This change make it possible to specify separate rules for native-bridge
case and non-native-bridge case.
Bug: http://b/153609531
Test: build 4arch product
Change-Id: I60145bbb9c94318f14af3cabd9f5960fc9ee62de
Updates the member snapshot creation code to support multiple os types.
It basically sorts the variants by os type, then applies the code to
optimize the arch properties and then it optimizes the properties that
are common across architectures and extracts any properties that are
common across os types.
The java and cc member types needed to be modified to make the location
of the generated files within the snapshot os type dependent when there
is more than one os type. That was done by adding an OsPrefix() method
to the SdkMemberPropertiesBase which returns the os prefix to use when
there is > 1 os type and otherwise returns an empty string.
Added three tests, one for cc shared libraries, one for cc binary and
one for java header libraries.
Bug: 150451422
Test: m nothing
Change-Id: I08f5fbdd7852b06c9a9a2f1cfdc364338a3d5bac
Adds a CommonOS variant for sdk that depends on the os specific
variants and is used to generate a single sdk for multiple OsTypes,
e.g. host linux and android.
At the minute the member types only support a single OsType but the
basic mechanism for managing the CommonOS variant and collating the
variants across all of them is there.
The only visible effect of this change is that the location of the
generated snapshot is changed, it is no longer os specific and instead
is in the same location irrespective of which os it is built for.
A lot of tests needed to be changed to specify "common_os" as the
variant type instead of the specific os type. As that is the same across
all tests it is hard coded in CheckSnapshot method.
Test: m nothing
Bug: 150451422
Change-Id: If36be39b06d6910453649f7c288c2d34f688b2f4
Rename the two prefix-in-list funcs so that their usages are clearer.
Also find and replace all the code that essentially does either.
This introduces additional loops in some places, but I think the added
readability and simplicity outweighs the performance degradation, which
should be negligible anyway.
Test: m nothing
Test: TreeHugger
Change-Id: I37e2276ca0c815105ed0031f23c0b1264b480e4f
This implements four modules (static/shared/header libraries, and
binaries) for vendor snapshot. These modules will override source
modules if BOARD_VNDK_VERSION != current.
Bug: 65377115
Test: 1) VNDK_SNAPSHOT_BUILD_ARTIFACTS=true m dist vndk vendor-snapshot
Test: 2) install snapshot under source tree
Test: 3) set BOARD_VNDK_VERSION and boot cuttlefish
Change-Id: I24ddb4c3aa6abeab60bbfd31bcbd8753e2592dc5
proptools.FilterPropertyStructSharded now takes a desired max size
of the name.
Fixes: 146234651
Test: all soong tests
Change-Id: Ifa10039e5f55a5d0f4a35172983d43f33af88e93
Remove more android struct tags from runtime created arch structs
to reduce the size of their names, which hit the 64kB limit in
runtime.StructOf if they are too long.
Bug: 146234651
Test: m checkbuild
Change-Id: I6362765275b93c8932eb0b1abbcb4be47031d9b1
Instead of creating a runtime generated struct type:
struct {
Arch struct {
Arm struct {
Armv7-a struct {
...
}
}
...
}
Multilib struct {
Lib32 struct {
...
}
...
}
Target struct {
Android struct {
...
}
...
}
}
Replace it with a top-level struct of empty interfaces:
type archPropRoot struct {
Arch, Multilib, Target interface{}
}
And then embed nil pointers to smaller structs inside it:
struct {
Arm struct {
Armv7-a struct {
...
}
}
...
}
This reduces the maximum number of times the archecture specific
properties are embedded in a runtime generated struct from 139
to 97, which makes it easier to avoid hitting the runtime generated
struct name size limit of 64kB.
Bug: 146234651
Test: all soong tests
Change-Id: I901c434ac9a1d99c74fc2370c5e6803b3682dc73
prebuilts.
This runs Soong in skip-make mode, using normal in-make mode only to query
platform versions.
The same ${OUT_DIR} cannot be used for both skip-make and in-make builds,
because Soong generates a smaller build.ninja file in in-make builds where
many build targets are expected to be provided by the mk files. Thus this
script avoids using ${OUT_DIR} if it's an in-make build, defaulting instead
to out-aml/.
The script is based on build-ndk-prebuilts.sh, but uses a separate Soong
variable Aml_abis to enable the appropriate target architectures for
Mainline modules. Aml_abis is very similar to Ndk_abis, except "armeabi-v7a"
is used instead of "armeabi", which is necessary to match prebuilt
dependencies, e.g. for LLVM.
Test: build/soong/scripts/build-aml-prebuilts.sh libart libdexfile_external
(verify that libraries for arm, arm64, x86, x86_64 are built)
Test: build/soong/scripts/build-aml-prebuilts.sh \
out-aml/soong/.intermediates/external/conscrypt/conscrypt-module-sdk/android_common/conscrypt-module-sdk-current.zip
(verify that the zip file contains libconscrypt_jni.so's for all four arches)
Test: build/soong/scripts/build-aml-prebuilts.sh com.android.art.{release,debug,testing,host}
(verify that the build completes)
Test: Two identical build/soong/scripts/build-aml-prebuilts.sh runs after each other
(verify that the 2nd run completes both Soong and ninja steps quickly without any building)
Change-Id: I35712f9f8f0b1cbb77107314c5927c6720e6c3bf
The logic in cc.Module.ImageMutatorBegin() to skip creating vendor
variants for native bridge modules stopped working when the image
mutator was moved before the arch mutator because the native
bridge variant hadn't been created yet. Move the logic to the arch
mutator to avoid creating native bridge variants for vendor variants.
Fixes: 146486214
Test: lunch aosp_cf_x86_pasan-userdebug && m SANITIZE_TARGET=address
Change-Id: I46cff4905f929c9989d005f0ceb01388d47323c1
Run the imageMutator between osMutator and archMutator so that
different arch variants can be set for the different partitions.
Bug: 142286466
Test: m checkbuild
Change-Id: I65d05714b75aa462bf9816da60fdc2deda4de593
Merged-In: I65d05714b75aa462bf9816da60fdc2deda4de593
(cherry picked from commit 9c8f687584)
Split the archMutator into two mutators, osMutator to create
android/linux/darwin variants, and archMutator to then further
split them into arm/arm64, etc. When combined with alias
variants in Blueprint this will allow us to depend on the
"android" variant without having to know what architectures
were used to build it.
Test: m checkbuild
Change-Id: I0f51d60fc440cf5ae503904ce7849ec797af5fe2
The arch variants are hardcoded in every module type. Refactor
them out into a Target.Variations() method in preparation for
splitting the arch mutator into two, which will require using
different variations.
Test: m checkbuild
Change-Id: I28ef7cd5168095ac888fe77f04e27f9ad81978c0
Translated second architectures now go in NativeBridgeArch instead
of DeviceSecondaryArch.
This reapplies I568046330abc002d4eed582cb999b62a5eaba790 with
ctx.Config().HasMulitlibConflict() added to fix the NDK build,
which has arm64, arm, x86_64, and x86 architectures enabled.
Test: m checkbuild
Test: OUT_DIR=out_ndk build/soong/scripts/build-ndk-prebuilts.sh
Test: no change to build.ninja or Android-aosp_cf_x86_phone.mk
Change-Id: Iadcafbd64bfb9579ae7c86914927c43a062b0c8e
This reverts commit fe6330c7d2.
Reason for revert: this was accidentally propping up NDK builds
Bug: 141242600
Change-Id: Ie9f8e31479a74e8e9e14d720a5936da6b1d5d113
Translated second architectures now go in NativeBridgeArch instead
of DeviceSecondaryArch.
Bug: 141242600
Test: m checkbuild
Change-Id: I568046330abc002d4eed582cb999b62a5eaba790
This change introduces NativeBridgeRelativePath and
NativeBridgeSecondaryRelativePath product variables to
make relative path to native_bridge binaries configurable.
It also removes moves information about native bridge
host architecture to Target and sets it during decodeTargetProductVariables
Test: make PRODUCT-cf_x86_phone-userdebug dist
Change-Id: Ie736e81eae507e1775566ce9f29135011b12af27
I3b23590b8ce7c8a1ea1139411d84a53163288da7 renamed AndroidModuleBase
to ModuleBase, but left the receivers called 'a'. Rename them to
'm' to match the type.
Test: m checkbuild
Change-Id: I0ebb738152febfd9a2a56ac0b197f19f9e01a1e7
Ignore native_bridge targets when detecting whether second
architecture is translated or not.
The concept of second translated architectures is almost deprected
and will be removed once all affected products are switched to
native_bridge support from the build system.
Bug: http://b/77159578
Test: make 4arch target
Change-Id: Ife93ccc60f5c1a4020f66b2cf862b709889d9d6b
This allows us to build guest libraries for the native bridge for
arm/arm64 architectures.
Bug: http://b/77159578
Test: make
Change-Id: I35520ca456105ddadd456c78a4eb1e6de39147c5
This patch adds new arch features avx2 and avx512 and
their supported archictectures for x86 and x86_64
Bug: 123376719
Test: m checkbuild
Change-Id: I3c079741815b39d7dbb4072e12ef7c9c2c15f5fe
Signed-off-by: Shalini Salomi Bodapati <shalini.salomi.bodapati@intel.com>
Unfortunately we still need to build the NDK sysroot as non-NEON
since that's still supported, though it's no longer the default.
This reverts commit f4e0601c7e.
Test: treehugger
Test: rebuilt NDK sysroot and imported into the NDK, ran NDK tests
Bug: None
Make ArchType implement the encoding.TextMarshaller and
encoding.TextUnmarshaller interfaces so that it can be used
as a value in the dexpreopt config structs that are passed
through JSON files.
Test: m checkbuild
Change-Id: Ie4c12443e7ee5fe43f42d5403bcde12d62f617e2
Add an opaque OnceKey type and use it for all calls to Once in
build/soong. A future patch will convert the arguments to
Once* to OnceKey once users outside build/soong have been updated.
Test: onceper_test.go
Change-Id: Ifcb338e6e603e804e507203c9508d30ffb2df966
There is no need for *[]string, []string can already hold a nil
value to specify "not set".
Test: all soong tests
Change-Id: I85a51b042c12aee1565a9287d62924feeeafd486
Runtime-created struct types have limit on the name of the type,
which is the full text of the fields in the struct. Avoid creating
runtime struct types when filtering produces an identical struct.
Also add tests and godoc for filterArchStruct.
Test: arch_test.go
Change-Id: If7eb27ea9e72073d5252e93f2110b08889ed76b1
This change adds Fuchsia as a valid OS. Future changes
will add proper toolchain support.
Bug: 119831161
Test: Compile walleye. Confirm that adding `fuchsia` to the
os-specific declaration of an arbitrary target does not throw
a build error.
Change-Id: I64eb928afb7512f3fbe32abb313b4c3efe16b169
This enables using arch-dependent property values in modules defined in
android/.
Bug: 122332178
Test: Soong tests + TreeHugger
Change-Id: I89869e395fabf0e69f505b77eab8a4221384124e
The minimum set of supported features are different between Android and
Linux with X86_64. So while the list of variants and features are still
technically correct, and may be shared more in the future, the default
needs to be configured differently.
Bug: 120208462
Test: add `host_supported: true` to libopus; m libopus
Change-Id: Ib79707d17b852f77341026085230c026b6386b78
Support Kryo 385 cpu variant which should be used for
Qualcomm Snapdragon 845.
Motivation:
kryo385 CPU *IS NOT* an Arm Cortex-A75.
Particular problem:
Cortex-A75 CPU must support optional ARMv8.2 dot product
feature; kryo385 CPU (Qualcomm Snapdragon 845) doesn't
support it.
Test: Builds when kryo385 is used as cpu variant.
Test: Pixel 3 boots to GUI with kryo385 as CPU variant
with this patch on AOSP 9.0.0 r16.
Bug: 119564566
Change-Id: I62ffb46b1977b48446c6c1ca1400b1b39f7a8457
Arch property struct types are created at runtime. Go has a limit
of 2**16 bytes for the name of a type, and the type of a struct
created at runtime is a string containing all the names and types
of its fields. To avoid going over the limit, split the runtime
created structs into multiple shards.
Also undo MoreBaseLinkerProperties now that it is no longer
required.
Bug: 80437643
Test: m checkbuild
Test: no change to out/soong/build.ninja
Change-Id: I035b20332ec63f3d4b1696855c5b0b0a810597b7
Add a missing toolchain method, don't require modules to opt in to
linux_bionic.
Bug: 31559095
Test: attempt to build host bionic
Change-Id: Ia5c8e2f127b88f56191b7abc389d5329cfdb0720
When building with two OSes of the same class (Host, etc) enabled,
independently produce variants of each module for each OS, instead of
making `compile_multilib: "first"` select just one of the OSes.
My use case this is for Linux + LinuxBionic, which are both considered
Host class, but this would also applied if we wanted to generate rules
for Linux+Darwin, or some combination of firmware bits (either in
addition to Android in the Device Class, or multiple firmwares in a new
Class)
Bug: 31559095
Test: enable host bionic, see a linux(_glibc) and linux_bionic
definition for binaries
Change-Id: I55e7f046758b3dc2a31da2c4b20274ec68ece0a0
Elliott Hughes