aboutsummaryrefslogtreecommitdiff
path: root/scripts
AgeCommit message (Collapse)AuthorFilesLines
2025-07-15march-to-cpu-opt: Add zvfh and zvfhminPaul-Antoine Arras1-0/+2
Add zvfh and zvfhmin to the list of ISA extension supported by QEMU.
2024-07-02Check UNRESOLVED in testsuite-filterCraig Blackmore1-1/+1
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
2024-03-27Update 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>
2023-12-19Address the comments.2023.12.20Pan Li1-4/+8
Signed-off-by: Pan Li <pan2.li@intel.com>
2023-12-19Fix one bug of spike varch elenPan Li1-1/+11
According to the ISA, the zve32/zve64 xdf decides the elen of the RVV. Thus we should pick up extension instead of xlen for spike varch generation. Signed-off-by: Pan Li <pan2.li@intel.com>
2023-12-13Fix one bug of generate target board2023.12.14Pan Li1-1/+1
The default value of subarray is the length, thus we should take `idx` instead of `idx - 1` here. Signed-off-by: Pan Li <pan2.li@intel.com>
2023-12-11Merge pull request #1387 from yanzhang-dev/yanzhang/bug-fixKito Cheng1-1/+2
Add misaligned memory access for spike.
2023-12-08Add misaligned memory access for spike.Yanzhang Wang1-1/+2
The spec does not forbid the misaligned memory access. And the gcc will generate vector load/store to access misaligned memory by default. Currently the QEMU supports this feature, so to support in spike is also make sense. Reference test case: gcc/gcc/testsuite/gcc.dg/vect/vect-align-1.c
2023-11-28Add OR operator for option '--with-extra-multilib-test'Pan Li1-7/+17
This patch allows you provide the flags in build flags array acts on arch-abi __respectively__, you can use ',' to separate them. For example: `rv64gcv-lp64d:--param=riscv-autovec-lmul=dynamic,--param=riscv-autovec-preference=fixed-vlmax` will be consider as two target boards same as below: ``` riscv-sim/-march=rv64gcv/-mabi=lp64d/-mcmodel=medlow/--param=riscv-autovec-preference=fixed-vlmax riscv-sim/-march=rv64gcv/-mabi=lp64d/-mcmodel=medlow/--param=riscv-autovec-lmul=dynamic ``` However, you can also leverage AND(`:`), OR(`,`) operator together but the OR(`,`) will always have the higher priority. For example: `rv64gcv-lp64d:--param=riscv-autovec-lmul=dynamic:--param=riscv-autovec-preference=fixed-vlmax,--param=riscv-autovec-lmul=m2` will be consider as tow target boars same as below: ``` riscv-sim/-march=rv64gcv/-mabi=lp64d/-mcmodel=medlow/--param=riscv-autovec-lmul=dynamic/--param=riscv-autovec-preference=fixed-vlmax riscv-sim/-march=rv64gcv/-mabi=lp64d/-mcmodel=medlow/--param=riscv-autovec-lmul=m2 ``` Signed-off-by: Pan Li <pan2.li@intel.com>
2023-11-27Merge pull request #1370 from Incarnation-p-lee/masterKito Cheng3-40/+110
Inject CPU options to spike run test from elf file
2023-11-22Fix CI run failure on rv64gcPan Li1-2/+6
Signed-off-by: Pan Li <pan2.li@intel.com>
2023-11-21Inject CPU options to spike run test from elf filePan Li3-40/+106
* Refactor CPU options for exacting common part between qemu and spike. * Inject CPU options to spike run command line. The below common CPU options are extracted to common part. * xlen * vlen * extensions The above CPU options will be leveraged to build the run arguments for both the qemu and spike, for example. QEMU: qemu-riscv64 -cpu rv64,v=true,vlen=1024,elen=64,vext_spec=v1.0 SPIKE: spike --isa=rv64imafdcv_zicsr_zifencei_zmmul --varch=vlen:1024,elen:64 Signed-off-by: Pan Li <pan2.li@intel.com>
2023-11-20Improve make report with extra test argumentsimprove-reportKito Cheng1-4/+9
Recently we have add the capability to adding extra compilation options for running GCC testsuite, but it seems not handle well on the `make report` part, it will ignore extra compilation options, and then...combine multiple report together. Now it will put in two different line, and the extra compilation options will listed in follwoing line, here is example output here: ========= Summary of gcc testsuite ========= | # of unexpected case / # of unique unexpected case | gcc | g++ | gfortran | rv64gc_zicfilp_zicfiss/ lp64d/ medlow |20350 / 3967 |10158 / 2583 |18374 / 3098 | rv64gc_zicfilp_zicfiss/ lp64d/ medlow | 86 / 19 | 15 / 9 | 51 / 51 | -static
2023-11-17Extend semantic for option '--with-extra-multilib-test'Pan Li1-0/+67
The `--with-extra-multilib-test` take the format like below for now. "<arch>-<abi>-[code-model][;<arch>-<abi>-[code-model]*" This patch would like to extend the sematic to support additional buil options for each combination. Aka: "<arch>-<abi>-[code-model][:opts]*[;<arch>-<abi>-[code-model][:opts]*]*" opts can be one or more build options splited by ':'. For example as below: * --param=riscv-autovec-lmul=m1 * --param=riscv-autovec-lmul=dynamic:--param=riscv-autovec-preference=fixed-vlmax Thus the full example of the option `--with-extra-multilib-test` will be: `rv64gcv_zvl128b-lp64d:--param=riscv-autovec-lmul=m1;rv64gcv_zvl256b-lp64d:- -param=riscv-autovec-lmul=dynamic:--param=riscv-autovec-preference=fixed-vlmax` Signed-off-by: Pan Li <pan2.li@intel.com>
2023-10-03use python3 in testsuite-filter ...Vineet Gupta1-1/+1
... just as in march-to-cpu-opt Signed-off-by: Vineet Gupta <vineetg@rivosinc.com>
2023-09-08Specify vector spec v1.0 in QEMU arguments.Patrick O'Neill1-1/+1
Some testcases in GCC's testsuite check stdout. Qemu will print a warning if the vector spec is not specified which will cause those testcases to fail. The warning: vector version is not specified, use the default value v1.0 Signed-off-by: Patrick O'Neill <patrick@rivosinc.com>
2023-05-18Merge pull request #1246 from patrick-rivos/make-report-ci2023.05.19Kito Cheng1-1/+1
Add ci jobs for make-report and make-report multilib
2023-05-11Return error when no testsuite results are detectedPatrick O'Neill1-1/+1
2023-04-24Add --get-elf-class for march-to-cpu-optKito Cheng2-4/+17
2023-04-24Change /usr/bin/python3 to /usr/bin/env python3Kito Cheng1-1/+1
2023-04-24Set qemu cpu option from ELF attributeKito Cheng2-2/+45
This could help multi-lib testing, but the price is slightly increase the testing time since it will need to extract ELF attribute from binary before running qemu. But I think the cost is acceptable compare to make build system more complicate, and actually we already use this approach in our internal stuffs for years.
2022-12-16Merge pull request #1168 from riscv-collab/march-to-cpu-opt-unittest2023.01.042023.01.032022.12.17Christoph Müllner1-1/+16
Add unittest to march-to-cpu-opt
2022-12-16Merge pull request #1162 from riscv-collab/fix-zhinxKito Cheng1-0/+9
Support testing zdinx/zfinx/zhinx.
2022-12-16Add unittest to march-to-cpu-optKito Cheng1-1/+16
https://github.com/riscv-collab/riscv-gnu-toolchain/pull/1167 has found some bug, and I realized the testing of march-to-cpu-opt is...not well, so spend some time to improve that a little bit, it's not complete testing, but at least it's a start :P
2022-12-16Fix wrong parse scriptKwanghoon Son1-2/+2
If test --with-arch=rv32imc_zve128x NameError: name 'arch' is not defined Signed-off-by: Kwanghoon Son <kwangson@yahoo.com>
2022-12-02Support testing zdinx/zfinx/zhinx.Kito Cheng1-0/+9
- We don't set correct qemu option for zdinx/zfinx/zhinx, this patch is fixing the march-to-cpu-opt, which is a script translate -march string to qemu options.
2022-06-03Pass right extension info to qemuKito Cheng2-1/+145
- Parse that from `--with-arch=` option
2022-05-23Add option '--with-sim' to set simulatorLiaoshihua1-2/+2
2021-04-01Specify the kernel version for qemu user mode.Kito Cheng1-1/+1
- RV32 require 5.10 which is too new for most linux distro.
2020-11-04Support run testsuite with spikeKito Cheng4-0/+10
2020-08-06Allowlist might contain XPASS and FAIL for same case in sametimeKito Cheng1-2/+5
e.g. The testcase already fail, the second XPASS is kind of false positive. FAIL: gcc.dg/analyzer/taint-1.c XPASS: gcc.dg/analyzer/taint-1.c
2020-07-27Check for dejagnu ERROR.Jim Wilson1-1/+2
2020-06-17Rename whitelist to allowlistKito Cheng1-1/+1
2019-10-18Merge pull request #528 from riscv/fix-python3Kito Cheng1-3/+3
Use items rather than iteritems for python 3 compatible
2019-10-17Use items rather than iteritems for python 3 compatibleKito Cheng1-3/+3
2019-10-03Add macos to travis-ciKito Cheng1-0/+6
2019-07-23Apply new whitelist scheme for binutilsKito Cheng1-36/+70
2019-07-14Rewrite whitelist filter and reorg the scheme of whitelistKito Cheng1-0/+263
2017-09-22Fail mail_tail when make failsPalmer Dabbelt1-0/+3
2017-09-13Allow tests to pass special QEMU argumentsPalmer Dabbelt1-1/+13
2017-08-20Run the GCC test suite on TravisPalmer Dabbelt1-0/+3
I don't run any multilib tests here because they take too long for Travis.
2017-08-20Link all the QEMU wrapper scripts togetherPalmer Dabbelt4-4/+6
2017-02-09Use qemu for check-gcc-newlibAndrew Waterman2-0/+2
gdb-sim accounted for nearly all test failures, so don't use it for now.
2016-09-29Add support for run testsuite with qemuKito Cheng2-0/+2
2015-04-11Avoid wrapping GNU sed/awk on systems where that's the defaultPalmer Dabbelt3-2/+3
I have no idea why, but the installed "sed" wrapper never terminates on RHEL. If I don't use the wrappen the the tools build fine, so this just uses the already set autoconf variables to attempt to determine if the system sed/awk are gsed/gawk and if they are then this avoids using the wrappers. There's a few oddities here: * I have no idea why the sed wrapper fails, as it seems super safe. * I haven't run into any awk problems, but I figured I'd treat it the same as it isn't any harder. * We shouldn't have to support 10 year old distributions. Hopefully this doesn't break anyone's builds...
2015-02-20Centralize GNU utility checks in autoconfAlbert Ou3-3/+5
Tests for program presence, such as those formerly embedded in the top-level Makefile for gawk and gsed, are better suited for autoconf. Note that it is not sufficient to merely export AWK and SED environment variables, as packages may still directly invoke awk(1) and sed(1) with non-standard features independent of the autotools framework. Wrapper scripts therefore remain necessary, although these are now generated by the configure script to avoid hard-coded paths. Do not assume the existence of /bin/bash on all systems.
2015-02-20scripts: Enhance portability of cp_sAlbert Ou1-16/+23
Eliminate bashisms; target the Bourne shell and a safer subset of POSIX utilities. Furthermore, properly handle pathnames with whitespace.
2015-01-09Update dejagnu board to link against pthreadsAndrew Waterman1-1/+1
generated by cgit v1.1 at 2026-01-26 04:01:56 +0000