runtime: remove non-reserved heap logic

Currently large sysReserve calls on some OSes don't actually reserve
the memory, but just check that it can be reserved. This was important
when we called sysReserve to "reserve" many gigabytes for the heap up
front, but now that we map memory in small increments as we need it,
this complication is no longer necessary.

This has one curious side benefit: currently, on Linux, allocations
that are large enough to be rejected by mmap wind up freezing the
application for a long time before it panics. This happens because
sysReserve doesn't reserve the memory, so sysMap calls mmap_fixed,
which calls mmap, which fails because the mapping is too large.
However, mmap_fixed doesn't inspect *why* mmap fails, so it falls back
to probing every page in the desired region individually with mincore
before performing an (otherwise dangerous) MAP_FIXED mapping, which
will also fail. This takes a long time for a large region. Now this
logic is gone, so the mmap failure leads to an immediate panic.

Updates #10460.

Change-Id: I8efe88c611871cdb14f99fadd09db83e0161ca2e
Reviewed-on: https://go-review.googlesource.com/85888
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rick Hudson <rlh@golang.org>

src/runtime/export_test.go

@@ -437,8 +437,7 @@ func MapNextArenaHint() (start, end uintptr) {
 	} else {
 		start, end = addr, addr+heapArenaBytes
 	}
-	var reserved bool
-	sysReserve(unsafe.Pointer(addr), physPageSize, &reserved)
+	sysReserve(unsafe.Pointer(addr), physPageSize)
 	return
 }
 

src/runtime/malloc.go

@@ -235,18 +235,12 @@ var physPageSize uintptr
 // SysReserve reserves address space without allocating memory.
 // If the pointer passed to it is non-nil, the caller wants the
 // reservation there, but SysReserve can still choose another
-// location if that one is unavailable. On some systems and in some
-// cases SysReserve will simply check that the address space is
-// available and not actually reserve it. If SysReserve returns
-// non-nil, it sets *reserved to true if the address space is
-// reserved, false if it has merely been checked.
+// location if that one is unavailable.
 // NOTE: SysReserve returns OS-aligned memory, but the heap allocator
 // may use larger alignment, so the caller must be careful to realign the
 // memory obtained by sysAlloc.
 //
 // SysMap maps previously reserved address space for use.
-// The reserved argument is true if the address space was really
-// reserved, not merely checked.
 //
 // SysFault marks a (already sysAlloc'd) region to fault
 // if accessed. Used only for debugging the runtime.
@@ -361,8 +355,7 @@ func mallocinit() {
 		// heap reservation.
 
 		const arenaMetaSize = unsafe.Sizeof(heapArena{}) * uintptr(len(*mheap_.arenas))
-		var reserved bool
-		meta := uintptr(sysReserve(nil, arenaMetaSize, &reserved))
+		meta := uintptr(sysReserve(nil, arenaMetaSize))
 		if meta != 0 {
 			mheap_.heapArenaAlloc.init(meta, arenaMetaSize)
 		}
@@ -399,7 +392,7 @@ func mallocinit() {
 			128 << 20,
 		}
 		for _, arenaSize := range arenaSizes {
-			a, size := sysReserveAligned(unsafe.Pointer(p), arenaSize, heapArenaBytes, &reserved)
+			a, size := sysReserveAligned(unsafe.Pointer(p), arenaSize, heapArenaBytes)
 			if a != nil {
 				mheap_.arena.init(uintptr(a), size)
 				p = uintptr(a) + size // For hint below
@@ -440,7 +433,7 @@ func (h *mheap) sysAlloc(n uintptr) (v unsafe.Pointer, size uintptr) {
 			// We can't use this, so don't ask.
 			v = nil
 		} else {
-			v = sysReserve(unsafe.Pointer(p), n, &h.arena_reserved)
+			v = sysReserve(unsafe.Pointer(p), n)
 		}
 		if p == uintptr(v) {
 			// Success. Update the hint.
@@ -468,7 +461,7 @@ func (h *mheap) sysAlloc(n uintptr) (v unsafe.Pointer, size uintptr) {
 		// All of the hints failed, so we'll take any
 		// (sufficiently aligned) address the kernel will give
 		// us.
-		v, size = sysReserveAligned(nil, n, heapArenaBytes, &h.arena_reserved)
+		v, size = sysReserveAligned(nil, n, heapArenaBytes)
 		if v == nil {
 			return nil, 0
 		}
@@ -494,7 +487,7 @@ func (h *mheap) sysAlloc(n uintptr) (v unsafe.Pointer, size uintptr) {
 	}
 
 	// Back the reservation.
-	sysMap(v, size, h.arena_reserved, &memstats.heap_sys)
+	sysMap(v, size, &memstats.heap_sys)
 
 mapped:
 	// Create arena metadata.
@@ -529,13 +522,13 @@ mapped:
 // sysReserveAligned is like sysReserve, but the returned pointer is
 // aligned to align bytes. It may reserve either n or n+align bytes,
 // so it returns the size that was reserved.
-func sysReserveAligned(v unsafe.Pointer, size, align uintptr, reserved *bool) (unsafe.Pointer, uintptr) {
+func sysReserveAligned(v unsafe.Pointer, size, align uintptr) (unsafe.Pointer, uintptr) {
 	// Since the alignment is rather large in uses of this
 	// function, we're not likely to get it by chance, so we ask
 	// for a larger region and remove the parts we don't need.
 	retries := 0
 retry:
-	p := uintptr(sysReserve(v, size+align, reserved))
+	p := uintptr(sysReserve(v, size+align))
 	switch {
 	case p == 0:
 		return nil, 0
@@ -550,7 +543,7 @@ retry:
 		// so we may have to try again.
 		sysFree(unsafe.Pointer(p), size+align, nil)
 		p = round(p, align)
-		p2 := sysReserve(unsafe.Pointer(p), size, reserved)
+		p2 := sysReserve(unsafe.Pointer(p), size)
 		if p != uintptr(p2) {
 			// Must have raced. Try again.
 			sysFree(p2, size, nil)
@@ -1095,7 +1088,7 @@ func (l *linearAlloc) alloc(size, align uintptr, sysStat *uint64) unsafe.Pointer
 	l.next = p + size
 	if pEnd := round(l.next-1, physPageSize); pEnd > l.mapped {
 		// We need to map more of the reserved space.
-		sysMap(unsafe.Pointer(l.mapped), pEnd-l.mapped, true, sysStat)
+		sysMap(unsafe.Pointer(l.mapped), pEnd-l.mapped, sysStat)
 		l.mapped = pEnd
 	}
 	return unsafe.Pointer(p)

src/runtime/mem_bsd.go

@@ -7,7 +7,6 @@
 package runtime
 
 import (
-	"runtime/internal/sys"
 	"unsafe"
 )
 
@@ -42,51 +41,20 @@ func sysFault(v unsafe.Pointer, n uintptr) {
 	mmap(v, n, _PROT_NONE, _MAP_ANON|_MAP_PRIVATE|_MAP_FIXED, -1, 0)
 }
 
-func sysReserve(v unsafe.Pointer, n uintptr, reserved *bool) unsafe.Pointer {
-	// On 64-bit, people with ulimit -v set complain if we reserve too
-	// much address space. Instead, assume that the reservation is okay
-	// and check the assumption in SysMap.
-	if sys.PtrSize == 8 && uint64(n) > 1<<32 || sys.GoosNacl != 0 {
-		*reserved = false
-		return v
-	}
-
+func sysReserve(v unsafe.Pointer, n uintptr) unsafe.Pointer {
 	p, err := mmap(v, n, _PROT_NONE, _MAP_ANON|_MAP_PRIVATE, -1, 0)
 	if err != 0 {
 		return nil
 	}
-	*reserved = true
 	return p
 }
 
 const _sunosEAGAIN = 11
 const _ENOMEM = 12
 
-func sysMap(v unsafe.Pointer, n uintptr, reserved bool, sysStat *uint64) {
+func sysMap(v unsafe.Pointer, n uintptr, sysStat *uint64) {
 	mSysStatInc(sysStat, n)
 
-	// On 64-bit, we don't actually have v reserved, so tread carefully.
-	if !reserved {
-		flags := int32(_MAP_ANON | _MAP_PRIVATE)
-		if GOOS == "dragonfly" {
-			// TODO(jsing): For some reason DragonFly seems to return
-			// memory at a different address than we requested, even when
-			// there should be no reason for it to do so. This can be
-			// avoided by using MAP_FIXED, but I'm not sure we should need
-			// to do this - we do not on other platforms.
-			flags |= _MAP_FIXED
-		}
-		p, err := mmap(v, n, _PROT_READ|_PROT_WRITE, flags, -1, 0)
-		if err == _ENOMEM || (GOOS == "solaris" && err == _sunosEAGAIN) {
-			throw("runtime: out of memory")
-		}
-		if p != v || err != 0 {
-			print("runtime: address space conflict: map(", v, ") = ", p, "(err ", err, ")\n")
-			throw("runtime: address space conflict")
-		}
-		return
-	}
-
 	p, err := mmap(v, n, _PROT_READ|_PROT_WRITE, _MAP_ANON|_MAP_FIXED|_MAP_PRIVATE, -1, 0)
 	if err == _ENOMEM || (GOOS == "solaris" && err == _sunosEAGAIN) {
 		throw("runtime: out of memory")

src/runtime/mem_darwin.go

@@ -38,8 +38,7 @@ func sysFault(v unsafe.Pointer, n uintptr) {
 	mmap(v, n, _PROT_NONE, _MAP_ANON|_MAP_PRIVATE|_MAP_FIXED, -1, 0)
 }
 
-func sysReserve(v unsafe.Pointer, n uintptr, reserved *bool) unsafe.Pointer {
-	*reserved = true
+func sysReserve(v unsafe.Pointer, n uintptr) unsafe.Pointer {
 	p, err := mmap(v, n, _PROT_NONE, _MAP_ANON|_MAP_PRIVATE, -1, 0)
 	if err != 0 {
 		return nil
@@ -51,7 +50,7 @@ const (
 	_ENOMEM = 12
 )
 
-func sysMap(v unsafe.Pointer, n uintptr, reserved bool, sysStat *uint64) {
+func sysMap(v unsafe.Pointer, n uintptr, sysStat *uint64) {
 	mSysStatInc(sysStat, n)
 	p, err := mmap(v, n, _PROT_READ|_PROT_WRITE, _MAP_ANON|_MAP_FIXED|_MAP_PRIVATE, -1, 0)
 	if err == _ENOMEM {

src/runtime/mem_linux.go

@@ -14,46 +14,6 @@ const (
 	_EINVAL = 22
 )
 
-// NOTE: vec must be just 1 byte long here.
-// Mincore returns ENOMEM if any of the pages are unmapped,
-// but we want to know that all of the pages are unmapped.
-// To make these the same, we can only ask about one page
-// at a time. See golang.org/issue/7476.
-var addrspace_vec [1]byte
-
-func addrspace_free(v unsafe.Pointer, n uintptr) bool {
-	for off := uintptr(0); off < n; off += physPageSize {
-		// Use a length of 1 byte, which the kernel will round
-		// up to one physical page regardless of the true
-		// physical page size.
-		errval := mincore(unsafe.Pointer(uintptr(v)+off), 1, &addrspace_vec[0])
-		if errval == -_EINVAL {
-			// Address is not a multiple of the physical
-			// page size. Shouldn't happen, but just ignore it.
-			continue
-		}
-		// ENOMEM means unmapped, which is what we want.
-		// Anything else we assume means the pages are mapped.
-		if errval != -_ENOMEM {
-			return false
-		}
-	}
-	return true
-}
-
-func mmap_fixed(v unsafe.Pointer, n uintptr, prot, flags, fd int32, offset uint32) (unsafe.Pointer, int) {
-	p, err := mmap(v, n, prot, flags, fd, offset)
-	// On some systems, mmap ignores v without
-	// MAP_FIXED, so retry if the address space is free.
-	if p != v && addrspace_free(v, n) {
-		if err == 0 {
-			munmap(p, n)
-		}
-		p, err = mmap(v, n, prot, flags|_MAP_FIXED, fd, offset)
-	}
-	return p, err
-}
-
 // Don't split the stack as this method may be invoked without a valid G, which
 // prevents us from allocating more stack.
 //go:nosplit
@@ -180,48 +140,17 @@ func sysFault(v unsafe.Pointer, n uintptr) {
 	mmap(v, n, _PROT_NONE, _MAP_ANON|_MAP_PRIVATE|_MAP_FIXED, -1, 0)
 }
 
-func sysReserve(v unsafe.Pointer, n uintptr, reserved *bool) unsafe.Pointer {
-	// On 64-bit, people with ulimit -v set complain if we reserve too
-	// much address space. Instead, assume that the reservation is okay
-	// if we can reserve at least 64K and check the assumption in SysMap.
-	// Only user-mode Linux (UML) rejects these requests.
-	if sys.PtrSize == 8 && uint64(n) > 1<<32 {
-		p, err := mmap_fixed(v, 64<<10, _PROT_NONE, _MAP_ANON|_MAP_PRIVATE, -1, 0)
-		if p != v || err != 0 {
-			if err == 0 {
-				munmap(p, 64<<10)
-			}
-			return nil
-		}
-		munmap(p, 64<<10)
-		*reserved = false
-		return v
-	}
-
+func sysReserve(v unsafe.Pointer, n uintptr) unsafe.Pointer {
 	p, err := mmap(v, n, _PROT_NONE, _MAP_ANON|_MAP_PRIVATE, -1, 0)
 	if err != 0 {
 		return nil
 	}
-	*reserved = true
 	return p
 }
 
-func sysMap(v unsafe.Pointer, n uintptr, reserved bool, sysStat *uint64) {
+func sysMap(v unsafe.Pointer, n uintptr, sysStat *uint64) {
 	mSysStatInc(sysStat, n)
 
-	// On 64-bit, we don't actually have v reserved, so tread carefully.
-	if !reserved {
-		p, err := mmap_fixed(v, n, _PROT_READ|_PROT_WRITE, _MAP_ANON|_MAP_PRIVATE, -1, 0)
-		if err == _ENOMEM {
-			throw("runtime: out of memory")
-		}
-		if p != v || err != 0 {
-			print("runtime: address space conflict: map(", v, ") = ", p, " (err ", err, ")\n")
-			throw("runtime: address space conflict")
-		}
-		return
-	}
-
 	p, err := mmap(v, n, _PROT_READ|_PROT_WRITE, _MAP_ANON|_MAP_FIXED|_MAP_PRIVATE, -1, 0)
 	if err == _ENOMEM {
 		throw("runtime: out of memory")

src/runtime/mem_plan9.go

@@ -160,7 +160,7 @@ func sysUnused(v unsafe.Pointer, n uintptr) {
 func sysUsed(v unsafe.Pointer, n uintptr) {
 }
 
-func sysMap(v unsafe.Pointer, n uintptr, reserved bool, sysStat *uint64) {
+func sysMap(v unsafe.Pointer, n uintptr, sysStat *uint64) {
 	// sysReserve has already allocated all heap memory,
 	// but has not adjusted stats.
 	mSysStatInc(sysStat, n)
@@ -169,8 +169,7 @@ func sysMap(v unsafe.Pointer, n uintptr, reserved bool, sysStat *uint64) {
 func sysFault(v unsafe.Pointer, n uintptr) {
 }
 
-func sysReserve(v unsafe.Pointer, n uintptr, reserved *bool) unsafe.Pointer {
-	*reserved = true
+func sysReserve(v unsafe.Pointer, n uintptr) unsafe.Pointer {
 	lock(&memlock)
 	p := memAlloc(n)
 	memCheck()

src/runtime/mem_windows.go

@@ -98,8 +98,7 @@ func sysFault(v unsafe.Pointer, n uintptr) {
 	sysUnused(v, n)
 }
 
-func sysReserve(v unsafe.Pointer, n uintptr, reserved *bool) unsafe.Pointer {
-	*reserved = true
+func sysReserve(v unsafe.Pointer, n uintptr) unsafe.Pointer {
 	// v is just a hint.
 	// First try at v.
 	// This will fail if any of [v, v+n) is already reserved.
@@ -112,7 +111,7 @@ func sysReserve(v unsafe.Pointer, n uintptr, reserved *bool) unsafe.Pointer {
 	return unsafe.Pointer(stdcall4(_VirtualAlloc, 0, n, _MEM_RESERVE, _PAGE_READWRITE))
 }
 
-func sysMap(v unsafe.Pointer, n uintptr, reserved bool, sysStat *uint64) {
+func sysMap(v unsafe.Pointer, n uintptr, sysStat *uint64) {
 	mSysStatInc(sysStat, n)
 	p := stdcall4(_VirtualAlloc, uintptr(v), n, _MEM_COMMIT, _PAGE_READWRITE)
 	if p != uintptr(v) {

src/runtime/mheap.go

@@ -96,15 +96,6 @@ type mheap struct {
 	nlargefree  uint64                  // number of frees for large objects (>maxsmallsize)
 	nsmallfree  [_NumSizeClasses]uint64 // number of frees for small objects (<=maxsmallsize)
 
-	// arena_reserved indicates that the memory [arena_alloc,
-	// arena_end) is reserved (e.g., mapped PROT_NONE). If this is
-	// false, we have to be careful not to clobber existing
-	// mappings here. If this is true, then we own the mapping
-	// here and *must* clobber it to use it.
-	//
-	// TODO(austin): Remove.
-	arena_reserved bool
-
 	// arenas is the heap arena index. arenas[va/heapArenaBytes]
 	// points to the metadata for the heap arena containing va.
 	//
@@ -135,7 +126,7 @@ type mheap struct {
 	// (the actual arenas). This is only used on 32-bit.
 	arena linearAlloc
 
-	_ uint32 // ensure 64-bit alignment of central
+	//_ uint32 // ensure 64-bit alignment of central
 
 	// central free lists for small size classes.
 	// the padding makes sure that the MCentrals are

src/runtime/os_linux.go

@@ -194,6 +194,8 @@ const (
 
 var procAuxv = []byte("/proc/self/auxv\x00")
 
+var addrspace_vec [1]byte
+
 func mincore(addr unsafe.Pointer, n uintptr, dst *byte) int32
 
 func sysargs(argc int32, argv **byte) {