Re: [PATCH 10/12] powerpc/entry32: Blacklist exception entry points for kprobe.

[Naveen N. Rao]

Christophe Leroy wrote:
> Le 30/03/2020 Ã 20:33, Christophe Leroy a ÃcritÂ:
> > Le 30/03/2020 Ã 19:08, Naveen N. Rao a ÃcritÂ:
> > > Christophe Leroy wrote:
> > > > kprobe does not handle events happening in real mode.
> > > >
> > > > As exception entry points are running with MMU disabled,
> > > > blacklist them.
> > > >
> > > > Signed-off-by: Christophe Leroy <christophe.leroy@xxxxxx>
> > > > ---
> > > > Âarch/powerpc/kernel/entry_32.S | 7 +++++++
> > > > Â1 file changed, 7 insertions(+)
> > > >
> > > > diff --git a/arch/powerpc/kernel/entry_32.S 
> > > > b/arch/powerpc/kernel/entry_32.S
> > > > index 94f78c03cb79..9a1a45d6038a 100644
> > > > --- a/arch/powerpc/kernel/entry_32.S
> > > > +++ b/arch/powerpc/kernel/entry_32.S
> > > > @@ -51,6 +51,7 @@ mcheck_transfer_to_handler:
> > > > ÂÂÂÂ mfsprÂÂÂ r0,SPRN_DSRR1
> > > > ÂÂÂÂ stwÂÂÂ r0,_DSRR1(r11)
> > > > ÂÂÂÂ /* fall through */
> > > > +_ASM_NOKPROBE_SYMBOL(mcheck_transfer_to_handler)
> > > >
> > > > ÂÂÂÂ .globlÂÂÂ debug_transfer_to_handler
> > > > Âdebug_transfer_to_handler:
> > > > @@ -59,6 +60,7 @@ debug_transfer_to_handler:
> > > > ÂÂÂÂ mfsprÂÂÂ r0,SPRN_CSRR1
> > > > ÂÂÂÂ stwÂÂÂ r0,_CSRR1(r11)
> > > > ÂÂÂÂ /* fall through */
> > > > +_ASM_NOKPROBE_SYMBOL(debug_transfer_to_handler)
> > > >
> > > > ÂÂÂÂ .globlÂÂÂ crit_transfer_to_handler
> > > > Âcrit_transfer_to_handler:
> > > > @@ -94,6 +96,7 @@ crit_transfer_to_handler:
> > > > ÂÂÂÂ rlwinmÂÂÂ r0,r1,0,0,(31 - THREAD_SHIFT)
> > > > ÂÂÂÂ stwÂÂÂ r0,KSP_LIMIT(r8)
> > > > ÂÂÂÂ /* fall through */
> > > > +_ASM_NOKPROBE_SYMBOL(crit_transfer_to_handler)
> > > > Â#endif
> > > >
> > > > Â#ifdef CONFIG_40x
> > > > @@ -115,6 +118,7 @@ crit_transfer_to_handler:
> > > > ÂÂÂÂ rlwinmÂÂÂ r0,r1,0,0,(31 - THREAD_SHIFT)
> > > > ÂÂÂÂ stwÂÂÂ r0,KSP_LIMIT(r8)
> > > > ÂÂÂÂ /* fall through */
> > > > +_ASM_NOKPROBE_SYMBOL(crit_transfer_to_handler)
> > > > Â#endif
> > > >
> > > > Â/*
> > > > @@ -127,6 +131,7 @@ crit_transfer_to_handler:
> > > > ÂÂÂÂ .globlÂÂÂ transfer_to_handler_full
> > > > Âtransfer_to_handler_full:
> > > > ÂÂÂÂ SAVE_NVGPRS(r11)
> > > > +_ASM_NOKPROBE_SYMBOL(transfer_to_handler_full)
> > > > ÂÂÂÂ /* fall through */
> > > >
> > > > ÂÂÂÂ .globlÂÂÂ transfer_to_handler
> > > > @@ -286,6 +291,8 @@ reenable_mmu:
> > > > ÂÂÂÂ lwzÂÂÂ r2, GPR2(r11)
> > > > ÂÂÂÂ bÂÂÂ fast_exception_return
> > > > Â#endif
> > > > +_ASM_NOKPROBE_SYMBOL(transfer_to_handler)
> > > > +_ASM_NOKPROBE_SYMBOL(transfer_to_handler_cont)
> > >
> > > These are added after 'reenable_mmu', which is itself not blacklisted. 
> > > Is that intentional?
> >
> > Yes I put it as the complete end of the entry part, ie just before 
> > stack_ovf which is a function by itself.
> >
> > Note that reenable_mmu is inside an #ifdef CONFIG_TRACE_IRQFLAGS.
> >
> > I'm not completely sure where to put the _ASM_NOKPROBE_SYMBOL()s, that's 
> > the reason why I put it close to the symbol itself in my first series.
> >
> > Could you have a look at the code and tell me what looks the most 
> > appropriate as a location to you ?
> >
> > https://elixir.bootlin.com/linux/v5.6/source/arch/powerpc/kernel/entry_32.S#L230 
>
> Ok, thinking about it once more, I guess we have a problem as everything 
> after that reenable_mmu will be visible.

I see that we reach reenable_mmu through a 'rfi' with MSR_KERNEL, which 
seems safe to me. So, I figured it can be probed without issues?

- Naveen