linux-kernel - Re: [PATCH] mm/memory_hotplug: Drop 'reason' argument from check_pfn

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]

Date:   Wed, 25 May 2022 15:08:03 +0530
From:   Anshuman Khandual <anshuman.khandual@....com>
To:     Oscar Salvador <osalvador@...e.de>
Cc:     linux-mm@...ck.org, David Hildenbrand <david@...hat.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] mm/memory_hotplug: Drop 'reason' argument from
 check_pfn_span()



On 5/25/22 13:55, Oscar Salvador wrote:
> On Wed, May 25, 2022 at 09:09:09AM +0530, Anshuman Khandual wrote:
>> In check_pfn_span(), a 'reason' string is being used to recreate the caller
>> function name, while printing the warning message. It is really unnecessary
>> as the warning message could just be printed inside the caller depending on
>> the return code. Currentlyy there are just two callers for check_pfn_span()
>                    Currently

Ahh, will fix.

>> i.e  __add_pages() and __remove_pages(). Let's clean this up.
>>
>> Cc: David Hildenbrand <david@...hat.com>
>> Cc: Andrew Morton <akpm@...ux-foundation.org>
>> Cc: linux-mm@...ck.org
>> Cc: linux-kernel@...r.kernel.org
>> Signed-off-by: Anshuman Khandual <anshuman.khandual@....com>
> 
> One could argue if this is really a cleanup.
> I kind of agree that the "reason" thingy is a bit shaky, but instead of having a
> single place where we call WARN(), we now do have two.

check_pfn_span() is basically ensuring minimum alignment for both pfn and
nr_pages. Resulting error message when this alignment check does not hold
true, is caller specific than not.

> 
>> ---
>>  mm/memory_hotplug.c | 19 +++++++++----------
>>  1 file changed, 9 insertions(+), 10 deletions(-)
>>
>> diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
>> index 416b38ca8def..9b3d7295ef93 100644
>> --- a/mm/memory_hotplug.c
>> +++ b/mm/memory_hotplug.c
>> @@ -220,8 +220,7 @@ static void release_memory_resource(struct resource *res)
>>  	kfree(res);
>>  }
>>  
>> -static int check_pfn_span(unsigned long pfn, unsigned long nr_pages,
>> -		const char *reason)
>> +static int check_pfn_span(unsigned long pfn, unsigned long nr_pages)
>>  {
>>  	/*
>>  	 * Disallow all operations smaller than a sub-section and only
>> @@ -238,12 +237,8 @@ static int check_pfn_span(unsigned long pfn, unsigned long nr_pages,
>>  		min_align = PAGES_PER_SUBSECTION;
>>  	else
>>  		min_align = PAGES_PER_SECTION;
>> -	if (!IS_ALIGNED(pfn, min_align)
>> -			|| !IS_ALIGNED(nr_pages, min_align)) {
>> -		WARN(1, "Misaligned __%s_pages start: %#lx end: #%lx\n",
>> -				reason, pfn, pfn + nr_pages - 1);
>> +	if (!IS_ALIGNED(pfn, min_align) || !IS_ALIGNED(nr_pages, min_align))
>>  		return -EINVAL;
>> -	}
>>  	return 0;
>>  }
>>  
>> @@ -320,9 +315,11 @@ int __ref __add_pages(int nid, unsigned long pfn, unsigned long nr_pages,
>>  		altmap->alloc = 0;
>>  	}
>>  
>> -	err = check_pfn_span(pfn, nr_pages, "add");
>> -	if (err)
>> +	err = check_pfn_span(pfn, nr_pages);
>> +	if (err) {
>> +		WARN(1, "Misaligned %s start: %#lx end: #%lx\n", __func__, pfn, pfn + nr_pages - 1);
>>  		return err;
>> +	}
> 
> If you want to further clean this up, I would just do 
> 
>  if (check_pfn_span()) {
>          WARN(....)
>          return -EINVAL;
>  }
> 
> here as we do in __remove_pages(). check_pfn_span() can either return 0 or -EINVAL,
> so I think it is fine.

Sure, will change.