[PATCH v4 5/7] arm64: add KASAN support

[Andrey Ryabinin]

On 07/27/2015 06:59 PM, Catalin Marinas wrote:
> On Fri, Jul 24, 2015 at 07:41:57PM +0300, Andrey Ryabinin wrote:
>> diff --git a/arch/arm64/Makefile b/arch/arm64/Makefile
>> index 4d2a925..2cacf55 100644
>> --- a/arch/arm64/Makefile
>> +++ b/arch/arm64/Makefile
>> @@ -40,6 +40,12 @@ else
>>  TEXT_OFFSET := 0x00080000
>>  endif
>>  
>> +# KASAN_SHADOW_OFFSET = VA_START + (1 << (VA_BITS - 3)) - (1 << 61)
>> +KASAN_SHADOW_OFFSET := $(shell printf "0x%x\n" $$(( \
>> +			(-1 << $(CONFIG_ARM64_VA_BITS)) \
>> +			+ (1 << ($(CONFIG_ARM64_VA_BITS) - 3)) \
>> +			- (1 << (64 - 3)) )) )
> 
> Does this work with any POSIX shell? Do we always have a 64-bit type?
> As I wasn't sure about this, I suggested awk (or perl).
>

Ok, It will be safer to use 32-bit arithmetic.
I've checked this on 32-bit bash, however this doesn't guarantee that it works with
any other version of bash or another shell.

>> +static void __init clear_pgds(unsigned long start,
>> +			unsigned long end)
>> +{
>> +	/*
>> +	 * Remove references to kasan page tables from
>> +	 * swapper_pg_dir. pgd_clear() can't be used
>> +	 * here because it's nop on 2,3-level pagetable setups
>> +	 */
>> +	for (; start && start < end; start += PGDIR_SIZE)
>> +		set_pgd(pgd_offset_k(start), __pgd(0));
>> +}
> 
> I don't think we need the "start" check, just "start < end". Do you
> expect a start == 0 (or overflow)?

Right, we don't need this.