aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2026-01-22Add scalar crypto bitmanip tests (#646)HEADmasterAndrew Waterman18-2/+873
* Add first version of tests for Krypto B Extension (Zbkb, Zbkc, Zbkx) * Fix build of Zbkb/Zbkx tests * Remove duplicate tests * Fix bugs in packh, packw tests --------- Co-authored-by: Furkan Aktas <f_rkan_aktas@proton.me>
2026-01-13Fall back to RV64 compiler if no RV32 compiler and vice versaAndrew Waterman1-0/+9
2026-01-13Use Zicclsm instead of --misalignedAndrew Waterman1-2/+2
Spike no longer supports the latter.
2026-01-07Make build compatible with distro tooling (#643)Heinrich Schuchardt1-19/+21
* isa: don't build 32bit if XLEN == 64 Only build 32bit binaries if XLEN != 64. Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com> * isa: use API lp64d for XLEN=64 The standard ABI used on 64bit is lp64d. With this change compilation on Ubuntu no longer complains the missing include gnu/stubs-lp64.h. Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com> --------- Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
2026-01-07benchmarks: illegal addend for R_RISCV_GOT_HI20 in crt.S (#639)Heinrich Schuchardt2-2/+4
* benchmarks: illegal addend for R_RISCV_GOT_HI20 in crt.S With Ubuntu's GCC 15.2.0-7ubuntu1 building results in an error: /tmp/ccfCpSVm.o: in function `_start': (.text.init+0x100): dangerous relocation: The addend isn't allowed for R_RISCV_GOT_HI20 Split loading _end + 63 into register tp into separate load and add instructions. https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/master/riscv-elf.adoc only describes R_RISCV_GOT_HI20 with offset 0 to a label. The topic of addends for R_RISCV_GOT_HI20 was discussed in https://github.com/riscv-non-isa/riscv-elf-psabi-doc/issues/184#issuecomment-830988778 Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com> * isa/rv64mi/illegal.S: illegal addend for R_RISCV_GOT_HI20 With Ubuntu's GCC 15.2.0-7ubuntu1 building results in an error: rv64mi-p-illegal-illegal.o: in function `reset_vector': (.text.init+0x1dc): dangerous relocation: The addend isn't allowed for R_RISCV_GOT_HI20 Split loading mtvec_handler + 1 into register t into separate load and add instructions. https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/master/riscv-elf.adoc only describes R_RISCV_GOT_HI20 with offset 0 to a label. The topic of addends for R_RISCV_GOT_HI20 was discussed in https://github.com/riscv-non-isa/riscv-elf-psabi-doc/issues/184#issuecomment-830988778 Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com> --------- Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
2026-01-07benchmarks: undefined reference to `__printf_chk' (#640)Heinrich Schuchardt1-1/+1
When building with Ubuntu's gcc 15.2.0-7ubuntu1 an error occurs: undefined reference to `__printf_chk' This is due to _FORTIFY_SOURCE being set by the toolchain defaults. Undefine the symbol. Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
2026-01-07Avoid RWX section (#641)Heinrich Schuchardt1-2/+7
Building leads to a warning riscv64-linux-gnu/bin/ld: warning: median.riscv has a LOAD segment with RWX permissions The text sections should we RX only. Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
2026-01-07Fix QEMU syscall support (#642)Heinrich Schuchardt2-20/+9
* benchmark: correct tohost, fromhost symbol size Running on the spike virtual machine qemu-system-riscv64 -M spike -nographic -bios none \ -kernel benchmarks/qsort.riscv fails with qemu-system-riscv64: HTIF tohost must be 8 bytes because the symbol size if the ELF file is 0. Move the symbols tohost and fromhost to C code. Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com> * syscalls: QEMU only implements HTIF_CONSOLE_CMD_PUTC The only console output syscall emulated by QEMU 10.2 via the HTIF interface is writing a single character. String output is not available. Rewrite printstr() as a loop using HTIF_CONSOLE_CMD_PUTC. Rewrite putchar() using HTIF_CONSOLE_CMD_PUTC. With this change the benchmarks provide output on the Spike virtual machine: $ qemu-system-riscv64 -M spike -nographic -bios none \ -kernel benchmarks/qsort.riscv mcycle = 752216 minstret = 749734 Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com> --------- Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
2026-01-06syscalls: use correct printf code for counters[] (#638)Heinrich Schuchardt1-1/+1
Building benchmarks/common/syscalls.c results in a warning: warning: format ‘%d’ expects argument of type ‘int’, but argument 4 has type ‘uintptr_t’ {aka ‘long unsigned int’} [-Wformat=] 118 | pbuf += sprintf(pbuf, "%s = %d\n", counter_names[i], counters[i]); | ~^ ~~~~~~~~~~~ | | | | int uintptr_t {aka long unsigned int} Use %lu for to print uintptr_t. Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
2025-12-02[debug] check triggers exist in icount/itrigger/etrigger test (#617)Zane2-1/+22
2025-11-24[debug] Misc debug fixes (#632)hirooih4-18/+19
* debug: test S and U are supported for tdata1 Set MCONTROL_S and/or MCONTROL_U bits only if the hart supports S-mode and/or U-mode. Signed-off-by: Hiroo HAYASHI <24754036+hirooih@users.noreply.github.com> * debug: test S is supported in TranslateTest TranslateTest depends on the S extension being supported. Without this change, the test fails on targets that do not support S extension. Signed-off-by: Hiroo HAYASHI <24754036+hirooih@users.noreply.github.com> * debug: rename honors_tdata1_hmode to honors_tdata1_dmode This is for the dmode bit in tdata1. Signed-off-by: Hiroo HAYASHI <24754036+hirooih@users.noreply.github.com> * debug: fixed obsoleted openocd-command names This suppresses warnings like: Warning: Command 'gdb_port' is deprecated, please use 'gdb port' instead. Warning: Command 'tcl_port' is deprecated, please use 'tcl port' instead. Warning: Command 'telnet_port' is deprecated, please use 'telnet port' instead. from debug/README.md: > openocd ..., which should be the latest from https://github.com/riscv/riscv-openocd.git. Signed-off-by: Hiroo HAYASHI <24754036+hirooih@users.noreply.github.com> * debug: not to throw exception on an expected error This change makes debug/testlib.py not throw an exception when an expected error occurs while creating the logs/ directory. Fixing this is useful when debugging with python debugger not to break at the exception every time. Signed-off-by: Hiroo HAYASHI <24754036+hirooih@users.noreply.github.com> * debug: call create() instead of target() U500Sim.py is the only example of a target that uses VcsSim class. It does not work without this fix. Signed-off-by: Hiroo HAYASHI <24754036+hirooih@users.noreply.github.com> * debug: VcsSim: make server start string configurable example: for https://github.com/fjullien/jtag_vpi/blob/master/jtag_common.c#L80 def create(self): return testlib.VcsSim(..., server_stared="^Starting jtag_vpi server:.*, port (\d+)/tcp") Signed-off-by: Hiroo HAYASHI <24754036+hirooih@users.noreply.github.com> --------- Signed-off-by: Hiroo HAYASHI <24754036+hirooih@users.noreply.github.com>
2025-11-24debug: test implements_page_virtual_memory before executing sfence.vma ↵hirooih1-7/+8
instruction (#636) * debug: test implements_page_virtual_memory before executing sfence.vma instruction cf. #537 Signed-off-by: Hiroo HAYASHI <24754036+hirooih@users.noreply.github.com> * define exec_sfence_vma() Signed-off-by: Hiroo HAYASHI <24754036+hirooih@users.noreply.github.com> --------- Signed-off-by: Hiroo HAYASHI <24754036+hirooih@users.noreply.github.com>
2025-11-18[debug] Add support_abstractauto option to Spike target (#635)Nadime Barhoumi1-2/+5
This option is enabled by default (true). When set to false, it disables the optional register 'abstractauto' by using the Spike command option `--dm-no-abstractauto`.
2025-11-16Clear mcountinhibit.ir in minstret test to ensure minstret increments. (#634)Luke Wren1-0/+9
2025-11-06[debug] Support mcontrol6 in TriggerDmode test (#620)Zane3-4/+29
2025-10-30Improve Ziccid test to handle in-order fetch ruleAndrew Waterman1-5/+61
2025-10-24Fix duplicate mulh test case in rv32m tests (#630)404allen4041-1/+1
2025-10-14Merge pull request #629 from aman4150/masterAman2-0/+4
rv64um: Add a new division test for a potential false-positive signed overflow case
2025-06-13Merge pull request #624 from riscv-software-src/ziccidAndrew Waterman3-2/+61
Add Ziccid test
2025-06-13Add Ziccid testAndrew Waterman3-2/+61
Requires latest Spike.
2025-06-12Replace floating magic constant by macro (#623)Andrew Waterman17-43/+61
* macro: define [sq]NaN[dfh] and [n]Inf[dfh] Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> * apply sNaNf Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> * apply sNaNd Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> * apply qNaNd Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> * apply qNaNf Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> * apply sNaNh Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> * apply Infh Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> * apply nInfh Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> * apply Infd Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> * apply nInfd Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> * apply Inff Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> * apply nInff Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com> --------- Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
2025-06-10apply nInffChih-Min Chao4-4/+4
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
2025-06-10apply InffChih-Min Chao5-5/+5
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
2025-06-10apply nInfdChih-Min Chao3-3/+3
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
2025-06-10apply InfdChih-Min Chao4-4/+4
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
2025-06-10apply nInfhChih-Min Chao1-1/+1
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
2025-06-10apply InfhChih-Min Chao2-2/+2
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
2025-06-10apply sNaNhChih-Min Chao2-2/+2
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
2025-06-10apply qNaNfChih-Min Chao4-9/+9
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
2025-06-10apply qNaNdChih-Min Chao4-5/+5
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
2025-06-10apply sNaNdChih-Min Chao3-5/+5
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
2025-06-10apply sNaNfChih-Min Chao3-5/+5
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
2025-06-10macro: define [sq]NaN[dfh] and [n]Inf[dfh]Chih-Min Chao1-0/+18
Signed-off-by: Chih-Min Chao <chihmin.chao@sifive.com>
2025-04-28CI: Bump GitHub actions to latest versions (#619)Andrew Waterman2-11/+11
Co-authored-by: liangzhen <zhen.liang@spacemit.com>
2025-04-18Fix rv32ud-p-fcvt_w and -recoding being redundant with rv32ufAndrew Waterman3-3/+7
Resolves #614
2025-04-18configure: do autoupdateAriel Xiong1-4/+4
Signed-off-by: Ariel Xiong <ArielHeleneto@outlook.com>
2025-04-15Remove use of qNaN/sNaN in a different way to fix LLVM build (#613)Andrew Waterman9-69/+87
* Fix qNaN replacement in a more scalable way * Fix sNaN replacement in same manner as qNaN
2025-04-15Fix rv64ua-amomaxu_w and rv64ua-amominu_w testcases for rv32 (#611)etterli2-0/+8
Test case 6 of both rv64ua-amomaxu_w and rv64ua-amominu_w test a rv64 specific behaviour that the upper 32 bits must be ignored. This test does not apply for rv32 and this commits disables these tests for rv32.
2025-04-14Remove use of qNaN/sNaN to fix LLVM build (#612)Andrew Waterman9-22/+35
* Remove use of qNaN to fix LLVM build * Remove use of sNaN to fix LLVM build
2025-04-10Add test for pmpaddr[G-1] edge cases (#609)Tim Hutt4-0/+164
This adds a test for pmpaddr[G-1] edge cases. This bit is unusual because it has read legalisation that depends on another CSR. If the pmpcfg mode is OFF/TOR then it reads as 0. Otherwise it reads as its underlying value. Irrespective of pmpcfg it is always writable so you can always write it even when it reads as zero. Another edge case is when CSRS/C are used to modify a *different* bit and the underlying value of pmpaddr[G-1] is 1 but it reads as 0. The spec says "All CSR instructions atomically read-modify-write a single CSR" so even if these instructions don't directly modify pmpaddr[G-1] they do read it and then write it back, which will change its underlying value since it reads as 0.
2025-04-07Add instret_overflow testAndrew Waterman4-0/+52
See https://github.com/riscv/riscv-isa-manual/issues/1255 cc @Timmmm
2025-04-07Moving env forward (#607)Ariel Xiong1-0/+0
2025-04-03Removed irrelevant test (#605)Kathlene Magnus2-71/+0
2025-02-28Merge pull request #603 from riscv-software-src/bmark_compileJerry Zhao1-4/+11
Compile non-vec benchmarks for rv64gc
2025-02-24Compile non-vec benchmarks for rv64gcJerry Zhao1-4/+11
2025-02-07Added instructions to handle Rs1 and Rd dependency in load-store bypass ↵splinedrive1-0/+14
sequences. (#602)
2025-01-29Add Load-Store and Store-Load Bypass Tests for Forwarding in Pipelined CPU ↵splinedrive7-5/+301
(#600) - Enhanced test_macros.h with TEST_LD_ST_BYPASS and TEST_ST_LD_BYPASS macros to validate both load-store and store-load bypassing logic.
2024-11-18Suppress implicit-int and implicit-function-declaration warnings in ↵Tommy Murphy1-1/+1
Dhrystone which are non-fatal warnings in GCC 13 but fatal errors in GCC 14 (#587)
2024-11-16Fix the typo in the Makefrag for the reference to 2-stage-translation.S (#596)Kun Lu1-1/+1
2024-11-11Add hypervisor 2-stage translation test (#558)heiyuen19993-1/+147
* Create hypervisor Signed-off-by: heiyuen1999 <110437689+heiyuen1999@users.noreply.github.com> * Delete isa/hypervisor Signed-off-by: heiyuen1999 <110437689+heiyuen1999@users.noreply.github.com> * Create Makefrag Signed-off-by: heiyuen1999 <110437689+heiyuen1999@users.noreply.github.com> * Create 2-stage_translation.S set 2-stage translation, do a simple load and store Signed-off-by: heiyuen1999 <110437689+heiyuen1999@users.noreply.github.com> * Update Makefile add hypervisor to Makefile Signed-off-by: heiyuen1999 <110437689+heiyuen1999@users.noreply.github.com> * Update 2-stage_translation.S fix some error Signed-off-by: heiyuen1999 <110437689+heiyuen1999@users.noreply.github.com> * Update 2-stage_translation.S Signed-off-by: heiyuen1999 <110437689+heiyuen1999@users.noreply.github.com> * Update Makefrag Signed-off-by: heiyuen1999 <110437689+heiyuen1999@users.noreply.github.com> * Update Makefile Signed-off-by: heiyuen1999 <110437689+heiyuen1999@users.noreply.github.com> * Modify address alignment Modify address alignment to allow the page table to store 2048 entries Signed-off-by: heiyuen1999 <110437689+heiyuen1999@users.noreply.github.com> --------- Signed-off-by: heiyuen1999 <110437689+heiyuen1999@users.noreply.github.com>
generated by cgit v1.1 at 2026-03-02 05:16:08 +0000