Re: [RFC PATCH 05/10] powerpc/fsl_booke/32: introduce reloc_kernel_entry() helper

From: Christophe Leroy <christophe.leroy@c-s.fr>
To: Jason Yan <yanaijie@huawei.com>,
	mpe@ellerman.id.au, linuxppc-dev@lists.ozlabs.org,
	diana.craciun@nxp.com, benh@kernel.crashing.org,
	paulus@samba.org, npiggin@gmail.com, keescook@chromium.org,
	kernel-hardening@lists.openwall.com
Cc: linux-kernel@vger.kernel.org, wangkefeng.wang@huawei.com,
	yebin10@huawei.com, thunder.leizhen@huawei.com,
	jingxiangfeng@huawei.com, fanchengyang@huawei.com
Subject: Re: [RFC PATCH 05/10] powerpc/fsl_booke/32: introduce reloc_kernel_entry() helper
Date: Mon, 29 Jul 2019 13:08:08 +0200	[thread overview]
Message-ID: <e4ccd015-a9c4-b0a6-e3ca-d37a04e29ec6@c-s.fr> (raw)
In-Reply-To: <20190717080621.40424-6-yanaijie@huawei.com>

Le 17/07/2019 à 10:06, Jason Yan a écrit :
> Add a new helper reloc_kernel_entry() to jump back to the start of the
> new kernel. After we put the new kernel in a randomized place we can use
> this new helper to enter the kernel and begin to relocate again.
> 
> Signed-off-by: Jason Yan <yanaijie@huawei.com>
> Cc: Diana Craciun <diana.craciun@nxp.com>
> Cc: Michael Ellerman <mpe@ellerman.id.au>
> Cc: Christophe Leroy <christophe.leroy@c-s.fr>
> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
> Cc: Paul Mackerras <paulus@samba.org>
> Cc: Nicholas Piggin <npiggin@gmail.com>
> Cc: Kees Cook <keescook@chromium.org>
> ---
>   arch/powerpc/kernel/head_fsl_booke.S | 16 ++++++++++++++++
>   arch/powerpc/mm/mmu_decl.h           |  1 +
>   2 files changed, 17 insertions(+)
> 
> diff --git a/arch/powerpc/kernel/head_fsl_booke.S b/arch/powerpc/kernel/head_fsl_booke.S
> index a57d44638031..ce40f96dae20 100644
> --- a/arch/powerpc/kernel/head_fsl_booke.S
> +++ b/arch/powerpc/kernel/head_fsl_booke.S
> @@ -1144,6 +1144,22 @@ _GLOBAL(create_tlb_entry)
>   	sync
>   	blr
>   
> +/*
> + * Return to the start of the relocated kernel and run again
> + * r3 - virtual address of fdt
> + * r4 - entry of the kernel
> + */
> +_GLOBAL(reloc_kernel_entry)
> +	mfmsr	r7
> +	li	r8,(MSR_IS | MSR_DS)
> +	andc	r7,r7,r8

Instead of the li/andc, what about the following:

rlwinm r7, r7, 0, ~(MSR_IS | MSR_DS)

> +
> +	mtspr	SPRN_SRR0,r4
> +	mtspr	SPRN_SRR1,r7
> +	isync
> +	sync
> +	rfi

Are the isync/sync really necessary ? AFAIK, rfi is context synchronising.

> +
>   /*
>    * Create a tlb entry with the same effective and physical address as
>    * the tlb entry used by the current running code. But set the TS to 1.
> diff --git a/arch/powerpc/mm/mmu_decl.h b/arch/powerpc/mm/mmu_decl.h
> index d7737cf97cee..dae8e9177574 100644
> --- a/arch/powerpc/mm/mmu_decl.h
> +++ b/arch/powerpc/mm/mmu_decl.h
> @@ -143,6 +143,7 @@ extern void adjust_total_lowmem(void);
>   extern int switch_to_as1(void);
>   extern void restore_to_as0(int esel, int offset, void *dt_ptr, int bootcpu);
>   extern void create_tlb_entry(phys_addr_t phys, unsigned long virt, int entry);
> +extern void reloc_kernel_entry(void *fdt, int addr);

No new 'extern' please, see 
https://openpower.xyz/job/snowpatch/job/snowpatch-linux-checkpatch/8125//artifact/linux/checkpatch.log

>   #endif
>   extern void loadcam_entry(unsigned int index);
>   extern void loadcam_multi(int first_idx, int num, int tmp_idx);
> 

Christophe