Source file src/runtime/mpagealloc.go

     1  // Copyright 2019 The Go Authors. All rights reserved.
     2  // Use of this source code is governed by a BSD-style
     3  // license that can be found in the LICENSE file.
     4  
     5  // Page allocator.
     6  //
     7  // The page allocator manages mapped pages (defined by pageSize, NOT
     8  // physPageSize) for allocation and re-use. It is embedded into mheap.
     9  //
    10  // Pages are managed using a bitmap that is sharded into chunks.
    11  // In the bitmap, 1 means in-use, and 0 means free. The bitmap spans the
    12  // process's address space. Chunks are managed in a sparse-array-style structure
    13  // similar to mheap.arenas, since the bitmap may be large on some systems.
    14  //
    15  // The bitmap is efficiently searched by using a radix tree in combination
    16  // with fast bit-wise intrinsics. Allocation is performed using an address-ordered
    17  // first-fit approach.
    18  //
    19  // Each entry in the radix tree is a summary that describes three properties of
    20  // a particular region of the address space: the number of contiguous free pages
    21  // at the start and end of the region it represents, and the maximum number of
    22  // contiguous free pages found anywhere in that region.
    23  //
    24  // Each level of the radix tree is stored as one contiguous array, which represents
    25  // a different granularity of subdivision of the processes' address space. Thus, this
    26  // radix tree is actually implicit in these large arrays, as opposed to having explicit
    27  // dynamically-allocated pointer-based node structures. Naturally, these arrays may be
    28  // quite large for system with large address spaces, so in these cases they are mapped
    29  // into memory as needed. The leaf summaries of the tree correspond to a bitmap chunk.
    30  //
    31  // The root level (referred to as L0 and index 0 in pageAlloc.summary) has each
    32  // summary represent the largest section of address space (16 GiB on 64-bit systems),
    33  // with each subsequent level representing successively smaller subsections until we
    34  // reach the finest granularity at the leaves, a chunk.
    35  //
    36  // More specifically, each summary in each level (except for leaf summaries)
    37  // represents some number of entries in the following level. For example, each
    38  // summary in the root level may represent a 16 GiB region of address space,
    39  // and in the next level there could be 8 corresponding entries which represent 2
    40  // GiB subsections of that 16 GiB region, each of which could correspond to 8
    41  // entries in the next level which each represent 256 MiB regions, and so on.
    42  //
    43  // Thus, this design only scales to heaps so large, but can always be extended to
    44  // larger heaps by simply adding levels to the radix tree, which mostly costs
    45  // additional virtual address space. The choice of managing large arrays also means
    46  // that a large amount of virtual address space may be reserved by the runtime.
    47  
    48  package runtime
    49  
    50  import (
    51  	"internal/runtime/atomic"
    52  	"unsafe"
    53  )
    54  
    55  const (
    56  	// The size of a bitmap chunk, i.e. the amount of bits (that is, pages) to consider
    57  	// in the bitmap at once.
    58  	pallocChunkPages    = 1 << logPallocChunkPages
    59  	pallocChunkBytes    = pallocChunkPages * pageSize
    60  	logPallocChunkPages = 9
    61  	logPallocChunkBytes = logPallocChunkPages + pageShift
    62  
    63  	// The number of radix bits for each level.
    64  	//
    65  	// The value of 3 is chosen such that the block of summaries we need to scan at
    66  	// each level fits in 64 bytes (2^3 summaries * 8 bytes per summary), which is
    67  	// close to the L1 cache line width on many systems. Also, a value of 3 fits 4 tree
    68  	// levels perfectly into the 21-bit pallocBits summary field at the root level.
    69  	//
    70  	// The following equation explains how each of the constants relate:
    71  	// summaryL0Bits + (summaryLevels-1)*summaryLevelBits + logPallocChunkBytes = heapAddrBits
    72  	//
    73  	// summaryLevels is an architecture-dependent value defined in mpagealloc_*.go.
    74  	summaryLevelBits = 3
    75  	summaryL0Bits    = heapAddrBits - logPallocChunkBytes - (summaryLevels-1)*summaryLevelBits
    76  
    77  	// pallocChunksL2Bits is the number of bits of the chunk index number
    78  	// covered by the second level of the chunks map.
    79  	//
    80  	// See (*pageAlloc).chunks for more details. Update the documentation
    81  	// there should this change.
    82  	pallocChunksL2Bits  = heapAddrBits - logPallocChunkBytes - pallocChunksL1Bits
    83  	pallocChunksL1Shift = pallocChunksL2Bits
    84  )
    85  
    86  // maxSearchAddr returns the maximum searchAddr value, which indicates
    87  // that the heap has no free space.
    88  //
    89  // This function exists just to make it clear that this is the maximum address
    90  // for the page allocator's search space. See maxOffAddr for details.
    91  //
    92  // It's a function (rather than a variable) because it needs to be
    93  // usable before package runtime's dynamic initialization is complete.
    94  // See #51913 for details.
    95  func maxSearchAddr() offAddr { return maxOffAddr }
    96  
    97  // Global chunk index.
    98  //
    99  // Represents an index into the leaf level of the radix tree.
   100  // Similar to arenaIndex, except instead of arenas, it divides the address
   101  // space into chunks.
   102  type chunkIdx uint
   103  
   104  // chunkIndex returns the global index of the palloc chunk containing the
   105  // pointer p.
   106  func chunkIndex(p uintptr) chunkIdx {
   107  	return chunkIdx((p - arenaBaseOffset) / pallocChunkBytes)
   108  }
   109  
   110  // chunkBase returns the base address of the palloc chunk at index ci.
   111  func chunkBase(ci chunkIdx) uintptr {
   112  	return uintptr(ci)*pallocChunkBytes + arenaBaseOffset
   113  }
   114  
   115  // chunkPageIndex computes the index of the page that contains p,
   116  // relative to the chunk which contains p.
   117  func chunkPageIndex(p uintptr) uint {
   118  	return uint(p % pallocChunkBytes / pageSize)
   119  }
   120  
   121  // l1 returns the index into the first level of (*pageAlloc).chunks.
   122  func (i chunkIdx) l1() uint {
   123  	if pallocChunksL1Bits == 0 {
   124  		// Let the compiler optimize this away if there's no
   125  		// L1 map.
   126  		return 0
   127  	} else {
   128  		return uint(i) >> pallocChunksL1Shift
   129  	}
   130  }
   131  
   132  // l2 returns the index into the second level of (*pageAlloc).chunks.
   133  func (i chunkIdx) l2() uint {
   134  	if pallocChunksL1Bits == 0 {
   135  		return uint(i)
   136  	} else {
   137  		return uint(i) & (1<<pallocChunksL2Bits - 1)
   138  	}
   139  }
   140  
   141  // offAddrToLevelIndex converts an address in the offset address space
   142  // to the index into summary[level] containing addr.
   143  func offAddrToLevelIndex(level int, addr offAddr) int {
   144  	return int((addr.a - arenaBaseOffset) >> levelShift[level])
   145  }
   146  
   147  // levelIndexToOffAddr converts an index into summary[level] into
   148  // the corresponding address in the offset address space.
   149  func levelIndexToOffAddr(level, idx int) offAddr {
   150  	return offAddr{(uintptr(idx) << levelShift[level]) + arenaBaseOffset}
   151  }
   152  
   153  // addrsToSummaryRange converts base and limit pointers into a range
   154  // of entries for the given summary level.
   155  //
   156  // The returned range is inclusive on the lower bound and exclusive on
   157  // the upper bound.
   158  func addrsToSummaryRange(level int, base, limit uintptr) (lo int, hi int) {
   159  	// This is slightly more nuanced than just a shift for the exclusive
   160  	// upper-bound. Note that the exclusive upper bound may be within a
   161  	// summary at this level, meaning if we just do the obvious computation
   162  	// hi will end up being an inclusive upper bound. Unfortunately, just
   163  	// adding 1 to that is too broad since we might be on the very edge
   164  	// of a summary's max page count boundary for this level
   165  	// (1 << levelLogPages[level]). So, make limit an inclusive upper bound
   166  	// then shift, then add 1, so we get an exclusive upper bound at the end.
   167  	lo = int((base - arenaBaseOffset) >> levelShift[level])
   168  	hi = int(((limit-1)-arenaBaseOffset)>>levelShift[level]) + 1
   169  	return
   170  }
   171  
   172  // blockAlignSummaryRange aligns indices into the given level to that
   173  // level's block width (1 << levelBits[level]). It assumes lo is inclusive
   174  // and hi is exclusive, and so aligns them down and up respectively.
   175  func blockAlignSummaryRange(level int, lo, hi int) (int, int) {
   176  	e := uintptr(1) << levelBits[level]
   177  	return int(alignDown(uintptr(lo), e)), int(alignUp(uintptr(hi), e))
   178  }
   179  
   180  type pageAlloc struct {
   181  	// Radix tree of summaries.
   182  	//
   183  	// Each slice's cap represents the whole memory reservation.
   184  	// Each slice's len reflects the allocator's maximum known
   185  	// mapped heap address for that level.
   186  	//
   187  	// The backing store of each summary level is reserved in init
   188  	// and may or may not be committed in grow (small address spaces
   189  	// may commit all the memory in init).
   190  	//
   191  	// The purpose of keeping len <= cap is to enforce bounds checks
   192  	// on the top end of the slice so that instead of an unknown
   193  	// runtime segmentation fault, we get a much friendlier out-of-bounds
   194  	// error.
   195  	//
   196  	// To iterate over a summary level, use inUse to determine which ranges
   197  	// are currently available. Otherwise one might try to access
   198  	// memory which is only Reserved which may result in a hard fault.
   199  	//
   200  	// We may still get segmentation faults < len since some of that
   201  	// memory may not be committed yet.
   202  	summary [summaryLevels][]pallocSum
   203  
   204  	// chunks is a slice of bitmap chunks.
   205  	//
   206  	// The total size of chunks is quite large on most 64-bit platforms
   207  	// (O(GiB) or more) if flattened, so rather than making one large mapping
   208  	// (which has problems on some platforms, even when PROT_NONE) we use a
   209  	// two-level sparse array approach similar to the arena index in mheap.
   210  	//
   211  	// To find the chunk containing a memory address `a`, do:
   212  	//   chunkOf(chunkIndex(a))
   213  	//
   214  	// Below is a table describing the configuration for chunks for various
   215  	// heapAddrBits supported by the runtime.
   216  	//
   217  	// heapAddrBits | L1 Bits | L2 Bits | L2 Entry Size
   218  	// ------------------------------------------------
   219  	// 32           | 0       | 10      | 128 KiB
   220  	// 33 (iOS)     | 0       | 11      | 256 KiB
   221  	// 48           | 13      | 13      | 1 MiB
   222  	//
   223  	// There's no reason to use the L1 part of chunks on 32-bit, the
   224  	// address space is small so the L2 is small. For platforms with a
   225  	// 48-bit address space, we pick the L1 such that the L2 is 1 MiB
   226  	// in size, which is a good balance between low granularity without
   227  	// making the impact on BSS too high (note the L1 is stored directly
   228  	// in pageAlloc).
   229  	//
   230  	// To iterate over the bitmap, use inUse to determine which ranges
   231  	// are currently available. Otherwise one might iterate over unused
   232  	// ranges.
   233  	//
   234  	// Protected by mheapLock.
   235  	//
   236  	// TODO(mknyszek): Consider changing the definition of the bitmap
   237  	// such that 1 means free and 0 means in-use so that summaries and
   238  	// the bitmaps align better on zero-values.
   239  	chunks [1 << pallocChunksL1Bits]*[1 << pallocChunksL2Bits]pallocData
   240  
   241  	// The address to start an allocation search with. It must never
   242  	// point to any memory that is not contained in inUse, i.e.
   243  	// inUse.contains(searchAddr.addr()) must always be true. The one
   244  	// exception to this rule is that it may take on the value of
   245  	// maxOffAddr to indicate that the heap is exhausted.
   246  	//
   247  	// We guarantee that all valid heap addresses below this value
   248  	// are allocated and not worth searching.
   249  	searchAddr offAddr
   250  
   251  	// start and end represent the chunk indices
   252  	// which pageAlloc knows about. It assumes
   253  	// chunks in the range [start, end) are
   254  	// currently ready to use.
   255  	start, end chunkIdx
   256  
   257  	// inUse is a slice of ranges of address space which are
   258  	// known by the page allocator to be currently in-use (passed
   259  	// to grow).
   260  	//
   261  	// We care much more about having a contiguous heap in these cases
   262  	// and take additional measures to ensure that, so in nearly all
   263  	// cases this should have just 1 element.
   264  	//
   265  	// All access is protected by the mheapLock.
   266  	inUse addrRanges
   267  
   268  	// scav stores the scavenger state.
   269  	scav struct {
   270  		// index is an efficient index of chunks that have pages available to
   271  		// scavenge.
   272  		index scavengeIndex
   273  
   274  		// releasedBg is the amount of memory released in the background this
   275  		// scavenge cycle.
   276  		releasedBg atomic.Uintptr
   277  
   278  		// releasedEager is the amount of memory released eagerly this scavenge
   279  		// cycle.
   280  		releasedEager atomic.Uintptr
   281  	}
   282  
   283  	// mheap_.lock. This level of indirection makes it possible
   284  	// to test pageAlloc independently of the runtime allocator.
   285  	mheapLock *mutex
   286  
   287  	// sysStat is the runtime memstat to update when new system
   288  	// memory is committed by the pageAlloc for allocation metadata.
   289  	sysStat *sysMemStat
   290  
   291  	// summaryMappedReady is the number of bytes mapped in the Ready state
   292  	// in the summary structure. Used only for testing currently.
   293  	//
   294  	// Protected by mheapLock.
   295  	summaryMappedReady uintptr
   296  
   297  	// chunkHugePages indicates whether page bitmap chunks should be backed
   298  	// by huge pages.
   299  	chunkHugePages bool
   300  
   301  	// Whether or not this struct is being used in tests.
   302  	test bool
   303  }
   304  
   305  func (p *pageAlloc) init(mheapLock *mutex, sysStat *sysMemStat, test bool) {
   306  	if levelLogPages[0] > logMaxPackedValue {
   307  		// We can't represent 1<<levelLogPages[0] pages, the maximum number
   308  		// of pages we need to represent at the root level, in a summary, which
   309  		// is a big problem. Throw.
   310  		print("runtime: root level max pages = ", 1<<levelLogPages[0], "\n")
   311  		print("runtime: summary max pages = ", maxPackedValue, "\n")
   312  		throw("root level max pages doesn't fit in summary")
   313  	}
   314  	p.sysStat = sysStat
   315  
   316  	// Initialize p.inUse.
   317  	p.inUse.init(sysStat)
   318  
   319  	// System-dependent initialization.
   320  	p.sysInit(test)
   321  
   322  	// Start with the searchAddr in a state indicating there's no free memory.
   323  	p.searchAddr = maxSearchAddr()
   324  
   325  	// Set the mheapLock.
   326  	p.mheapLock = mheapLock
   327  
   328  	// Initialize the scavenge index.
   329  	p.summaryMappedReady += p.scav.index.init(test, sysStat)
   330  
   331  	// Set if we're in a test.
   332  	p.test = test
   333  }
   334  
   335  // tryChunkOf returns the bitmap data for the given chunk.
   336  //
   337  // Returns nil if the chunk data has not been mapped.
   338  func (p *pageAlloc) tryChunkOf(ci chunkIdx) *pallocData {
   339  	l2 := p.chunks[ci.l1()]
   340  	if l2 == nil {
   341  		return nil
   342  	}
   343  	return &l2[ci.l2()]
   344  }
   345  
   346  // chunkOf returns the chunk at the given chunk index.
   347  //
   348  // The chunk index must be valid or this method may throw.
   349  func (p *pageAlloc) chunkOf(ci chunkIdx) *pallocData {
   350  	return &p.chunks[ci.l1()][ci.l2()]
   351  }
   352  
   353  // grow sets up the metadata for the address range [base, base+size).
   354  // It may allocate metadata, in which case *p.sysStat will be updated.
   355  //
   356  // p.mheapLock must be held.
   357  func (p *pageAlloc) grow(base, size uintptr) {
   358  	assertLockHeld(p.mheapLock)
   359  
   360  	// Round up to chunks, since we can't deal with increments smaller
   361  	// than chunks. Also, sysGrow expects aligned values.
   362  	limit := alignUp(base+size, pallocChunkBytes)
   363  	base = alignDown(base, pallocChunkBytes)
   364  
   365  	// Grow the summary levels in a system-dependent manner.
   366  	// We just update a bunch of additional metadata here.
   367  	p.sysGrow(base, limit)
   368  
   369  	// Grow the scavenge index.
   370  	p.summaryMappedReady += p.scav.index.grow(base, limit, p.sysStat)
   371  
   372  	// Update p.start and p.end.
   373  	// If no growth happened yet, start == 0. This is generally
   374  	// safe since the zero page is unmapped.
   375  	firstGrowth := p.start == 0
   376  	start, end := chunkIndex(base), chunkIndex(limit)
   377  	if firstGrowth || start < p.start {
   378  		p.start = start
   379  	}
   380  	if end > p.end {
   381  		p.end = end
   382  	}
   383  	// Note that [base, limit) will never overlap with any existing
   384  	// range inUse because grow only ever adds never-used memory
   385  	// regions to the page allocator.
   386  	p.inUse.add(makeAddrRange(base, limit))
   387  
   388  	// A grow operation is a lot like a free operation, so if our
   389  	// chunk ends up below p.searchAddr, update p.searchAddr to the
   390  	// new address, just like in free.
   391  	if b := (offAddr{base}); b.lessThan(p.searchAddr) {
   392  		p.searchAddr = b
   393  	}
   394  
   395  	// Add entries into chunks, which is sparse, if needed. Then,
   396  	// initialize the bitmap.
   397  	//
   398  	// Newly-grown memory is always considered scavenged.
   399  	// Set all the bits in the scavenged bitmaps high.
   400  	for c := chunkIndex(base); c < chunkIndex(limit); c++ {
   401  		if p.chunks[c.l1()] == nil {
   402  			// Create the necessary l2 entry.
   403  			const l2Size = unsafe.Sizeof(*p.chunks[0])
   404  			r := sysAlloc(l2Size, p.sysStat)
   405  			if r == nil {
   406  				throw("pageAlloc: out of memory")
   407  			}
   408  			if !p.test {
   409  				// Make the chunk mapping eligible or ineligible
   410  				// for huge pages, depending on what our current
   411  				// state is.
   412  				if p.chunkHugePages {
   413  					sysHugePage(r, l2Size)
   414  				} else {
   415  					sysNoHugePage(r, l2Size)
   416  				}
   417  			}
   418  			// Store the new chunk block but avoid a write barrier.
   419  			// grow is used in call chains that disallow write barriers.
   420  			*(*uintptr)(unsafe.Pointer(&p.chunks[c.l1()])) = uintptr(r)
   421  		}
   422  		p.chunkOf(c).scavenged.setRange(0, pallocChunkPages)
   423  	}
   424  
   425  	// Update summaries accordingly. The grow acts like a free, so
   426  	// we need to ensure this newly-free memory is visible in the
   427  	// summaries.
   428  	p.update(base, size/pageSize, true, false)
   429  }
   430  
   431  // enableChunkHugePages enables huge pages for the chunk bitmap mappings (disabled by default).
   432  //
   433  // This function is idempotent.
   434  //
   435  // A note on latency: for sufficiently small heaps (<10s of GiB) this function will take constant
   436  // time, but may take time proportional to the size of the mapped heap beyond that.
   437  //
   438  // The heap lock must not be held over this operation, since it will briefly acquire
   439  // the heap lock.
   440  //
   441  // Must be called on the system stack because it acquires the heap lock.
   442  //
   443  //go:systemstack
   444  func (p *pageAlloc) enableChunkHugePages() {
   445  	// Grab the heap lock to turn on huge pages for new chunks and clone the current
   446  	// heap address space ranges.
   447  	//
   448  	// After the lock is released, we can be sure that bitmaps for any new chunks may
   449  	// be backed with huge pages, and we have the address space for the rest of the
   450  	// chunks. At the end of this function, all chunk metadata should be backed by huge
   451  	// pages.
   452  	lock(&mheap_.lock)
   453  	if p.chunkHugePages {
   454  		unlock(&mheap_.lock)
   455  		return
   456  	}
   457  	p.chunkHugePages = true
   458  	var inUse addrRanges
   459  	inUse.sysStat = p.sysStat
   460  	p.inUse.cloneInto(&inUse)
   461  	unlock(&mheap_.lock)
   462  
   463  	// This might seem like a lot of work, but all these loops are for generality.
   464  	//
   465  	// For a 1 GiB contiguous heap, a 48-bit address space, 13 L1 bits, a palloc chunk size
   466  	// of 4 MiB, and adherence to the default set of heap address hints, this will result in
   467  	// exactly 1 call to sysHugePage.
   468  	for _, r := range p.inUse.ranges {
   469  		for i := chunkIndex(r.base.addr()).l1(); i < chunkIndex(r.limit.addr()-1).l1(); i++ {
   470  			// N.B. We can assume that p.chunks[i] is non-nil and in a mapped part of p.chunks
   471  			// because it's derived from inUse, which never shrinks.
   472  			sysHugePage(unsafe.Pointer(p.chunks[i]), unsafe.Sizeof(*p.chunks[0]))
   473  		}
   474  	}
   475  }
   476  
   477  // update updates heap metadata. It must be called each time the bitmap
   478  // is updated.
   479  //
   480  // If contig is true, update does some optimizations assuming that there was
   481  // a contiguous allocation or free between addr and addr+npages. alloc indicates
   482  // whether the operation performed was an allocation or a free.
   483  //
   484  // p.mheapLock must be held.
   485  func (p *pageAlloc) update(base, npages uintptr, contig, alloc bool) {
   486  	assertLockHeld(p.mheapLock)
   487  
   488  	// base, limit, start, and end are inclusive.
   489  	limit := base + npages*pageSize - 1
   490  	sc, ec := chunkIndex(base), chunkIndex(limit)
   491  
   492  	// Handle updating the lowest level first.
   493  	if sc == ec {
   494  		// Fast path: the allocation doesn't span more than one chunk,
   495  		// so update this one and if the summary didn't change, return.
   496  		x := p.summary[len(p.summary)-1][sc]
   497  		y := p.chunkOf(sc).summarize()
   498  		if x == y {
   499  			return
   500  		}
   501  		p.summary[len(p.summary)-1][sc] = y
   502  	} else if contig {
   503  		// Slow contiguous path: the allocation spans more than one chunk
   504  		// and at least one summary is guaranteed to change.
   505  		summary := p.summary[len(p.summary)-1]
   506  
   507  		// Update the summary for chunk sc.
   508  		summary[sc] = p.chunkOf(sc).summarize()
   509  
   510  		// Update the summaries for chunks in between, which are
   511  		// either totally allocated or freed.
   512  		whole := p.summary[len(p.summary)-1][sc+1 : ec]
   513  		if alloc {
   514  			clear(whole)
   515  		} else {
   516  			for i := range whole {
   517  				whole[i] = freeChunkSum
   518  			}
   519  		}
   520  
   521  		// Update the summary for chunk ec.
   522  		summary[ec] = p.chunkOf(ec).summarize()
   523  	} else {
   524  		// Slow general path: the allocation spans more than one chunk
   525  		// and at least one summary is guaranteed to change.
   526  		//
   527  		// We can't assume a contiguous allocation happened, so walk over
   528  		// every chunk in the range and manually recompute the summary.
   529  		summary := p.summary[len(p.summary)-1]
   530  		for c := sc; c <= ec; c++ {
   531  			summary[c] = p.chunkOf(c).summarize()
   532  		}
   533  	}
   534  
   535  	// Walk up the radix tree and update the summaries appropriately.
   536  	changed := true
   537  	for l := len(p.summary) - 2; l >= 0 && changed; l-- {
   538  		// Update summaries at level l from summaries at level l+1.
   539  		changed = false
   540  
   541  		// "Constants" for the previous level which we
   542  		// need to compute the summary from that level.
   543  		logEntriesPerBlock := levelBits[l+1]
   544  		logMaxPages := levelLogPages[l+1]
   545  
   546  		// lo and hi describe all the parts of the level we need to look at.
   547  		lo, hi := addrsToSummaryRange(l, base, limit+1)
   548  
   549  		// Iterate over each block, updating the corresponding summary in the less-granular level.
   550  		for i := lo; i < hi; i++ {
   551  			children := p.summary[l+1][i<<logEntriesPerBlock : (i+1)<<logEntriesPerBlock]
   552  			sum := mergeSummaries(children, logMaxPages)
   553  			old := p.summary[l][i]
   554  			if old != sum {
   555  				changed = true
   556  				p.summary[l][i] = sum
   557  			}
   558  		}
   559  	}
   560  }
   561  
   562  // allocRange marks the range of memory [base, base+npages*pageSize) as
   563  // allocated. It also updates the summaries to reflect the newly-updated
   564  // bitmap.
   565  //
   566  // Returns the amount of scavenged memory in bytes present in the
   567  // allocated range.
   568  //
   569  // p.mheapLock must be held.
   570  func (p *pageAlloc) allocRange(base, npages uintptr) uintptr {
   571  	assertLockHeld(p.mheapLock)
   572  
   573  	limit := base + npages*pageSize - 1
   574  	sc, ec := chunkIndex(base), chunkIndex(limit)
   575  	si, ei := chunkPageIndex(base), chunkPageIndex(limit)
   576  
   577  	scav := uint(0)
   578  	if sc == ec {
   579  		// The range doesn't cross any chunk boundaries.
   580  		chunk := p.chunkOf(sc)
   581  		scav += chunk.scavenged.popcntRange(si, ei+1-si)
   582  		chunk.allocRange(si, ei+1-si)
   583  		p.scav.index.alloc(sc, ei+1-si)
   584  	} else {
   585  		// The range crosses at least one chunk boundary.
   586  		chunk := p.chunkOf(sc)
   587  		scav += chunk.scavenged.popcntRange(si, pallocChunkPages-si)
   588  		chunk.allocRange(si, pallocChunkPages-si)
   589  		p.scav.index.alloc(sc, pallocChunkPages-si)
   590  		for c := sc + 1; c < ec; c++ {
   591  			chunk := p.chunkOf(c)
   592  			scav += chunk.scavenged.popcntRange(0, pallocChunkPages)
   593  			chunk.allocAll()
   594  			p.scav.index.alloc(c, pallocChunkPages)
   595  		}
   596  		chunk = p.chunkOf(ec)
   597  		scav += chunk.scavenged.popcntRange(0, ei+1)
   598  		chunk.allocRange(0, ei+1)
   599  		p.scav.index.alloc(ec, ei+1)
   600  	}
   601  	p.update(base, npages, true, true)
   602  	return uintptr(scav) * pageSize
   603  }
   604  
   605  // findMappedAddr returns the smallest mapped offAddr that is
   606  // >= addr. That is, if addr refers to mapped memory, then it is
   607  // returned. If addr is higher than any mapped region, then
   608  // it returns maxOffAddr.
   609  //
   610  // p.mheapLock must be held.
   611  func (p *pageAlloc) findMappedAddr(addr offAddr) offAddr {
   612  	assertLockHeld(p.mheapLock)
   613  
   614  	// If we're not in a test, validate first by checking mheap_.arenas.
   615  	// This is a fast path which is only safe to use outside of testing.
   616  	ai := arenaIndex(addr.addr())
   617  	if p.test || mheap_.arenas[ai.l1()] == nil || mheap_.arenas[ai.l1()][ai.l2()] == nil {
   618  		vAddr, ok := p.inUse.findAddrGreaterEqual(addr.addr())
   619  		if ok {
   620  			return offAddr{vAddr}
   621  		} else {
   622  			// The candidate search address is greater than any
   623  			// known address, which means we definitely have no
   624  			// free memory left.
   625  			return maxOffAddr
   626  		}
   627  	}
   628  	return addr
   629  }
   630  
   631  // find searches for the first (address-ordered) contiguous free region of
   632  // npages in size and returns a base address for that region.
   633  //
   634  // It uses p.searchAddr to prune its search and assumes that no palloc chunks
   635  // below chunkIndex(p.searchAddr) contain any free memory at all.
   636  //
   637  // find also computes and returns a candidate p.searchAddr, which may or
   638  // may not prune more of the address space than p.searchAddr already does.
   639  // This candidate is always a valid p.searchAddr.
   640  //
   641  // find represents the slow path and the full radix tree search.
   642  //
   643  // Returns a base address of 0 on failure, in which case the candidate
   644  // searchAddr returned is invalid and must be ignored.
   645  //
   646  // p.mheapLock must be held.
   647  func (p *pageAlloc) find(npages uintptr) (uintptr, offAddr) {
   648  	assertLockHeld(p.mheapLock)
   649  
   650  	// Search algorithm.
   651  	//
   652  	// This algorithm walks each level l of the radix tree from the root level
   653  	// to the leaf level. It iterates over at most 1 << levelBits[l] of entries
   654  	// in a given level in the radix tree, and uses the summary information to
   655  	// find either:
   656  	//  1) That a given subtree contains a large enough contiguous region, at
   657  	//     which point it continues iterating on the next level, or
   658  	//  2) That there are enough contiguous boundary-crossing bits to satisfy
   659  	//     the allocation, at which point it knows exactly where to start
   660  	//     allocating from.
   661  	//
   662  	// i tracks the index into the current level l's structure for the
   663  	// contiguous 1 << levelBits[l] entries we're actually interested in.
   664  	//
   665  	// NOTE: Technically this search could allocate a region which crosses
   666  	// the arenaBaseOffset boundary, which when arenaBaseOffset != 0, is
   667  	// a discontinuity. However, the only way this could happen is if the
   668  	// page at the zero address is mapped, and this is impossible on
   669  	// every system we support where arenaBaseOffset != 0. So, the
   670  	// discontinuity is already encoded in the fact that the OS will never
   671  	// map the zero page for us, and this function doesn't try to handle
   672  	// this case in any way.
   673  
   674  	// i is the beginning of the block of entries we're searching at the
   675  	// current level.
   676  	i := 0
   677  
   678  	// firstFree is the region of address space that we are certain to
   679  	// find the first free page in the heap. base and bound are the inclusive
   680  	// bounds of this window, and both are addresses in the linearized, contiguous
   681  	// view of the address space (with arenaBaseOffset pre-added). At each level,
   682  	// this window is narrowed as we find the memory region containing the
   683  	// first free page of memory. To begin with, the range reflects the
   684  	// full process address space.
   685  	//
   686  	// firstFree is updated by calling foundFree each time free space in the
   687  	// heap is discovered.
   688  	//
   689  	// At the end of the search, base.addr() is the best new
   690  	// searchAddr we could deduce in this search.
   691  	firstFree := struct {
   692  		base, bound offAddr
   693  	}{
   694  		base:  minOffAddr,
   695  		bound: maxOffAddr,
   696  	}
   697  	// foundFree takes the given address range [addr, addr+size) and
   698  	// updates firstFree if it is a narrower range. The input range must
   699  	// either be fully contained within firstFree or not overlap with it
   700  	// at all.
   701  	//
   702  	// This way, we'll record the first summary we find with any free
   703  	// pages on the root level and narrow that down if we descend into
   704  	// that summary. But as soon as we need to iterate beyond that summary
   705  	// in a level to find a large enough range, we'll stop narrowing.
   706  	foundFree := func(addr offAddr, size uintptr) {
   707  		if firstFree.base.lessEqual(addr) && addr.add(size-1).lessEqual(firstFree.bound) {
   708  			// This range fits within the current firstFree window, so narrow
   709  			// down the firstFree window to the base and bound of this range.
   710  			firstFree.base = addr
   711  			firstFree.bound = addr.add(size - 1)
   712  		} else if !(addr.add(size-1).lessThan(firstFree.base) || firstFree.bound.lessThan(addr)) {
   713  			// This range only partially overlaps with the firstFree range,
   714  			// so throw.
   715  			print("runtime: addr = ", hex(addr.addr()), ", size = ", size, "\n")
   716  			print("runtime: base = ", hex(firstFree.base.addr()), ", bound = ", hex(firstFree.bound.addr()), "\n")
   717  			throw("range partially overlaps")
   718  		}
   719  	}
   720  
   721  	// lastSum is the summary which we saw on the previous level that made us
   722  	// move on to the next level. Used to print additional information in the
   723  	// case of a catastrophic failure.
   724  	// lastSumIdx is that summary's index in the previous level.
   725  	lastSum := packPallocSum(0, 0, 0)
   726  	lastSumIdx := -1
   727  
   728  nextLevel:
   729  	for l := 0; l < len(p.summary); l++ {
   730  		// For the root level, entriesPerBlock is the whole level.
   731  		entriesPerBlock := 1 << levelBits[l]
   732  		logMaxPages := levelLogPages[l]
   733  
   734  		// We've moved into a new level, so let's update i to our new
   735  		// starting index. This is a no-op for level 0.
   736  		i <<= levelBits[l]
   737  
   738  		// Slice out the block of entries we care about.
   739  		entries := p.summary[l][i : i+entriesPerBlock]
   740  
   741  		// Determine j0, the first index we should start iterating from.
   742  		// The searchAddr may help us eliminate iterations if we followed the
   743  		// searchAddr on the previous level or we're on the root level, in which
   744  		// case the searchAddr should be the same as i after levelShift.
   745  		j0 := 0
   746  		if searchIdx := offAddrToLevelIndex(l, p.searchAddr); searchIdx&^(entriesPerBlock-1) == i {
   747  			j0 = searchIdx & (entriesPerBlock - 1)
   748  		}
   749  
   750  		// Run over the level entries looking for
   751  		// a contiguous run of at least npages either
   752  		// within an entry or across entries.
   753  		//
   754  		// base contains the page index (relative to
   755  		// the first entry's first page) of the currently
   756  		// considered run of consecutive pages.
   757  		//
   758  		// size contains the size of the currently considered
   759  		// run of consecutive pages.
   760  		var base, size uint
   761  		for j := j0; j < len(entries); j++ {
   762  			sum := entries[j]
   763  			if sum == 0 {
   764  				// A full entry means we broke any streak and
   765  				// that we should skip it altogether.
   766  				size = 0
   767  				continue
   768  			}
   769  
   770  			// We've encountered a non-zero summary which means
   771  			// free memory, so update firstFree.
   772  			foundFree(levelIndexToOffAddr(l, i+j), (uintptr(1)<<logMaxPages)*pageSize)
   773  
   774  			s := sum.start()
   775  			if size+s >= uint(npages) {
   776  				// If size == 0 we don't have a run yet,
   777  				// which means base isn't valid. So, set
   778  				// base to the first page in this block.
   779  				if size == 0 {
   780  					base = uint(j) << logMaxPages
   781  				}
   782  				// We hit npages; we're done!
   783  				size += s
   784  				break
   785  			}
   786  			if sum.max() >= uint(npages) {
   787  				// The entry itself contains npages contiguous
   788  				// free pages, so continue on the next level
   789  				// to find that run.
   790  				i += j
   791  				lastSumIdx = i
   792  				lastSum = sum
   793  				continue nextLevel
   794  			}
   795  			if size == 0 || s < 1<<logMaxPages {
   796  				// We either don't have a current run started, or this entry
   797  				// isn't totally free (meaning we can't continue the current
   798  				// one), so try to begin a new run by setting size and base
   799  				// based on sum.end.
   800  				size = sum.end()
   801  				base = uint(j+1)<<logMaxPages - size
   802  				continue
   803  			}
   804  			// The entry is completely free, so continue the run.
   805  			size += 1 << logMaxPages
   806  		}
   807  		if size >= uint(npages) {
   808  			// We found a sufficiently large run of free pages straddling
   809  			// some boundary, so compute the address and return it.
   810  			addr := levelIndexToOffAddr(l, i).add(uintptr(base) * pageSize).addr()
   811  			return addr, p.findMappedAddr(firstFree.base)
   812  		}
   813  		if l == 0 {
   814  			// We're at level zero, so that means we've exhausted our search.
   815  			return 0, maxSearchAddr()
   816  		}
   817  
   818  		// We're not at level zero, and we exhausted the level we were looking in.
   819  		// This means that either our calculations were wrong or the level above
   820  		// lied to us. In either case, dump some useful state and throw.
   821  		print("runtime: summary[", l-1, "][", lastSumIdx, "] = ", lastSum.start(), ", ", lastSum.max(), ", ", lastSum.end(), "\n")
   822  		print("runtime: level = ", l, ", npages = ", npages, ", j0 = ", j0, "\n")
   823  		print("runtime: p.searchAddr = ", hex(p.searchAddr.addr()), ", i = ", i, "\n")
   824  		print("runtime: levelShift[level] = ", levelShift[l], ", levelBits[level] = ", levelBits[l], "\n")
   825  		for j := 0; j < len(entries); j++ {
   826  			sum := entries[j]
   827  			print("runtime: summary[", l, "][", i+j, "] = (", sum.start(), ", ", sum.max(), ", ", sum.end(), ")\n")
   828  		}
   829  		throw("bad summary data")
   830  	}
   831  
   832  	// Since we've gotten to this point, that means we haven't found a
   833  	// sufficiently-sized free region straddling some boundary (chunk or larger).
   834  	// This means the last summary we inspected must have had a large enough "max"
   835  	// value, so look inside the chunk to find a suitable run.
   836  	//
   837  	// After iterating over all levels, i must contain a chunk index which
   838  	// is what the final level represents.
   839  	ci := chunkIdx(i)
   840  	j, searchIdx := p.chunkOf(ci).find(npages, 0)
   841  	if j == ^uint(0) {
   842  		// We couldn't find any space in this chunk despite the summaries telling
   843  		// us it should be there. There's likely a bug, so dump some state and throw.
   844  		sum := p.summary[len(p.summary)-1][i]
   845  		print("runtime: summary[", len(p.summary)-1, "][", i, "] = (", sum.start(), ", ", sum.max(), ", ", sum.end(), ")\n")
   846  		print("runtime: npages = ", npages, "\n")
   847  		throw("bad summary data")
   848  	}
   849  
   850  	// Compute the address at which the free space starts.
   851  	addr := chunkBase(ci) + uintptr(j)*pageSize
   852  
   853  	// Since we actually searched the chunk, we may have
   854  	// found an even narrower free window.
   855  	searchAddr := chunkBase(ci) + uintptr(searchIdx)*pageSize
   856  	foundFree(offAddr{searchAddr}, chunkBase(ci+1)-searchAddr)
   857  	return addr, p.findMappedAddr(firstFree.base)
   858  }
   859  
   860  // alloc allocates npages worth of memory from the page heap, returning the base
   861  // address for the allocation and the amount of scavenged memory in bytes
   862  // contained in the region [base address, base address + npages*pageSize).
   863  //
   864  // Returns a 0 base address on failure, in which case other returned values
   865  // should be ignored.
   866  //
   867  // p.mheapLock must be held.
   868  //
   869  // Must run on the system stack because p.mheapLock must be held.
   870  //
   871  //go:systemstack
   872  func (p *pageAlloc) alloc(npages uintptr) (addr uintptr, scav uintptr) {
   873  	assertLockHeld(p.mheapLock)
   874  
   875  	// If the searchAddr refers to a region which has a higher address than
   876  	// any known chunk, then we know we're out of memory.
   877  	if chunkIndex(p.searchAddr.addr()) >= p.end {
   878  		return 0, 0
   879  	}
   880  
   881  	// If npages has a chance of fitting in the chunk where the searchAddr is,
   882  	// search it directly.
   883  	searchAddr := minOffAddr
   884  	if pallocChunkPages-chunkPageIndex(p.searchAddr.addr()) >= uint(npages) {
   885  		// npages is guaranteed to be no greater than pallocChunkPages here.
   886  		i := chunkIndex(p.searchAddr.addr())
   887  		if max := p.summary[len(p.summary)-1][i].max(); max >= uint(npages) {
   888  			j, searchIdx := p.chunkOf(i).find(npages, chunkPageIndex(p.searchAddr.addr()))
   889  			if j == ^uint(0) {
   890  				print("runtime: max = ", max, ", npages = ", npages, "\n")
   891  				print("runtime: searchIdx = ", chunkPageIndex(p.searchAddr.addr()), ", p.searchAddr = ", hex(p.searchAddr.addr()), "\n")
   892  				throw("bad summary data")
   893  			}
   894  			addr = chunkBase(i) + uintptr(j)*pageSize
   895  			searchAddr = offAddr{chunkBase(i) + uintptr(searchIdx)*pageSize}
   896  			goto Found
   897  		}
   898  	}
   899  	// We failed to use a searchAddr for one reason or another, so try
   900  	// the slow path.
   901  	addr, searchAddr = p.find(npages)
   902  	if addr == 0 {
   903  		if npages == 1 {
   904  			// We failed to find a single free page, the smallest unit
   905  			// of allocation. This means we know the heap is completely
   906  			// exhausted. Otherwise, the heap still might have free
   907  			// space in it, just not enough contiguous space to
   908  			// accommodate npages.
   909  			p.searchAddr = maxSearchAddr()
   910  		}
   911  		return 0, 0
   912  	}
   913  Found:
   914  	// Go ahead and actually mark the bits now that we have an address.
   915  	scav = p.allocRange(addr, npages)
   916  
   917  	// If we found a higher searchAddr, we know that all the
   918  	// heap memory before that searchAddr in an offset address space is
   919  	// allocated, so bump p.searchAddr up to the new one.
   920  	if p.searchAddr.lessThan(searchAddr) {
   921  		p.searchAddr = searchAddr
   922  	}
   923  	return addr, scav
   924  }
   925  
   926  // free returns npages worth of memory starting at base back to the page heap.
   927  //
   928  // p.mheapLock must be held.
   929  //
   930  // Must run on the system stack because p.mheapLock must be held.
   931  //
   932  //go:systemstack
   933  func (p *pageAlloc) free(base, npages uintptr) {
   934  	assertLockHeld(p.mheapLock)
   935  
   936  	// If we're freeing pages below the p.searchAddr, update searchAddr.
   937  	if b := (offAddr{base}); b.lessThan(p.searchAddr) {
   938  		p.searchAddr = b
   939  	}
   940  	limit := base + npages*pageSize - 1
   941  	if npages == 1 {
   942  		// Fast path: we're clearing a single bit, and we know exactly
   943  		// where it is, so mark it directly.
   944  		i := chunkIndex(base)
   945  		pi := chunkPageIndex(base)
   946  		p.chunkOf(i).free1(pi)
   947  		p.scav.index.free(i, pi, 1)
   948  	} else {
   949  		// Slow path: we're clearing more bits so we may need to iterate.
   950  		sc, ec := chunkIndex(base), chunkIndex(limit)
   951  		si, ei := chunkPageIndex(base), chunkPageIndex(limit)
   952  
   953  		if sc == ec {
   954  			// The range doesn't cross any chunk boundaries.
   955  			p.chunkOf(sc).free(si, ei+1-si)
   956  			p.scav.index.free(sc, si, ei+1-si)
   957  		} else {
   958  			// The range crosses at least one chunk boundary.
   959  			p.chunkOf(sc).free(si, pallocChunkPages-si)
   960  			p.scav.index.free(sc, si, pallocChunkPages-si)
   961  			for c := sc + 1; c < ec; c++ {
   962  				p.chunkOf(c).freeAll()
   963  				p.scav.index.free(c, 0, pallocChunkPages)
   964  			}
   965  			p.chunkOf(ec).free(0, ei+1)
   966  			p.scav.index.free(ec, 0, ei+1)
   967  		}
   968  	}
   969  	p.update(base, npages, true, false)
   970  }
   971  
   972  const (
   973  	pallocSumBytes = unsafe.Sizeof(pallocSum(0))
   974  
   975  	// maxPackedValue is the maximum value that any of the three fields in
   976  	// the pallocSum may take on.
   977  	maxPackedValue    = 1 << logMaxPackedValue
   978  	logMaxPackedValue = logPallocChunkPages + (summaryLevels-1)*summaryLevelBits
   979  
   980  	freeChunkSum = pallocSum(uint64(pallocChunkPages) |
   981  		uint64(pallocChunkPages<<logMaxPackedValue) |
   982  		uint64(pallocChunkPages<<(2*logMaxPackedValue)))
   983  )
   984  
   985  // pallocSum is a packed summary type which packs three numbers: start, max,
   986  // and end into a single 8-byte value. Each of these values are a summary of
   987  // a bitmap and are thus counts, each of which may have a maximum value of
   988  // 2^21 - 1, or all three may be equal to 2^21. The latter case is represented
   989  // by just setting the 64th bit.
   990  type pallocSum uint64
   991  
   992  // packPallocSum takes a start, max, and end value and produces a pallocSum.
   993  func packPallocSum(start, max, end uint) pallocSum {
   994  	if max == maxPackedValue {
   995  		return pallocSum(uint64(1 << 63))
   996  	}
   997  	return pallocSum((uint64(start) & (maxPackedValue - 1)) |
   998  		((uint64(max) & (maxPackedValue - 1)) << logMaxPackedValue) |
   999  		((uint64(end) & (maxPackedValue - 1)) << (2 * logMaxPackedValue)))
  1000  }
  1001  
  1002  // start extracts the start value from a packed sum.
  1003  func (p pallocSum) start() uint {
  1004  	if uint64(p)&uint64(1<<63) != 0 {
  1005  		return maxPackedValue
  1006  	}
  1007  	return uint(uint64(p) & (maxPackedValue - 1))
  1008  }
  1009  
  1010  // max extracts the max value from a packed sum.
  1011  func (p pallocSum) max() uint {
  1012  	if uint64(p)&uint64(1<<63) != 0 {
  1013  		return maxPackedValue
  1014  	}
  1015  	return uint((uint64(p) >> logMaxPackedValue) & (maxPackedValue - 1))
  1016  }
  1017  
  1018  // end extracts the end value from a packed sum.
  1019  func (p pallocSum) end() uint {
  1020  	if uint64(p)&uint64(1<<63) != 0 {
  1021  		return maxPackedValue
  1022  	}
  1023  	return uint((uint64(p) >> (2 * logMaxPackedValue)) & (maxPackedValue - 1))
  1024  }
  1025  
  1026  // unpack unpacks all three values from the summary.
  1027  func (p pallocSum) unpack() (uint, uint, uint) {
  1028  	if uint64(p)&uint64(1<<63) != 0 {
  1029  		return maxPackedValue, maxPackedValue, maxPackedValue
  1030  	}
  1031  	return uint(uint64(p) & (maxPackedValue - 1)),
  1032  		uint((uint64(p) >> logMaxPackedValue) & (maxPackedValue - 1)),
  1033  		uint((uint64(p) >> (2 * logMaxPackedValue)) & (maxPackedValue - 1))
  1034  }
  1035  
  1036  // mergeSummaries merges consecutive summaries which may each represent at
  1037  // most 1 << logMaxPagesPerSum pages each together into one.
  1038  func mergeSummaries(sums []pallocSum, logMaxPagesPerSum uint) pallocSum {
  1039  	// Merge the summaries in sums into one.
  1040  	//
  1041  	// We do this by keeping a running summary representing the merged
  1042  	// summaries of sums[:i] in start, most, and end.
  1043  	start, most, end := sums[0].unpack()
  1044  	for i := 1; i < len(sums); i++ {
  1045  		// Merge in sums[i].
  1046  		si, mi, ei := sums[i].unpack()
  1047  
  1048  		// Merge in sums[i].start only if the running summary is
  1049  		// completely free, otherwise this summary's start
  1050  		// plays no role in the combined sum.
  1051  		if start == uint(i)<<logMaxPagesPerSum {
  1052  			start += si
  1053  		}
  1054  
  1055  		// Recompute the max value of the running sum by looking
  1056  		// across the boundary between the running sum and sums[i]
  1057  		// and at the max sums[i], taking the greatest of those two
  1058  		// and the max of the running sum.
  1059  		most = max(most, end+si, mi)
  1060  
  1061  		// Merge in end by checking if this new summary is totally
  1062  		// free. If it is, then we want to extend the running sum's
  1063  		// end by the new summary. If not, then we have some alloc'd
  1064  		// pages in there and we just want to take the end value in
  1065  		// sums[i].
  1066  		if ei == 1<<logMaxPagesPerSum {
  1067  			end += 1 << logMaxPagesPerSum
  1068  		} else {
  1069  			end = ei
  1070  		}
  1071  	}
  1072  	return packPallocSum(start, most, end)
  1073  }
  1074  

View as plain text