linux-kernel - Re: [PATCH] mm/hugetlb: Get rid of NODEMASK

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: <20190403075236.GB31828@MiWiFi-R3L-srv>
Date:   Wed, 3 Apr 2019 15:52:36 +0800
From:   Baoquan He <bhe@...hat.com>
To:     Oscar Salvador <osalvador@...e.de>
Cc:     akpm@...ux-foundation.org, mike.kravetz@...cle.com,
        n-horiguchi@...jp.nec.com, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] mm/hugetlb: Get rid of NODEMASK_ALLOC

On 04/02/19 at 03:34pm, Oscar Salvador wrote:
> NODEMASK_ALLOC is used to allocate a nodemask bitmap, ant it does it by
                                                         ~ and
> first determining whether it should be allocated in the stack or dinamically
                                                              dynamically^^
> depending on NODES_SHIFT.
> Right now, it goes the dynamic path whenever the nodemask_t is above 32
> bytes.
> 
> Although we could bump it to a reasonable value, the largest a nodemask_t
> can get is 128 bytes, so since __nr_hugepages_store_common is called from
> a rather shore stack we can just get rid of the NODEMASK_ALLOC call here.
> 
> This reduces some code churn and complexity.
> 
> Signed-off-by: Oscar Salvador <osalvador@...e.de>
> ---
>  mm/hugetlb.c | 36 +++++++++++-------------------------
>  1 file changed, 11 insertions(+), 25 deletions(-)
> 
> diff --git a/mm/hugetlb.c b/mm/hugetlb.c
> index f79ae4e42159..9cb2f91af897 100644
> --- a/mm/hugetlb.c
> +++ b/mm/hugetlb.c
> @@ -2447,44 +2447,30 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
>  					   unsigned long count, size_t len)
>  {
>  	int err;
> -	NODEMASK_ALLOC(nodemask_t, nodes_allowed, GFP_KERNEL | __GFP_NORETRY);
> +	nodemask_t nodes_allowed, *n_mask;
>  
> -	if (hstate_is_gigantic(h) && !gigantic_page_runtime_supported()) {
> -		err = -EINVAL;
> -		goto out;
> -	}
> +	if (hstate_is_gigantic(h) && !gigantic_page_runtime_supported())
> +		return -EINVAL;
>  
>  	if (nid == NUMA_NO_NODE) {
>  		/*
>  		 * global hstate attribute
>  		 */
>  		if (!(obey_mempolicy &&
> -				init_nodemask_of_mempolicy(nodes_allowed))) {
> -			NODEMASK_FREE(nodes_allowed);
> -			nodes_allowed = &node_states[N_MEMORY];
> -		}
> -	} else if (nodes_allowed) {
> +				init_nodemask_of_mempolicy(&nodes_allowed)))
> +			n_mask = &node_states[N_MEMORY];
> +		else
> +			n_mask = &nodes_allowed;
> +	} else {
>  		/*
>  		 * Node specific request.  count adjustment happens in
>  		 * set_max_huge_pages() after acquiring hugetlb_lock.
>  		 */
> -		init_nodemask_of_node(nodes_allowed, nid);
> -	} else {
> -		/*
> -		 * Node specific request, but we could not allocate the few
> -		 * words required for a node mask.  We are unlikely to hit
> -		 * this condition.  Since we can not pass down the appropriate
> -		 * node mask, just return ENOMEM.
> -		 */
> -		err = -ENOMEM;
> -		goto out;
> +		init_nodemask_of_node(&nodes_allowed, nid);
> +		n_mask = &nodes_allowed;
>  	}
>  
> -	err = set_max_huge_pages(h, count, nid, nodes_allowed);
> -
> -out:
> -	if (nodes_allowed != &node_states[N_MEMORY])
> -		NODEMASK_FREE(nodes_allowed);
> +	err = set_max_huge_pages(h, count, nid, n_mask);
>  
>  	return err ? err : len;
>  }
> -- 
> 2.13.7
>