Linux Virtualization - Nov 2016 - [PATCH kernel v4 7/7] virtio-balloon: tell host vm's unused page info

Please squish this and patch 5 together.  It makes no sense to separate
them.
> +static void send_unused_pages_info(struct virtio_balloon *vb,
> +				unsigned long req_id)
> +{
> +	struct scatterlist sg_in;
> +	unsigned long pfn = 0, bmap_len, pfn_limit, last_pfn, nr_pfn;
> +	struct virtqueue *vq = vb->req_vq;
> +	struct virtio_balloon_resp_hdr *hdr = vb->resp_hdr;
> +	int ret = 1, used_nr_bmap = 0, i;
> +
> +	if (virtio_has_feature(vb->vdev, VIRTIO_BALLOON_F_PAGE_BITMAP)
&&
> +		vb->nr_page_bmap == 1)
> +		extend_page_bitmap(vb);
> +
> +	pfn_limit = PFNS_PER_BMAP * vb->nr_page_bmap;
> +	mutex_lock(&vb->balloon_lock);
> +	last_pfn = get_max_pfn();
> +
> +	while (ret) {
> +		clear_page_bitmap(vb);
> +		ret = get_unused_pages(pfn, pfn + pfn_limit, vb->page_bitmap,
> +			 PFNS_PER_BMAP, vb->nr_page_bmap);
This changed the underlying data structure without changing the way that
the structure is populated.

This algorithm picks a "PFNS_PER_BMAP * vb->nr_page_bmap"-sized set
of
pfns, allocates a bitmap for them, the loops through all zones looking
for pages in any free list that are in that range.

Unpacking all the indirection, it looks like this:

for (pfn = 0; pfn < get_max_pfn(); pfn += BITMAP_SIZE_IN_PFNS)
	for_each_populated_zone(zone)
		for_each_migratetype_order(order, t)
			list_for_each(..., &zone->free_area[order])...

Let's say we do a 32k bitmap that can hold ~1M pages.  That's 4GB of
RAM.  On a 1TB system, that's 256 passes through the top-level loop.
The bottom-level lists have tens of thousands of pages in them, even on
my laptop.  Only 1/256 of these pages will get consumed in a given pass.

That's an awfully inefficient way of doing it.  This patch essentially
changed the data structure without changing the algorithm to populate it.

Please change the *algorithm* to use the new data structure efficiently.
 Such a change would only do a single pass through each freelist, and
would choose whether to use the extent-based (pfn -> range) or
bitmap-based approach based on the contents of the free lists.

You should not be using get_max_pfn().  Any patch set that continues to
use it is not likely to be using a proper algorithm.

> Please squish this and patch 5 together.  It makes no sense to separate
them.
> 
OK.
> > +static void send_unused_pages_info(struct virtio_balloon *vb,
> > +				unsigned long req_id)
> > +{
> > +	struct scatterlist sg_in;
> > +	unsigned long pfn = 0, bmap_len, pfn_limit, last_pfn, nr_pfn;
> > +	struct virtqueue *vq = vb->req_vq;
> > +	struct virtio_balloon_resp_hdr *hdr = vb->resp_hdr;
> > +	int ret = 1, used_nr_bmap = 0, i;
> > +
> > +	if (virtio_has_feature(vb->vdev,
> VIRTIO_BALLOON_F_PAGE_BITMAP) &&
> > +		vb->nr_page_bmap == 1)
> > +		extend_page_bitmap(vb);
> > +
> > +	pfn_limit = PFNS_PER_BMAP * vb->nr_page_bmap;
> > +	mutex_lock(&vb->balloon_lock);
> > +	last_pfn = get_max_pfn();
> > +
> > +	while (ret) {
> > +		clear_page_bitmap(vb);
> > +		ret = get_unused_pages(pfn, pfn + pfn_limit, vb-
> >page_bitmap,
> > +			 PFNS_PER_BMAP, vb->nr_page_bmap);
> 
> This changed the underlying data structure without changing the way that
> the structure is populated.
> 
> This algorithm picks a "PFNS_PER_BMAP *
vb->nr_page_bmap"-sized set of
> pfns, allocates a bitmap for them, the loops through all zones looking for
> pages in any free list that are in that range.
> 
> Unpacking all the indirection, it looks like this:
> 
> for (pfn = 0; pfn < get_max_pfn(); pfn += BITMAP_SIZE_IN_PFNS)
> 	for_each_populated_zone(zone)
> 		for_each_migratetype_order(order, t)
> 			list_for_each(..., &zone->free_area[order])...
> 
> Let's say we do a 32k bitmap that can hold ~1M pages.  That's 4GB
of RAM.
> On a 1TB system, that's 256 passes through the top-level loop.
> The bottom-level lists have tens of thousands of pages in them, even on my
> laptop.  Only 1/256 of these pages will get consumed in a given pass.
> 
Your description is not exactly.
A 32k bitmap is used only when there is few free memory left in the system and
when
the extend_page_bitmap() failed to allocate more memory for the bitmap. Or
dozens of
32k split bitmap will be used, this version limit the bitmap count to 32, it
means we can use
at most 32*32 kB for the bitmap, which can cover 128GB for RAM. We can increase
the bitmap
count limit to a larger value if 32 is not big enough.
> That's an awfully inefficient way of doing it.  This patch essentially
changed
> the data structure without changing the algorithm to populate it.
> 
> Please change the *algorithm* to use the new data structure efficiently.
>  Such a change would only do a single pass through each freelist, and would
> choose whether to use the extent-based (pfn -> range) or bitmap-based
> approach based on the contents of the free lists.
Save the free page info to a raw bitmap first and then process the raw bitmap to
get the proper ' extent-based ' and  'bitmap-based' is the most
efficient way I can
come up with to save the virtio data transmission.  Do you have some better
idea?


In the QEMU, no matter how we encode the bitmap, the raw format bitmap will be
used in the end.  But what I did in this version is:
   kernel: get the raw bitmap  --> encode the bitmap
   QEMU: decode the bitmap --> get the raw bitmap

Is it worth to do this kind of job here? we can save the virtio data
transmission, but at the
same time, we did extra work.

It seems the benefit we get for this feature is not as big as that in fast
balloon inflating/deflating.
> 
> You should not be using get_max_pfn().  Any patch set that continues to use
> it is not likely to be using a proper algorithm.
Do you have any suggestion about how to avoid it?

Thanks!
Liang

On 11/06/2016 07:37 PM, Li, Liang Z wrote:
>> Let's say we do a 32k bitmap that can hold ~1M pages.  That's
4GB of RAM.
>> On a 1TB system, that's 256 passes through the top-level loop.
>> The bottom-level lists have tens of thousands of pages in them, even on
my
>> laptop.  Only 1/256 of these pages will get consumed in a given pass.
>>
> Your description is not exactly.
> A 32k bitmap is used only when there is few free memory left in the system
and when
> the extend_page_bitmap() failed to allocate more memory for the bitmap. Or
dozens of
> 32k split bitmap will be used, this version limit the bitmap count to 32,
it means we can use
> at most 32*32 kB for the bitmap, which can cover 128GB for RAM. We can
increase the bitmap
> count limit to a larger value if 32 is not big enough.
OK, so it tries to allocate a large bitmap.  But, if it fails, it will
try to work with a smaller bitmap.  Correct?

So, what's the _worst_ case?  It sounds like it is even worse than I was
positing.
>> That's an awfully inefficient way of doing it.  This patch
essentially changed
>> the data structure without changing the algorithm to populate it.
>>
>> Please change the *algorithm* to use the new data structure
efficiently.
>>  Such a change would only do a single pass through each freelist, and
would
>> choose whether to use the extent-based (pfn -> range) or
bitmap-based
>> approach based on the contents of the free lists.
> 
> Save the free page info to a raw bitmap first and then process the raw
bitmap to
> get the proper ' extent-based ' and  'bitmap-based' is the
most efficient way I can
> come up with to save the virtio data transmission.  Do you have some better
idea?
That's kinda my point.  This patch *does* processing to try to pack the
bitmaps full of pages from the various pfn ranges.  It's a form of
processing that gets *REALLY*, *REALLY* bad in some (admittedly obscure)
cases.

Let's not pretend that making an essentially unlimited number of passes
over the free lists is not processing.

1. Allocate as large of a bitmap as you can. (what you already do)
2. Iterate from the largest freelist order.  Store those pages in the
   bitmap.
3. If you can no longer fit pages in the bitmap, return the list that
   you have.
4. Make an approximation about where the bitmap does not make any more,
   and fall back to listing individual PFNs.  This would make sens, for
   instance in a large zone with very few free order-0 pages left.
			
> It seems the benefit we get for this feature is not as big as that in fast
balloon inflating/deflating.
>>
>> You should not be using get_max_pfn().  Any patch set that continues to
use
>> it is not likely to be using a proper algorithm.
> 
> Do you have any suggestion about how to avoid it?
Yes: get the pfns from the page free lists alone.  Don't derive them
from the pfn limits of the system or zones.