linux-kernel - Re: [PATCH] mips/vdso: Support mremap() for vDSO

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] [day] [month] [year] [list]

Message-ID: <20200115000815.narpydp4unkkuboj@pburton-laptop>
Date:   Tue, 14 Jan 2020 16:08:15 -0800
From:   Paul Burton <paulburton@...nel.org>
To:     Guoyun Sun <sunguoyun@...ngson.cn>
Cc:     Ralf Baechle <ralf@...ux-mips.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Vincenzo Frascino <vincenzo.frascino@....com>,
        Allison Randal <allison@...utok.net>,
        linux-mips@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH]  mips/vdso: Support mremap() for vDSO

Hi Guoyun,

On Tue, Jan 14, 2020 at 01:13:28PM +0800, Guoyun Sun wrote:
> diff --git a/arch/mips/kernel/vdso.c b/arch/mips/kernel/vdso.c
> index bc35f84..eac3982 100644
> --- a/arch/mips/kernel/vdso.c
> +++ b/arch/mips/kernel/vdso.c
> @@ -38,6 +38,28 @@ static struct vm_special_mapping vdso_vvar_mapping = {
>  	.pages = no_pages,
>  };
>  
> +int vdso_mremap(const struct vm_special_mapping *sm,
> +		       struct vm_area_struct *new_vma)
> +{
> +	unsigned long new_size = new_vma->vm_end - new_vma->vm_start;
> +
> +#ifdef CONFIG_MIPS32_O32
> +	if (vdso_image_o32.size != new_size)
> +		return -EINVAL;
> +#endif
> +
> +#ifdef CONFIG_MIPS32_N32
> +	if (vdso_image_n32.size != new_size)
> +		return -EINVAL;
> +#endif
> +
> +	if (vdso_image.size != new_size)
> +		return -EINVAL;
> +
> +	current->mm->context.vdso = (void __user *)(new_vma->vm_start);
> +	return 0;
> +}
> +

Thanks for the patch; this bit doesn't seem right though. It requires
that all the VDSOs be the same size, which seems questionable. I think
we probably need separate mremap callbacks for each VDSO, so that we can
check the appropriate VDSO size.

Thanks,
    Paul