linux-kernel - Re: [PATCH v2 02/14] arm64, hibernate: create_safe_exec

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 19 Aug 2019 16:50:14 +0100
From:   Mark Rutland <mark.rutland@....com>
To:     Pavel Tatashin <pasha.tatashin@...een.com>
Cc:     jmorris@...ei.org, sashal@...nel.org, ebiederm@...ssion.com,
        kexec@...ts.infradead.org, linux-kernel@...r.kernel.org,
        corbet@....net, catalin.marinas@....com, will@...nel.org,
        linux-arm-kernel@...ts.infradead.org, marc.zyngier@....com,
        james.morse@....com, vladimir.murzin@....com,
        matthias.bgg@...il.com, bhsharma@...hat.com, linux-mm@...ck.org
Subject: Re: [PATCH v2 02/14] arm64, hibernate: create_safe_exec_page cleanup

On Fri, Aug 16, 2019 at 10:46:17PM -0400, Pavel Tatashin wrote:
> create_safe_exec_page() is going to be split into two parts in preparation
> of moving page table handling code out of hibernate.c
> 
> Remove allocator parameter, and rename dst to page. Also, remove the
> goto's, as we can return directly without cleanups.

It would be nice if you could do the goto/allocator/rename changes as
separate patches, since it's vastly easier to verify each change in
isolation that way.

What's the point of the rename? It's inconsistent with the phys_dst_addr
that you leave as-is, so I'm not sure that's worthwhile.

> 
> Signed-off-by: Pavel Tatashin <pasha.tatashin@...een.com>
> ---
>  arch/arm64/kernel/hibernate.c | 60 +++++++++++++++--------------------
>  1 file changed, 26 insertions(+), 34 deletions(-)
> 
> diff --git a/arch/arm64/kernel/hibernate.c b/arch/arm64/kernel/hibernate.c
> index 9341fcc6e809..96b6f8da7e49 100644
> --- a/arch/arm64/kernel/hibernate.c
> +++ b/arch/arm64/kernel/hibernate.c
> @@ -196,57 +196,51 @@ EXPORT_SYMBOL(arch_hibernation_header_restore);
>   */
>  static int create_safe_exec_page(void *src_start, size_t length,
>  				 unsigned long dst_addr,
> -				 phys_addr_t *phys_dst_addr,
> -				 void *(*allocator)(gfp_t mask),
> -				 gfp_t mask)
> +				 phys_addr_t *phys_dst_addr)
>  {
> -	int rc = 0;
> +	void *page = (void *)get_safe_page(GFP_ATOMIC);
> +	pgd_t *trans_table;

The addition of this trans_table variable wasn't mentioned in the commit
message...

> +	trans_table = (void *)get_safe_page(GFP_ATOMIC);
> +	if (!trans_table)
> +		return -ENOMEM;
>  
> -	pgdp = pgd_offset_raw(allocator(mask), dst_addr);
> +	pgdp = pgd_offset_raw(trans_table, dst_addr);

> -	write_sysreg(phys_to_ttbr(virt_to_phys(pgdp)), ttbr0_el1);
> +	write_sysreg(phys_to_ttbr(virt_to_phys(trans_table)), ttbr0_el1);


... and I guess you're trying to ensure that we program the TTBR with
the correct base address, without the offset of whatever pgd entry we
happen to have plumbed in?

I think that's a fix, and should come before any other cleanup or
rework.

If you can respin that specific change with s/trans_table/pgdir/, that
would make sense to me.

Thanks,
Mark.