linux-kernel - Re: [PATCH] arm64: fix build error when use rodata

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:   Wed, 5 Jan 2022 14:51:05 +0530
From:   Anshuman Khandual <anshuman.khandual@....com>
To:     AliOS system security <alios_sys_security@...ux.alibaba.com>,
        catalin.marinas@....com, will@...nel.org
Cc:     linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] arm64: fix build error when use rodata_enabled

Hello,

On 1/5/22 8:37 AM, AliOS system security wrote:
> rodata_enabled should be used when CONFIG_STRICT_KERNEL_RWX
> or CONFIG_STRICT_MODULE_RWX is selected

Both these configs get selected invariably with CONFIG_ARM64 in the
platform config file (arch/arm64/Kconfig). I guess there can not be
any such situation, where both configs will be missing/not selected
given ARCH_OPTIONAL_KERNEL_RWX[or _DEFAULT] is not enabled on arm64.

config ARM64
        def_bool y
        select ACPI_CCA_REQUIRED if ACPI
	.....
        select ARCH_HAS_STRICT_KERNEL_RWX
        select ARCH_HAS_STRICT_MODULE_RWX
	.....

Hence for all practical purpose, rodata_enabled could be considered
always available. I am sure there other similar situations as well,
where code elements are not wrapped around if the config option is
always present.

> 
> Signed-off-by: AliOS system security <alios_sys_security@...ux.alibaba.com>

Also please refer Documentation/process/submitting-patches.rst for
the rules regarding names, that can be used for a commit sign off.

------------------------------------------------------------------------
then you just add a line saying::

        Signed-off-by: Random J Developer <random@...eloper.example.org>

using your real name (sorry, no pseudonyms or anonymous contributions.)
------------------------------------------------------------------------

- Anshuman

> ---
>  arch/arm64/mm/mmu.c | 14 ++++++++++++--
>  1 file changed, 12 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c
> index acfae9b..47f8754 100644
> --- a/arch/arm64/mm/mmu.c
> +++ b/arch/arm64/mm/mmu.c
> @@ -596,6 +596,7 @@ static void __init map_kernel_segment(pgd_t *pgdp, void *va_start, void *va_end,
>  	vm_area_add_early(vma);
>  }
>  
> +#if defined(CONFIG_STRICT_KERNEL_RWX) || defined(CONFIG_STRICT_MODULE_RWX)
>  static int __init parse_rodata(char *arg)
>  {
>  	int ret = strtobool(arg, &rodata_enabled);
> @@ -613,11 +614,16 @@ static int __init parse_rodata(char *arg)
>  	return 0;
>  }
>  early_param("rodata", parse_rodata);
> +#endif
>  
>  #ifdef CONFIG_UNMAP_KERNEL_AT_EL0
>  static int __init map_entry_trampoline(void)
>  {
> -	pgprot_t prot = rodata_enabled ? PAGE_KERNEL_ROX : PAGE_KERNEL_EXEC;
> +	pgprot_t prot = PAGE_KERNEL_EXEC;
> +#if defined(CONFIG_STRICT_KERNEL_RWX) || defined(CONFIG_STRICT_MODULE_RWX)
> +	if (rodata_enabled)
> +		prot = PAGE_KERNEL_ROX;
> +#endif
>  	phys_addr_t pa_start = __pa_symbol(__entry_tramp_text_start);
>  
>  	/* The trampoline is always mapped and can therefore be global */
> @@ -672,7 +678,11 @@ static void __init map_kernel(pgd_t *pgdp)
>  	 * mapping to install SW breakpoints. Allow this (only) when
>  	 * explicitly requested with rodata=off.
>  	 */
> -	pgprot_t text_prot = rodata_enabled ? PAGE_KERNEL_ROX : PAGE_KERNEL_EXEC;
> +	pgprot_t text_prot = PAGE_KERNEL_EXEC;
> +#if defined(CONFIG_STRICT_KERNEL_RWX) || defined(CONFIG_STRICT_MODULE_RWX)
> +	if (rodata_enabled)
> +		text_prot = PAGE_KERNEL_ROX;
> +#endif
>  
>  	/*
>  	 * If we have a CPU that supports BTI and a kernel built for
>