaketir1-1/+1
Signed-off-by: Osman Karaketir <osmankaraketir@gmail.com>
2024-08-08fix: shared lib suffix LLVMgold macOSRodolphe CARGNELLO1-2/+6
2024-08-05Bump binutils to 2.43 release.2024.08.06Jiawei1-0/+0
2024-08-02Bump LLVM to 18.1.82024.08.03Kito Cheng1-0/+0
2024-08-01musl: support build gdbMark Zhuang1-0/+25
2024-07-19Build openmp libraries for Clang/LLVM on RV64wangpc1-0/+41
For Clang/LLVM, the default OpenMP library is `libomp`. However, we don't build openmp library, which will cause a linker error: ``` riscv64-unknown-linux-gnu-ld: cannot find -lomp: No such file or directory clang: error: linker command failed with exit code 1 (use -v to see invocation) ``` So we build `libomp.so/libomp.a` manually.
2024-07-17Add --enable-default-pie optionKito Cheng4-0/+42
`--enable-default-pie` is used to control the default PIE enablement for Linux GCC, which is disabled by default. NOTE: baremetal toolchain isn't affected by this option.
2024-07-14Bump Spike from commit 20a2b6d to the latest commit f7d0dba. Addresses ↵Tommy Murphy1-0/+0
https://github.com/riscv-collab/riscv-gnu-toolchain/issues/1507
2024-07-14Address Kito's commentsNick Kossifidis1-2/+2
Signed-off-by: Nick Kossifidis <mick@ics.forth.gr>
2024-07-02Check UNRESOLVED in testsuite-filterCraig Blackmore3-1/+13
The resulting UNRESOLVEDs have been added to the allowlists. Tested with report-linux and report-newlib.
2024-07-02Add filling-one for tail/mask agnostic on qemuKito Cheng1-0/+4
Default behavior of qemu is treat tail/mask agnostic as tail/mask undisturbed, and this may hidding some problem, one example is [PR115725](https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115725) for GCC.
2024-07-02Qemu has deprecated extension name with capital 'Z'Kito Cheng1-4/+4
See this for detail: https://www.qemu.org/docs/master/about/deprecated.html#risc-v-cpu-properties-which-start-with-capital-z-since-8-2
2024-06-26Add NEWLIB_TARGET_FLAGS_EXTRAChristoph Müllner2-4/+7
This adds NEWLIB_TARGET_FLAGS_EXTRA, which works the same like GLIBC_TARGET_FLAGS_EXTRA. The flags are used for configuring newlib or newlib-nano. Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
2024-06-19Bump QEMU from 8.2.2 to 9.0.0Akihiko Odaki1-0/+0
Test results: ``` ========= Summary of gcc testsuite ========= | # of unexpected case / # of unique unexpected case | gcc | g++ | gfortran | rv64g/ lp64d/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | ``` Signed-off-by: Akihiko Odaki <akihiko.odaki@gmail.com>
2024-05-16Fix typo in configure.ac: host_gcc -> host-gccKito Cheng1-1/+1
I found this issue when I regen the configure script, but seems like configure is right...
2024-05-11Add uClibc-ng support for NOMMU Linux buildsNick Kossifidis5-856/+1728
This adds support for building uClibc-ng toolchain so that we can play with NOMMU Linux. This is still experimental (ABI is still WiP) but it works for e.g. building busybox, and since crosstool-ng and others don't have support for it yet, let's add it here so that people can play with it. I've tested this on QEMU and real hardware (only busybox), it generates ELF (not FLAT) binaries, and in order for the FDPIC loader to work, the static-pie loader (/lib/ld-uClibc.so.0) should also be included in rootfs with busybox compiled as dynamic/pie binary. No multilib for now, but we can add it in the future if you think it would be useful. Signed-off-by: Nick Kossifidis <mick@ics.forth.gr>
2024-05-11Fix check for gsed/gawk on MakefileNick Kossifidis1-2/+2
We end up using sed/awk wrappers even when we don't need to. Instead of having hardcoded paths (/bin/sed, /usr/bin/gawk), use the which command to check how the shell would resolve sed/awk before we tamper with PATH. Also handle the case where sed/awk are symbolic links to gsed/gawk, using realpath. Signed-off-by: Nick Kossifidis <mick@ics.forth.gr>
2024-04-26dejagnu: Bump version from c298959 to ca371cfChristoph Müllner1-0/+0
Test results (with --enable-multilib): ``` ========= Summary of gcc testsuite ========= | # of unexpected case / # of unique unexpected case | gcc | g++ | gfortran | rv32imac/ ilp32/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv32imafdc/ ilp32d/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv64imac/ lp64/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv64imafdc/ lp64d/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | ``` Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
2024-04-26spike: Bump version from d1efcdf to 20a2b6dChristoph Müllner1-0/+0
``` ========= Summary of gcc testsuite ========= | # of unexpected case / # of unique unexpected case | gcc | g++ | gfortran | rv64gc/ lp64d/ medlow | 0 / 0 | 0 / 0 | - | rv32imafc/ ilp32f/ medlow | 0 / 0 | 0 / 0 | - | ``` Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
2024-04-24Bump QEMU from 8.2.1 to 8.2.2Christoph Müllner1-0/+0
Test results (rv64/rv32): ``` ========= Summary of gcc testsuite ========= | # of unexpected case / # of unique unexpected case | gcc | g++ | gfortran | rv64gc/ lp64d/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | rv32gc/ ilp32d/ medlow | 0 / 0 | 0 / 0 | 0 / 0 | ``` Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
2024-04-24gdb: Bump version from 14.1 to 14.2Christoph Müllner1-0/+0
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
2024-04-11Merge pull request #1440 from ff520git/ff520git-patch-12024.04.12Kito Cheng1-3/+8
RISCV: Add target_board generation for multiple arch_abi parameters
2024-04-10Enable user mode network backend support in qemuwangpc2-5/+5
If `libslirp` is found during configuration, it will be enabled. So we just need install `libslirp` before building.
2024-04-06Bump newlib to 4.4.0Christoph Müllner1-0/+0
This commit bumps newlib from 4.3.0 to 4.4.0. Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
2024-04-05Add double quotation marks for build-arch-abiff520git1-3/+3
In order to generate target board info for multilib, the parameters of build-arch-abi should be enclosed in double quotation marks. And the the python script generate_target_board will process it correctly Signed-off-by: ff520git <wangfeng@eswincomputing.com>
2024-04-05Update README with a short summary of the --with-cmodel configure option (#1414)Valen1-0/+12
Update README.md to indicate core model option Signed-off-by: Valen <18544814+SpiccyMayonnaise@users.noreply.github.com> Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu> Co-authored-by: Christoph Müllner <christoph.muellner@vrull.eu>
2024-03-27Update generate_target_boardff520git1-3/+8
RISCV: Add target_board generation for multiple arch_abi parameters Currently this script just can handle one parameter for arch_abi,but there is more than one arch_abi in "newlib_multilib_names" which comes from the configure. Execute the following command: generate_target_board --sim-name riscv-sim --cmodel medany --build-arch-abi "rv32imafc-ilp32f rv64imafdc-lp64d" Before this patch the result is: riscv-sim/-march=rv32imafc/-mabi=ilp32f rv64imafdc/-mcmodel=medany After this patch,the result is: riscv-sim/-march=rv32imafc/-mabi=ilp32f/-mcmodel=medany riscv-sim/-march=rv64imafdc/-mabi=lp64d/-mcmodel=medany At the same time,there is small modification for Makefile.in,the parameters of build-arch-abi should be enclosed in double quotation marks. --build-arch-abi "$(NEWLIB_MULTILIB_NAMES)" Signed-off-by: ff520git <wangfeng@eswincomputing.com>
2024-02-29Remove --enable-gold and simplify include search2024.03.01wangpc1-8/+5
generated by cgit v1.1 at 2025-09-18 21:24:52 +0000