Messages in this thread | | | Date | Mon, 17 Mar 2014 12:49:04 -0700 | Subject | Re: [PATCH] improve_stack: make stack dump output useful again | From | Linus Torvalds <> |
| |
On Mon, Mar 17, 2014 at 12:31 PM, Sasha Levin <sasha.levin@oracle.com> wrote: > > What about duplicate symbol names? Without the full hex address we lose > the ability to correctly translate from symbol name to file/line.
Yeah, and we should strive to avoid that issue.
The thing is, think about address randomization. We can pretty much assume that is going to be the standard behavior due to it being really cheap (just a trivial boot time cost), and helping some security issues.
So we might have a few ambiguous cases, that's still better than the alternative.
Linus
| |