Re: [PATCH v4 6/7] mm: truncate: split huge page cache page to a non-zero order if possible.

[Ryan Roberts <ryan.roberts@arm.com>]

On 14/02/2024 16:19, Zi Yan wrote:
> On 14 Feb 2024, at 5:43, Ryan Roberts wrote:
> 
>> On 13/02/2024 21:55, Zi Yan wrote:
>>> From: Zi Yan <ziy@nvidia.com>
>>>
>>> To minimize the number of pages after a huge page truncation, we do not
>>> need to split it all the way down to order-0. The huge page has at most
>>> three parts, the part before offset, the part to be truncated, the part
>>> remaining at the end. Find the greatest common divisor of them to
>>> calculate the new page order from it, so we can split the huge
>>> page to this order and keep the remaining pages as large and as few as
>>> possible.
>>>
>>> Signed-off-by: Zi Yan <ziy@nvidia.com>
>>> ---
>>>  mm/truncate.c | 21 +++++++++++++++++++--
>>>  1 file changed, 19 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/mm/truncate.c b/mm/truncate.c
>>> index 725b150e47ac..49ddbbf7a617 100644
>>> --- a/mm/truncate.c
>>> +++ b/mm/truncate.c
>>> @@ -21,6 +21,7 @@
>>>  #include <linux/task_io_accounting_ops.h>
>>>  #include <linux/shmem_fs.h>
>>>  #include <linux/rmap.h>
>>> +#include <linux/gcd.h>
>>>  #include "internal.h"
>>>
>>>  /*
>>> @@ -210,7 +211,8 @@ int truncate_inode_folio(struct address_space *mapping, struct folio *folio)
>>>  bool truncate_inode_partial_folio(struct folio *folio, loff_t start, loff_t end)
>>>  {
>>>  	loff_t pos = folio_pos(folio);
>>> -	unsigned int offset, length;
>>> +	unsigned int offset, length, remaining;
>>> +	unsigned int new_order = folio_order(folio);
>>>
>>>  	if (pos < start)
>>>  		offset = start - pos;
>>> @@ -221,6 +223,7 @@ bool truncate_inode_partial_folio(struct folio *folio, loff_t start, loff_t end)
>>>  		length = length - offset;
>>>  	else
>>>  		length = end + 1 - pos - offset;
>>> +	remaining = folio_size(folio) - offset - length;
>>>
>>>  	folio_wait_writeback(folio);
>>>  	if (length == folio_size(folio)) {
>>> @@ -235,11 +238,25 @@ bool truncate_inode_partial_folio(struct folio *folio, loff_t start, loff_t end)
>>>  	 */
>>>  	folio_zero_range(folio, offset, length);
>>>
>>> +	/*
>>> +	 * Use the greatest common divisor of offset, length, and remaining
>>> +	 * as the smallest page size and compute the new order from it. So we
>>> +	 * can truncate a subpage as large as possible. Round up gcd to
>>> +	 * PAGE_SIZE, otherwise ilog2 can give -1 when gcd/PAGE_SIZE is 0.
>>> +	 */
>>> +	new_order = ilog2(round_up(gcd(gcd(offset, length), remaining),
>>> +				   PAGE_SIZE) / PAGE_SIZE);
>>
>> Given you have up to 2 regions remaining, isn't it possible that you want a
>> different order for both those regions (or even multiple orders within the same
>> region)? I guess you just choose gcd for simplicity?
> 
> Right. You raise the same concern as Hugh[1]. I am minimizing the call of
> split_huge_page_to_list_to_order() and you and Hugh want to minimize the
> number of folios after the split. Yours will give better outcome after split,
> but requires either multiple calls or a more sophisticated implementation
> of page split[2]. We probably can revisit this once splitting to any order
> gets wider use.

Yeah, fair enough. Sorry hadn't read Hugh's original feedback.

> 
>>> +
>>> +	/* order-1 THP not supported, downgrade to order-0 */
>>> +	if (new_order == 1)
>>> +		new_order = 0;
>>
>> I guess this would need to change if supporting order-1 file folios?
> 
> Right.
> 
>>> +
>>> +
>>>  	if (folio_has_private(folio))
>>>  		folio_invalidate(folio, offset, length);
>>>  	if (!folio_test_large(folio))
>>>  		return true;
>>> -	if (split_folio(folio) == 0)
>>> +	if (split_huge_page_to_list_to_order(&folio->page, NULL, new_order) == 0)
>>
>> I know you are discussing removing this patch, but since you created
>> split_folio_to_order() wouldn't that be better here?
> 
> Sure. Will change the patch locally.
> 
> [1] https://lore.kernel.org/linux-mm/9dd96da-efa2-5123-20d4-4992136ef3ad@google.com/
> [2] https://lore.kernel.org/linux-mm/0AC0520E-1BD2-497E-A7ED-05394400BFC9@nvidia.com/
> 
> --
> Best Regards,
> Yan, Zi