Re: [PATCH v2 2/2] mm: speed up mremap by 500x on large regions

[Anton Ivanov <anton.ivanov@kot-begemot.co.uk>]

On 12/10/2018 15:37, Kirill A. Shutemov wrote:
> On Fri, Oct 12, 2018 at 03:09:49PM +0100, Anton Ivanov wrote:
>> On 10/12/18 2:37 AM, Joel Fernandes (Google) wrote:
>>> Android needs to mremap large regions of memory during memory management
>>> related operations. The mremap system call can be really slow if THP is
>>> not enabled. The bottleneck is move_page_tables, which is copying each
>>> pte at a time, and can be really slow across a large map. Turning on THP
>>> may not be a viable option, and is not for us. This patch speeds up the
>>> performance for non-THP system by copying at the PMD level when possible.
>>>
>>> The speed up is three orders of magnitude. On a 1GB mremap, the mremap
>>> completion times drops from 160-250 millesconds to 380-400 microseconds.
>>>
>>> Before:
>>> Total mremap time for 1GB data: 242321014 nanoseconds.
>>> Total mremap time for 1GB data: 196842467 nanoseconds.
>>> Total mremap time for 1GB data: 167051162 nanoseconds.
>>>
>>> After:
>>> Total mremap time for 1GB data: 385781 nanoseconds.
>>> Total mremap time for 1GB data: 388959 nanoseconds.
>>> Total mremap time for 1GB data: 402813 nanoseconds.
>>>
>>> Incase THP is enabled, the optimization is skipped. I also flush the
>>> tlb every time we do this optimization since I couldn't find a way to
>>> determine if the low-level PTEs are dirty. It is seen that the cost of
>>> doing so is not much compared the improvement, on both x86-64 and arm64.
>>>
>>> Cc: minchan@kernel.org
>>> Cc: pantin@google.com
>>> Cc: hughd@google.com
>>> Cc: lokeshgidra@google.com
>>> Cc: dancol@google.com
>>> Cc: mhocko@kernel.org
>>> Cc: kirill@shutemov.name
>>> Cc: akpm@linux-foundation.org
>>> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
>>> ---
>>>    mm/mremap.c | 62 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>>>    1 file changed, 62 insertions(+)
>>>
>>> diff --git a/mm/mremap.c b/mm/mremap.c
>>> index 9e68a02a52b1..d82c485822ef 100644
>>> --- a/mm/mremap.c
>>> +++ b/mm/mremap.c
>>> @@ -191,6 +191,54 @@ static void move_ptes(struct vm_area_struct *vma, pmd_t *old_pmd,
>>>    		drop_rmap_locks(vma);
>>>    }
>>> +static bool move_normal_pmd(struct vm_area_struct *vma, unsigned long old_addr,
>>> +		  unsigned long new_addr, unsigned long old_end,
>>> +		  pmd_t *old_pmd, pmd_t *new_pmd, bool *need_flush)
>>> +{
>>> +	spinlock_t *old_ptl, *new_ptl;
>>> +	struct mm_struct *mm = vma->vm_mm;
>>> +
>>> +	if ((old_addr & ~PMD_MASK) || (new_addr & ~PMD_MASK)
>>> +	    || old_end - old_addr < PMD_SIZE)
>>> +		return false;
>>> +
>>> +	/*
>>> +	 * The destination pmd shouldn't be established, free_pgtables()
>>> +	 * should have release it.
>>> +	 */
>>> +	if (WARN_ON(!pmd_none(*new_pmd)))
>>> +		return false;
>>> +
>>> +	/*
>>> +	 * We don't have to worry about the ordering of src and dst
>>> +	 * ptlocks because exclusive mmap_sem prevents deadlock.
>>> +	 */
>>> +	old_ptl = pmd_lock(vma->vm_mm, old_pmd);
>>> +	if (old_ptl) {
>>> +		pmd_t pmd;
>>> +
>>> +		new_ptl = pmd_lockptr(mm, new_pmd);
>>> +		if (new_ptl != old_ptl)
>>> +			spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING);
>>> +
>>> +		/* Clear the pmd */
>>> +		pmd = *old_pmd;
>>> +		pmd_clear(old_pmd);
>>> +
>>> +		VM_BUG_ON(!pmd_none(*new_pmd));
>>> +
>>> +		/* Set the new pmd */
>>> +		set_pmd_at(mm, new_addr, new_pmd, pmd);
>> UML does not have set_pmd_at at all
> Every architecture does. :)

I tried to build it patching vs 4.19-rc before I made this statement and 
ran into that.

Presently it does not.

https://elixir.bootlin.com/linux/v4.19-rc7/ident/set_pmd_at - UML is not 
on the list.

>
> But it may come not from the arch code.

There is no generic definition as far as I can see. All 12 defines in 
4.19 are in arch specific code. Unless i am missing something...

>
>> If I read the code right, MIPS completely ignores the address argument so
>> set_pmd_at there may not have the effect which this patch is trying to
>> achieve.
> Ignoring address is fine. Most architectures do that..
> The ideas is to move page table to the new pmd slot. It's nothing to do
> with the address passed to set_pmd_at().

If that is it's only function, then I am going to appropriate the code 
out of the MIPS tree for further uml testing. It does exactly that - 
just move the pmd the new slot.

>
A.