Re: [PATCH 3/7] dump_stack: consolidate dump_stack() implementations and unify their behaviors

[Tejun Heo <tj@kernel.org>]

Hello,

On Fri, Apr 12, 2013 at 04:39:01PM -0400, Chris Metcalf wrote:
> On 4/3/2013 3:14 PM, Tejun Heo wrote:
> > This patch expands the dummy fallback dump_stack() implementation in
> > lib/dump_stack.c such that it prints out debug information (taken from
> > x86) and invokes show_stack(NULL, NULL) and drops arch-specific
> > dump_stack() implementations in all archs except blackfin.  Blackfin's
> > dump_stack() does something wonky that I don't understand.
> 
> arch/tile has a dump_stack() entry point in assembly (see
> arch/tile/kernel/entry.S) that passes to _dump_stack(), which passes
> to dump_stack_regs(), which passes to tile_show_stack().  Similarly,
> show_stack() passes to tile_show_stack(), which does lots of work to
> cross into userspace and continue showing the stack if possible,
> print symbol info for both kernel and userspace, etc.  I don't know
> if that's all supported fully in Tejun's generic version.

I see.  In that case, this patchset shouldn't change anything for tile
as the generic dump_stack() is linked iff arch implementation doesn't
exist.

> Would it make sense for me to look at this after it's merged up into
> -next and see what tile needs?  Unfortunately I'm taking off for a
> week's vacation so won't be able to do much until after the 23rd.

It'd be nice to get it inline with other archs but I don't think
updating tile to print the common debug info is a priority, so no
problem with dealing with it later.

Thanks.

Thanks.

-- 
tejun