Re: [PATCH 10/13] x86/dumpstack: Try harder to get a call trace on stack overflow

2016-06-16 Thread Josh Poimboeuf
On Thu, Jun 16, 2016 at 11:37:07AM -0700, Andy Lutomirski wrote: > On Thu, Jun 16, 2016 at 11:33 AM, Josh Poimboeuf wrote: > > On Thu, Jun 16, 2016 at 11:22:14AM -0700, Andy Lutomirski wrote: > >> On Thu, Jun 16, 2016 at 11:16 AM, Josh Poimboeuf > >> wrote: > >> > On Wed, Jun 15, 2016 at 05:28:3

Re: [PATCH 10/13] x86/dumpstack: Try harder to get a call trace on stack overflow

2016-06-16 Thread Andy Lutomirski
On Thu, Jun 16, 2016 at 11:33 AM, Josh Poimboeuf wrote: > On Thu, Jun 16, 2016 at 11:22:14AM -0700, Andy Lutomirski wrote: >> On Thu, Jun 16, 2016 at 11:16 AM, Josh Poimboeuf wrote: >> > On Wed, Jun 15, 2016 at 05:28:32PM -0700, Andy Lutomirski wrote: >> >> If we overflow the stack, print_context

Re: [PATCH 10/13] x86/dumpstack: Try harder to get a call trace on stack overflow

2016-06-16 Thread Josh Poimboeuf
On Thu, Jun 16, 2016 at 11:22:14AM -0700, Andy Lutomirski wrote: > On Thu, Jun 16, 2016 at 11:16 AM, Josh Poimboeuf wrote: > > On Wed, Jun 15, 2016 at 05:28:32PM -0700, Andy Lutomirski wrote: > >> If we overflow the stack, print_context_stack will abort. Detect > >> this case and rewind back into

Re: [PATCH 10/13] x86/dumpstack: Try harder to get a call trace on stack overflow

2016-06-16 Thread Andy Lutomirski
On Thu, Jun 16, 2016 at 11:16 AM, Josh Poimboeuf wrote: > On Wed, Jun 15, 2016 at 05:28:32PM -0700, Andy Lutomirski wrote: >> If we overflow the stack, print_context_stack will abort. Detect >> this case and rewind back into the valid part of the stack so that >> we can trace it. >> >> Signed-off

Re: [PATCH 10/13] x86/dumpstack: Try harder to get a call trace on stack overflow

2016-06-16 Thread Josh Poimboeuf
On Wed, Jun 15, 2016 at 05:28:32PM -0700, Andy Lutomirski wrote: > If we overflow the stack, print_context_stack will abort. Detect > this case and rewind back into the valid part of the stack so that > we can trace it. > > Signed-off-by: Andy Lutomirski > --- > arch/x86/kernel/dumpstack.c | 7

[PATCH 10/13] x86/dumpstack: Try harder to get a call trace on stack overflow

2016-06-15 Thread Andy Lutomirski
If we overflow the stack, print_context_stack will abort. Detect this case and rewind back into the valid part of the stack so that we can trace it. Signed-off-by: Andy Lutomirski --- arch/x86/kernel/dumpstack.c | 7 +++ 1 file changed, 7 insertions(+) diff --git a/arch/x86/kernel/dumpstac