linux-kernel - Re: [RESEND] [PATCH 1/2] x86/mm: Add an option to change the padding used for the physical memory mapping.

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] [day] [month] [year] [list]

Date:   Tue, 21 Aug 2018 12:03:58 -0400
From:   Masayoshi Mizuma <msys.mizuma@...il.com>
To:     bhe@...hat.com
Cc:     tglx@...utronix.de, mingo@...hat.com, hpa@...or.com,
        x86@...nel.org, m.mizuma@...fujitsu.com,
        linux-kernel@...r.kernel.org
Subject: Re: [RESEND] [PATCH 1/2] x86/mm: Add an option to change the padding
 used for the physical memory mapping.

Hi Baoquan,

On 08/21/2018 10:51 AM, Baoquan He wrote:
> Hi Masa,
> 
> On 08/21/18 at 09:24am, Masayoshi Mizuma wrote:
>> From: Masayoshi Mizuma <m.mizuma@...fujitsu.com>
>>
>> There are some exceptional cases that the padding used for the physical
>> memory mapping section is not enough.
>>
>> For example of the cases:
>> - As Baoquan reported in the following, SGI UV system.
>>   https://lkml.org/lkml/2017/9/7/87
>> - Each node of physical memory layout has huge space for hotplug.
>>   For exapmle of the layout:
>>     SRAT: Node 6 PXM 4 [mem 0x100000000000-0x13ffffffffff] hotplug
>>     SRAT: Node 7 PXM 5 [mem 0x140000000000-0x17ffffffffff] hotplug
>>     SRAT: Node 2 PXM 6 [mem 0x180000000000-0x1bffffffffff] hotplug
>>     SRAT: Node 3 PXM 7 [mem 0x1c0000000000-0x1fffffffffff] hotplug
>>
>> We can increase the padding by CONFIG_RANDOMIZE_MEMORY_PHYSICAL_PADDING,
>> however, the available entropy is decreased if the padding is increased.
>> So the config change is not very good for the other most of systems.
>> And, the needed padding size depends on the system environment, for
>> example physical memory layout, so the kernel option is better than
>> changing the config.
>>
>> Signed-off-by: Masayoshi Mizuma <m.mizuma@...fujitsu.com>
>> ---
>>  arch/x86/mm/kaslr.c | 20 +++++++++++++++++++-
>>  1 file changed, 19 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/x86/mm/kaslr.c b/arch/x86/mm/kaslr.c
>> index 61db77b..8c7b4f6 100644
>> --- a/arch/x86/mm/kaslr.c
>> +++ b/arch/x86/mm/kaslr.c
>> @@ -33,6 +33,9 @@
>>  
>>  #define TB_SHIFT 40
> 
> I think this fix makes sense. Since we usually compiled in hotplug code
> by default, while may not really hot plug a memory board on most of
> systems. However on those systems which really need do hot plug, and
> might plug memory board onto physical position above 10TB, this will
> cause issues.
> 
> One tiny concern is that we have below definition for 5-level, means we
> can only have 4PB of system RAM for now. Not sure if one day it will be
> enlarged to 32PB as Documentation/x86/x86_64/mm.txt tell.
> 
> # define MAX_PHYSMEM_BITS       (pgtable_l5_enabled() ? 52 : 46)

Thank you for pointing it out.
I will change the max padding as follows and post v2 patch.

static int __init rand_mem_physical_padding_setup(char *str)
{
       int max_padding = (1 <<(MAX_PHYSMEM_BITS - TB_SHIFT)) - 1;

> Otherwise this patchset looks good to me.

Thank you for your review!

- Masa

> 
> Thanks
> Baoquan
> 
>>  
>> +#define MAX_PADDING_L4 63
>> +#define MAX_PADDING_L5 32767
>> +
>>  /*
>>   * The end address could depend on more configuration options to make the
>>   * highest amount of space for randomization available, but that's too hard
>> @@ -40,6 +43,7 @@
>>   */
>>  static const unsigned long vaddr_end = CPU_ENTRY_AREA_BASE;
>>  
>> +static int __initdata rand_mem_physical_padding = CONFIG_RANDOMIZE_MEMORY_PHYSICAL_PADDING;
>>  /*
>>   * Memory regions randomized by KASLR (except modules that use a separate logic
>>   * earlier during boot). The list is ordered based on virtual addresses. This
>> @@ -69,6 +73,20 @@ static inline bool kaslr_memory_enabled(void)
>>  	return kaslr_enabled() && !IS_ENABLED(CONFIG_KASAN);
>>  }
>>  
>> +static int __init rand_mem_physical_padding_setup(char *str)
>> +{
>> +	int max_padding = pgtable_l5_enabled() ? MAX_PADDING_L5 : MAX_PADDING_L4;
>> +
>> +	get_option(&str, &rand_mem_physical_padding);
>> +	if (rand_mem_physical_padding < 0)
>> +		rand_mem_physical_padding = 0;
>> +	else if (rand_mem_physical_padding > max_padding)
>> +		rand_mem_physical_padding = max_padding;
>> +
>> +	return 0;
>> +}
>> +early_param("rand_mem_physical_padding", rand_mem_physical_padding_setup);
>> +
>>  /* Initialize base and padding for each memory region randomized with KASLR */
>>  void __init kernel_randomize_memory(void)
>>  {
>> @@ -102,7 +120,7 @@ void __init kernel_randomize_memory(void)
>>  	 */
>>  	BUG_ON(kaslr_regions[0].base != &page_offset_base);
>>  	memory_tb = DIV_ROUND_UP(max_pfn << PAGE_SHIFT, 1UL << TB_SHIFT) +
>> -		CONFIG_RANDOMIZE_MEMORY_PHYSICAL_PADDING;
>> +		rand_mem_physical_padding;
>>  
>>  	/* Adapt phyiscal memory region size based on available memory */
>>  	if (memory_tb < kaslr_regions[0].size_tb)
>> -- 
>> 2.18.0
>>