linux-kernel - Re: [PATCH v2 1/2] mm/mremap: Clean up vma_to

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 for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <8ea11588-02ac-419f-b531-90e8cc3b330f@lucifer.local>
Date: Mon, 21 Oct 2024 11:03:30 +0100
From: Lorenzo Stoakes <lorenzo.stoakes@...cle.com>
To: "Liam R. Howlett" <Liam.Howlett@...cle.com>
Cc: Andrew Morton <akpm@...ux-foundation.org>, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org, Jann Horn <jannh@...gle.com>,
        David Hildenbrand <david@...hat.com>,
        Qi Zheng <zhengqi.arch@...edance.com>,
        Kefeng Wang <wangkefeng.wang@...wei.com>,
        Jeff Xu <jeffxu@...omium.org>, Pedro Falcato <pedro.falcato@...il.com>
Subject: Re: [PATCH v2 1/2] mm/mremap: Clean up vma_to_resize()

On Fri, Oct 18, 2024 at 01:41:13PM -0400, Liam R. Howlett wrote:
> From: "Liam R. Howlett" <Liam.Howlett@...cle.com>
>
> vma_to_resize() is used in two locations to find and validate the vma
> for the mremap location.  One of the two locations already has the vma,
> which is then re-found to validate the same vma.
>
> This code can be simplified by moving the vma_lookup() from
> vma_to_resize() to mremap_to() and changing the return type to an int
> error.
>
> Since the function now just validates the vma, the function is renamed
> to resize_is_valid() to better reflect what it is doing.
>
> This commit also adds documentation about the function.
>
> Signed-off-by: Liam R. Howlett <Liam.Howlett@...cle.com>

git-icdiff helped here a lot :)

This LGTM,

Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@...cle.com>

Generally a really nice clean up in any case, thanks for doing this!

> ---
>  mm/mremap.c | 53 +++++++++++++++++++++++++++++++----------------------
>  1 file changed, 31 insertions(+), 22 deletions(-)
>
> diff --git a/mm/mremap.c b/mm/mremap.c
> index 5917feafe8cc..e781ec4573ca 100644
> --- a/mm/mremap.c
> +++ b/mm/mremap.c
> @@ -826,17 +826,24 @@ static unsigned long move_vma(struct vm_area_struct *vma,
>  	return new_addr;
>  }
>
> -static struct vm_area_struct *vma_to_resize(unsigned long addr,
> +/*
> + * resize_is_valid() - Ensure the vma can be resized to the new length at the give
> + * address.
> + *
> + * @vma: The vma to resize
> + * @addr: The old address
> + * @old_len: The current size
> + * @new_len: The desired size
> + * @flags: The vma flags
> + *
> + * Return 0 on success, error otherwise.
> + */
> +static int resize_is_valid(struct vm_area_struct *vma, unsigned long addr,
>  	unsigned long old_len, unsigned long new_len, unsigned long flags)
>  {
>  	struct mm_struct *mm = current->mm;
> -	struct vm_area_struct *vma;
>  	unsigned long pgoff;
>
> -	vma = vma_lookup(mm, addr);
> -	if (!vma)
> -		return ERR_PTR(-EFAULT);
> -
>  	/*
>  	 * !old_len is a special case where an attempt is made to 'duplicate'
>  	 * a mapping.  This makes no sense for private mappings as it will
> @@ -847,37 +854,37 @@ static struct vm_area_struct *vma_to_resize(unsigned long addr,
>  	 */
>  	if (!old_len && !(vma->vm_flags & (VM_SHARED | VM_MAYSHARE))) {
>  		pr_warn_once("%s (%d): attempted to duplicate a private mapping with mremap.  This is not supported.\n", current->comm, current->pid);
> -		return ERR_PTR(-EINVAL);
> +		return -EINVAL;
>  	}
>
>  	if ((flags & MREMAP_DONTUNMAP) &&
>  			(vma->vm_flags & (VM_DONTEXPAND | VM_PFNMAP)))
> -		return ERR_PTR(-EINVAL);
> +		return -EINVAL;
>
>  	/* We can't remap across vm area boundaries */
>  	if (old_len > vma->vm_end - addr)
> -		return ERR_PTR(-EFAULT);
> +		return -EFAULT;
>
>  	if (new_len == old_len)
> -		return vma;
> +		return 0;
>
>  	/* Need to be careful about a growing mapping */
>  	pgoff = (addr - vma->vm_start) >> PAGE_SHIFT;
>  	pgoff += vma->vm_pgoff;
>  	if (pgoff + (new_len >> PAGE_SHIFT) < pgoff)
> -		return ERR_PTR(-EINVAL);
> +		return -EINVAL;
>
>  	if (vma->vm_flags & (VM_DONTEXPAND | VM_PFNMAP))
> -		return ERR_PTR(-EFAULT);
> +		return -EFAULT;
>
>  	if (!mlock_future_ok(mm, vma->vm_flags, new_len - old_len))
> -		return ERR_PTR(-EAGAIN);
> +		return -EAGAIN;
>
>  	if (!may_expand_vm(mm, vma->vm_flags,
>  				(new_len - old_len) >> PAGE_SHIFT))
> -		return ERR_PTR(-ENOMEM);
> +		return -ENOMEM;
>
> -	return vma;
> +	return 0;
>  }
>
>  static unsigned long mremap_to(unsigned long addr, unsigned long old_len,
> @@ -936,12 +943,16 @@ static unsigned long mremap_to(unsigned long addr, unsigned long old_len,
>  		old_len = new_len;
>  	}
>
> -	vma = vma_to_resize(addr, old_len, new_len, flags);
> -	if (IS_ERR(vma)) {
> -		ret = PTR_ERR(vma);
> +	vma = vma_lookup(mm, addr);
> +	if (!vma) {
> +		ret = -EFAULT;
>  		goto out;
>  	}
>
> +	ret = resize_is_valid(vma, addr, old_len, new_len, flags);
> +	if (ret)
> +		goto out;
> +
>  	/* MREMAP_DONTUNMAP expands by old_len since old_len == new_len */
>  	if (flags & MREMAP_DONTUNMAP &&
>  		!may_expand_vm(mm, vma->vm_flags, old_len >> PAGE_SHIFT)) {
> @@ -1114,11 +1125,9 @@ SYSCALL_DEFINE5(mremap, unsigned long, addr, unsigned long, old_len,
>  	/*
>  	 * Ok, we need to grow..
>  	 */
> -	vma = vma_to_resize(addr, old_len, new_len, flags);
> -	if (IS_ERR(vma)) {
> -		ret = PTR_ERR(vma);
> +	ret = resize_is_valid(vma, addr, old_len, new_len, flags);
> +	if (ret)
>  		goto out;
> -	}
>
>  	/* old_len exactly to the end of the area..
>  	 */
> --
> 2.43.0
>