Re: [PATCH v5 1/5] mm/memblock: Tag memblocks with crypto capabilities

Re: [PATCH v5 1/5] mm/memblock: Tag memblocks with crypto capabilities

[Mike Rapoport]

On Thu, Jan 13, 2022 at 06:30:23PM -0300, Martin Fernandez wrote:
> Add the capability to mark regions of the memory memory_type able of
> hardware memory encryption.
> 
> Also add the capability to query if all regions of a memory node are
> able to do hardware memory encryption to call it when initializing the
> nodes.
> 
> Signed-off-by: Martin Fernandez <martin.fernandez@eclypsium.com>
> ---
>  include/linux/memblock.h |  5 ++++
>  mm/memblock.c            | 49 ++++++++++++++++++++++++++++++++++++++++
>  2 files changed, 54 insertions(+)
> 
> diff --git a/include/linux/memblock.h b/include/linux/memblock.h
> index 9dc7cb239d21..374c03e10b2e 100644
> --- a/include/linux/memblock.h
> +++ b/include/linux/memblock.h
> @@ -41,6 +41,7 @@ extern unsigned long long max_possible_pfn;
>   * via a driver, and never indicated in the firmware-provided memory map as
>   * system RAM. This corresponds to IORESOURCE_SYSRAM_DRIVER_MANAGED in the
>   * kernel resource tree.
> + * @MEMBLOCK_CRYPTO_CAPABLE: capable of hardware encryption
>   */
>  enum memblock_flags {
>  	MEMBLOCK_NONE		= 0x0,	/* No special request */
> @@ -48,6 +49,7 @@ enum memblock_flags {
>  	MEMBLOCK_MIRROR		= 0x2,	/* mirrored region */
>  	MEMBLOCK_NOMAP		= 0x4,	/* don't add to kernel direct mapping */
>  	MEMBLOCK_DRIVER_MANAGED = 0x8,	/* always detected via a driver */
> +	MEMBLOCK_CRYPTO_CAPABLE = 0x10,  /* capable of hardware encryption */

Nit: please keep the comments aligned with TAB.

>  };
>  
>  /**
> @@ -121,6 +123,9 @@ int memblock_physmem_add(phys_addr_t base, phys_addr_t size);
>  void memblock_trim_memory(phys_addr_t align);
>  bool memblock_overlaps_region(struct memblock_type *type,
>  			      phys_addr_t base, phys_addr_t size);
> +bool memblock_node_is_crypto_capable(int nid);
> +int memblock_mark_crypto_capable(phys_addr_t base, phys_addr_t size);
> +int memblock_clear_crypto_capable(phys_addr_t base, phys_addr_t size);
>  int memblock_mark_hotplug(phys_addr_t base, phys_addr_t size);
>  int memblock_clear_hotplug(phys_addr_t base, phys_addr_t size);
>  int memblock_mark_mirror(phys_addr_t base, phys_addr_t size);
> diff --git a/mm/memblock.c b/mm/memblock.c
> index 1018e50566f3..61ec50647469 100644
> --- a/mm/memblock.c
> +++ b/mm/memblock.c
> @@ -191,6 +191,27 @@ bool __init_memblock memblock_overlaps_region(struct memblock_type *type,
>  	return i < type->cnt;
>  }
>  
> +/**
> + * memblock_node_is_crypto_capable - get if whole node is capable
> + * of encryption
> + * @nid: number of node
> + *
> + * Iterate over all memory memblock_type and find if all regions under
> + * node @nid are capable of hardware encryption.

Please add Return: description, otherwise kernel-doc is unhappy

> + */
> +bool __init_memblock memblock_node_is_crypto_capable(int nid)
> +{
> +	struct memblock_region *region;
> +
> +	for_each_mem_region(region) {
> +		if ((memblock_get_region_node(region) == nid) &&
> +		    !(region->flags & MEMBLOCK_CRYPTO_CAPABLE))
> +			return false;
> +	}

As we discussed on v3, please add a printk if the same node has both
crypto-capable and not crypto-capable regions.

https://lore.kernel.org/all/Ya++1FwWzKr2wYQH@kernel.org/

> +
> +	return true;
> +}
> +
>  /**
>   * __memblock_find_range_bottom_up - find free area utility in bottom-up
>   * @start: start of candidate range
> @@ -885,6 +906,34 @@ static int __init_memblock memblock_setclr_flag(phys_addr_t base,
>  	return 0;
>  }
>  
> +/**
> + * memblock_mark_crypto_capable - Mark memory regions capable of hardware
> + * encryption with flag MEMBLOCK_CRYPTO_CAPABLE.
> + * @base: the base phys addr of the region
> + * @size: the size of the region
> + *
> + * Return: 0 on success, -errno on failure.
> + */
> +int __init_memblock memblock_mark_crypto_capable(phys_addr_t base,
> +						 phys_addr_t size)
> +{
> +	return memblock_setclr_flag(base, size, 1, MEMBLOCK_CRYPTO_CAPABLE);
> +}
> +
> +/**
> + * memblock_clear_crypto_capable - Clear flag MEMBLOCK_CRYPTO for a
> + * specified region.
> + * @base: the base phys addr of the region
> + * @size: the size of the region
> + *
> + * Return: 0 on success, -errno on failure.
> + */
> +int __init_memblock memblock_clear_crypto_capable(phys_addr_t base,
> +						  phys_addr_t size)
> +{
> +	return memblock_setclr_flag(base, size, 0, MEMBLOCK_CRYPTO_CAPABLE);
> +}
> +
>  /**
>   * memblock_mark_hotplug - Mark hotpluggable memory with flag MEMBLOCK_HOTPLUG.
>   * @base: the base phys addr of the region
> -- 
> 2.30.2
> 

-- 
Sincerely yours,
Mike.

Re: [PATCH v5 1/5] mm/memblock: Tag memblocks with crypto capabilities

[Martin Fernandez]

On 1/14/22, Mike Rapoport <rppt@kernel.org> wrote:
> On Thu, Jan 13, 2022 at 06:30:23PM -0300, Martin Fernandez wrote:
>> Add the capability to mark regions of the memory memory_type able of
>> hardware memory encryption.
>>
>> Also add the capability to query if all regions of a memory node are
>> able to do hardware memory encryption to call it when initializing the
>> nodes.
>>
>> Signed-off-by: Martin Fernandez <martin.fernandez@eclypsium.com>
>> ---
>>  include/linux/memblock.h |  5 ++++
>>  mm/memblock.c            | 49 ++++++++++++++++++++++++++++++++++++++++
>>  2 files changed, 54 insertions(+)
>>
>> diff --git a/include/linux/memblock.h b/include/linux/memblock.h
>> index 9dc7cb239d21..374c03e10b2e 100644
>> --- a/include/linux/memblock.h
>> +++ b/include/linux/memblock.h
>> @@ -41,6 +41,7 @@ extern unsigned long long max_possible_pfn;
>>   * via a driver, and never indicated in the firmware-provided memory map
>> as
>>   * system RAM. This corresponds to IORESOURCE_SYSRAM_DRIVER_MANAGED in
>> the
>>   * kernel resource tree.
>> + * @MEMBLOCK_CRYPTO_CAPABLE: capable of hardware encryption
>>   */
>>  enum memblock_flags {
>>  	MEMBLOCK_NONE		= 0x0,	/* No special request */
>> @@ -48,6 +49,7 @@ enum memblock_flags {
>>  	MEMBLOCK_MIRROR		= 0x2,	/* mirrored region */
>>  	MEMBLOCK_NOMAP		= 0x4,	/* don't add to kernel direct mapping */
>>  	MEMBLOCK_DRIVER_MANAGED = 0x8,	/* always detected via a driver */
>> +	MEMBLOCK_CRYPTO_CAPABLE = 0x10,  /* capable of hardware encryption */
>
> Nit: please keep the comments aligned with TAB.
>
>>  };
>>
>>  /**
>> @@ -121,6 +123,9 @@ int memblock_physmem_add(phys_addr_t base, phys_addr_t
>> size);
>>  void memblock_trim_memory(phys_addr_t align);
>>  bool memblock_overlaps_region(struct memblock_type *type,
>>  			      phys_addr_t base, phys_addr_t size);
>> +bool memblock_node_is_crypto_capable(int nid);
>> +int memblock_mark_crypto_capable(phys_addr_t base, phys_addr_t size);
>> +int memblock_clear_crypto_capable(phys_addr_t base, phys_addr_t size);
>>  int memblock_mark_hotplug(phys_addr_t base, phys_addr_t size);
>>  int memblock_clear_hotplug(phys_addr_t base, phys_addr_t size);
>>  int memblock_mark_mirror(phys_addr_t base, phys_addr_t size);
>> diff --git a/mm/memblock.c b/mm/memblock.c
>> index 1018e50566f3..61ec50647469 100644
>> --- a/mm/memblock.c
>> +++ b/mm/memblock.c
>> @@ -191,6 +191,27 @@ bool __init_memblock memblock_overlaps_region(struct
>> memblock_type *type,
>>  	return i < type->cnt;
>>  }
>>
>> +/**
>> + * memblock_node_is_crypto_capable - get if whole node is capable
>> + * of encryption
>> + * @nid: number of node
>> + *
>> + * Iterate over all memory memblock_type and find if all regions under
>> + * node @nid are capable of hardware encryption.
>
> Please add Return: description, otherwise kernel-doc is unhappy
>
>> + */
>> +bool __init_memblock memblock_node_is_crypto_capable(int nid)
>> +{
>> +	struct memblock_region *region;
>> +
>> +	for_each_mem_region(region) {
>> +		if ((memblock_get_region_node(region) == nid) &&
>> +		    !(region->flags & MEMBLOCK_CRYPTO_CAPABLE))
>> +			return false;
>> +	}
>
> As we discussed on v3, please add a printk if the same node has both
> crypto-capable and not crypto-capable regions.
>
> https://lore.kernel.org/all/Ya++1FwWzKr2wYQH@kernel.org/
>
>> +
>> +	return true;
>> +}
>> +
>>  /**
>>   * __memblock_find_range_bottom_up - find free area utility in bottom-up
>>   * @start: start of candidate range
>> @@ -885,6 +906,34 @@ static int __init_memblock
>> memblock_setclr_flag(phys_addr_t base,
>>  	return 0;
>>  }
>>
>> +/**
>> + * memblock_mark_crypto_capable - Mark memory regions capable of
>> hardware
>> + * encryption with flag MEMBLOCK_CRYPTO_CAPABLE.
>> + * @base: the base phys addr of the region
>> + * @size: the size of the region
>> + *
>> + * Return: 0 on success, -errno on failure.
>> + */
>> +int __init_memblock memblock_mark_crypto_capable(phys_addr_t base,
>> +						 phys_addr_t size)
>> +{
>> +	return memblock_setclr_flag(base, size, 1, MEMBLOCK_CRYPTO_CAPABLE);
>> +}
>> +
>> +/**
>> + * memblock_clear_crypto_capable - Clear flag MEMBLOCK_CRYPTO for a
>> + * specified region.
>> + * @base: the base phys addr of the region
>> + * @size: the size of the region
>> + *
>> + * Return: 0 on success, -errno on failure.
>> + */
>> +int __init_memblock memblock_clear_crypto_capable(phys_addr_t base,
>> +						  phys_addr_t size)
>> +{
>> +	return memblock_setclr_flag(base, size, 0, MEMBLOCK_CRYPTO_CAPABLE);
>> +}
>> +
>>  /**
>>   * memblock_mark_hotplug - Mark hotpluggable memory with flag
>> MEMBLOCK_HOTPLUG.
>>   * @base: the base phys addr of the region
>> --
>> 2.30.2
>>
>
> --
> Sincerely yours,
> Mike.
>

Ok, will do. Thanks for the feedback.

Re: [PATCH v5 3/5] x86/e820: Tag e820_entry with crypto capabilities

[Dave Hansen]

On 1/13/22 1:30 PM, Martin Fernandez wrote:
> +/*
> + * Update crypto capabilities in a range
> + */
> +static u64 __init __e820__range_update_crypto(struct e820_table *table,
> +					      u64 start, u64 size,
> +					      enum e820_crypto_capabilities crypto_capable)

This looks like an almost pure copy-and-paste of a 70-line function.
That's not the end of the world, but it does seem like a place where
refactoring is in order.

The rest of this series looks pretty straightforward, but this part in
particular is the most worrying.

Re: [PATCH v5 3/5] x86/e820: Tag e820_entry with crypto capabilities

[Martin Fernandez]

On 1/14/22, Dave Hansen <dave.hansen@intel.com> wrote:
> On 1/13/22 1:30 PM, Martin Fernandez wrote:
>> +/*
>> + * Update crypto capabilities in a range
>> + */
>> +static u64 __init __e820__range_update_crypto(struct e820_table *table,
>> +					      u64 start, u64 size,
>> +					      enum e820_crypto_capabilities crypto_capable)
>
> This looks like an almost pure copy-and-paste of a 70-line function.
> That's not the end of the world, but it does seem like a place where
> refactoring is in order.
>
> The rest of this series looks pretty straightforward, but this part in
> particular is the most worrying.
>

Yep, I'm not super happy with this either. I'll check it out to see
what we can do. Btw e820__range_remove is quite similiar too.

Re: [PATCH v5 3/5] x86/e820: Tag e820_entry with crypto capabilities

[Martin Fernandez]

> On 1/14/22, Dave Hansen <dave.hansen@intel.com> wrote:
>> On 1/13/22 1:30 PM, Martin Fernandez wrote:
>>> +/*
>>> + * Update crypto capabilities in a range
>>> + */
>>> +static u64 __init __e820__range_update_crypto(struct e820_table *table,
>>> +					      u64 start, u64 size,
>>> +					      enum e820_crypto_capabilities crypto_capable)
>>
>> This looks like an almost pure copy-and-paste of a 70-line function.
>> That's not the end of the world, but it does seem like a place where
>> refactoring is in order.

Introducing a for_each to iterate over the table is one of the
improvements I'm thinking of. Do you think it's a good idea to
introduce it in this patch (changing all the relevant for loops for
this new for_each) or should I do it in a separate patch?