zircon/system/core/devmgr/fshost/block-device.cpp

// 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 <fcntl.h>
#include <inttypes.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>

#include <fbl/algorithm.h>
#include <fbl/auto_call.h>
#include <fbl/string_buffer.h>
#include <fbl/unique_fd.h>
#include <fs-management/mount.h>
#include <fuchsia/device/c/fidl.h>
#include <fuchsia/hardware/block/c/fidl.h>
#include <fuchsia/hardware/block/partition/c/fidl.h>
#include <gpt/gpt.h>
#include <lib/fdio/directory.h>
#include <lib/fdio/fd.h>
#include <lib/fdio/fdio.h>
#include <lib/fdio/unsafe.h>
#include <lib/fdio/watcher.h>
#include <lib/fzl/fdio.h>
#include <lib/fzl/time.h>
#include <lib/zx/channel.h>
#include <lib/zx/time.h>
#include <loader-service/loader-service.h>
#include <minfs/fsck.h>
#include <minfs/minfs.h>
#include <zircon/device/block.h>
#include <zircon/processargs.h>
#include <zircon/status.h>
#include <zircon/syscalls.h>
#include <zxcrypt/fdio-volume.h>

#include <utility>

#include "block-device.h"
#include "block-watcher.h"
#include "pkgfs-launcher.h"

namespace devmgr {
namespace {

// Attempt to mount the device pointed to be the file descriptor at a known
// location.
//
// Returns ZX_ERR_ALREADY_BOUND if the device could be mounted, but something
// is already mounted at that location. Returns ZX_ERR_INVALID_ARGS if the
// GUID of the device does not match a known valid one. Returns
// ZX_ERR_NOT_SUPPORTED if the GUID is a system GUID. Returns ZX_OK if an
// attempt to mount is made, without checking mount success.
zx_status_t MountMinfs(FilesystemMounter* mounter, fbl::unique_fd fd, mount_options_t* options) {
    fuchsia_hardware_block_partition_GUID type_guid;
    {
        fzl::UnownedFdioCaller disk_connection(fd.get());
        zx::unowned_channel channel(disk_connection.borrow_channel());
        zx_status_t io_status, status;
        io_status = fuchsia_hardware_block_partition_PartitionGetTypeGuid(channel->get(), &status,
                                                                          &type_guid);
        if (io_status != ZX_OK)
            return io_status;
        if (status != ZX_OK)
            return status;
    }

    if (gpt_is_sys_guid(type_guid.value, GPT_GUID_LEN)) {
        return ZX_ERR_NOT_SUPPORTED;
    } else if (gpt_is_data_guid(type_guid.value, GPT_GUID_LEN)) {
        return mounter->MountData(std::move(fd), options);
    } else if (gpt_is_install_guid(type_guid.value, GPT_GUID_LEN)) {
        return mounter->MountInstall(std::move(fd), options);
    }
    printf("fshost: Unrecognized partition GUID for minfs; not mounting\n");
    return ZX_ERR_WRONG_TYPE;
}

// return value is ignored
int UnsealZxcryptThread(void* arg) {
    std::unique_ptr<int> fd_ptr(static_cast<int*>(arg));
    fbl::unique_fd fd(*fd_ptr);
    fbl::unique_fd devfs_root(open("/dev", O_RDONLY));

    zx_status_t rc;
    std::unique_ptr<zxcrypt::FdioVolume> zxcrypt_volume;
    if ((rc = zxcrypt::FdioVolume::Init(std::move(fd), std::move(devfs_root),
                                        &zxcrypt_volume)) != ZX_OK) {
        printf("fshost: couldn't open zxcrypt fdio volume\n");
        return ZX_OK;
    }

    zx::channel zxcrypt_volume_manager_chan;
    if ((rc = zxcrypt_volume->OpenManager(zx::sec(2),
                                          zxcrypt_volume_manager_chan.reset_and_get_address())) !=
        ZX_OK) {
        printf("fshost: couldn't open zxcrypt manager device\n");
        return 0;
    }

    zxcrypt::FdioVolumeManager zxcrypt_volume_manager(std::move(zxcrypt_volume_manager_chan));
    uint8_t slot = 0;
    if ((rc = zxcrypt_volume_manager.UnsealWithDeviceKey(slot)) != ZX_OK) {
        printf("fshost: couldn't unseal zxcrypt manager device\n");
        return 0;
    }

    return 0;
}

} // namespace

BlockDevice::BlockDevice(FilesystemMounter* mounter, fbl::unique_fd fd)
    : mounter_(mounter), fd_(std::move(fd)),
      format_(detect_disk_format(fd_.get())) {}

disk_format_t BlockDevice::GetFormat() {
    return format_;
}

void BlockDevice::SetFormat(disk_format_t format) {
    format_ = format;
}

bool BlockDevice::Netbooting() {
    return mounter_->Netbooting();
}

zx_status_t BlockDevice::GetInfo(fuchsia_hardware_block_BlockInfo* out_info) {
    if (info_.has_value()) {
        memcpy(out_info, &*info_, sizeof(*out_info));
        return ZX_OK;
    }
    fzl::UnownedFdioCaller connection(fd_.get());
    zx_status_t io_status, call_status;
    io_status = fuchsia_hardware_block_BlockGetInfo(connection.borrow_channel(), &call_status,
                                                    out_info);
    if (io_status != ZX_OK) {
        return io_status;
    }
    info_ = *out_info;
    return call_status;
}

zx_status_t BlockDevice::GetTypeGUID(fuchsia_hardware_block_partition_GUID* out_guid) {
    fzl::UnownedFdioCaller connection(fd_.get());
    zx_status_t io_status, call_status;
    io_status = fuchsia_hardware_block_partition_PartitionGetTypeGuid(connection.borrow_channel(),
                                                                      &call_status,
                                                                      out_guid);
    if (io_status != ZX_OK) {
        return io_status;
    }
    return call_status;
}

zx_status_t BlockDevice::AttachDriver(const fbl::StringPiece& driver) {
    printf("fshost: Binding: %.*s\n", static_cast<int>(driver.length()), driver.data());
    fzl::UnownedFdioCaller connection(fd_.get());
    zx_status_t io_status, call_status;
    io_status = fuchsia_device_ControllerBind(connection.borrow_channel(), driver.data(),
                                              driver.length(), &call_status);
    if (io_status != ZX_OK) {
        return io_status;
    }
    return call_status;
}

zx_status_t BlockDevice::UnsealZxcrypt() {
    printf("fshost: unsealing zxcrypt\n");
    // Bind and unseal the driver from a separate thread, since we
    // have to wait for a number of devices to do I/O and settle,
    // and we don't want to block block-watcher for any nontrivial
    // length of time.

    // We transfer fd to the spawned thread.  Since it's UB to cast
    // ints to pointers and back, we allocate the fd on the heap.
    int loose_fd = fd_.release();
    int* raw_fd_ptr = new int(loose_fd);
    thrd_t th;
    int err = thrd_create_with_name(&th, &UnsealZxcryptThread, raw_fd_ptr, "zxcrypt-unseal");
    if (err != thrd_success) {
        printf("fshost: failed to spawn zxcrypt unseal thread");
        close(loose_fd);
        delete raw_fd_ptr;
    } else {
        thrd_detach(th);
    }
    return ZX_OK;
}

zx_status_t BlockDevice::IsUnsealedZxcrypt(bool* is_unsealed_zxcrypt) {
    zx_status_t call_status;
    fbl::StringBuffer<PATH_MAX> path;
    path.Resize(path.capacity());
    size_t path_len;
    fzl::UnownedFdioCaller disk_connection(fd_.get());
    // Both the zxcrypt and minfs partitions have the same gpt guid, so here we
    // determine which it actually is. We do this by looking up the topological
    // path.
    if (fuchsia_device_ControllerGetTopologicalPath(disk_connection.borrow_channel(), &call_status,
                                                    path.data(), path.capacity(),
                                                    &path_len) != ZX_OK) {
        return ZX_ERR_NOT_FOUND;
    }
    if (call_status != ZX_OK) {
        return call_status;
    }
    const fbl::StringPiece kZxcryptPath("/zxcrypt/unsealed/block");
    if (path_len < kZxcryptPath.length()) {
        *is_unsealed_zxcrypt = false;
    } else {
        *is_unsealed_zxcrypt = fbl::StringPiece(path.begin() + path_len -
                                                kZxcryptPath.length())
                                   .compare(kZxcryptPath) == 0;
    }
    return ZX_OK;
}

zx_status_t BlockDevice::FormatZxcrypt() {
  fbl::unique_fd devfs_root_fd(open("/dev", O_RDONLY));
  if (!devfs_root_fd) {
      return ZX_ERR_NOT_FOUND;
  }
  return zxcrypt::FdioVolume::CreateWithDeviceKey(fd_.duplicate(), std::move(devfs_root_fd),
                                                  nullptr);
}

bool BlockDevice::ShouldCheckFilesystems() {
    return mounter_->ShouldCheckFilesystems();
}

zx_status_t BlockDevice::CheckFilesystem() {
    if (!ShouldCheckFilesystems()) {
        return ZX_OK;
    }

    zx_status_t status;
    fuchsia_hardware_block_BlockInfo info;
    if ((status = GetInfo(&info)) != ZX_OK) {
        return status;
    }

    switch (format_) {
    case DISK_FORMAT_BLOBFS: {
        fprintf(stderr, "fshost: Skipping blobfs consistency checker.\n");
        return ZX_OK;
    }
    case DISK_FORMAT_MINFS: {
        zx::ticks before = zx::ticks::now();
        auto timer = fbl::MakeAutoCall([before]() {
            auto after = zx::ticks::now();
            auto duration = fzl::TicksToNs(after - before);
            printf("fshost: fsck took %" PRId64 ".%" PRId64 " seconds\n", duration.to_secs(),
                   duration.to_msecs() % 1000);
        });
        printf("fshost: fsck of %s started\n", disk_format_string_[format_]);
        uint64_t device_size = info.block_size * info.block_count / minfs::kMinfsBlockSize;
        std::unique_ptr<minfs::Bcache> bc;
        zx_status_t status;
        if ((status = minfs::Bcache::Create(&bc, fd_.duplicate(),
                                            static_cast<uint32_t>(device_size))) != ZX_OK) {
            fprintf(stderr, "fshost: Could not initialize minfs bcache.\n");
            return status;
        }
        status = minfs::Fsck(std::move(bc));

        if (status != ZX_OK) {
            fprintf(stderr, "--------------------------------------------------------------\n");
            fprintf(stderr, "|                                                             \n");
            fprintf(stderr, "|   WARNING: fshost fsck failure!                             \n");
            fprintf(stderr, "|   Corrupt %s filesystem\n", disk_format_string_[format_]);
            fprintf(stderr, "|                                                             \n");
            fprintf(stderr, "|   If your system encountered power-loss due to an unclean   \n");
            fprintf(stderr, "|   shutdown, this error was expected. Journaling in minfs    \n");
            fprintf(stderr, "|   is being tracked by ZX-2093. Re-paving will reset your    \n");
            fprintf(stderr, "|   device.                                                   \n");
            fprintf(stderr, "|                                                             \n");
            fprintf(stderr, "|   If your system was shutdown cleanly (via 'dm poweroff'    \n");
            fprintf(stderr, "|   or an OTA), report this device to the local-storage       \n");
            fprintf(stderr, "|   team. Please file bugs with logs before and after reboot. \n");
            fprintf(stderr, "|   Please use the 'filesystem' and 'minfs' component tag.    \n");
            fprintf(stderr, "|                                                             \n");
            fprintf(stderr, "--------------------------------------------------------------\n");
        } else {
            printf("fshost: fsck of %s completed OK\n", disk_format_string_[format_]);
        }
        return status;
    }
    default:
        fprintf(stderr, "fshost: Not checking unknown filesystem\n");
        return ZX_ERR_NOT_SUPPORTED;
    }
}

zx_status_t BlockDevice::FormatFilesystem() {
    zx_status_t status;
    fuchsia_hardware_block_BlockInfo info;
    if ((status = GetInfo(&info)) != ZX_OK) {
        return status;
    }

    switch (format_) {
    case DISK_FORMAT_BLOBFS: {
        fprintf(stderr, "fshost: Not formatting blobfs.\n");
        return ZX_ERR_NOT_SUPPORTED;
    }
    case DISK_FORMAT_MINFS: {
        fprintf(stderr, "fshost: Formatting minfs.\n");
        uint64_t blocks = info.block_size * info.block_count / minfs::kMinfsBlockSize;
        std::unique_ptr<minfs::Bcache> bc;
        zx_status_t status;
        if ((status = minfs::Bcache::Create(&bc, fd_.duplicate(),
                                            static_cast<uint32_t>(blocks))) != ZX_OK) {
            fprintf(stderr, "fshost: Could not initialize minfs bcache.\n");
            return status;
        }
        minfs::MountOptions options = {};
        if ((status = minfs::Mkfs(options, std::move(bc))) != ZX_OK) {
            fprintf(stderr, "fshost: Could not format minfs filesystem.\n");
            return status;
        }
        printf("fshost: Minfs filesystem re-formatted. Expect data loss.\n");
        return ZX_OK;
    }
    default:
        fprintf(stderr, "fshost: Not formatting unknown filesystem.\n");
        return ZX_ERR_NOT_SUPPORTED;
    }
}

zx_status_t BlockDevice::MountFilesystem() {
    // Go through the song-and-dance of cloning our reference to |fd_|
    // so we can hand off a cloned connection to the mount functions.
    // The mount functions are very possessive of their fds, and don't like
    // operating on dup-ed descriptors.
    //
    // In the future, this could be simplified by passing channels directly,
    // and avoiding file descriptors altogether.
    fbl::unique_fd cloned_fd;
    {
        fzl::UnownedFdioCaller disk_connection(fd_.get());
        zx::unowned_channel channel(disk_connection.borrow_channel());
        zx::channel cloned_channel(fdio_service_clone(channel->get()));
        fdio_t* io;
        zx_status_t status = fdio_create(cloned_channel.release(), &io);
        if (status != ZX_OK) {
            return status;
        }
        cloned_fd.reset(fdio_bind_to_fd(io, -1, 0));
        if (!cloned_fd) {
            return ZX_ERR_BAD_STATE;
        }
    }

    switch (format_) {
    case DISK_FORMAT_BLOBFS: {
        fprintf(stderr, "fshost: BlockDevice::MountFilesystem(blobfs)\n");
        mount_options_t options = default_mount_options;
        options.enable_journal = true;
        options.collect_metrics = true;
        zx_status_t status = mounter_->MountBlob(std::move(cloned_fd), &options);
        if (status != ZX_OK) {
            printf("fshost: Failed to mount blobfs partition: %s.\n",
                   zx_status_get_string(status));
            return status;
        } else {
            LaunchBlobInit(mounter_);
        }
        return ZX_OK;
    }
    case DISK_FORMAT_MINFS: {
        mount_options_t options = default_mount_options;
        fprintf(stderr, "fshost: BlockDevice::MountFilesystem(minfs)\n");
        return MountMinfs(mounter_, std::move(cloned_fd), &options);
    }
    default:
        fprintf(stderr, "fshost: BlockDevice::MountFilesystem(unknown)\n");
        return ZX_ERR_NOT_SUPPORTED;
    }
}

zx_status_t BlockDeviceInterface::Add() {
    disk_format_t df = GetFormat();
    fuchsia_hardware_block_BlockInfo info;
    zx_status_t status;
    if ((status = GetInfo(&info)) != ZX_OK) {
        return status;
    }

    if (info.flags & BLOCK_FLAG_BOOTPART) {
        return AttachDriver(kBootpartDriverPath);
    }

    switch (df) {
    case DISK_FORMAT_GPT: {
        return AttachDriver(kGPTDriverPath);
    }
    case DISK_FORMAT_FVM: {
        return AttachDriver(kFVMDriverPath);
    }
    case DISK_FORMAT_MBR: {
        return AttachDriver(kMBRDriverPath);
    }
    case DISK_FORMAT_ZXCRYPT: {
        if (!Netbooting()) {
            return UnsealZxcrypt();
        }
        return ZX_OK;
    }
    default:
        break;
    }

    fuchsia_hardware_block_partition_GUID guid;
    if ((status = GetTypeGUID(&guid)) != ZX_OK) {
        return status;
    }

    // If we're in netbooting mode, then only bind drivers for partition
    // containers and the install partition, not regular filesystems.
    if (Netbooting()) {
        if (gpt_is_install_guid(guid.value, GPT_GUID_LEN)) {
            printf("fshost: mounting install partition\n");
            return MountFilesystem();
        }
        return ZX_OK;
    }

    switch (df) {
    case DISK_FORMAT_BLOBFS: {
        const uint8_t expected_guid[GPT_GUID_LEN] = GUID_BLOB_VALUE;

        if (memcmp(guid.value, expected_guid, GPT_GUID_LEN)) {
            return ZX_ERR_INVALID_ARGS;
        }
        if ((status = CheckFilesystem()) != ZX_OK) {
            return status;
        }

        return MountFilesystem();
    }
    case DISK_FORMAT_MINFS: {
        printf("fshost: mounting minfs\n");
        if (CheckFilesystem() != ZX_OK) {
            if ((status = FormatFilesystem()) != ZX_OK) {
                return status;
            }
        }
        status = MountFilesystem();
        if (status != ZX_OK) {
            printf("fshost: failed to mount filesystem: %s\n", zx_status_get_string(status));
            return status;
        }
        return ZX_OK;
    }
    default:
        // If the disk format is unknown but we know it should be the data
        // partition, format the disk properly.
        if (gpt_is_data_guid(guid.value, GPT_GUID_LEN)) {
            printf("fshost: Data partition has unknown format\n");
            bool is_unsealed_zxcrypt;
            if (IsUnsealedZxcrypt(&is_unsealed_zxcrypt) != ZX_OK) {
                return ZX_ERR_NOT_SUPPORTED;
            }
            if (is_unsealed_zxcrypt) {
                printf("fshost: Formatting as minfs partition\n");
                SetFormat(DISK_FORMAT_MINFS);
                status = FormatFilesystem();
                if (status != ZX_OK) {
                  return status;
                }
            } else {
                printf("fshost: Formatting as zxcrypt partition\n");
                SetFormat(DISK_FORMAT_ZXCRYPT);
                status = FormatZxcrypt();
                if (status != ZX_OK) {
                  return status;
                }
            }
            return Add();
        }
        return ZX_ERR_NOT_SUPPORTED;
    }
}

} // namespace devmgr