linux-kernel - Re: [PATCH 5/5] KVM: Add hugepage support for dedicated guest memory

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]

Message-ID: <Z9P74jUTsWo3CHGy@yzhao56-desk.sh.intel.com>
Date: Fri, 14 Mar 2025 17:50:26 +0800
From: Yan Zhao <yan.y.zhao@...el.com>
To: Michael Roth <michael.roth@....com>
CC: <kvm@...r.kernel.org>, <linux-coco@...ts.linux.dev>, <linux-mm@...ck.org>,
	<linux-kernel@...r.kernel.org>, <jroedel@...e.de>, <thomas.lendacky@....com>,
	<pbonzini@...hat.com>, <seanjc@...gle.com>, <vbabka@...e.cz>,
	<amit.shah@....com>, <pratikrajesh.sampat@....com>, <ashish.kalra@....com>,
	<liam.merwick@...cle.com>, <david@...hat.com>, <vannapurve@...gle.com>,
	<ackerleytng@...gle.com>, <quic_eberman@...cinc.com>
Subject: Re: [PATCH 5/5] KVM: Add hugepage support for dedicated guest memory

> +static struct folio *kvm_gmem_get_huge_folio(struct inode *inode, pgoff_t index,
> +					     unsigned int order)
> +{
> +	pgoff_t npages = 1UL << order;
> +	pgoff_t huge_index = round_down(index, npages);
> +	struct address_space *mapping  = inode->i_mapping;
> +	gfp_t gfp = mapping_gfp_mask(mapping) | __GFP_NOWARN;
> +	loff_t size = i_size_read(inode);
> +	struct folio *folio;
> +
> +	/* Make sure hugepages would be fully-contained by inode */
> +	if ((huge_index + npages) * PAGE_SIZE > size)
> +		return NULL;
> +
> +	if (filemap_range_has_page(mapping, (loff_t)huge_index << PAGE_SHIFT,
> +				   (loff_t)(huge_index + npages - 1) << PAGE_SHIFT))
> +		return NULL;
> +
> +	folio = filemap_alloc_folio(gfp, order);
> +	if (!folio)
> +		return NULL;
Instead of returning NULL here, what about invoking __filemap_get_folio()
directly as below?

> +	if (filemap_add_folio(mapping, folio, huge_index, gfp)) {
> +		folio_put(folio);
> +		return NULL;
> +	}
> +
> +	return folio;
> +}
> +
>  /*
>   * Returns a locked folio on success.  The caller is responsible for
>   * setting the up-to-date flag before the memory is mapped into the guest.
> @@ -284,8 +314,15 @@ static int kvm_gmem_prepare_folio(struct kvm *kvm, struct file *file,
>   */
>  static struct folio *kvm_gmem_get_folio(struct inode *inode, pgoff_t index)
>  {
> -	/* TODO: Support huge pages. */
> -	return filemap_grab_folio(inode->i_mapping, index);
> +	struct folio *folio = NULL;
> +
> +	if (gmem_2m_enabled)
> +		folio = kvm_gmem_get_huge_folio(inode, index, PMD_ORDER);
> +
> +	if (!folio)
Also need to check IS_ERR(folio).

> +		folio = filemap_grab_folio(inode->i_mapping, index);
> +
> +	return folio;
>  }
Could we introduce a common helper to calculate max_order by checking for
gfn/index alignment and ensuring memory attributes in range are uniform?

Then we can pass in the max_order to kvm_gmem_get_folio() and only allocate huge
folio when it's necessary.

static struct folio *kvm_gmem_get_folio(struct inode *inode, pgoff_t index, int max_order)
{                                                                                
        struct folio *folio = NULL;                                              
                                                                                 
        if (max_order >= PMD_ORDER) {                                            
                fgf_t fgp_flags = FGP_LOCK | FGP_ACCESSED | FGP_CREAT;           
                                                                                 
                fgp_flags |= fgf_set_order(1U << (PAGE_SHIFT + PMD_ORDER));      
                folio = __filemap_get_folio(inode->i_mapping, index, fgp_flags,  
                        mapping_gfp_mask(inode->i_mapping));                     
        }                                                                        
                                                                                 
        if (!folio || IS_ERR(folio))                                             
                folio = filemap_grab_folio(inode->i_mapping, index);             
                                                                                 
        return folio;                                                            
}