Commit graph

20 commits

Author SHA1 Message Date
Daniel Rosenberg
cc874804dd Set media folder +F for adopted storage as well
We previously only set +F for /data/media, but adopted storage needs
this as well. Instead we add support for adding attrs to PrepareDir.

Bug: 163453310
Test: sm set-virtual-disk true
      follow UI setup and confirm +F on /mnt/expand/*/media
Change-Id: I08f13b57a4de3538e88b38eb95b0ac115a5a5ce8
Merged-In: I08f13b57a4de3538e88b38eb95b0ac115a5a5ce8
2020-10-20 18:55:54 -07:00
Alistair Delva
c671731075 Handle virtio in private fs mapping
When the vold core decides if a device is SD or USB, it checks for MMC
or virtio, however when the filesystem type is decided, it does not
check for virtio, only MMC. This causes virtio SD cards to be formatted
with ext4 unconditionally.

This fix is independently correct, but it incidentally gets adopted
storage working on cuttlefish (and Android Emulator) because f2fs can
support fscrypt and casefolding at the same time; ext4 currently cannot.

Bug: 156286088
Change-Id: I0b41670d5f76b2506dad437917c2276f8e0aaccf
2020-05-19 17:42:20 -07:00
Martijn Coenen
5ec8658abc Introduce postMount() VolumeBase helper.
When we're mounting a private volume, we create stacked emulated volumes
on top of it. Due to the ordering there, we would broadcast the emulated
volumes being created *before* the "mounted" status update. This in turn
could cause us to try and mount these emulated volumes before the
underlying private volume is really mounted. This is problematic in
particular on devices that support a filesystem keyring, where we need
to do some additional setup before the devices can be used.

While we could modify StorageManagerService to delay the mount, a safer
fix at this stage of the release is to just fix the ordering of these
events. To achieve that, add a simple postMount() helper, that is called
after a succesful mount. This allows us to setup the volume properly
before trying to mount any stacked volumes.

Bug: 151079464
Test: atest AdoptableHostTest
Change-Id: I2cc4113d4d71d89aa629bb9c0fa9be441355c079
2020-05-04 15:09:32 +02:00
Martijn Coenen
6c695ef1d9 Prefer f2fs for virtual (private) storage volumes.
Since ext4 currently doesn't have the required kernel patches in place
on cuttlefish.

Bug: 150935323
Test: sm set-virtual-disk true
      sm partition disk:7,xyz private
      inspect mount output
Change-Id: Ief5bd9ace9d39bdfbae8d3857044a2143801f6be
2020-03-11 15:36:14 +01:00
Automerger Merge Worker
7489ab6961 Merge changes from topics "metadata_wrapped_key_aosp", "volume_metadata" am: 36fd1ebfae am: 6891eb7e2d am: c14f46d114
Change-Id: I89f51bfaeb61c235aeccbe8a5a5a447ab14c46cb
2020-02-19 22:19:26 +00:00
Paul Crowley
886e572009 On newer devices, use dm-default-key to encrypt SD cards
The dm-crypt solution requires a kernel patch that won't be present in
the GKI kernel, while the new metadata encryption system in the GKI
kernel solves this problem in a much cleaner way.

Test: create private volume on Cuttlefish, setting property both ways.
Bug: 147814592
Change-Id: Ie02bd647c38d8101af2bbc47637f65845d312cea
2020-02-18 13:01:00 -08:00
Ricky Wai
9eb4367165 Retry deleting dm devices.
For some reason this can be racy; until we understand the root cause,
retry to unblock AdoptableHostTest.

Bug: 149396179
Test: atest AdoptableHostTest no longer hangs
Change-Id: I162ff8ad305535e7a4fab3d88f38b687b50cf4a3
2020-02-15 01:15:42 +00:00
Automerger Merge Worker
50f7de29be Merge "Use DM layer directly to manage private DM volumes" am: dd12ea5bd2 am: 47aff8772d am: e3609aa837
Change-Id: I663a0ad6990acfc2cd936b52e39b346c2e48b6ed
2020-02-13 19:08:54 +00:00
Automerger Merge Worker
395edfede5 Merge "Pass volume key as a KeyBuffer" am: 334a684557 am: 4d37b4f980 am: aebcc399ea
Change-Id: I7f410a56f2c301557657f9357494bcf97c7cf92d
2020-02-13 05:41:52 +00:00
Paul Crowley
659b63fe00 Use DM layer directly to manage private DM volumes
Abolish cryptfs_revert_ext_volume, handle in caller. This allows us to
use DeleteDeviceIfExists, avoiding a spurious error message.

Test: create private volume on Cuttlefish, eject, check logs
Bug: 147814592
Change-Id: I836d8bd11b29e32da0863aaa75144543bb9cab9c
2020-02-12 14:26:38 -08:00
Paul Crowley
3d98f5d159 Pass volume key as a KeyBuffer
Not for security, but for consistency with the way we handle other
keys, and to move the length check to where it belongs.

Test: create private volume on Cuttlefish
Bug: 147814592
Change-Id: I10fc4896183d050ce25ff174faf78f525cf62930
2020-02-12 14:26:38 -08:00
Automerger Merge Worker
3fcbdbff5f Merge "Use std::string to return crypto device, not char *" am: 80731b0975 am: 2381810b99 am: 1541f9f32b
Change-Id: I68cf7bfc6ce642d46d874bb44694109b1f43258e
2020-02-11 01:18:30 +00:00
Paul Crowley
81796e9dce Use std::string to return crypto device, not char *
Bug: 147814592
Test: can create private volume on Cuttlefish
Change-Id: Ic2bca81c0f0319e1b988e9204a2f4e91af57d157
2020-02-07 11:27:49 -08:00
Zim
a438b24368 Fix multi-user and multi-storage with FUSE
Up until now, the FUSE mount logic has made two assumptions:
1. The primary external volume is an emulated volume on /data/media
2. Only the primary user is running, as user zero

These assumptions are fixed by the following changes
creating an EmulatedVolume per Android user and changing the
VolumeBase id format to append the user to the id, so
s/emulated/emulated-0/. This allows us mount separate volumes per user

Some additional refactorings to re-use/clean up code.

Test: adb shell sm set-virtual-disk and partition disk operations work
even after setting up a work profile
Bug: 135341433

Change-Id: Ifabaa12368e5a591fbcdce4ee71c83ff35fdac6b
2019-11-11 20:24:49 +00:00
Paul Crowley
14c8c0765a clang-format many files.
Test: Format-only changes; treehugger suffices.
Change-Id: I23cde3f0bbcac13bef555d13514e922c79d5ad48
2018-09-18 15:41:22 -07:00
Greg Kaiser
57f9af6af4 cryptfs: Require ext disk crypt to match code
Our external partitions have no crypto header/footer, so we
only get the keysize and key.  Our code has been implicitly
assuming that this keysize off of disk matches the crypto
type we have in our code (and thus matches the keysize our
code is using as well).  We now make this assumption
explicit, and check for this and no longer allow external
code to pass a keysize in to cryptfs.

Bug: 73079191
Test: Compiled and tested in combination with other CLs.
Change-Id: I1a1996187e1aaad6f103982652b1bcdfd5be33ce
2018-02-16 15:23:56 -08:00
Jeff Sharkey
3472e52fc2 Move to modern utility methods from android::base.
Moves away from crufty char* operations to std::string utility
methods, including android::base methods for splitting/parsing.

Rewrite of how Process handles scanning procfs for filesystem
references; now uses fts(3) for more sane traversal.

Replace sscanf() with new FindValue() method, also has unit tests.

Remove some unused methods.  Switch almost everyone over to using
modern logging library.

Test: cts-tradefed run commandAndExit cts-dev -m CtsOsTestCases -t android.os.storage.cts.StorageManagerTest
Test: cts-tradefed run commandAndExit cts-dev --abi armeabi-v7a -m CtsAppSecurityHostTestCases -t android.appsecurity.cts.AdoptableHostTest
Bug: 67041047
Change-Id: I70dc512f21459d1e25b187f24289002b2c7bc7af
2017-10-17 12:40:51 -06:00
Jeff Sharkey
cbe69fc060 Destroy vold socket interface completely.
Long live Binder.

Test: yes
Bug: 13758960
Change-Id: If6be379b5a873f1b0c66dd1522b87413ad10fc46
2017-09-18 16:00:14 -06:00
Jeff Sharkey
814e9d308e Move unsolicited vold events to Binder.
Create IVoldListener and move most unsolicited vold events over to
this new interface.  The remaining events will be routed through
method-specific listeners instead of a global one.

Move to upstream DISALLOW_COPY_AND_ASSIGN macro.

Test: cts-tradefed run commandAndExit cts-dev -m CtsAppSecurityHostTestCases -t android.appsecurity.cts.DirectBootHostTest
Test: cts-tradefed run commandAndExit cts-dev -m CtsAppSecurityHostTestCases -t android.appsecurity.cts.AdoptableHostTest
Test: cts-tradefed run commandAndExit cts-dev -m CtsOsTestCases -t android.os.storage.cts.StorageManagerTest
Bug: 13758960
Change-Id: Ib9293487db2d525a76b9b9c2e9ac18d98601c6cf
2017-09-13 11:52:53 -06:00
Jeff Sharkey
11c2d380a7 Move even more vold commands over to Binder.
This moves fstrim, obb and appfuse commands over to the new Binder
interface.  This change also separates creating/destroying and
mounting/unmounting of OBB volumes, which means they finally flow
nicely into the modern VolumeInfo/VolumeBase design.

We now generate unique identifiers for all OBB volumes, instead of
using a shady MD5 hash.

Change all "loop" and "dm" devices to tag the kernel resources with
a vold-specific prefix so that we can clean them up if vold crashes;
there are new destroyAll() methods that handle this cleanup.

Move appfuse mounting/unmounting into VolumeManager so it can be
shared.  Move various model objects into a separate directory to
tidy things up.

Test: cts-tradefed run commandAndExit cts-dev -m CtsOsTestCases -t android.os.storage.cts.StorageManagerTest
Bug: 13758960
Change-Id: I7294e32b3fb6efe07cb3b77bd20166e70b66958f
2017-09-11 18:44:17 -06:00
Renamed from PrivateVolume.cpp (Browse further)