Package: luajit Version: 2.1.0+openresty20251030-1 Severity: important X-Debbugs-Cc: [email protected], [email protected] User: [email protected] Usertags: riscv64
Dear maintainer, while debugging a autopkgtest regression in python-mpv after switching mpv to luajit, I tried to run some commands with luajit directly to identify the issue. At some point I came across a comment on Github https://github.com/LuaJIT/LuaJIT/pull/1267#issuecomment-2495376529 And indeed, luajit -e 'print(pcall(pcall))' on riscv64 causes a SIGSEGV. On amd64 it produces: $ luajit -e 'print(pcall(pcall))' false bad argument #1 to '?' (value expected) >From gdb on ricci.d.o/riscv64: Starting program: /usr/bin/luajit -e print\(pcall\(pcall\)\) [Thread debugging using libthread_db enabled] Using host libthread_db library "/usr/lib/riscv64-linux-gnu/libthread_db.so.1". Program received signal SIGSEGV, Segmentation fault. lj_err_callermsg (L=L@entry=0x3ff7d88380, msg=0x3ff7d8a7b8 "bad argument #1 to '?' (value expected)") at ./src/lj_err.c:1022 warning: 1022 ./src/lj_err.c: No such file or directory (gdb) bt #0 lj_err_callermsg (L=L@entry=0x3ff7d88380, msg=0x3ff7d8a7b8 "bad argument #1 to '?' (value expected)") at ./src/lj_err.c:1022 #1 0x0000002aaaadf7ae in err_argmsg (L=0x3ff7d88380, narg=<optimized out>, narg@entry=1, msg=<optimized out>) at ./src/lj_err.c:1074 #2 0x0000002aaaadf7e0 in lj_err_arg (L=<optimized out>, narg=narg@entry=1, em=em@entry=LJ_ERR_NOVAL) at ./src/lj_err.c:1091 #3 0x0000002aaaae04a8 in lj_lib_checkany (L=<optimized out>, narg=1) at ./src/lj_lib.c:192 #4 lj_ffh_pcall (L=<optimized out>) at ./src/lib_base.c:357 #5 0x0000002aaab0c154 in lj_fff_fallback () at ./src/buildvm_riscv.dasc:2931 #6 0x0000002aaaafd304 in lua_pcall (L=L@entry=0x3ff7d88380, nargs=nargs@entry=0, nresults=nresults@entry=0, errfunc=errfunc@entry=2) at ./src/lj_api.c:1151 #7 0x0000002aaaab24b0 in docall (L=L@entry=0x3ff7d88380, narg=narg@entry=0, clear=clear@entry=1) at ./src/luajit.c:137 #8 0x0000002aaaab2742 in dostring (L=L@entry=0x3ff7d88380, s=0x3ffffffe33 "print(pcall(pcall))", name=name@entry=0x2aaab0cfb0 "=(command line)") at ./src/luajit.c:191 #9 0x0000002aaaab2c96 in runargs (L=<optimized out>, argv=<optimized out>, argn=<optimized out>) at ./src/luajit.c:473 #10 pmain (L=0x3ff7d88380) at ./src/luajit.c:560 #11 0x0000002aaab0a05c in lj_BC_FUNCC () at ./src/buildvm_riscv.dasc:1279 #12 0x0000002aaaafd428 in lua_cpcall (L=L@entry=0x3ff7d88380, func=func@entry=0x2aaaab2906 <pmain>, ud=ud@entry=0x0) at ./src/lj_api.c:1179 #13 0x0000002aaaab1fe6 in main (argc=3, argv=0x3ffffffb78) at ./src/luajit.c:595 I am not sure if this is anyway related to the issue I was debugging in mpv, but I think it would be good to have it fixed anway. Cheers (data below is from reporting reportbug on my amd64 machine, so s not reall helpful) -- System Information: Debian Release: forky/sid APT prefers unstable-debug APT policy: (650, 'unstable-debug'), (650, 'unstable'), (601, 'testing'), (600, 'experimental-debug'), (600, 'experimental') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 6.17.13+deb14-amd64 (SMP w/8 CPU threads; PREEMPT) Kernel taint flags: TAINT_PROPRIETARY_MODULE, TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE=en_US:en Shell: /bin/sh linked to /usr/bin/dash Init: systemd (via /run/systemd/system) Versions of packages luajit depends on: ii libc6 2.43-7 ii libgcc-s1 16.2.0-4 ii libluajit-5.1-2 2.1.0+openresty20251030-1+b2 ii libluajit-5.1-common 2.1.0+openresty20251030-1 luajit recommends no packages. luajit suggests no packages. -- Sebastian Ramacher

