Re: [PATCH v1 1/3] arm64: mm: Fix rodata=full block mapping support for realm guests

[Ryan Roberts]

On 23/03/2026 16:52, Kevin Brodsky wrote:
> On 23/03/2026 14:03, Ryan Roberts wrote:
>> [...]
>>
>> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
>> index 96711b8578fd0..b9b248d24fd10 100644
>> --- a/arch/arm64/mm/init.c
>> +++ b/arch/arm64/mm/init.c
>> @@ -350,7 +350,6 @@ void __init arch_mm_preinit(void)
>>  	}
>>  
>>  	swiotlb_init(swiotlb, flags);
>> -	swiotlb_update_mem_attributes();
>>  
>>  	/*
>>  	 * Check boundaries twice: Some fundamental inconsistencies can be
>> @@ -377,6 +376,14 @@ void __init arch_mm_preinit(void)
>>  	}
>>  }
>>  
>> +bool page_alloc_available __ro_after_init;
>> +
>> +void __init mem_init(void)
>> +{
>> +	page_alloc_available = true;
>> +	swiotlb_update_mem_attributes();
> 
> The move seems reasonable, x86 calls this function even later (from
> arch_cpu_finalize_init()).
> 
>> +}
>> +
>>  void free_initmem(void)
>>  {
>>  	void *lm_init_begin = lm_alias(__init_begin);
>> diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c
>> index a6a00accf4f93..5b6a8d53e64b7 100644
>> --- a/arch/arm64/mm/mmu.c
>> +++ b/arch/arm64/mm/mmu.c
>> @@ -773,14 +773,33 @@ int split_kernel_leaf_mapping(unsigned long start, unsigned long end)
>>  {
>>  	int ret;
>>  
>> -	/*
>> -	 * !BBML2_NOABORT systems should not be trying to change permissions on
>> -	 * anything that is not pte-mapped in the first place. Just return early
>> -	 * and let the permission change code raise a warning if not already
>> -	 * pte-mapped.
>> -	 */
>> -	if (!system_supports_bbml2_noabort())
>> -		return 0;
>> +	if (!system_supports_bbml2_noabort()) {
>> +		/*
>> +		 * !BBML2_NOABORT systems should not be trying to change
>> +		 * permissions on anything that is not pte-mapped in the first
>> +		 * place. Just return early and let the permission change code
>> +		 * raise a warning if not already pte-mapped.
>> +		 */
>> +		if (system_capabilities_finalized() ||
>> +		    !cpu_supports_bbml2_noabort())
>> +			return 0;
>> +
>> +		/*
>> +		 * Boot-time: split_kernel_leaf_mapping_locked() allocates from
>> +		 * page allocator. Can't split until it's available.
>> +		 */
>> +		extern bool page_alloc_available;
> 
> Could we at least have the declaration in say <asm/mmu.h>? x86 defines a
> similar global so we could eventually have a generic global (defined
> before mem_init() is called).

Yeah, fair enough. I was being lazy. I'll move it to the header for v2.

> 
> Looks good otherwise:
> 
> Reviewed-by: Kevin Brodsky <kevin.brodsky@xxxxxxx>
> 
>> +		if (WARN_ON(!page_alloc_available))
>> +			return -EBUSY;
>> +
>> +		/*
>> +		 * Boot-time: Started secondary cpus but don't know if they
>> +		 * support BBML2_NOABORT yet. Can't allow splitting in this
>> +		 * window in case they don't.
>> +		 */
>> +		if (WARN_ON(num_online_cpus() > 1))
>> +			return -EBUSY;
>> +	}
>>  
>>  	/*
>>  	 * If the region is within a pte-mapped area, there is no need to try to