On Wed, Jul 27, 2022 at 11:15:06AM +0800, Chen Zhongjin wrote: > This reverts commit f1d9a2abff66aa8156fbc1493abed468db63ea48. > > When CONFIG_GCOV_PROFILE_ALL is enabled, show_stack() and related > functions (e.g. dump_stack) will break for x86 ORC unwinder. > > Call Trace: > <TASK> > ? dump_stack_lvl+0x83/0xb7 > ? schedule+0x1/0x190 > ? dump_stack+0x13/0x1f > ? handler_pre0+0x3f/0x53 [kp_unwind] > ... > > show_trace_log_lvl() searches text address on stack to validate > whether unwind results are reliable. The code: Hi, Thanks for the patch. The change itself makes sense, though I'm having trouble recreating the bug described in the patch description. I enabled CONFIG_GCOV_PROFILE_ALL and did 'echo l > /proc/sysrq-trigger', but I got a valid stack trace: # echo l > /proc/sysrq-trigger [ 343.916728] sysrq: Show backtrace of all active CPUs [ 343.917459] NMI backtrace for cpu 2 [ 343.917884] CPU: 2 PID: 1007 Comm: bash Not tainted 5.19.0-rc8+ #68 [ 343.918562] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.0-1.fc36 04/01/2014 [ 343.919534] Call Trace: [ 343.919904] <TASK> [ 343.920226] dump_stack_lvl+0xcc/0x11b [ 343.920742] dump_stack+0x17/0x24 [ 343.921199] nmi_cpu_backtrace.cold+0xb5/0x10d [ 343.921795] ? lapic_can_unplug_cpu+0xa0/0xa0 [ 343.922375] nmi_trigger_cpumask_backtrace+0x171/0x200 [ 343.923053] arch_trigger_cpumask_backtrace+0x21/0x30 [ 343.923599] sysrq_handle_showallcpus+0x1f/0x30 [ 343.924104] __handle_sysrq.cold+0x81/0x234 [ 343.924576] write_sysrq_trigger+0x6a/0x90 [ 343.925098] proc_reg_write+0x99/0x120 [ 343.925587] vfs_write+0x16c/0x350 [ 343.926093] ksys_write+0x8c/0x180 [ 343.926597] __x64_sys_write+0x21/0x30 [ 343.927144] do_syscall_64+0x76/0x100 [ 343.927676] entry_SYSCALL_64_after_hwframe+0x63/0xcd [ 343.928343] RIP: 0033:0x7f555e8f7b50 [ 343.928822] Code: 73 01 c3 48 8b 0d 38 83 2c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 0f 1f 44 00 00 83 3d 79 db 2c 00 00 75 10 b8 01 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 31 c3 48 83 ec 08 e8 1e e3 01 00 48 89 04 24 [ 343.930801] RSP: 002b:00007ffdd20d0978 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 [ 343.931761] RAX: ffffffffffffffda RBX: 0000000000000002 RCX: 00007f555e8f7b50 [ 343.932551] RDX: 0000000000000002 RSI: 000055ff471bfe20 RDI: 0000000000000001 [ 343.933347] RBP: 000055ff471bfe20 R08: 00007f555ebc2740 R09: 00007f555f80f700 [ 343.934155] R10: 0000000000000073 R11: 0000000000000246 R12: 0000000000000002 [ 343.934962] R13: 0000000000000001 R14: 00007f555ebc15e0 R15: 00007f555ebbd3c0 [ 343.935798] </TASK> [ 343.936341] Sending NMI from CPU 2 to CPUs 0-1,3: [ 343.937163] NMI backtrace for cpu 3 [ 343.937167] CPU: 3 PID: 600 Comm: systemd-journal Not tainted 5.19.0-rc8+ #68 Was this with an upstream kernel? Can you share the config and toolchain versions? -- Josh
