summary refs log tree commit diff stats
path: root/results/classifier/118/all/1703
diff options
context:
space:
mode:
authorChristian Krinitsin <mail@krinitsin.com>2025-06-16 16:59:00 +0000
committerChristian Krinitsin <mail@krinitsin.com>2025-06-16 16:59:33 +0000
commit9aba81d8eb048db908c94a3c40c25a5fde0caee6 (patch)
treeb765e7fb5e9a3c2143c68b0414e0055adb70e785 /results/classifier/118/all/1703
parentb89a938452613061c0f1f23e710281cf5c83cb29 (diff)
downloadqemu-analysis-9aba81d8eb048db908c94a3c40c25a5fde0caee6.tar.gz
qemu-analysis-9aba81d8eb048db908c94a3c40c25a5fde0caee6.zip
add 18th iteration of classifier
Diffstat (limited to 'results/classifier/118/all/1703')
-rw-r--r--results/classifier/118/all/170375
1 files changed, 75 insertions, 0 deletions
diff --git a/results/classifier/118/all/1703 b/results/classifier/118/all/1703
new file mode 100644
index 000000000..32699dae8
--- /dev/null
+++ b/results/classifier/118/all/1703
@@ -0,0 +1,75 @@
+semantic: 0.947
+permissions: 0.942
+mistranslation: 0.935
+arm: 0.926
+performance: 0.921
+architecture: 0.918
+debug: 0.916
+graphic: 0.914
+user-level: 0.910
+assembly: 0.907
+register: 0.903
+risc-v: 0.901
+TCG: 0.900
+device: 0.900
+virtual: 0.892
+PID: 0.889
+peripherals: 0.865
+kernel: 0.861
+VMM: 0.859
+ppc: 0.851
+network: 0.846
+socket: 0.837
+boot: 0.837
+hypervisor: 0.832
+KVM: 0.831
+vnc: 0.809
+x86: 0.803
+files: 0.758
+i386: 0.728
+
+Undefined behaviour when running guest with -enable-kvm and attached debugger
+Description of problem:
+When attaching a debugger to a Qemu instance with `-enable-kvm` my linux kernel panics on (f.e.) module load.
+I am not sure if this is a Qemu bug, however the issue is not occurring if I a) do not attach the debugger (even though Qemu is listening for one) or b) I do not pass `-enable-kvm` (and attach a debugger).
+The issue seems to relate to the `lx-symbols` command provided by the Linux kernel gdb script suite.
+Every time a module is loaded this script will reload the symbols for said module which may take some time, so maybe there is some race involved?
+The issue does not reproduce if you do not run `lx-symbols` prior to continuing (it will however run automatically after first module load as it adds a breakpoint to kernel/module/main.c:do_init_module, so the kernel will crash after the second module load)
+Steps to reproduce:
+1. Start kernel with some img
+2. Attach gdb debugger
+3. Run the `lx-symbols` command provided by the Linux kernel gdb scripts in gdb, run `continue` in gdb
+3. Load a kernel module
+Additional information:
+This is the kernel stack trace:
+```
+[   22.930691] invalid opcode: 0000 [#1] PREEMPT SMP NOPTI
+[   22.931174] CPU: 2 PID: 241 Comm: modprobe Tainted: G            E      6.1.31+ #2
+[   22.931675] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.2-1.fc37 04/01/2014
+[   22.931675] RIP: 0010:do_init_module+0x1/0x210
+[   22.931675] Code: 74 0c 48 8b 78 08 48 89 de e8 8b df ff ff 65 ff 0d 84 94 ef 7e 0f 85 e5 fe ff ff 0f 1f 44 00 008
+[   22.931675] RSP: 0018:ffffc90000593e40 EFLAGS: 00010246
+[   22.931675] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 000000000006e202
+[   22.931675] RDX: 000000000006e002 RSI: 5b4504de76578f76 RDI: ffffffffc024e180
+[   22.931675] RBP: ffffc90000593e50 R08: ffffea0000174a88 R09: ffffea0000174ac0
+[   22.931675] R10: ffff888006a9c270 R11: 0000000000000100 R12: 0000562f9087b4a0
+[   22.931675] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
+[   22.931675] FS:  00007f0dbc5a4040(0000) GS:ffff88801f500000(0000) knlGS:0000000000000000
+[   22.931675] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
+[   22.931675] CR2: 00007ffdc94bc3f8 CR3: 0000000006f8e000 CR4: 00000000003506e0
+[   22.931675] Call Trace:
+[   22.931675]  <TASK>
+[   22.931675]  ? die+0x32/0x80
+[   22.931675]  ? do_trap+0xd6/0x100
+[   22.931675]  ? do_init_module+0x1/0x210
+[   22.931675]  ? do_error_trap+0x6a/0x90
+[   22.931675]  ? do_init_module+0x1/0x210
+[   22.931675]  ? exc_invalid_op+0x4c/0x60
+[   22.931675]  ? do_init_module+0x1/0x210
+[   22.931675]  ? asm_exc_invalid_op+0x16/0x20
+[   22.931675]  ? do_init_module+0x1/0x210
+[   22.931675]  __do_sys_finit_module+0x9e/0xf0
+[   22.931675]  do_syscall_64+0x63/0x90
+[   22.931675]  ? exit_to_user_mode_prepare+0x1a/0x120
+[   22.931675]  entry_SYSCALL_64_after_hwframe+0x63/0xcd
+```