Messages in this thread | | | Date | Fri, 29 Apr 2016 09:00:39 -0500 | From | Josh Poimboeuf <> | Subject | Re: [RFC PATCH v2 06/18] x86: dump_trace() error handling |
| |
On Fri, Apr 29, 2016 at 09:45:58PM +0800, Minfei Huang wrote: > On 04/28/16 at 03:44P, Josh Poimboeuf wrote: > > In preparation for being able to determine whether a given stack trace > > is reliable, allow the stacktrace_ops functions to propagate errors to > > dump_trace(). > > Hi, Josh. > > Have you considered to make walk_stack function as non-return function, > since there is no obvious error during detecting the frame points?
If you look at the next patch 07/18, there are several cases where walk_stack (print_context_stack_reliable) returns an error.
For example, if a function gets preempted before it gets a chance to save the frame pointer, the function's caller would get skipped on the stack trace. So for preempted tasks, we always have to consider their stacks unreliable.
-- Josh
| |