Re: [RFC] how to perform a safe NMI stack trace on all CPUs on x86?

2015-05-14 Thread long.wanglong
On 2015/5/13 22:26, Jiri Kosina wrote: > On Wed, 13 May 2015, 王龙 wrote: > >> Hi all, >> >> In kernel before 3.19, when trigger_all_cpu_backtrace() is called on x86, >> it will trigger an NMI on each CPU and call show_regs(). But this can lead >> to a hard lock up if the NMI comes in on another pr

Re: [RFC] how to perform a safe NMI stack trace on all CPUs on x86?

2015-05-14 Thread long.wanglong
On 2015/5/13 22:22, Steven Rostedt wrote: > On Wed, 13 May 2015 22:14:54 +0800 > "王龙" wrote: > > >> context. But how do we fix this problem in older version of kernel(eg, 3.10 >> stable)? >> The 3.10 stable has no "switch printk routine" and "seq_buf" infrastructures. >> >> Could anyone give m

Re: [RFC] how to perform a safe NMI stack trace on all CPUs on x86?

2015-05-13 Thread Jiri Kosina
On Wed, 13 May 2015, 王龙 wrote: > Hi all, > > In kernel before 3.19, when trigger_all_cpu_backtrace() is called on x86, > it will trigger an NMI on each CPU and call show_regs(). But this can lead > to a hard lock up if the NMI comes in on another printk(). > > The commit a9edc88093287183ac934be

[RFC] how to perform a safe NMI stack trace on all CPUs on x86?

2015-05-13 Thread 王龙
Hi all, In kernel before 3.19, when trigger_all_cpu_backtrace() is called on x86, it will trigger an NMI on each CPU and call show_regs(). But this can lead to a hard lock up if the NMI comes in on another printk(). The commit a9edc88093287183ac934be44f295f183b2c62dd (x86/nmi: Perform a safe NM

Re: [RFC] how to perform a safe NMI stack trace on all CPUs on x86?

2015-05-13 Thread Steven Rostedt
On Wed, 13 May 2015 22:14:54 +0800 "王龙" wrote: > context. But how do we fix this problem in older version of kernel(eg, 3.10 > stable)? > The 3.10 stable has no "switch printk routine" and "seq_buf" infrastructures. > > Could anyone give me some ideas? > Backport the necessary patches. --