Re: [PATCH] Make system_reset_pSeries relocatable

[Benjamin Herrenschmidt <benh@kernel.crashing.org>]

On Wed, 2016-07-27 at 17:32 +1000, Balbir Singh wrote:
> From: Balbir Singh <bsingharora@gmail.com>
> 
> Currently the power management bits are broken w.r.t. relocation.
> There are direct branches from system_reset_pSeries to
> power7_wakeup_*.

Side track: we should really get rid of the _pSeries suffix for these
things :-)

>  The correct way to do it is to do what
> the slb miss handler does, which is jump to a small stub within
> the first 64k of the relocated address and then jump to the
> actual location.
> 
> The code has been lightly tested (not the kvm bits), I would highly
> appreciate a review of the code. I suspect there might be easy
> to find bugs :)
> 
> Cc: benh@kernel.crashing.org
> Cc: mpe@ellerman.id.au
> Cc: paulus@samba.org
> Cc: npiggin@gmail.com
> Cc: svaidy@linux.vnet.ibm.com
> 
> Signed-off-by: Balbir Singh <bsingharora@gmail.com>
> ---
>  arch/powerpc/kernel/exceptions-64s.S | 82 ++++++++++++++++++++++--
> ------------
>  1 file changed, 51 insertions(+), 31 deletions(-)
> 
> diff --git a/arch/powerpc/kernel/exceptions-64s.S
> b/arch/powerpc/kernel/exceptions-64s.S
> index 8bcc1b4..64f9650 100644
> --- a/arch/powerpc/kernel/exceptions-64s.S
> +++ b/arch/powerpc/kernel/exceptions-64s.S
> @@ -118,39 +118,21 @@ BEGIN_FTR_SECTION
>  	cmpwi	cr4,r5,1
>  	mtspr	SPRN_HSPRG0,r13
>  
> -	lbz	r0,PACA_THREAD_IDLE_STATE(r13)
> -	cmpwi   cr2,r0,PNV_THREAD_NAP
> -	bgt     cr2,8f				/* Either
> sleep or Winkle */
> -
> -	/* Waking up from nap should not cause hypervisor state loss
> */
> -	bgt	cr3,.
> -
> -	/* Waking up from nap */
> -	li	r0,PNV_THREAD_RUNNING
> -	stb	r0,PACA_THREAD_IDLE_STATE(r13)	/* Clear
> thread state */
> -
> -#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
> -	li	r0,KVM_HWTHREAD_IN_KERNEL
> -	stb	r0,HSTATE_HWTHREAD_STATE(r13)
> -	/* Order setting hwthread_state vs. testing hwthread_req */
> -	sync
> -	lbz	r0,HSTATE_HWTHREAD_REQ(r13)
> -	cmpwi	r0,0
> -	beq	1f
> -	b	kvm_start_guest
> -1:
> +#ifndef CONFIG_RELOCATABLE
> +	b	power7_wakeup_common
> +#else
> +	/*
> +	 * We can't just use a direct branch to power7_wakeup_common
> +	 * because the distance from here to there depends on where
> +	 * the kernel ends up being put.
> +	 */
> +	mfctr	r11
> +	ld	r10, PACAKBASE(r13)
> +	LOAD_HANDLER(r10, power7_wakeup_common)
> +	mtctr	r10
> +	bctr
>  #endif
>  
> -	/* Return SRR1 from power7_nap() */
> -	mfspr	r3,SPRN_SRR1
> -	beq	cr3,2f
> -	b	power7_wakeup_noloss
> -2:	b	power7_wakeup_loss
> -
> -	/* Fast Sleep wakeup on PowerNV */
> -8:	GET_PACA(r13)
> -	b 	power7_wakeup_tb_loss
> -
>  9:
>  END_FTR_SECTION_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206)
>  #endif /* CONFIG_PPC_P7_NAP */
> @@ -1448,6 +1430,44 @@ power4_fixup_nap:
>  	blr
>  #endif
>  
> +	.align 7
> +_GLOBAL(power7_wakeup_common)
> +#ifdef CONFIG_RELOCATABLE
> +	mtctr	r11
> +#endif
> +	lbz	r0,PACA_THREAD_IDLE_STATE(r13)
> +	cmpwi   cr2,r0,PNV_THREAD_NAP
> +	bgt     cr2,8f				/* Either
> sleep or Winkle */
> +
> +	/* Waking up from nap should not cause hypervisor state loss
> */
> +	bgt	cr3,.
> +
> +	/* Waking up from nap */
> +	li	r0,PNV_THREAD_RUNNING
> +	stb	r0,PACA_THREAD_IDLE_STATE(r13)	/* Clear
> thread state */
> +
> +#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
> +	li	r0,KVM_HWTHREAD_IN_KERNEL
> +	stb	r0,HSTATE_HWTHREAD_STATE(r13)
> +	/* Order setting hwthread_state vs. testing hwthread_req */
> +	sync
> +	lbz	r0,HSTATE_HWTHREAD_REQ(r13)
> +	cmpwi	r0,0
> +	beq	1f
> +	b	kvm_start_guest
> +1:
> +#endif
> +
> +	/* Return SRR1 from power7_nap() */
> +	mfspr	r3,SPRN_SRR1
> +	beq	cr3,2f
> +	b	power7_wakeup_noloss
> +2:	b	power7_wakeup_loss
> +
> +	/* Fast Sleep wakeup on PowerNV */
> +8:	GET_PACA(r13)
> +	b 	power7_wakeup_tb_loss
> +
>  /*
>   * Hash table stuff
>   */