linux-kernel - Re: [PATCH v1 3/5] s390/uv: convert PG_arch

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: <c9bfaf90-d744-4371-8f6f-5739f3373072@redhat.com>
Date: Thu, 11 Apr 2024 12:38:17 +0200
From: David Hildenbrand <david@...hat.com>
To: Claudio Imbrenda <imbrenda@...ux.ibm.com>
Cc: linux-kernel@...r.kernel.org, linux-mm@...ck.org,
 linux-s390@...r.kernel.org, kvm@...r.kernel.org,
 Matthew Wilcox <willy@...radead.org>, Heiko Carstens <hca@...ux.ibm.com>,
 Vasily Gorbik <gor@...ux.ibm.com>, Alexander Gordeev
 <agordeev@...ux.ibm.com>, Christian Borntraeger <borntraeger@...ux.ibm.com>,
 Sven Schnelle <svens@...ux.ibm.com>, Janosch Frank <frankja@...ux.ibm.com>,
 Gerald Schaefer <gerald.schaefer@...ux.ibm.com>,
 Thomas Huth <thuth@...hat.com>
Subject: Re: [PATCH v1 3/5] s390/uv: convert PG_arch_1 users to only work on
 small folios

On 10.04.24 19:42, Claudio Imbrenda wrote:
> On Thu,  4 Apr 2024 18:36:40 +0200
> David Hildenbrand <david@...hat.com> wrote:
> 
>> Now that make_folio_secure() may only set PG_arch_1 for small folios,
>> let's convert relevant remaining UV code to only work on (small) folios
>> and simply reject large folios early. This way, we'll never end up
>> touching PG_arch_1 on tail pages of a large folio in UV code.
>>
>> The folio_get()/folio_put() for functions that are documented to already
>> hold a folio reference look weird and it should probably be removed.
>> Similarly, uv_destroy_owned_page() and uv_convert_owned_from_secure()
>> should really consume a folio reference instead. But these are cleanups for
>> another day.
>>
>> Signed-off-by: David Hildenbrand <david@...hat.com>
>> ---
>>   arch/s390/include/asm/page.h |  1 +
>>   arch/s390/kernel/uv.c        | 39 +++++++++++++++++++++---------------
>>   2 files changed, 24 insertions(+), 16 deletions(-)
>>
>> diff --git a/arch/s390/include/asm/page.h b/arch/s390/include/asm/page.h
>> index 54d015bcd8e3..b64384872c0f 100644
>> --- a/arch/s390/include/asm/page.h
>> +++ b/arch/s390/include/asm/page.h
>> @@ -214,6 +214,7 @@ static inline unsigned long __phys_addr(unsigned long x, bool is_31bit)
>>   #define pfn_to_phys(pfn)	((pfn) << PAGE_SHIFT)
>>   
>>   #define phys_to_page(phys)	pfn_to_page(phys_to_pfn(phys))
>> +#define phys_to_folio(phys)	page_folio(phys_to_page(phys))
>>   #define page_to_phys(page)	pfn_to_phys(page_to_pfn(page))
>>   #define folio_to_phys(page)	pfn_to_phys(folio_pfn(folio))
>>   
>> diff --git a/arch/s390/kernel/uv.c b/arch/s390/kernel/uv.c
>> index adcbd4b13035..9c0113b26735 100644
>> --- a/arch/s390/kernel/uv.c
>> +++ b/arch/s390/kernel/uv.c
>> @@ -134,14 +134,17 @@ static int uv_destroy_page(unsigned long paddr)
>>    */
>>   int uv_destroy_owned_page(unsigned long paddr)
>>   {
>> -	struct page *page = phys_to_page(paddr);
>> +	struct folio *folio = phys_to_folio(paddr);
>>   	int rc;
>>   
>> -	get_page(page);
>> +	if (unlikely(folio_test_large(folio)))
>> +		return 0;
> 
> please add a comment here to explain why it's ok to just return 0
> here...


Will use

"/* See gmap_make_secure(): large folios cannot be protected */"

> 
>> +
>> +	folio_get(folio);
>>   	rc = uv_destroy_page(paddr);
>>   	if (!rc)
>> -		clear_bit(PG_arch_1, &page->flags);
>> -	put_page(page);
>> +		clear_bit(PG_arch_1, &folio->flags);
>> +	folio_put(folio);
>>   	return rc;
>>   }
>>   
>> @@ -169,14 +172,17 @@ int uv_convert_from_secure(unsigned long paddr)
>>    */
>>   int uv_convert_owned_from_secure(unsigned long paddr)
>>   {
>> -	struct page *page = phys_to_page(paddr);
>> +	struct folio *folio = phys_to_folio(paddr);
>>   	int rc;
>>   
>> -	get_page(page);
>> +	if (unlikely(folio_test_large(folio)))
>> +		return 0;
> 
> ... and here

here as well and ...

> 
>> +
>> +	folio_get(folio);
>>   	rc = uv_convert_from_secure(paddr);
>>   	if (!rc)
>> -		clear_bit(PG_arch_1, &page->flags);
>> -	put_page(page);
>> +		clear_bit(PG_arch_1, &folio->flags);
>> +	folio_put(folio);
>>   	return rc;
>>   }
>>   
>> @@ -457,33 +463,34 @@ EXPORT_SYMBOL_GPL(gmap_destroy_page);
>>    */
>>   int arch_make_page_accessible(struct page *page)
>>   {
>> +	struct folio *folio = page_folio(page);
>>   	int rc = 0;
>>   
>> -	/* Hugepage cannot be protected, so nothing to do */
>> -	if (PageHuge(page))
>> +	/* Large folios cannot be protected, so nothing to do */

^ here as well.

>> +	if (unlikely(folio_test_large(folio)))
>>   		return 0;
>>   

-- 
Cheers,

David / dhildenb