Re: [PATCH v2 4/4] arm64: drop pfn_valid_within() and simplify pfn_valid()

[Anshuman Khandual]

On 4/21/21 5:54 PM, Mike Rapoport wrote:
> On Wed, Apr 21, 2021 at 04:36:46PM +0530, Anshuman Khandual wrote:
>>
>> On 4/21/21 12:21 PM, Mike Rapoport wrote:
>>> From: Mike Rapoport <rppt@xxxxxxxxxxxxx>
>>>
>>> The arm64's version of pfn_valid() differs from the generic because of two
>>> reasons:
>>>
>>> * Parts of the memory map are freed during boot. This makes it necessary to
>>>   verify that there is actual physical memory that corresponds to a pfn
>>>   which is done by querying memblock.
>>>
>>> * There are NOMAP memory regions. These regions are not mapped in the
>>>   linear map and until the previous commit the struct pages representing
>>>   these areas had default values.
>>>
>>> As the consequence of absence of the special treatment of NOMAP regions in
>>> the memory map it was necessary to use memblock_is_map_memory() in
>>> pfn_valid() and to have pfn_valid_within() aliased to pfn_valid() so that
>>> generic mm functionality would not treat a NOMAP page as a normal page.
>>>
>>> Since the NOMAP regions are now marked as PageReserved(), pfn walkers and
>>> the rest of core mm will treat them as unusable memory and thus
>>> pfn_valid_within() is no longer required at all and can be disabled by
>>> removing CONFIG_HOLES_IN_ZONE on arm64.
>>
>> This makes sense.
>>
>>>
>>> pfn_valid() can be slightly simplified by replacing
>>> memblock_is_map_memory() with memblock_is_memory().
>>>
>>> Signed-off-by: Mike Rapoport <rppt@xxxxxxxxxxxxx>
>>> ---
>>>  arch/arm64/Kconfig   | 3 ---
>>>  arch/arm64/mm/init.c | 4 ++--
>>>  2 files changed, 2 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
>>> index e4e1b6550115..58e439046d05 100644
>>> --- a/arch/arm64/Kconfig
>>> +++ b/arch/arm64/Kconfig
>>> @@ -1040,9 +1040,6 @@ config NEED_PER_CPU_EMBED_FIRST_CHUNK
>>>  	def_bool y
>>>  	depends on NUMA
>>>  
>>> -config HOLES_IN_ZONE
>>> -	def_bool y
>>> -
>>
>> Right.
>>
>>>  source "kernel/Kconfig.hz"
>>>  
>>>  config ARCH_SPARSEMEM_ENABLE
>>> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
>>> index dc03bdc12c0f..eb3f56fb8c7c 100644
>>> --- a/arch/arm64/mm/init.c
>>> +++ b/arch/arm64/mm/init.c
>>> @@ -243,7 +243,7 @@ int pfn_valid(unsigned long pfn)
>>>  
>>>  	/*
>>>  	 * ZONE_DEVICE memory does not have the memblock entries.
>>> -	 * memblock_is_map_memory() check for ZONE_DEVICE based
>>> +	 * memblock_is_memory() check for ZONE_DEVICE based
>>>  	 * addresses will always fail. Even the normal hotplugged
>>>  	 * memory will never have MEMBLOCK_NOMAP flag set in their
>>>  	 * memblock entries. Skip memblock search for all non early
>>> @@ -254,7 +254,7 @@ int pfn_valid(unsigned long pfn)
>>>  		return pfn_section_valid(ms, pfn);
>>>  }
>>>  #endif
>>> -	return memblock_is_map_memory(addr);
>>> +	return memblock_is_memory(addr);
>>
>> Wondering if MEMBLOCK_NOMAP is now being treated similarly to other
>> memory pfns for page table walking purpose but with PageReserved(),
>> why memblock_is_memory() is still required ? At this point, should
>> not we just return valid for early_section() memory. As pfn_valid()
>> now just implies that pfn has a struct page backing which has been
>> already verified with valid_section() etc.
> 
> memblock_is_memory() is required because arm64 frees unused parts of the
> memory map. So, for instance, if we have 64M out of 128M populated in a
> section the section based calculation would return 1 for a pfn in the
> second half of the section, but there would be no memory map there.

Understood.