diff options
Diffstat (limited to '')
| -rw-r--r-- | results/classifier/118/files/1917 | 81 | ||||
| -rw-r--r-- | results/classifier/118/files/1917940 | 68 |
2 files changed, 149 insertions, 0 deletions
diff --git a/results/classifier/118/files/1917 b/results/classifier/118/files/1917 new file mode 100644 index 00000000..c1b2e67c --- /dev/null +++ b/results/classifier/118/files/1917 @@ -0,0 +1,81 @@ +files: 0.891 +architecture: 0.868 +graphic: 0.864 +network: 0.820 +user-level: 0.815 +ppc: 0.812 +permissions: 0.797 +virtual: 0.793 +semantic: 0.781 +assembly: 0.778 +socket: 0.764 +performance: 0.763 +hypervisor: 0.753 +PID: 0.749 +register: 0.748 +arm: 0.748 +boot: 0.732 +TCG: 0.726 +vnc: 0.724 +mistranslation: 0.722 +device: 0.722 +peripherals: 0.709 +risc-v: 0.708 +debug: 0.708 +KVM: 0.680 +VMM: 0.636 +kernel: 0.627 +x86: 0.599 +i386: 0.484 + +cargo on ppc64 fails: invalid instruction: *EDIT*: on ntpdate as well +Description of problem: +Machine boots, +but when compiling a rust library, the following issue appears: +``` +cargo build --release --manifest-path rust-src/Cargo.toml +cargo[376]: illegal instruction (4) at 12e0933c0 nip 12e0933c0 lr 12dd7c768 code 1 in cargo[12dc10000+956000] +cargo[376]: code: 00000000 00000000 00000000 7c0802a6 f8010010 f821ff11 fba100d8 60000000 +cargo[376]: code: fbc100e0 8862cd50 28030000 418200d4 <104214c4> 38810070 3bc00000 7c4021ce +make: *** [Makefile:133: rust-src/target/release/libbcachefs_rust.a] Illegal instruction (core dumped) +make: *** Waiting for unfinished jobs.... +ar[375]: illegal instruction (4) at 3fff9b2a4dac nip 3fff9b2a4dac lr 3fff9b2a4da4 code 1 in libLLVM-16.so.1[3fff99f10000+792f000] +ar[375]: code: f87d0028 7fa3eb78 b29d0090 f89d0020 4810a8c1 60000000 7f83e378 7fa4eb78 +ar[375]: code: 7fc5f378 49bfd3e1 e8410028 60000000 <104214c4> 38810070 fb610080 eba29fe0 +make: *** [Makefile:129: libbcachefs.a] Illegal instruction (core dumped) +make: *** Deleting file 'libbcachefs.a' +``` +the core dump files of cargo and ar are attached + +~~I have no clue whether this is a rustc or qemu bug, so please let me know if this issue should be forwarded to rust devs~~ +EDIT: as this happens with ntpdate as well, I think it's an emulator issue: + +``` +ntpdig[1179]: illegal instruction (4) at 102382c4 nip 102382c4 lr 102382a8 code 1 in python3.11[10000000+63e000] +ntpdig[1179]: code: 3d22ffdd c8094448 fc1e0000 41c2022c 4bde9b5d e8410028 3d42ffdd 39200000 +ntpdig[1179]: code: c80a4450 91230000 fc1e0000 41c001a4 <ffe0f02c> fc1ff800 41c30280 3d22ffdd +``` +Steps to reproduce: +1. create a debian ppc64 root image using debian sid & debootstrap +2. install rust using rustup +3. compile bcachefs-tools in ppc64 + +2b. Install ntpdate using apt-get ntpdate +3b. run ntpdate +Additional information: +Core dump command: +``` +cat /proc/sys/kernel/core_pattern +|/bin/cp --sparse=always /dev/stdin /host//repos/janpieter/linux/bcachefs/ktest-out/core.%e.PID%p.SIG%s.TIME%t +``` + + +[core.ar.PID374.SIG4.TIME1696070088.xz](/uploads/6a540c4d13351871b1e22153ad87ab99/core.ar.PID374.SIG4.TIME1696070088.xz) AR core dump + +[core.ar.PID375.SIG4.TIME1696070088.xz](/uploads/7c314eba58c2190e3a9fbd88f8eb1242/core.ar.PID375.SIG4.TIME1696070088.xz) AR core dump + +[core.cargo.PID375.SIG4.TIME1696070087.xz](/uploads/0097d457eb2d25e0123874b59405647a/core.cargo.PID375.SIG4.TIME1696070087.xz) cargo core dump + +[core.cargo.PID376.SIG4.TIME1696070087.xz](/uploads/53834fa9608036d6de9dafc3f778f165/core.cargo.PID376.SIG4.TIME1696070087.xz) cargo core dump + +[core.ntpdig.PID1171.SIG4.TIME1696070657.xz](/uploads/8a96d86338d7c6bebe39657a24f570d8/core.ntpdig.PID1171.SIG4.TIME1696070657.xz) ntpdig core dump diff --git a/results/classifier/118/files/1917940 b/results/classifier/118/files/1917940 new file mode 100644 index 00000000..a743cbba --- /dev/null +++ b/results/classifier/118/files/1917940 @@ -0,0 +1,68 @@ +architecture: 0.952 +x86: 0.951 +files: 0.935 +user-level: 0.907 +performance: 0.902 +semantic: 0.894 +device: 0.886 +boot: 0.848 +graphic: 0.822 +mistranslation: 0.752 +permissions: 0.742 +debug: 0.732 +hypervisor: 0.669 +peripherals: 0.651 +PID: 0.632 +assembly: 0.625 +register: 0.612 +network: 0.609 +ppc: 0.607 +arm: 0.601 +kernel: 0.577 +i386: 0.555 +risc-v: 0.513 +virtual: 0.512 +KVM: 0.509 +socket: 0.496 +TCG: 0.479 +VMM: 0.467 +vnc: 0.461 + +-bios edk2-$arch-code doesn't work for x86 + +Whilst creating a flash device is recommended, -bios <file> is extremely useful in many cases as it automatically searches $PREFIX/share/qemu rather than requiring the caller (be it a human or a script) to work out where that directory is for the QEMU being called and prepend it to the file name. + +Currently, all the x86 EDK2 FD code files are 3653632 bytes in size, or 0x37c000 bytes. However, for some reason I cannot find the answer to (I traced the code back to 7587cf44019d593bb12703e7046bd7738996c55c), x86's -bios only allows files that are multiples of 64K in size (x86_bios_rom_init), which would require the EDK2 ROMs to be rounded up to 0x380000 bytes. If I delete the check, QEMU is able to load the only-16K-multiple-sized EDK2 and boot an OS just fine. If I pad EDK2 with 16K of zeroes at the *start* (since the ROM gets mapped counting backwards), it also works just fine (but padding at the *end* doesn't). Please therefore either relax the check in x86_bios_rom_init or ensure the EDK2 binary is suitably padded. + +The QEMU project is currently moving its bug tracking to another system. +For this we need to know which bugs are still valid and which could be +closed already. Thus we are setting the bug state to "Incomplete" now. + +If the bug has already been fixed in the latest upstream version of QEMU, +then please close this ticket as "Fix released". + +If it is not fixed yet and you think that this bug report here is still +valid, then you have two options: + +1) If you already have an account on gitlab.com, please open a new ticket +for this problem in our new tracker here: + + https://gitlab.com/qemu-project/qemu/-/issues + +and then close this ticket here on Launchpad (or let it expire auto- +matically after 60 days). Please mention the URL of this bug ticket on +Launchpad in the new ticket on GitLab. + +2) If you don't have an account on gitlab.com and don't intend to get +one, but still would like to keep this ticket opened, then please switch +the state back to "New" or "Confirmed" within the next 60 days (other- +wise it will get closed as "Expired"). We will then eventually migrate +the ticket automatically to the new system (but you won't be the reporter +of the bug in the new system and thus you won't get notified on changes +anymore). + +Thank you and sorry for the inconvenience. + + +[Expired for QEMU because there has been no activity for 60 days.] + |