The function used to be serving system and vendor partitions only (as
they were the only partitions using sparse image at the point). The code
itself doesn't rely on anything specific to system/vendor.
Test: python -m unittest test_common
Change-Id: Ia4ecdeedb262f3d9db082128eaf9bab299983333
The signature size will be 512 bytes when signing the payload
with 4096 bits key. This cl determines the key size with
"openssl rsa -modulus"
The new key in testdata is generated by
"openssl genrsa -out testkey 4096"
Bug: 129163830
Test: generate and verify an OTA package
Change-Id: I6662b0a0c553dc0fd84711312a1256b887e332fd
* changes:
Only assert-max-image-size for static partitions.
sparse_img.py --get_partition_size return size of partition
Revert "Fix dynamic partition size check for devices with recovery"
If TARGET_USERIMAGES_SPARSE_EXT_DISABLED is set, don't provide
--sparse to lpmake, so that a non-sparse super image is built.
Test: build with the flag set.
Bug: 120041578
Change-Id: I5a26e4c793b0e2ddc89e9c38c8828ac21044e78a
Merged-In: I5a26e4c793b0e2ddc89e9c38c8828ac21044e78a
assert-max-image-size doesn't make sense for
dynamic partitions, as build_image.py always find the
right size for the output image. Hence:
- build_image.py no longer need to write generated_*_info.txt
(which contains the size of the image).
- assert-max-image-size on the static BOARD_*IMAGE_PARTITION_SIZE. If
a partition is dynamic, that variable isn't set, and
assert-max-image-size becomes a no-op. If the partition is static,
assert-max-image-size checks the static partition size as it used
to be.
- Fix read-size-of-partitions to use the size of the partition by
reading the image directly (instead of using generated_*_info.txt).
For devices without AVB, with DAP enabled, and does not have
RESERVED_SIZE for partitions, because of right sizing, the original
code always warns about approaching size limits. Since such checks
doesn't make sense for dynamic partitions, remove them.
Test: builds on device with dynamic partitions
Test: builds on cuttlefish with DAP enabled (without AVB), no
more size limit warnings:
WARNING: out/target/product/vsoc_x86/vendor.img approaching size limit (X now; limit X)
Fixes: 122377935
Change-Id: I75e1b8322197cb18cf397d02aefd49d777bb6405
Also, move code from build_super_image.py to sparse_img.py.
Test: sparse_img.py on sparse and non-sparse images
Bug: 122377935
Change-Id: Ie91fdfdbb54298ea27eb20d1b5363aeb1470356e
If TARGET_USERIMAGES_SPARSE_EXT_DISABLED is set, don't provide
--sparse to lpmake, so that a non-sparse super image is built.
Test: build with the flag set.
Fixes: 120041578
Change-Id: I5a26e4c793b0e2ddc89e9c38c8828ac21044e78a
For non-A/B, you must supply the --system-item-list, --other-item-list, and
--system-misc-info-keys parameters approrpriate for merging two partial target
files. Additionally, you must supply the --rebuild-recovery option to correctly
generate the filesystem config and file contexts. With all of these parameters,
the script will generate a merged target files containing the correct recovery
files.
Also fix the --keep-tmp option to be consistent with the other options.
Bug: 122813742
Test: verify that merged target image boots and can perform OTA.
Change-Id: I5a942ac0cd9924fec419a686794a2340304594c8
This reverts commit 9788b4ed31. All the
blocking issues have been addressed.
Fixes: 120517892
Test: Run validate_target_files.py on crosshatch signed
target_files.zip.
Change-Id: I95de241e159998e002dedddafea65953b1a1b263
Previously it was following a wrong order by doing `zipalign` after
calling SignApk, which effectively compromised the signature. This CL
corrects the logic, and follows the same flow as in build system:
- Pack APEX file;
- `zipalign -f 4096`;
- Call SignApk to sign the container with `-a 4096` flag.
Bug: 129148142
Test: Run sign_target_files_apks.py on taimen target_files.zip. Boot the
image after signing.
Change-Id: I91bd3dce4f45c1891c5e122212a699f4808618fa
(cherry picked from commit 0e06cb0a8b)
Previously it was following a wrong order by doing `zipalign` after
calling SignApk, which effectively compromised the signature. This CL
corrects the logic, and follows the same flow as in build system:
- Pack APEX file;
- `zipalign -f 4096`;
- Call SignApk to sign the container with `-a 4096` flag.
Bug: 129148142
Test: Run sign_target_files_apks.py on taimen target_files.zip. Boot the
image after signing.
Change-Id: I91bd3dce4f45c1891c5e122212a699f4808618fa
To build a complete list of the dynamic partitions and partitions
groups, we need to merge the contribution from the system and other
target files.
Bug: 127687287
Test: Running merge_target_files.py and observing partition lists are
merged as expected.
Change-Id: I5bb9bd0e3179d48c9bfacdb3aca8253158f61cf6
For PRESIGNED APEXes, we should keep carrying the matching public keys
at /system/etc/security/apex.
Bug: 129148142
Test: Run sign_target_files_apks.py on a target_files.zip with presigned
APEXes. Check the output zip.
Change-Id: I2e941fd9b10e99d2db9df1e5308cbbe8c760177b
(cherry picked from commit bf3fb024cd)
For PRESIGNED APEXes, we should keep carrying the matching public keys
at /system/etc/security/apex.
Bug: 129148142
Test: Run sign_target_files_apks.py on a target_files.zip with presigned
APEXes. Check the output zip.
Change-Id: I2e941fd9b10e99d2db9df1e5308cbbe8c760177b
This change changes auto-generated RROs from DEVICE_PACKAGE_OVERLAYS
to be generated in the vendor partition, as opposed to /product where
they were generated in the past.
Note that PRODUCT_PACKAGE_OVERLAYS continue generating RRO packages
to /product, which means that a single app can be overlayed from
different partitions. These RROs have been given module and package
names based on their location.
Bug: 127758779
Test: verify noop on presubmit targets
Change-Id: I5cee70e28e3969e67b2d83eaf25d9c6e3a11102d
This reverts commit 5516d37f41.
The previous issue in unzipping non-matching files has been addressed
with commit a49054ca2f2959f50f3188914ec0faebc90ebcbe. This CL rolls
forward to allow dumping container certifcates for APEXes.
Bug: 128848294
Test: Run check_target_files_signatures.py on target_files.zips w/ and
w/o APEX files.
Change-Id: I662aab3d96fc40ac8e5e206e32b73ac763220b70
common.UnzipTemp() calls `unzip` to do the unzipping, which will
complain if there's non-existent names in the given list. Prior to this
CL, callers had to do the work to remove non-existent entries. This CL
filters out the given patterns in common.UnzipTemp()/common.UnzipToDir()
to make callers' works easier.
Bug: 128848294
Test: `m dist` with aosp_taimen-userdebug (which calls
ota_from_target_files.py on a target_files.zip that doesn't
contain RADIO/*).
Test: `python -m unittest test_common.CommonZipTest`
Change-Id: I5e741c27ea8d0b8126c398a7e1b56a8deb4a3d7f
When TARGET_VNDK_USE_CORE_VARIANT is set to true, the vendor variant of
VNDK libraries are by default not installed. Instead, the core variant
will be used by vendor binaries at runtime.
To ensure the core variant of VNDK libraries are installed, we also add
a flag LOCAL_VNDK_DEPEND_ON_CORE_VARIANT to indicate that the vendor
variant module depends on the core variant module. This flag should be
set by Soong for all VNDK libraries without the vendor variant
installed. When the flag is set, the vendor variant binary is also
compared against the core variant binary to ensure they are
functionally identical.
As we are merging the two variants for some libraries, we need a new
link type to denote a module is usable as both native:vndk and
native:platform. We add native:platform_vndk for this.
Bug: 119423884
Test: With the corresponding Soong change, build with
TARGET_VNDK_USE_CORE_VARIANT set to true.
Test: Add a dummy VNDK library and a dummy vendor binary that depends
on it. Build with no-vendor-variant VNDK and check the core
variant is installed.
Test: Add conditional compilation based on __ANDROID_VNDK__ in the
dummy VNDK library and check build fails.
Change-Id: I40000f2728e8193212113c1ee950e9d697f2d40d
Currently system_other AVB public key is placed in system.img.
However, this makes it's harder to have a *generic* system.img
across different product configs. Moving the key to /product
partition to allow more product-specific AVB keys.
Device board config can add /product/etc/fstab.postinstall,
to mount system_other with this key in /product. It can specify
different mount options, file systems, verity settings, etc., in
this product-specific fstab as well.
Bug: 123611926
Test: `make productimage` checks the following is generated.
$OUT/product/etc/security/avb/system_other.avbpubkey
Also checks it's included in $OUT/installed-files-product.{json, txt}
Test: run the following command and checks that
PRODUCT/etc/security/avb/system_other.avbpubkey is updated:
./build/tools/releasetools/sign_target_files_apks \
--avb_system_other_algorithm SHA256_RSA2048 \
--avb_system_other_key external/avb/test/data/testkey_rsa2048.pem \
out/dist/*-target_files-*.zip signed-target_files.zip
Change-Id: I6804f29941bec54375d80bd68a5aedb5c23b842e
This commit updates how `check_elf_file.py` parses the symbol name
because the `llvm-readobj` (from clang-r353983) does not print "@" if
the symbol is not versioned.
See also. https://reviews.llvm.org/D56319
Bug: 128959554
Test: CHECK_ELF_FILES=true make check-elf-files
Change-Id: I0dee5e505225e57750a2c86cf0d25a151c218eb1
This CL adds support that allows treating an APEX as pre-signed. We can
skip signing an APEX with `-e <apex-name>=` and
`--extra_apex_payload_key <apex-name>=`. Note that the payload_key and
container_key must be in consistent state - either they're both
PRESIGNED or none of them is. CheckApkAndApexKeysAvailable() has been
updated to perform the sanity check.
Bug: 123716522
Test: Run sign_target_files_apks.py with the above flags.
Test: python -m unittest test_sign_target_files_apks
Change-Id: Id1e2f3f2facd4a97a385983cc9b78c028f7e7e73
This validation is to help ensure that any usage of custom merge config
files does not accidentally exclude any item that has been added to the
default config lists.
Bug: 124197349
Test: Run merge_target_files with custom merge config files.
Change-Id: I34c51cb75212368146a2944d37621f311060d24d
This reverts commit d8469727bc. The script
is broken on target_files.zip that don't contain any APEX.
Bug: 128848294
Test: Run check_target_files_signatures.py on target_files.zip w/o APEX.
OTA tools should pick up the avbtool, as listed in dict['avb_avbtool'],
from the current PATH (plus bin/ under the dir specified via `--path`),
the same way as handling all other host tools.
Test: `m dist`
Change-Id: I3eb4d2c61979b03d9c23b2403d9a38cf052d87ea
Also makes AddSystem check that an output_zip exists before attempting
to add the recovery patch to the output zip.
Bug: 128838154
Test: Running merge_target_files with --rebuild_recovery and verifying
it passes --rebuild_recovery to add_img_to_target_files.
Change-Id: I19347b2c0dabf29b7196045b18551b5d0687df2c
The keys_info in the touched code is a tuple, which is immutable.
Bug: 123716522
Test: Run sign_target_files_apks.py with '-e foo.apex=bar' that replaces
the APEX container key.
Change-Id: I4e57e46c93a56b7f6646764d021ebb42c19bf7f5
A few tweaks to make it easier to extend to generating RROs in
multiple partitions:
- deduce the module name inside generate_enforce_rro
- dedup rule definition
- tweak framework-res check to use source module name instead
Bug: 127758779
Test: verify noop on presubmit targets
Change-Id: I2f0d6270b21f5427c372c04a5c6e7fb712e72a9a
Bug: 123716522
Test: Run sign_target_files_apks.py to sign a target_files with APEXes.
Test: Run check_target_files_signatures.py on signed artifact.
Test: python -m unittest test_sign_target_files_apks
Change-Id: I3fa13e3d9461cf5e0838e0572d436e218164fe41
(cherry picked from commit aa7e993a22)
Only the container certs will be checked and reported. For the payload
within an APEX, we can't easily extract the cert info.
It needs to go along a longer path, if ever needed, by:
- extracting public keys from all the available certs;
- using each of them to verify against an APEX payload to find a match
(`avbtool verify_image --image payload --key public_key`).
Bug: 123716522
Test: Run check_target_files_signatures.py on target_files with APEXes.
Change-Id: I2ef318e05433d2d65ab84e2dff9e01fb6ee3373d
(cherry picked from commit d8469727bc)
Bug: 123716522
Test: Run sign_target_files_apks.py to sign a target_files with APEXes.
Test: Run check_target_files_signatures.py on signed artifact.
Test: python -m unittest test_sign_target_files_apks
Change-Id: I3fa13e3d9461cf5e0838e0572d436e218164fe41
Only the container certs will be checked and reported. For the payload
within an APEX, we can't easily extract the cert info.
It needs to go along a longer path, if ever needed, by:
- extracting public keys from all the available certs;
- using each of them to verify against an APEX payload to find a match
(`avbtool verify_image --image payload --key public_key`).
Bug: 123716522
Test: Run check_target_files_signatures.py on target_files with APEXes.
Change-Id: I2ef318e05433d2d65ab84e2dff9e01fb6ee3373d
Other modules have switched to logging module. sign_target_files_apks.py
needs to init the logger to get the logs.
Test: Run `sign_target_files_apks.py -v`. Check outputs.
Test: Run `check_target_files_signatures.py -v`.
Change-Id: Ic68c019f6fb14840561885f1194ad6efdfdb7d82
superimage-nodeps and supernod depends
on images from $(ANDROID_PRODUCT_OUT) (not from
target files package). It doesn't rebuild source
images if they are present.
A typical workflow is:
m -j
# change code in system
m snod -j
m supernod -j
Test: For non retrofit, run:
`m snod -j; m supernod -j`
Fixes: 128321505
Change-Id: Ib8c011cadb9c0cd334234aef39f19be6a48fee62
A followup change will make it not always equal TARGET_DEVICE,
so that name is unsuitable. Make its name follow the other product
variables.
Bug: 110206836
Test: presubmit
Change-Id: Icb1422ec5e7af658c5cc3070993c472e99805c6b
Bug: 124467065
Test: Running `python merge_target_files.py` using the three new flags
and observing that their contents are passed to the merge_target_files() function.
Change-Id: I4de46f041f5ae8bc8be2730313ce873a952bf78e
sign_target_files_apks script looks for the signapk.jar inside the out dir.
If the our dir is set to a different directory via OUT_DIR_COMMON_BASE the script does not work properly.
From now script checks if the OUT_DIR_COMMON_BASE is set, then searches the jar in the proper path.
If OUT_DIR_COMMON_BASE is unset, searches in "out" like it did before.
Test: Build with OUT_DIR_COMMON_BASE set and unset and verify that sign_target_files_apks works in both cases
Change-Id: I9218b98ff79526184f8353705640193405afac9e
For retrofit updates:
ab_partitions -= dynamic_partition_list
ab_partitions += super_block_devices
For example,
if super_block_devices == ["system", "vendor", "odm"] and
dynamic_partition_list == ["system", "vendor", "product"],
product partition needs to be removed from ab_partitions.txt
otherwise brillo_update_payload will generate a payload with
product partition in it.
Test: retrofit update that adds a partition
Fixes: 127425410
Change-Id: Id79a410cee3c611ac50d27f14282916aea34f938
This change adds another utility function to common.py: UnzipToDir, which is
generally useful. Refactor merge_target_files.py to use it, and also refactor
other uses in common.py to use it.
Test: ota_from_target_files.py, validate_target_files.py, test_common.py
Bug: 124464492
Change-Id: Ia571070bceb7d3c8002304836bdf688485bf0dd9
Then refactor the code in merge_target_files.py to adapt to this semantic
change. This makes the code more consistent with existing releasetools code,
and it's easier to follow.
Test: Failure cases (verify exception), success cases (merged target generated)
Bug: 124521133
Change-Id: I56f04e360d8ff8ffcd6245359cdeb79f4565a9c4
The merge_target_files.py script needs fc_sort and sefcontext_compile, so
include these tools into otatools.zip via core/Makefile.
Modify tools/releasetools/merge_target_files.py to use the otatools common argv
processing to take advantage of the '--path' option so that we add point the
'--path' option to an extracted otatools.zip package to gain access to fc_sort,
sefcontext_compile, and soong_zip (previously included in otatools.zip).
Bug: 123600124
Test: extract otatools.zip, use --path option to point to it, verify result
Change-Id: I7d84525981b8741c6bdbcac9984256920fc7f417
The validate_target_files.py checks the 'incomplete' field of the range
in file_map. And range has already considered the shared blocks and
could be smaller in size than the original file range. Therefore, the
'incomplete' flag was set on the original range in common.py; and we
should switch to use the original range also during validation.
I also checked another flag usage in CanUseImgdiff(), and it has
explicitly rejected cases of shared blocks.
Bug: 124868891
Test: unit tests pass
Change-Id: I03959625d7b81fd83420db98f01d23f54064bcd2
Add a flag to write a copy of the metadata to a separate file. Therefore,
users can read the post build fingerprint without extracting the OTA package.
Bug: 124783265
Test: Check the dumped metadata file after generate A/B and non-A/B OTA.
Change-Id: I8918aec87bb81906ef0a7eee774178e9f689d91d
This serves as a workaround to skip the compatibility check for devices
with incompatible kernels.
Bug: 114240221
Test: generate and check the OTA package for wear device
Change-Id: I65b523a66648af7a77fc3ea79176764fe8ae8d02
This script takes as input two partial target files (one contains system bits,
and the other contains non-system, or other, bits). The script merges the
contents of the two partial target files packages to produce a complete target
files package.
Bug: 123430711
Test: Build two partial target files, merge, compare with full target files.
Test: Validate merged target files via validate_target_files.py.
Change-Id: Ic24acf43b86fc703fb4c970688b006291a1861f8
We want to remove target specific host tools and since
fs_config_generate is compiled with a target specific header file, we
instead remove fs_config_generate entirely and allow python to build
the fs_config_files/dirs files directly from config.fs files and
parsed C headers.
Test: associated unit tests and new end to end test
Test: aosp_sailfish, aosp_crosshatch build produces valid fs_config files
Test: aosp_cf_x86_phone build correctly produces empty fs_config files
Change-Id: Idbc63ff56c0979e1e4c17721371de9d9d02dc8ff
This was deprecated along with TARGET_ANDROID_FILESYSTEM_CONFIG_H so
warn if it still exists.
Test: successful error if it exists, no error if it does not
Change-Id: I7dde3c88aaf7ecec23ffe018cbe8b42ebadb2bf8
tools/releasetools/add_img_to_target_files.py: This patch excludes the
inclusion of the system path into the target files if it does not exist (which
it will not if we are not building the system image).
Bug: 123430711
Test: Disable building system image, verify that target files builds without system.
Change-Id: Iaf964ede2b1df5ea4e004b572fd91187a366e75e
This commit extracts the AVB key used to sign system_other.img into
system.img, for init to verify system_other's AVB metadata.
The extracted key will locate in:
/system/etc/security/avb/system_other.avbpubkey
Bug: 123611926
Test: build and checks the following is generated
$OUT/system/etc/security/avb/system_other.avbpubkey
Change-Id: Icdc703ff5a0d50f8140bb652507b9b4cbc8a2118
This change allows ro.product.[brand|device|manufacturer|model|name] and
ro.build.fingerprint to be derived at boot time (and in the OTA
generation scripts) from partition-specific properties.
Test: booted system image, verified properties
Test: booted recovery image, verified properties
Test: unpacked OTA package, verified build fingerprint
Bug: 120123525
Change-Id: Iadd230a0577f35c7c37b0f911e91a5c2863ed1fe
TARGET_FS_CONFIG_GEN has existed as the preferred mechanism for two
releases, so we finally deprecate TARGET_ANDROID_FILESYSTEM_CONFIG_H.
Test: build
Change-Id: I299a4d1d1d893ac16d6e6ce2ec4659bfcdc19095
Add kernel configs / version to verified_assembled_vendor_manifest.xml
so that the kernel of the incoming package can be checked against
the framework. Previously, the running kernel was used instead.
Bug: 111125947
Test: test_extract_kernel
Test: manual OTA on Pixel 3 from build:
Android P (kernel version 4.9.96)
to ToT build:
device kernel version = (manually modified) framework requirement = latest,
PRODUCT_OTA_ENFORCE_VINTF_KERNEL_REQUIREMENTS = true
Change-Id: Id524a58e94bdb6bba348ca461c9d33614ce451a9
For dynamic partitions in retrofit devices, system partition will be
a logical partition but system_other is not. However, current
build system use the same settings (logical) for both system.img and
system_other.img, leading AVB unable to locate the footer from the end
of system_other partition.
This commit support building system_other.img with correct partition size
while building system.img as a dynamic image.
Bug: 123506156
Test: check there is "system_other_size=2952790016" in file
$OUT/obj/PACKAGING/system_other_intermediates/system_other_image_info.txt
Test: build system_other.img, then
`simg2img $OUT/system_other.img system_other.img.raw`, checks the
raw image size.
Change-Id: I748320a7770c694d06f06f4a35bfceb622849aa8
- Add hashtree_info to EmptyImage so that BlockDifference.Compute()
can accept EmptyImage() as target image, which is the case when
a partition is removed.
- BlockDifference also checks source_info_dict to determine
whether a partition is dynamic. When a partition is removed,
its name does not appear in target_info_dict.
- Add tests to ensure DynamicPartitionDifference() still works.
Test: DynamicPartitionDifferenceTest
Test: test_blockimgdiff
Change-Id: Iadb1db075f5dc344db6d5ade358c83b01231e443
This is not used by anyone and the other half of the code to compare
against it is already functionally dead, so remove this.
Test: build
Change-Id: I44ed087cb7735bbc23e30b6c310c80eb3b7b6488
Bug: 111136242
Test: When BOARD_PREBUILT_DTB_DIR is set correctly,
generated $OUT/boot.img contains the DTB image.
Change-Id: I282e31b04cc60383377b9e9b54f8fe64a8140242
Support signing system_other.img but shouldn't include it into the
top-level vbmeta.img. system_other verifiation will not be included
in /vbmeta chains and will be done separately.
Bug: 112103720
Test: avbtool info_image --image $OUT/system_other.img
Test: avbtool info_image --image $OUT/vbmeta.img, checks 'system_other' is NOT included.
Test: Checks $OUT/obj/PACKAGING/system_other_intermediates/system_other_image_info.txt
See the following:
avb_system_other_hashtree_enable=true
avb_system_other_add_hashtree_footer_args=--rollback_index 1551744000
avb_system_other_key_path=external/avb/test/data/testkey_rsa4096.pem
avb_system_other_algorithm=SHA256_RSA4096
Change-Id: Ia152aaab1387dcf556a42222adb39ea76881263a
This commit introduces a prebuilt ELF binaries checker. The checker
will check:
1. Whether all DT_NEEDED shared libraries are specified in
`shared_libs` (Android.bp) or `LOCAL_SHARED_LIBRARIES` (Android.mk).
2. Whether all undefined symbols in the prebuilt binary can be resolved
to defined symbols exported by its dependencies.
This ensures that prebuilt binaries won't silently become ABI
incompatible.
To check the prebuilt binaries, all of the dependencies must be
specified in `shared_libs` (Android.bp) or `LOCAL_SHARED_LIBRARIES`
(Android.mk).
If your prebuilt binaries cannot be checked for some reason, you may add
the following property to Android.bp:
check_elf_files: false,
Or, add the following setting to Android.mk:
LOCAL_CHECK_ELF_FILES := false
Bug: 119084334
Test: CHECK_ELF_FILES=true make check-elf-files
Change-Id: I523d3083f22fd4053c096d26f61f8375800281c8
- When removing a partition, BlockDifference() object
will have tgt=EmptyImage(). Fix the asserts accordingly.
Also, BlockDifference object now allow tgt=None case.
- When adding a partition, BlockDifference() object
will have src=None. Fix the asserts accordingly.
Also, add unit tests to DynamicPartitionsDifference.
Test: create incremental OTA
Test: test_common.DynamicPartitionsDifferenceTest
Bug: 111801737
Change-Id: I3a35378ecf93111b8f44545cff6ae9696b6b4851
We cannot simultaneously stash more blocks than the size limit imposed by
the cache size. As a result, some 'diff' commands will be inevitably
converted to new. We used to do this conversion blindly when iterating
through the transfer list. This leads to an unintended large package.
In order to choose the right transfers to convert, we calculate the size
of the compressed data, and build a heuristic about the package size
increase to remove each stash blocks. After the process, the given
package size for the watch device further reduces from 186M->155M.
In some rare cases, the removed stashed blocks don't directly contribute
to the maximum simultaneously stashed size. For example,
stash A: 10 blocks
stash B: 5 blocks
free B: 5 blocks <-- stash B has been freed before we reach max stashed blocks
stash C: 10 blocks
Converting these blocks lead to an uncertain result. On one hand, patches
are generally smaller than the new data; while on the other hand, the
regenerated graph may have fewer order violation and thus give some size
reduction. But these cases are rare and it seems an overkill to consider all
possible scenarios here.
Bug: 120561199
Test: build non-A/B incrementals and check the size
(p.s. it can be tested on all target files with customed cache threshold)
Change-Id: I599420a91b80f1a1d83d22ee1b336b699050cfb4
To address problems creating real tiny filesystem placeholders,
make sure extra spare inodes margin is greater than 0.
For initial estimate we add 4% of total we request at least 8.
For second pass estimate we add 0.2% of total we request at least 1.
We bumped up the margin for zie on the second pass to 0.3% as the
value was too close on one of the builds.
Test: build
Bug: 122328872
Change-Id: I41707bb6fcc8bbfbdda143a9ce62446cad9c1533
Test: sideload full OTA on cuttlefish
Test: sideload incremental OTA on cuttlefish (that grows
system, shrinks vendor, and move vendor to group foo)
Bug: 111801737
Change-Id: Ie8a267a90b4df9e9e0a2fbcc1b582ab2e353df52
We used to create the intermediate output file as a tempfile, when
adding images to a given zip file. This CL changes OutputFile to write
intermediate files under the given input dir (i.e. OPTIONS.input_tmp
that holds the unzipped files), if the final destination is a zip
archive. This allows image building codes finding the generated images
at consistent locations. The intermediate files will be cleaned up as
part of OPTIONS.input_tmp.
Bug: 122608028
Test: `m dist`
Test: Delete OTA/super*.img from a target_files.zip. Run
`add_img_to_target_files.py -a` that re-generates split super
images.
Change-Id: I97903a59fcc0ca5e43bb9b07c3a0b25e9baa92f9
Bug: 122608028
Test: Run sign_target_files_apks.py on a target-files zip that has split
super images (e.g. OTA/super_system.img).
Change-Id: Iaf7263790961a897ea3f339f5af6b18cf253b946
Allow right-size to support images that are not sparse.
This is in support of cuttlefish.
Test: build
Bug: 122328872
Change-Id: Ic8ef968e750203dffde7044bc6dfc71c1e283158
To address flakiness in image creation, add a .2% margin of safety
for both inodes and space for ext filesystems.
Test: build
Bug: 122328872
Change-Id: I6665bd6fe642291b825dba58dfd09cc55628230f
inode-size will jump around based on filesystem size, however
readonly Android system partitions have a specific pattern of
xattr associated with sepolicy and 256 is the most efficient at
absorbing the content.
Test: manual
Bug: 122328872
Change-Id: I06dd6a503067ab6477293b386d56a89dd86b0e83
We used to build retrofit full OTA package as long as
PRODUCT_RETROFIT_DYNAMIC_PARTITIONS is true. This doesn't work with AOSP
targets that have the flag set but without any available vendor image.
This CL detects such a condition and uses a separate flag to guard the
retrofit OTA building as well as the split super images generation.
Bug: 120852744
Test: `m -j dist` on blueline (w/ vendor images) and aosp_blueline (w/o
vendor images).
Change-Id: I65726f24f8fc546be6802941a6a06590a3804c16
soong_ui is passing the Android.mk file list to Kati, and it ignores
Android.mk files found via symlinks, so we don't need the workarounds
for build/make/target/board/Android.mk and build/target/board/Android.mk
both existing.
We've got dist support in Soong now, so we can remove the signapk
Android.mk
Bug: 122332221
Test: m dist out/dist/signapk.jar out/dist/libconscrypt_openjdk_jni.so
Change-Id: I2ac9016b04146340c5ac80f7090d00e186023574
These replace $USER and `hostname`, which will soon become stable values
due to sandboxing the build on Linux.
Bug: 122270019
Test: check build.prop
Change-Id: I7493e45a2a2defbdba5d82596cb91d68480f0187
This is just a hash of the current.txt and system-current.txt right now,
though it may expand to include other API surfaces in the future.
Once prebuilts/sdk/current is populated with api_fingerprint.txt files,
we can use those for unbundled builds, but until then, just use the
PLATFORM_PREVIEW_SDK_VERSION as a placeholder.
MD5 was just the most convenient, since we don't have a sha tool that we
can use on Mac currently. I'm hoping we can get a toybox-based tool in
the future that standardizes that, but we aren't currently using sha*sum
from toybox on Linux, much less on Darwin yet.
Test: m dist out/dist/api_fingerprint.txt
Test: m out/target/product/.../system/build.prop
Change-Id: If69f270560d05135cb81a9bb2d1b208ea78f86df
Package size will be unintended large if we stash more blocks than the
stash limit specified by the cache size. To reduce the maximum size of
simultaneous stashed blocks, we will inevitably convert some 'diff'
commands to 'new' commands.
To mitigate the impact, we add a new function to smartly select the
transfers to convert based on their patch size and compressed size.
This cl converts the transfers that have a larger patch size than the
compressed target sizes. And there's a slightly improvement in the
final package size: from 194M -> 185M.
Bug: 120561199
Test: build a non-A/B incremental package, run simulator
Change-Id: Id73ff736ba4e6901d245ad5549d42310d0740284
This change excludes retrofit devices from getting a third metadata
slot, until code is in place to correctly upgrade those devices through
fastbootd/update_engine.
Bug: 121149247
Test: lpdump super_empty.img
Change-Id: Id147c42cee93a1f6d61ee0edc9cddc0c0540a279
The current increase of 4 % is insufficient for some
targets.
Test: mmm
Bug: 119115481
Change-Id: Idcba8025b913da9b70794bfc7464d15b4d99ad34
(cherry picked from commit 3e02e34090)
Dan Willemsen