system/ulib/minfs/writeback.cpp - zircon/ - Git at Google

 // Copyright 2017 The Fuchsia Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.

 #include <inttypes.h>

 #ifdef __Fuchsia__
 #include <fbl/auto_lock.h>
 #include <fbl/mutex.h>
 #include <fbl/vector.h>
 #include <lib/fzl/owned-vmo-mapper.h>
 #include <lib/zx/vmo.h>
 #endif

 #include <fbl/algorithm.h>
 #include <fbl/intrusive_hash_table.h>
 #include <fbl/intrusive_single_list.h>
 #include <fbl/macros.h>
 #include <fbl/ref_ptr.h>
 #include <fbl/unique_ptr.h>
 #include <fs/block-txn.h>
 #include <fs/vfs.h>

 #include "minfs-private.h"
 #include <minfs/writeback.h>

 #include <utility>

 namespace minfs {

 #ifdef __Fuchsia__

 void WriteTxn::Enqueue(zx_handle_t vmo, uint64_t vmo_offset, uint64_t dev_offset,
                        uint64_t nblocks) {
     ValidateVmoSize(vmo, static_cast<blk_t>(vmo_offset));
     for (size_t i = 0; i < requests_.size(); i++) {
         if (requests_[i].vmo != vmo) {
             continue;
         }

         if (requests_[i].vmo_offset == vmo_offset) {
             // Take the longer of the operations (if operating on the same
             // blocks).
             requests_[i].length = (requests_[i].length > nblocks) ? requests_[i].length : nblocks;
             return;
         } else if ((requests_[i].vmo_offset + requests_[i].length == vmo_offset) &&
                    (requests_[i].dev_offset + requests_[i].length == dev_offset)) {
             // Combine with the previous request, if immediately following.
             requests_[i].length += nblocks;
             return;
         }
     }

     WriteRequest request;
     request.vmo = vmo;
     // NOTE: It's easier to compare everything when dealing
     // with blocks (not offsets!) so the following are described in
     // terms of blocks until we Flush().
     request.vmo_offset = vmo_offset;
     request.dev_offset = dev_offset;
     request.length = nblocks;
     requests_.push_back(std::move(request));
 }

 zx_status_t WriteTxn::Flush(zx_handle_t vmo, vmoid_t vmoid) {
     ZX_DEBUG_ASSERT(vmo != ZX_HANDLE_INVALID);
     ZX_DEBUG_ASSERT(vmoid != VMOID_INVALID);

     // Update all the outgoing transactions to be in "disk blocks",
     // not "Minfs blocks".
     block_fifo_request_t blk_reqs[requests_.size()];
     const uint32_t kDiskBlocksPerMinfsBlock = kMinfsBlockSize / bc_->DeviceBlockSize();
     for (size_t i = 0; i < requests_.size(); i++) {
         blk_reqs[i].group = bc_->BlockGroupID();
         blk_reqs[i].vmoid = vmoid;
         blk_reqs[i].opcode = BLOCKIO_WRITE;
         blk_reqs[i].vmo_offset = requests_[i].vmo_offset * kDiskBlocksPerMinfsBlock;
         blk_reqs[i].dev_offset = requests_[i].dev_offset * kDiskBlocksPerMinfsBlock;
         // TODO(ZX-2253): Remove this assertion.
         uint64_t length = requests_[i].length * kDiskBlocksPerMinfsBlock;
         ZX_ASSERT_MSG(length < UINT32_MAX, "Too many blocks");
         blk_reqs[i].length = static_cast<uint32_t>(length);
     }

     // Actually send the operations to the underlying block device.
     zx_status_t status = bc_->Transaction(blk_reqs, requests_.size());

     requests_.reset();
     return status;
 }

 size_t WriteTxn::BlkCount() const {
     size_t blocks_needed = 0;
     for (size_t i = 0; i < requests_.size(); i++) {
         blocks_needed += requests_[i].length;
     }
     return blocks_needed;
 }

 #endif  // __Fuchsia__

 WritebackWork::WritebackWork(Bcache* bc) : WriteTxn(bc),
 #ifdef __Fuchsia__
     closure_(nullptr),
 #endif
     node_count_(0) {}

 void WritebackWork::Reset() {
 #ifdef __Fuchsia__
     ZX_DEBUG_ASSERT(Requests().size() == 0);
     closure_ = nullptr;
 #endif
     while (0 < node_count_) {
         vn_[--node_count_] = nullptr;
     }
 }

 #ifdef __Fuchsia__
 // Returns the number of blocks of the writeback buffer that have been
 // consumed
 size_t WritebackWork::Complete(zx_handle_t vmo, vmoid_t vmoid) {
     size_t blk_count = BlkCount();
     zx_status_t status = Flush(vmo, vmoid);
     if (closure_) {
         closure_(status);
     }
     Reset();
     return blk_count;
 }

 void WritebackWork::SetClosure(SyncCallback closure) {
     ZX_DEBUG_ASSERT(!closure_);
     closure_ = std::move(closure);
 }
 #else
 void WritebackWork::Complete() {
     Transact();
     Reset();
 }
 #endif  // __Fuchsia__

 // Allow "pinning" Vnodes so they aren't destroyed while we're completing
 // this writeback operation.
 void WritebackWork::PinVnode(fbl::RefPtr<VnodeMinfs> vn) {
     for (size_t i = 0; i < node_count_; i++) {
         if (vn_[i].get() == vn.get()) {
             // Already pinned
             return;
         }
     }
     ZX_DEBUG_ASSERT(node_count_ < fbl::count_of(vn_));
     vn_[node_count_++] = std::move(vn);
 }

 #ifdef __Fuchsia__

 zx_status_t WritebackBuffer::Create(Bcache* bc, fzl::OwnedVmoMapper mapper,
                                     fbl::unique_ptr<WritebackBuffer>* out) {
     fbl::unique_ptr<WritebackBuffer> wb(new WritebackBuffer(bc, std::move(mapper)));
     if (wb->mapper_.size() % kMinfsBlockSize != 0) {
         return ZX_ERR_INVALID_ARGS;
     } else if (cnd_init(&wb->consumer_cvar_) != thrd_success) {
         return ZX_ERR_NO_RESOURCES;
     } else if (cnd_init(&wb->producer_cvar_) != thrd_success) {
         return ZX_ERR_NO_RESOURCES;
     } else if (thrd_create_with_name(&wb->writeback_thrd_,
                                      WritebackBuffer::WritebackThread, wb.get(),
                                      "minfs-writeback") != thrd_success) {
         return ZX_ERR_NO_RESOURCES;
     }
     zx_status_t status = wb->bc_->AttachVmo(wb->mapper_.vmo(), &wb->buffer_vmoid_);
     if (status != ZX_OK) {
         return status;
     }

     *out = std::move(wb);
     return ZX_OK;
 }

 WritebackBuffer::WritebackBuffer(Bcache* bc, fzl::OwnedVmoMapper mapper) :
     bc_(bc), unmounting_(false), mapper_(std::move(mapper)),
     cap_(mapper_.size() / kMinfsBlockSize) {}

 WritebackBuffer::~WritebackBuffer() {
     // Block until the background thread completes itself.
     {
         fbl::AutoLock lock(&writeback_lock_);
         unmounting_ = true;
         cnd_signal(&consumer_cvar_);
     }
     int r;
     thrd_join(writeback_thrd_, &r);

     if (buffer_vmoid_ != VMOID_INVALID) {
         block_fifo_request_t request;
         request.group = bc_->BlockGroupID();
         request.vmoid = buffer_vmoid_;
         request.opcode = BLOCKIO_CLOSE_VMO;
         bc_->Transaction(&request, 1);
     }
 }

 zx_status_t WritebackBuffer::EnsureSpaceLocked(size_t blocks) {
     if (blocks > cap_) {
         // There will never be enough room in the writeback buffer
         // for this request.
         return ZX_ERR_NO_RESOURCES;
     }
     while (len_ + blocks > cap_) {
         // Not enough room to write back work, yet. Wait until
         // room is available.
         Waiter w;
         producer_queue_.push(&w);

         do {
             cnd_wait(&producer_cvar_, writeback_lock_.GetInternal());
         } while ((&producer_queue_.front() != &w) && // We are first in line to enqueue...
                  (len_ + blocks > cap_)); // ... and there is enough space for us.

         producer_queue_.pop();
     }
     return ZX_OK;
 }

 void WritebackBuffer::CopyToBufferLocked(WriteTxn* txn) {
     auto& reqs = txn->Requests();
     // Write back to the buffer
     for (size_t i = 0; i < reqs.size(); i++) {
         size_t vmo_offset = reqs[i].vmo_offset;
         size_t dev_offset = reqs[i].dev_offset;
         const size_t vmo_len = reqs[i].length;
         ZX_DEBUG_ASSERT(vmo_len > 0);
         size_t wb_offset = (start_ + len_) % cap_;
         size_t wb_len = (wb_offset + vmo_len > cap_) ? cap_ - wb_offset : vmo_len;
         ZX_DEBUG_ASSERT(wb_len <= vmo_len);
         ZX_DEBUG_ASSERT(wb_offset < cap_);
         zx_handle_t vmo = reqs[i].vmo;

         void* ptr = (void*)((uintptr_t)(mapper_.start()) +
                             (uintptr_t)(wb_offset * kMinfsBlockSize));
         zx_status_t status;
         ZX_DEBUG_ASSERT((start_ <= wb_offset) ?
                         (start_ < wb_offset + wb_len) :
                         (wb_offset + wb_len <= start_)); // Wraparound
         ZX_ASSERT_MSG((status = zx_vmo_read(vmo, ptr, vmo_offset * kMinfsBlockSize,
                       wb_len * kMinfsBlockSize)) == ZX_OK, "VMO Read Fail: %d", status);
         len_ += wb_len;

         // Update the WriteRequest to transfer from the writeback buffer
         // out to disk, rather than the supplied VMO
         reqs[i].vmo_offset = wb_offset;
         reqs[i].length = wb_len;

         if (wb_len != vmo_len) {
             // We wrapped around; write what remains from this request
             vmo_offset += wb_len;
             dev_offset += wb_len;
             wb_len = vmo_len - wb_len;
             ptr = mapper_.start();
             ZX_DEBUG_ASSERT((start_ == 0) ?  (start_ < wb_len) : (wb_len <= start_)); // Wraparound
             ZX_ASSERT(zx_vmo_read(vmo, ptr, vmo_offset * kMinfsBlockSize,
                                   wb_len * kMinfsBlockSize) == ZX_OK);
             len_ += wb_len;

             // Shift down all following write requests
             static_assert(fbl::is_pod<WriteRequest>::value, "Can't memmove non-POD");

             // Insert the "new" request, which is the latter half of
             // the request we wrote out earlier
             WriteRequest request;
             request.vmo = reqs[i].vmo;
             request.vmo_offset = 0;
             request.dev_offset = dev_offset;
             request.length = wb_len;
             i++;
             reqs.insert(i, request);
         }
     }
 }

 void WritebackBuffer::Enqueue(fbl::unique_ptr<WritebackWork> work) {
     TRACE_DURATION("minfs", "WritebackBuffer::Enqueue");
     TRACE_FLOW_BEGIN("minfs", "writeback", reinterpret_cast<trace_flow_id_t>(work.get()));
     fbl::AutoLock lock(&writeback_lock_);

     {
         TRACE_DURATION("minfs", "Allocating Writeback space");
         size_t blocks = work->BlkCount();
         // TODO(smklein): Experimentally, all filesystem operations cause between
         // 0 and 10 blocks to be updated, though the writeback buffer has space
         // for thousands of blocks.
         //
         // Hypothetically, an operation (most likely, an enormous write) could
         // cause a single operation to exceed the size of the writeback buffer,
         // but this is currently impossible as our writes are broken into 8KB
         // chunks.
         //
         // Regardless, there should either (1) exist a fallback mechanism for these
         // extremely large operations, or (2) the worst-case operation should be
         // calculated, and it should be proven that it will always fit within
         // the allocated writeback buffer.
         ZX_ASSERT_MSG(EnsureSpaceLocked(blocks) == ZX_OK,
                       "Requested txn (%zu blocks) larger than writeback buffer", blocks);
     }

     {
         TRACE_DURATION("minfs", "Copying to Writeback buffer");
         CopyToBufferLocked(work.get());
     }

     work_queue_.push(std::move(work));
     cnd_signal(&consumer_cvar_);
 }

 int WritebackBuffer::WritebackThread(void* arg) {
     WritebackBuffer* b = reinterpret_cast<WritebackBuffer*>(arg);

     b->writeback_lock_.Acquire();
     while (true) {
         while (!b->work_queue_.is_empty()) {
             auto work = b->work_queue_.pop();
             TRACE_DURATION("minfs", "WritebackBuffer::WritebackThread");

             // Stay unlocked while processing a unit of work
             b->writeback_lock_.Release();

             // TODO(smklein): We could add additional validation that the blocks
             // in "work" are contiguous and in the range of [start_, len_) (including
             // wraparound).
             size_t blks_consumed = work->Complete(b->mapper_.vmo().get(), b->buffer_vmoid_);
             TRACE_FLOW_END("minfs", "writeback", reinterpret_cast<trace_flow_id_t>(work.get()));
             work = nullptr;

             // Relock before checking the state of the queue
             b->writeback_lock_.Acquire();
             b->start_ = (b->start_ + blks_consumed) % b->cap_;
             b->len_ -= blks_consumed;
             cnd_signal(&b->producer_cvar_);
         }

         // Before waiting, we should check if we're unmounting.
         if (b->unmounting_) {
             b->writeback_lock_.Release();
             return 0;
         }
         cnd_wait(&b->consumer_cvar_, b->writeback_lock_.GetInternal());
     }
 }

 #endif  // __Fuchsia__

 } // namespace minfs
	// Copyright 2017 The Fuchsia Authors. All rights reserved.
	// Use of this source code is governed by a BSD-style license that can be
	// found in the LICENSE file.

	#include <inttypes.h>

	#ifdef __Fuchsia__
	#include <fbl/auto_lock.h>
	#include <fbl/mutex.h>
	#include <fbl/vector.h>
	#include <lib/fzl/owned-vmo-mapper.h>
	#include <lib/zx/vmo.h>
	#endif

	#include <fbl/algorithm.h>
	#include <fbl/intrusive_hash_table.h>
	#include <fbl/intrusive_single_list.h>
	#include <fbl/macros.h>
	#include <fbl/ref_ptr.h>
	#include <fbl/unique_ptr.h>
	#include <fs/block-txn.h>
	#include <fs/vfs.h>

	#include "minfs-private.h"
	#include <minfs/writeback.h>

	#include <utility>

	namespace minfs {

	#ifdef __Fuchsia__

	void WriteTxn::Enqueue(zx_handle_t vmo, uint64_t vmo_offset, uint64_t dev_offset,
	uint64_t nblocks) {
	ValidateVmoSize(vmo, static_cast<blk_t>(vmo_offset));
	for (size_t i = 0; i < requests_.size(); i++) {
	if (requests_[i].vmo != vmo) {
	continue;
	}

	if (requests_[i].vmo_offset == vmo_offset) {
	// Take the longer of the operations (if operating on the same
	// blocks).
	requests_[i].length = (requests_[i].length > nblocks) ? requests_[i].length : nblocks;
	return;
	} else if ((requests_[i].vmo_offset + requests_[i].length == vmo_offset) &&
	(requests_[i].dev_offset + requests_[i].length == dev_offset)) {
	// Combine with the previous request, if immediately following.
	requests_[i].length += nblocks;
	return;
	}
	}

	WriteRequest request;
	request.vmo = vmo;
	// NOTE: It's easier to compare everything when dealing
	// with blocks (not offsets!) so the following are described in
	// terms of blocks until we Flush().
	request.vmo_offset = vmo_offset;
	request.dev_offset = dev_offset;
	request.length = nblocks;
	requests_.push_back(std::move(request));
	}

	zx_status_t WriteTxn::Flush(zx_handle_t vmo, vmoid_t vmoid) {
	ZX_DEBUG_ASSERT(vmo != ZX_HANDLE_INVALID);
	ZX_DEBUG_ASSERT(vmoid != VMOID_INVALID);

	// Update all the outgoing transactions to be in "disk blocks",
	// not "Minfs blocks".
	block_fifo_request_t blk_reqs[requests_.size()];
	const uint32_t kDiskBlocksPerMinfsBlock = kMinfsBlockSize / bc_->DeviceBlockSize();
	for (size_t i = 0; i < requests_.size(); i++) {
	blk_reqs[i].group = bc_->BlockGroupID();
	blk_reqs[i].vmoid = vmoid;
	blk_reqs[i].opcode = BLOCKIO_WRITE;
	blk_reqs[i].vmo_offset = requests_[i].vmo_offset * kDiskBlocksPerMinfsBlock;
	blk_reqs[i].dev_offset = requests_[i].dev_offset * kDiskBlocksPerMinfsBlock;
	// TODO(ZX-2253): Remove this assertion.
	uint64_t length = requests_[i].length * kDiskBlocksPerMinfsBlock;
	ZX_ASSERT_MSG(length < UINT32_MAX, "Too many blocks");
	blk_reqs[i].length = static_cast<uint32_t>(length);
	}

	// Actually send the operations to the underlying block device.
	zx_status_t status = bc_->Transaction(blk_reqs, requests_.size());

	requests_.reset();
	return status;
	}

	size_t WriteTxn::BlkCount() const {
	size_t blocks_needed = 0;
	for (size_t i = 0; i < requests_.size(); i++) {
	blocks_needed += requests_[i].length;
	}
	return blocks_needed;
	}

	#endif // __Fuchsia__

	WritebackWork::WritebackWork(Bcache* bc) : WriteTxn(bc),
	#ifdef __Fuchsia__
	closure_(nullptr),
	#endif
	node_count_(0) {}

	void WritebackWork::Reset() {
	#ifdef __Fuchsia__
	ZX_DEBUG_ASSERT(Requests().size() == 0);
	closure_ = nullptr;
	#endif
	while (0 < node_count_) {
	vn_[--node_count_] = nullptr;
	}
	}

	#ifdef __Fuchsia__
	// Returns the number of blocks of the writeback buffer that have been
	// consumed
	size_t WritebackWork::Complete(zx_handle_t vmo, vmoid_t vmoid) {
	size_t blk_count = BlkCount();
	zx_status_t status = Flush(vmo, vmoid);
	if (closure_) {
	closure_(status);
	}
	Reset();
	return blk_count;
	}

	void WritebackWork::SetClosure(SyncCallback closure) {
	ZX_DEBUG_ASSERT(!closure_);
	closure_ = std::move(closure);
	}
	#else
	void WritebackWork::Complete() {
	Transact();
	Reset();
	}
	#endif // __Fuchsia__

	// Allow "pinning" Vnodes so they aren't destroyed while we're completing
	// this writeback operation.
	void WritebackWork::PinVnode(fbl::RefPtr<VnodeMinfs> vn) {
	for (size_t i = 0; i < node_count_; i++) {
	if (vn_[i].get() == vn.get()) {
	// Already pinned
	return;
	}
	}
	ZX_DEBUG_ASSERT(node_count_ < fbl::count_of(vn_));
	vn_[node_count_++] = std::move(vn);
	}

	#ifdef __Fuchsia__

	zx_status_t WritebackBuffer::Create(Bcache* bc, fzl::OwnedVmoMapper mapper,
	fbl::unique_ptr<WritebackBuffer>* out) {
	fbl::unique_ptr<WritebackBuffer> wb(new WritebackBuffer(bc, std::move(mapper)));
	if (wb->mapper_.size() % kMinfsBlockSize != 0) {
	return ZX_ERR_INVALID_ARGS;
	} else if (cnd_init(&wb->consumer_cvar_) != thrd_success) {
	return ZX_ERR_NO_RESOURCES;
	} else if (cnd_init(&wb->producer_cvar_) != thrd_success) {
	return ZX_ERR_NO_RESOURCES;
	} else if (thrd_create_with_name(&wb->writeback_thrd_,
	WritebackBuffer::WritebackThread, wb.get(),
	"minfs-writeback") != thrd_success) {
	return ZX_ERR_NO_RESOURCES;
	}
	zx_status_t status = wb->bc_->AttachVmo(wb->mapper_.vmo(), &wb->buffer_vmoid_);
	if (status != ZX_OK) {
	return status;
	}

	*out = std::move(wb);
	return ZX_OK;
	}

	WritebackBuffer::WritebackBuffer(Bcache* bc, fzl::OwnedVmoMapper mapper) :
	bc_(bc), unmounting_(false), mapper_(std::move(mapper)),
	cap_(mapper_.size() / kMinfsBlockSize) {}

	WritebackBuffer::~WritebackBuffer() {
	// Block until the background thread completes itself.
	{
	fbl::AutoLock lock(&writeback_lock_);
	unmounting_ = true;
	cnd_signal(&consumer_cvar_);
	}
	int r;
	thrd_join(writeback_thrd_, &r);

	if (buffer_vmoid_ != VMOID_INVALID) {
	block_fifo_request_t request;
	request.group = bc_->BlockGroupID();
	request.vmoid = buffer_vmoid_;
	request.opcode = BLOCKIO_CLOSE_VMO;
	bc_->Transaction(&request, 1);
	}
	}

	zx_status_t WritebackBuffer::EnsureSpaceLocked(size_t blocks) {
	if (blocks > cap_) {
	// There will never be enough room in the writeback buffer
	// for this request.
	return ZX_ERR_NO_RESOURCES;
	}
	while (len_ + blocks > cap_) {
	// Not enough room to write back work, yet. Wait until
	// room is available.
	Waiter w;
	producer_queue_.push(&w);

	do {
	cnd_wait(&producer_cvar_, writeback_lock_.GetInternal());
	} while ((&producer_queue_.front() != &w) && // We are first in line to enqueue...
	(len_ + blocks > cap_)); // ... and there is enough space for us.

	producer_queue_.pop();
	}
	return ZX_OK;
	}

	void WritebackBuffer::CopyToBufferLocked(WriteTxn* txn) {
	auto& reqs = txn->Requests();
	// Write back to the buffer
	for (size_t i = 0; i < reqs.size(); i++) {
	size_t vmo_offset = reqs[i].vmo_offset;
	size_t dev_offset = reqs[i].dev_offset;
	const size_t vmo_len = reqs[i].length;
	ZX_DEBUG_ASSERT(vmo_len > 0);
	size_t wb_offset = (start_ + len_) % cap_;
	size_t wb_len = (wb_offset + vmo_len > cap_) ? cap_ - wb_offset : vmo_len;
	ZX_DEBUG_ASSERT(wb_len <= vmo_len);
	ZX_DEBUG_ASSERT(wb_offset < cap_);
	zx_handle_t vmo = reqs[i].vmo;

	void* ptr = (void*)((uintptr_t)(mapper_.start()) +
	(uintptr_t)(wb_offset * kMinfsBlockSize));
	zx_status_t status;
	ZX_DEBUG_ASSERT((start_ <= wb_offset) ?
	(start_ < wb_offset + wb_len) :
	(wb_offset + wb_len <= start_)); // Wraparound
	ZX_ASSERT_MSG((status = zx_vmo_read(vmo, ptr, vmo_offset * kMinfsBlockSize,
	wb_len * kMinfsBlockSize)) == ZX_OK, "VMO Read Fail: %d", status);
	len_ += wb_len;

	// Update the WriteRequest to transfer from the writeback buffer
	// out to disk, rather than the supplied VMO
	reqs[i].vmo_offset = wb_offset;
	reqs[i].length = wb_len;

	if (wb_len != vmo_len) {
	// We wrapped around; write what remains from this request
	vmo_offset += wb_len;
	dev_offset += wb_len;
	wb_len = vmo_len - wb_len;
	ptr = mapper_.start();
	ZX_DEBUG_ASSERT((start_ == 0) ? (start_ < wb_len) : (wb_len <= start_)); // Wraparound
	ZX_ASSERT(zx_vmo_read(vmo, ptr, vmo_offset * kMinfsBlockSize,
	wb_len * kMinfsBlockSize) == ZX_OK);
	len_ += wb_len;

	// Shift down all following write requests
	static_assert(fbl::is_pod<WriteRequest>::value, "Can't memmove non-POD");

	// Insert the "new" request, which is the latter half of
	// the request we wrote out earlier
	WriteRequest request;
	request.vmo = reqs[i].vmo;
	request.vmo_offset = 0;
	request.dev_offset = dev_offset;
	request.length = wb_len;
	i++;
	reqs.insert(i, request);
	}
	}
	}

	void WritebackBuffer::Enqueue(fbl::unique_ptr<WritebackWork> work) {
	TRACE_DURATION("minfs", "WritebackBuffer::Enqueue");
	TRACE_FLOW_BEGIN("minfs", "writeback", reinterpret_cast<trace_flow_id_t>(work.get()));
	fbl::AutoLock lock(&writeback_lock_);

	{
	TRACE_DURATION("minfs", "Allocating Writeback space");
	size_t blocks = work->BlkCount();
	// TODO(smklein): Experimentally, all filesystem operations cause between
	// 0 and 10 blocks to be updated, though the writeback buffer has space
	// for thousands of blocks.
	//
	// Hypothetically, an operation (most likely, an enormous write) could
	// cause a single operation to exceed the size of the writeback buffer,
	// but this is currently impossible as our writes are broken into 8KB
	// chunks.
	//
	// Regardless, there should either (1) exist a fallback mechanism for these
	// extremely large operations, or (2) the worst-case operation should be
	// calculated, and it should be proven that it will always fit within
	// the allocated writeback buffer.
	ZX_ASSERT_MSG(EnsureSpaceLocked(blocks) == ZX_OK,
	"Requested txn (%zu blocks) larger than writeback buffer", blocks);
	}

	{
	TRACE_DURATION("minfs", "Copying to Writeback buffer");
	CopyToBufferLocked(work.get());
	}

	work_queue_.push(std::move(work));
	cnd_signal(&consumer_cvar_);
	}

	int WritebackBuffer::WritebackThread(void* arg) {
	WritebackBuffer* b = reinterpret_cast<WritebackBuffer*>(arg);

	b->writeback_lock_.Acquire();
	while (true) {
	while (!b->work_queue_.is_empty()) {
	auto work = b->work_queue_.pop();
	TRACE_DURATION("minfs", "WritebackBuffer::WritebackThread");

	// Stay unlocked while processing a unit of work
	b->writeback_lock_.Release();

	// TODO(smklein): We could add additional validation that the blocks
	// in "work" are contiguous and in the range of [start_, len_) (including
	// wraparound).
	size_t blks_consumed = work->Complete(b->mapper_.vmo().get(), b->buffer_vmoid_);
	TRACE_FLOW_END("minfs", "writeback", reinterpret_cast<trace_flow_id_t>(work.get()));
	work = nullptr;

	// Relock before checking the state of the queue
	b->writeback_lock_.Acquire();
	b->start_ = (b->start_ + blks_consumed) % b->cap_;
	b->len_ -= blks_consumed;
	cnd_signal(&b->producer_cvar_);
	}

	// Before waiting, we should check if we're unmounting.
	if (b->unmounting_) {
	b->writeback_lock_.Release();
	return 0;
	}
	cnd_wait(&b->consumer_cvar_, b->writeback_lock_.GetInternal());
	}
	}

	#endif // __Fuchsia__

	} // namespace minfs