Re: [PATCH 05/12] powerpc/ftrace: ftrace_graph_caller() needs to save/restore toc

[Torsten Duwe <duwe@lst.de>]

On Thu, Feb 25, 2016 at 11:30:38AM +1100, Balbir Singh wrote:
> On 25/02/16 01:28, Michael Ellerman wrote:
> > Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
> > ---
> >  arch/powerpc/kernel/entry_64.S | 8 ++++++++
> >  1 file changed, 8 insertions(+)
> >

Ah, -mprofile-kernel, DYNAMIC_FTRACE but without REGS.
Hadn't considered that, thanks!

> >
> > diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S
> > index 2a7313cfbc7d..9e77a2c8f218 100644
> > --- a/arch/powerpc/kernel/entry_64.S
> > +++ b/arch/powerpc/kernel/entry_64.S
> > @@ -1237,6 +1237,11 @@ _GLOBAL(ftrace_graph_caller)
> >  	std	r5, 64(r1)
> >  	std	r4, 56(r1)
> >  	std	r3, 48(r1)
> > +
> > +	/* Save callee's TOC in the ABI compliant location */
> > +	std	r2, 24(r1)
> R2_STACK_OFFSET for readability?

I have encountered LRSAVE vs. PPC_LR_STKOFF and
STK_GOT vs. R2_STACK_OFFSET, some usable in assembler source, some in C.

> > +	ld	r2, PACATOC(r13)	/* get kernel TOC in r2 */
> > +
> >  	mfctr	r4		/* ftrace_caller has moved local addr here */
> >  	std	r4, 40(r1)
> >  	mflr	r3		/* ftrace_caller has restored LR from stack */
> > @@ -1262,6 +1267,9 @@ _GLOBAL(ftrace_graph_caller)
> >  	ld	r4, 56(r1)
> >  	ld	r3, 48(r1)
> >  
> > +	/* Restore callee's TOC */
> > +	ld	r2, 24(r1)
> > +
> >  	addi	r1, r1, 112
> >  	mflr	r0
> >  	std	r0, LRSAVE(r1)
> 
> Reviewed-by: Balbir Singh <bsingharora@gmail.com>

Reviewed-by: Torsten Duwe <duwe@suse.de>

	Torsten