mariadb/storage/innobase/buf/buf0buddy.cc

/*****************************************************************************

Copyright (c) 2006, 2016, Oracle and/or its affiliates. All Rights Reserved.
Copyright (c) 2018, 2021, MariaDB Corporation.

This program is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free Software
Foundation; version 2 of the License.

This program is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with
this program; if not, write to the Free Software Foundation, Inc.,
51 Franklin Street, Fifth Floor, Boston, MA 02110-1335 USA

*****************************************************************************/

/**************************************************//**
@file buf/buf0buddy.cc
Binary buddy allocator for compressed pages

Created December 2006 by Marko Makela
*******************************************************/

#include "buf0buddy.h"
#include "buf0buf.h"
#include "buf0lru.h"
#include "buf0flu.h"
#include "page0zip.h"
#include "srv0start.h"

/** When freeing a buf we attempt to coalesce by looking at its buddy
and deciding whether it is free or not. To ascertain if the buddy is
free we look for BUF_BUDDY_STAMP_FREE at BUF_BUDDY_STAMP_OFFSET
within the buddy. The question is how we can be sure that it is
safe to look at BUF_BUDDY_STAMP_OFFSET.
The answer lies in following invariants:
* All blocks allocated by buddy allocator are used for compressed
page frame.
* A compressed table always have space_id < SRV_SPACE_ID_UPPER_BOUND
* BUF_BUDDY_STAMP_OFFSET always points to the space_id field in
a frame.
  -- The above is true because we look at these fields when the
     corresponding buddy block is free which implies that:
     * The block we are looking at must have an address aligned at
       the same size that its free buddy has. For example, if we have
       a free block of 8K then its buddy's address must be aligned at
       8K as well.
     * It is possible that the block we are looking at may have been
       further divided into smaller sized blocks but its starting
       address must still remain the start of a page frame i.e.: it
       cannot be middle of a block. For example, if we have a free
       block of size 8K then its buddy may be divided into blocks
       of, say, 1K, 1K, 2K, 4K but the buddy's address will still be
       the starting address of first 1K compressed page.
     * What is important to note is that for any given block, the
       buddy's address cannot be in the middle of a larger block i.e.:
       in above example, our 8K block cannot have a buddy whose address
       is aligned on 8K but it is part of a larger 16K block.
*/

/** Offset within buf_buddy_free_t where free or non_free stamps
are written.*/
#define BUF_BUDDY_STAMP_OFFSET	FIL_PAGE_ARCH_LOG_NO_OR_SPACE_ID

/** Value that we stamp on all buffers that are currently on the zip_free
list. This value is stamped at BUF_BUDDY_STAMP_OFFSET offset */
#define BUF_BUDDY_STAMP_FREE	 SRV_SPACE_ID_UPPER_BOUND

/** Stamp value for non-free buffers. Will be overwritten by a non-zero
value by the consumer of the block */
#define BUF_BUDDY_STAMP_NONFREE	0XFFFFFFFFUL

/** Return type of buf_buddy_is_free() */
enum buf_buddy_state_t {
	BUF_BUDDY_STATE_FREE,	/*!< If the buddy to completely free */
	BUF_BUDDY_STATE_USED,	/*!< Buddy currently in used */
	BUF_BUDDY_STATE_PARTIALLY_USED/*!< Some sub-blocks in the buddy
				are in use */
};

/**********************************************************************//**
Invalidate memory area that we won't access while page is free */
UNIV_INLINE
void
buf_buddy_mem_invalid(
/*==================*/
	buf_buddy_free_t*	buf,	/*!< in: block to check */
	ulint			i)	/*!< in: index of zip_free[] */
{
  ut_ad(i <= BUF_BUDDY_SIZES);

  MEM_CHECK_ADDRESSABLE(buf, BUF_BUDDY_LOW << i);
  MEM_UNDEFINED(buf, BUF_BUDDY_LOW << i);
}

/**********************************************************************//**
Check if a buddy is stamped free.
@return whether the buddy is free */
UNIV_INLINE MY_ATTRIBUTE((warn_unused_result))
bool
buf_buddy_stamp_is_free(
/*====================*/
	const buf_buddy_free_t*	buf)	/*!< in: block to check */
{
	compile_time_assert(BUF_BUDDY_STAMP_FREE < BUF_BUDDY_STAMP_NONFREE);
	return(mach_read_from_4(buf->stamp.bytes + BUF_BUDDY_STAMP_OFFSET)
	       == BUF_BUDDY_STAMP_FREE);
}

/**********************************************************************//**
Stamps a buddy free. */
UNIV_INLINE
void
buf_buddy_stamp_free(
/*=================*/
	buf_buddy_free_t*	buf,	/*!< in/out: block to stamp */
	ulint			i)	/*!< in: block size */
{
	ut_d(memset(&buf->stamp.bytes, int(i), BUF_BUDDY_LOW << i));
	buf_buddy_mem_invalid(buf, i);
	mach_write_to_4(buf->stamp.bytes + BUF_BUDDY_STAMP_OFFSET,
			BUF_BUDDY_STAMP_FREE);
	buf->stamp.size = i;
}

/**********************************************************************//**
Stamps a buddy nonfree.
@param[in,out]	buf	block to stamp
@param[in]	i	block size */
static inline void buf_buddy_stamp_nonfree(buf_buddy_free_t* buf, ulint i)
{
	buf_buddy_mem_invalid(buf, i);
	compile_time_assert(BUF_BUDDY_STAMP_NONFREE == 0xffffffffU);
	memset(buf->stamp.bytes + BUF_BUDDY_STAMP_OFFSET, 0xff, 4);
}

/**********************************************************************//**
Get the offset of the buddy of a compressed page frame.
@return the buddy relative of page */
UNIV_INLINE
void*
buf_buddy_get(
/*==========*/
	byte*	page,	/*!< in: compressed page */
	ulint	size)	/*!< in: page size in bytes */
{
	ut_ad(ut_is_2pow(size));
	ut_ad(size >= BUF_BUDDY_LOW);
	ut_ad(BUF_BUDDY_LOW <= UNIV_ZIP_SIZE_MIN);
	ut_ad(size < BUF_BUDDY_HIGH);
	ut_ad(BUF_BUDDY_HIGH == srv_page_size);
	ut_ad(!ut_align_offset(page, size));

	if (((ulint) page) & size) {
		return(page - size);
	} else {
		return(page + size);
	}
}

#ifdef UNIV_DEBUG
const buf_block_t *buf_pool_t::contains_zip(const void *data, size_t shift)
  const noexcept
{
  const size_t d= size_t(data) >> shift;

  for (size_t i= 0; i < n_blocks; i++)
  {
    const buf_block_t *block= get_nth_page(i);
    if (size_t(block->page.zip.data) >> shift == d)
      return block;
  }
  return nullptr;
}

/** Validate a given zip_free list. */
struct	CheckZipFree {
	CheckZipFree(ulint i) : m_i(i) {}

	void operator()(const buf_buddy_free_t* elem) const
	{
		ut_ad(buf_buddy_stamp_is_free(elem));
		ut_ad(elem->stamp.size <= m_i);
	}

	const ulint m_i;
};

/** Validate a buddy list.
@param[in]	i		buddy size to validate */
static void buf_buddy_list_validate(ulint i)
{
	ut_list_validate(buf_pool.zip_free[i], CheckZipFree(i));
}

/**********************************************************************//**
Debug function to validate that a buffer is indeed free i.e.: in the
zip_free[].
@param[in]	buf		block to check
@param[in]	i		index of buf_pool.zip_free[]
@return true if free */
static bool buf_buddy_check_free(const buf_buddy_free_t* buf, ulint i)
{
	const ulint	size	= BUF_BUDDY_LOW << i;

	mysql_mutex_assert_owner(&buf_pool.mutex);
	ut_ad(!ut_align_offset(buf, size));
	ut_ad(i >= buf_buddy_get_slot(UNIV_ZIP_SIZE_MIN));

	buf_buddy_free_t* itr;

	for (itr = UT_LIST_GET_FIRST(buf_pool.zip_free[i]);
	     itr && itr != buf;
	     itr = UT_LIST_GET_NEXT(list, itr)) {
	}

	return(itr == buf);
}
#endif /* UNIV_DEBUG */

/**********************************************************************//**
Checks if a buf is free i.e.: in the zip_free[].
@retval BUF_BUDDY_STATE_FREE if fully free
@retval BUF_BUDDY_STATE_USED if currently in use
@retval BUF_BUDDY_STATE_PARTIALLY_USED if partially in use. */
static  MY_ATTRIBUTE((warn_unused_result))
buf_buddy_state_t
buf_buddy_is_free(
/*==============*/
	buf_buddy_free_t*	buf,	/*!< in: block to check */
	ulint			i)	/*!< in: index of
					buf_pool.zip_free[] */
{
#ifdef UNIV_DEBUG
	const ulint	size	= BUF_BUDDY_LOW << i;
	ut_ad(!ut_align_offset(buf, size));
	ut_ad(i >= buf_buddy_get_slot(UNIV_ZIP_SIZE_MIN));
#endif /* UNIV_DEBUG */

	/* We assume that all memory from buf_buddy_alloc()
	is used for compressed page frames. */

	/* We look inside the allocated objects returned by
	buf_buddy_alloc() and assume that each block is a compressed
	page that contains one of the following in space_id.
	* BUF_BUDDY_STAMP_FREE if the block is in a zip_free list or
	* BUF_BUDDY_STAMP_NONFREE if the block has been allocated but
	not initialized yet or
	* A valid space_id of a compressed tablespace

	The call below attempts to read from free memory.  The memory
	is "owned" by the buddy allocator (and it has been allocated
	from the buffer pool), so there is nothing wrong about this. */
	if (!buf_buddy_stamp_is_free(buf)) {
		return(BUF_BUDDY_STATE_USED);
	}

	/* A block may be free but a fragment of it may still be in use.
	To guard against that we write the free block size in terms of
	zip_free index at start of stamped block. Note that we can
	safely rely on this value only if the buf is free. */
	ut_ad(buf->stamp.size <= i);
	return(buf->stamp.size == i
	       ? BUF_BUDDY_STATE_FREE
	       : BUF_BUDDY_STATE_PARTIALLY_USED);
}

/** Add a block to the head of the appropriate buddy free list.
@param[in,out]	buf		block to be freed
@param[in]	i		index of buf_pool.zip_free[] */
static void buf_buddy_add_to_free(buf_buddy_free_t *buf, ulint i)
{
	mysql_mutex_assert_owner(&buf_pool.mutex);
	ut_ad(buf_pool.zip_free[i].start != buf);
	buf_buddy_stamp_free(buf, i);
	UT_LIST_ADD_FIRST(buf_pool.zip_free[i], buf);
	ut_d(buf_buddy_list_validate(i));
}

/** Remove a block from the appropriate buddy free list.
@param[in,out]	buf		block to be freed
@param[in]	i		index of buf_pool.zip_free[] */
static void buf_buddy_remove_from_free(buf_buddy_free_t *buf, ulint i)
{
	mysql_mutex_assert_owner(&buf_pool.mutex);
	ut_ad(buf_buddy_check_free(buf, i));

	UT_LIST_REMOVE(buf_pool.zip_free[i], buf);
	buf_buddy_stamp_nonfree(buf, i);
}

/** Try to allocate a block from buf_pool.zip_free[].
@param[in]	i		index of buf_pool.zip_free[]
@return allocated block, or NULL if buf_pool.zip_free[] was empty */
static buf_buddy_free_t* buf_buddy_alloc_zip(ulint i)
{
	buf_buddy_free_t*	buf;

	mysql_mutex_assert_owner(&buf_pool.mutex);
	ut_a(i < BUF_BUDDY_SIZES);
	ut_a(i >= buf_buddy_get_slot(UNIV_ZIP_SIZE_MIN));

	ut_d(buf_buddy_list_validate(i));

	buf = UT_LIST_GET_FIRST(buf_pool.zip_free[i]);

	if (size_t size = buf_pool.shrinking_size()) {
		while (buf != NULL
		       && buf_pool.will_be_withdrawn(
			       reinterpret_cast<byte*>(buf), size)) {
			/* This should be withdrawn, not to be allocated */
			buf = UT_LIST_GET_NEXT(list, buf);
		}
	}

	if (buf) {
		buf_buddy_remove_from_free(buf, i);
		ut_ad(!buf_pool.contains_zip(buf, BUF_BUDDY_LOW_SHIFT + i));
	} else if (i + 1 < BUF_BUDDY_SIZES) {
		/* Attempt to split. */
		buf = buf_buddy_alloc_zip(i + 1);

		if (buf) {
			buf_buddy_free_t* buddy =
				reinterpret_cast<buf_buddy_free_t*>(
					reinterpret_cast<byte*>(buf)
					+ (BUF_BUDDY_LOW << i));
			buf_buddy_add_to_free(buddy, i);
		}
	}

	if (buf) {
		/* Trash the page other than the BUF_BUDDY_STAMP_NONFREE. */
		MEM_UNDEFINED(buf, BUF_BUDDY_STAMP_OFFSET);
		MEM_UNDEFINED(BUF_BUDDY_STAMP_OFFSET + 4 + buf->stamp.bytes,
			      (BUF_BUDDY_LOW << i)
			      - (BUF_BUDDY_STAMP_OFFSET + 4));
		ut_ad(mach_read_from_4(buf->stamp.bytes
				       + BUF_BUDDY_STAMP_OFFSET)
		      == BUF_BUDDY_STAMP_NONFREE);
	}

	return(buf);
}

#ifdef UNIV_DEBUG
/** number of blocks allocated to the buddy system */
static size_t buf_buddy_n_frames;
#endif

/** Deallocate a buffer frame of srv_page_size.
@param buf    buffer frame to deallocate */
static void buf_buddy_block_free(void *buf) noexcept
{
  mysql_mutex_assert_owner(&buf_pool.mutex);
  buf_block_t *block= buf_pool.block_from(buf);
  ut_ad(block->page.state() == buf_page_t::MEMORY);
  ut_ad(block->page.frame == buf);
  ut_ad(!buf_pool.contains_zip(buf, srv_page_size_shift));
  ut_d(memset(buf, 0, srv_page_size));
  MEM_UNDEFINED(buf, srv_page_size);
  buf_LRU_block_free_non_file_page(block);
  ut_ad(buf_buddy_n_frames > 0);
  ut_d(buf_buddy_n_frames--);
}

/** Allocate a buffer block to the buddy allocator.
@param block   buffer block to register */
static void buf_buddy_block_register(buf_block_t *block) noexcept
{
  ut_ad(buf_pool.is_uncompressed_current(block));
  ut_ad(block->page.state() == buf_page_t::MEMORY);
  ut_d(buf_buddy_n_frames++);
}

/** Allocate a block from a bigger object.
@param[in]	buf		a block that is free to use
@param[in]	i		index of buf_pool.zip_free[]
@return allocated block */
static void *buf_buddy_alloc_from(void *buf, ulint i)
{
	ut_ad(i >= buf_buddy_get_slot(UNIV_ZIP_SIZE_MIN));
	ut_ad(i <= BUF_BUDDY_SIZES);
	ut_ad(!ut_align_offset(buf, srv_page_size));
	ut_ad(!buf_pool.contains_zip(buf, srv_page_size));

	/* Add the unused parts of the block to the free lists. */
	for (ulint j = BUF_BUDDY_SIZES, offs = srv_page_size; j-- > i; ) {
		buf_buddy_free_t*	zip_buf;

		offs >>= 1;

		zip_buf = reinterpret_cast<buf_buddy_free_t*>(
			reinterpret_cast<byte*>(buf) + offs);
		buf_buddy_add_to_free(zip_buf, j);
	}

	buf_buddy_stamp_nonfree(reinterpret_cast<buf_buddy_free_t*>(buf), i);
	return(buf);
}

/** Allocate a ROW_FORMAT=COMPRESSED block.
@param i      index of buf_pool.zip_free[] or BUF_BUDDY_SIZES
@param lru    assigned to true if buf_pool.mutex was temporarily released
@return allocated block, never NULL */
byte *buf_buddy_alloc_low(ulint i, bool *lru) noexcept
{
	buf_block_t*	block;

	mysql_mutex_assert_owner(&buf_pool.mutex);
	ut_ad(i >= buf_buddy_get_slot(UNIV_ZIP_SIZE_MIN));

	if (i < BUF_BUDDY_SIZES) {
		/* Try to allocate from the buddy system. */
		block = (buf_block_t*) buf_buddy_alloc_zip(i);

		if (block) {
			goto func_exit;
		}
	}

	/* Try allocating from the buf_pool.free list. */
	block = buf_pool.allocate();

	if (block) {
		goto alloc_big;
	}

	/* Try replacing an uncompressed page in the buffer pool. */
	block = buf_LRU_get_free_block(true);
	if (lru) {
		*lru = true;
	}

alloc_big:
	buf_buddy_block_register(block);

	block = reinterpret_cast<buf_block_t*>(
		buf_buddy_alloc_from(block->page.frame, i));

func_exit:
	buf_pool.buddy_stat[i].used++;
	return reinterpret_cast<byte*>(block);
}

/** Try to relocate a block.
@param[in]	src		block to relocate
@param[in]	dst		free block to relocated to
@param[in]	i		index of buf_pool.zip_free[]
@param[in]	force		true if we must relocated always
@return true if relocated */
static bool buf_buddy_relocate(void *src, void *dst, ulint i, bool force)
  noexcept
{
	buf_page_t*	bpage;
	const ulint	size = BUF_BUDDY_LOW << i;

	mysql_mutex_assert_owner(&buf_pool.mutex);
	ut_ad(!ut_align_offset(src, size));
	ut_ad(!ut_align_offset(dst, size));
	ut_ad(i >= buf_buddy_get_slot(UNIV_ZIP_SIZE_MIN));
	MEM_CHECK_ADDRESSABLE(dst, size);

	uint32_t space = mach_read_from_4(static_cast<const byte*>(src)
					  + FIL_PAGE_ARCH_LOG_NO_OR_SPACE_ID);
	uint32_t offset = mach_read_from_4(static_cast<const byte*>(src)
					   + FIL_PAGE_OFFSET);

	/* Suppress Valgrind or MSAN warnings. */
	MEM_MAKE_DEFINED(&space, sizeof space);
	MEM_MAKE_DEFINED(&offset, sizeof offset);

	ut_ad(space != BUF_BUDDY_STAMP_FREE);

	const page_id_t	page_id(space, offset);
	/* FIXME: we are computing this while holding buf_pool.mutex */
	auto &cell= buf_pool.page_hash.cell_get(page_id.fold());

	bpage = buf_pool.page_hash.get(page_id, cell);

	if (!bpage || bpage->zip.data != src) {
		/* The block has probably been freshly
		allocated by buf_LRU_get_free_block() but not
		added to buf_pool.page_hash yet.  Obviously,
		it cannot be relocated. */

		if (!force || space != 0 || offset != 0) {
			return(false);
		}

		/* It might be just uninitialized page.
		We should search from LRU list also. */

		bpage = UT_LIST_GET_FIRST(buf_pool.LRU);
		while (bpage != NULL) {
			if (bpage->zip.data == src) {
				ut_ad(bpage->id() == page_id);
				break;
			}
			bpage = UT_LIST_GET_NEXT(LRU, bpage);
		}

		if (bpage == NULL) {
			return(false);
		}
	}

	if (page_zip_get_size(&bpage->zip) != size) {
		/* The block is of different size.  We would
		have to relocate all blocks covered by src.
		For the sake of simplicity, give up. */
		ut_ad(page_zip_get_size(&bpage->zip) < size);
		return(false);
	}

	/* The block must have been allocated, but it may
	contain uninitialized data. */
	MEM_CHECK_ADDRESSABLE(src, size);

	if (!bpage->can_relocate()) {
		return false;
	}

	page_hash_latch &hash_lock = buf_pool.page_hash.lock_get(cell);
	/* It does not make sense to use transactional_lock_guard here,
	because the memcpy() of 1024 to 16384 bytes would likely make the
	memory transaction too large. */
	hash_lock.lock();

	if (bpage->can_relocate()) {
		/* Relocate the compressed page. */
		const ulonglong ns = my_interval_timer();

		ut_a(bpage->zip.data == src);

		memcpy(dst, src, size);
		bpage->zip.data = reinterpret_cast<page_zip_t*>(dst);

		hash_lock.unlock();

		buf_buddy_mem_invalid(
			reinterpret_cast<buf_buddy_free_t*>(src), i);

		buf_buddy_stat_t*	buddy_stat = &buf_pool.buddy_stat[i];
		buddy_stat->relocated++;
		buddy_stat->relocated_usec+= (my_interval_timer() - ns) / 1000;
		return(true);
	}

	hash_lock.unlock();

	return(false);
}

/** Deallocate a block.
@param[in]	buf	block to be freed, must not be pointed to
			by the buffer pool
@param[in]	i	index of buf_pool.zip_free[], or BUF_BUDDY_SIZES */
void buf_buddy_free_low(void* buf, ulint i) noexcept
{
	buf_buddy_free_t*	buddy;

	mysql_mutex_assert_owner(&buf_pool.mutex);
	ut_ad(i <= BUF_BUDDY_SIZES);
	ut_ad(i >= buf_buddy_get_slot(UNIV_ZIP_SIZE_MIN));
	ut_ad(buf_pool.buddy_stat[i].used > 0);

	buf_pool.buddy_stat[i].used--;
recombine:
	MEM_UNDEFINED(buf, BUF_BUDDY_LOW << i);

	if (i == BUF_BUDDY_SIZES) {
		buf_buddy_block_free(buf);
		return;
	}

	ut_ad(i < BUF_BUDDY_SIZES);
	ut_ad(buf == ut_align_down(buf, BUF_BUDDY_LOW << i));
	ut_ad(!buf_pool.contains_zip(buf, BUF_BUDDY_LOW_SHIFT + i));

	/* Do not recombine blocks if there are few free blocks.
	We may waste up to 15360*max_len bytes to free blocks
	(1024 + 2048 + 4096 + 8192 = 15360) */
	if (UT_LIST_GET_LEN(buf_pool.zip_free[i]) < 16) {
		goto func_exit;
	}

	/* Try to combine adjacent blocks. */
	buddy = reinterpret_cast<buf_buddy_free_t*>(
		buf_buddy_get(reinterpret_cast<byte*>(buf),
			      BUF_BUDDY_LOW << i));

	switch (buf_buddy_is_free(buddy, i)) {
	case BUF_BUDDY_STATE_FREE:
		/* The buddy is free: recombine */
		buf_buddy_remove_from_free(buddy, i);
buddy_is_free:
		i++;
		buf = ut_align_down(buf, BUF_BUDDY_LOW << i);
		ut_ad(!buf_pool.contains_zip(buf, BUF_BUDDY_LOW_SHIFT + i));
		goto recombine;

	case BUF_BUDDY_STATE_USED:
		ut_d(buf_buddy_list_validate(i));

		/* The buddy is not free. Is there a free block of
		this size? */
		if (buf_buddy_free_t* zip_buf =
			UT_LIST_GET_FIRST(buf_pool.zip_free[i])) {

			/* Remove the block from the free list, because
			a successful buf_buddy_relocate() will overwrite
			zip_free->list. */
			buf_buddy_remove_from_free(zip_buf, i);

			/* Try to relocate the buddy of buf to the free
			block. */
			if (buf_buddy_relocate(buddy, zip_buf, i, false)) {
				goto buddy_is_free;
			}

			buf_buddy_add_to_free(zip_buf, i);
		}

		break;
	case BUF_BUDDY_STATE_PARTIALLY_USED:
		/* Some sub-blocks in the buddy are still in use.
		Relocation will fail. No need to try. */
		break;
	}

func_exit:
	/* Free the block to the buddy list. */
	buf_buddy_add_to_free(reinterpret_cast<buf_buddy_free_t*>(buf), i);
}

/** Reallocate a ROW_FORMAT=COMPRESSED page frame during buf_pool_t::resize().
@param bpage page descriptor covering a ROW_FORMAT=COMPRESSED page
@param block uncompressed block for storage
@return block
@retval nullptr if the block was consumed */
ATTRIBUTE_COLD
buf_block_t *buf_buddy_shrink(buf_page_t *bpage, buf_block_t *block) noexcept
{
  ut_ad(bpage->zip.data);

  void *dst= nullptr;
  ulint size= page_zip_get_size(&bpage->zip);
  ulint i= buf_buddy_get_slot(size);

  ut_ad(buf_pool.will_be_withdrawn(bpage->zip.data, size));
  ut_ad(bpage->can_relocate());
  ut_ad(i <= BUF_BUDDY_SIZES);
  ut_ad(i >= buf_buddy_get_slot(UNIV_ZIP_SIZE_MIN));

  if (UNIV_LIKELY(i < BUF_BUDDY_SIZES))
    dst= buf_buddy_alloc_zip(i);

  if (!dst)
  {
    buf_buddy_block_register(block);
    dst= buf_buddy_alloc_from(block->page.frame, i);
    ut_ad(dst);
    block= nullptr;
  }

  void *src= bpage->zip.data;
  memcpy_aligned<UNIV_ZIP_SIZE_MIN>(dst, src, size);
  bpage->zip.data= static_cast<page_zip_t*>(dst);
  buf_pool.buddy_stat[i].relocated++;

  for (;;)
  {
    MEM_UNDEFINED(src, BUF_BUDDY_LOW << i);
    ut_ad(i < BUF_BUDDY_SIZES);
    /* Try to combine adjacent blocks. */
    buf_buddy_free_t *buddy= reinterpret_cast<buf_buddy_free_t*>
      (buf_buddy_get(static_cast<byte*>(src), BUF_BUDDY_LOW << i));

    if (buf_buddy_is_free(buddy, i) != BUF_BUDDY_STATE_FREE)
    {
      ut_ad(!buf_pool.contains_zip(src, BUF_BUDDY_LOW_SHIFT + i));
      buf_buddy_add_to_free(static_cast<buf_buddy_free_t*>(src), i);
      break;
    }

    /* The buddy is free: recombine */
    buf_buddy_remove_from_free(buddy, i);
    i++;
    src= ut_align_down(src, BUF_BUDDY_LOW << i);
    if (i == BUF_BUDDY_SIZES)
    {
      buf_buddy_block_free(src);
      break;
    }
  }

  return block;
}

/** Combine all pairs of free buddies.
@param size  the target innodb_buffer_pool_size */
ATTRIBUTE_COLD void buf_buddy_condense_free(size_t size) noexcept
{
  ut_ad(size);
  ut_ad(size == buf_pool.shrinking_size());

  for (ulint i= 0; i < array_elements(buf_pool.zip_free); i++)
  {
    buf_buddy_free_t *buf= UT_LIST_GET_FIRST(buf_pool.zip_free[i]);

    /* seek to withdraw target */
    while (buf &&
           !buf_pool.will_be_withdrawn(reinterpret_cast<byte*>(buf), size))
      buf= UT_LIST_GET_NEXT(list, buf);

    for (buf_buddy_free_t *next= buf; buf; buf= next)
    {
      buf_buddy_free_t *buddy= reinterpret_cast<buf_buddy_free_t*>
        (buf_buddy_get(reinterpret_cast<byte*>(buf), BUF_BUDDY_LOW << i));

      /* seek to the next withdraw target */
      do
      {
        while ((next= UT_LIST_GET_NEXT(list, next)) &&
               !buf_pool.will_be_withdrawn(reinterpret_cast<byte*>(next),
                                           size)) {}
      }
      while (buddy == next);

      if (buf_buddy_is_free(buddy, i) != BUF_BUDDY_STATE_FREE)
        continue;

      buf_buddy_remove_from_free(buf, i);
      ulint j= i;
    recombine:
      buf_buddy_remove_from_free(buddy, j);
      j++;
      buf= static_cast<buf_buddy_free_t*>
        (ut_align_down(buf, BUF_BUDDY_LOW << j));
      MEM_UNDEFINED(buf, BUF_BUDDY_LOW << j);

      if (j == BUF_BUDDY_SIZES)
      {
        buf_buddy_block_free(buf);
        continue;
      }

      buddy= reinterpret_cast<buf_buddy_free_t*>
        (buf_buddy_get(reinterpret_cast<byte*>(buf), BUF_BUDDY_LOW << j));
      if (buf_buddy_is_free(buddy, j) == BUF_BUDDY_STATE_FREE)
        goto recombine;

      ut_ad(!buf_pool.contains_zip(buf, BUF_BUDDY_LOW_SHIFT + j));
      buf_buddy_add_to_free(buf, j);
    }
  }
}