Following 99bdb2ab4fd5168a71a20aecf10611425be47ec4 in build/blueprint,
move more methods into BaseModuleContext.
Test: m checkbuild
Change-Id: I9f8df94f1ae2b55d3cccf7b9468247f3e7cb2ebd
This reverts commit 6d8c0a50bc.
The switchover to lld appears to be making adb crash on exit for unclear
reasons. Revert it as a quick fix pending further investigation.
Bug: http://b/134613180
Test: adb.exe under windbg
Change-Id: Ibdf5981c3c6828b3684974b7c30a7c96449c64c5
blueprint.BaseModuleContext is the set of methods available to all
module-specific calls (GenerateBuildActions or mutators). The
android package split the same functionality across baseContext (nee
androidBaseContext), BaseModuleContext, and BaseContext.
Consolidate all of them into android.BaseModuleContext.
Test: m checkbuild
Change-Id: I2d7f5c56fd4424032cb93edff6dc730ff33e4f1e
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
This jlink plugin is intended to optimise startup times for Java
runtimes by embedding a pre-processed and pre-validated module graph
in the system image.
This provides no benefit on Android, since the Android runtime does
not make use of the module graph it produces. (It doesn't seem useful
on Android anyway, since the system image only contains one module,
namely java.base.)
Furthermore, the plugin causes the jlink invocation to fail when using
the jlink tool from OpenJDK 11. The issue here is the plugin uses
classes in the jdk.internal.module package to describe the module
graph; that package is not part of libcore and therefore not listed in
the module-info.java for java.base on Android; but the plugin has the
side-effect of adding the package to java.base; this causes jlink to
subsequently fail with an error "Module java.base's descriptor
indicates the set of packages is : <lots of packages>, but module
contains packages: <same packages plus jdk.internal.module>". (The
implementation of the plugin changed significantly in OpenJDK 10,
which is presumably why this issue does not occur using OpenJDK 9's
jlink.)
Therefore, it is safe and beneficial to disable the plugin.
Test: EXPERIMENTAL_JAVA_LANGUAGE_LEVEL_9=true make core-all-system-modules, using an OpenJDK 11 toolchain via OVERRIDE_ANDROID_JAVA_HOME and changing the jmod create invocation to use --module-version 11
Test: EXPERIMENTAL_JAVA_LANGUAGE_LEVEL_9=true make droid, flash to device & sanity check
Bug: 131683177
Change-Id: I52333f32c88aa85cd3652ad91d50d9927ff61daf
The system_modules defn used with core.current.stubs
was core-system-modules which doesn't match. Instead,
this change uses core-current-stubs-system-modules.
Test: build
Change-Id: Ifb9e77b02bbb04c1cb39c6b9f97932c20d44d8ba
As of [1], the Google format, which we default to, uses IncludeBlocks:
Regroup, which sorts all includes with <> and without .h into the top
group reserved for libc headers, e.g.:
+#include <android-base/properties.h>
+#include <android-base/strings.h>
#include <ctype.h>
#include <stdlib.h>
#include <sys/system_properties.h>
#include <iostream>
-#include <android-base/properties.h>
-#include <android-base/strings.h>
This is accurate to the Google C++ style guide, which states that
other library headers should be included with "", however we have a
strong tendency to use <> for these headers for Android, so we must
revert this change.
Note, clang-format allows regexes for creating the blocks as well,
however, short of including all possible libc headers, I don't think
there's a way to differentiate between <sys/wait.h> and
<other_lib/header.h>.
[1]:
62e3198c4f (diff-b689ebff2e51e675b346a88289e74542)
Test: clang-format doesn't do the above.
Change-Id: Ifef605ffd6b4c8a36512377644e64ddd26b68a21
There are a couple of makefiles that have etc with $(TARGET_OUT)/etc/firmware,
$(TARGET_OUT_VENDOR)/firmare as the local module path. Added the conversion in
androidmk for target.
Bug: b/133711895
Test: Unit test cases.
Change-Id: I74aca18a7eb8aa4dd692cde680fa4e112edcf6d1
When it is set to c++_shared, the added dependency should be
libc++_shared, not libc++.
Fixes: 130891985
Test: app_test.go, atest 'CtsNdkBinderTestCases' w/ conversion CL
Change-Id: I0e50e0d5bf511aa6e52d16cd715a185721011255
Allow java modules to be referenced as ":module{.jar}" to get
the classes .jar.
Bug: 122333007
Test: m checkbuild
Change-Id: I2d07f774ae8ed5a39206059c4102727c6802d26e
There are cases where a module needs to refer to an intermediate
output of another module instead of its final output. For example,
a module may want to use the .jar containing .class files from
another module whose final output is a .jar containing classes.dex
files. Support a new ":module{.tag}" format in any property that
is annotated with `android:"path"`, which will query the target
module for its ".tag" output(s).
Test: path_properties_test.go, paths_test.go
Test: no unexpected changes in build.ninja
Change-Id: Icd3c9b0d83ff125771767c04046fcffb9fc3f65a
no_libgcc is no longer needed anywhere. Move all occurances to no_libcrt
and remove no_libgcc.
Test: build
Change-Id: I6dd49db71d05d7685aa90cc837627f65e6742d6d
Sort the list of libraries when iterating over the vndk modules
map to avoid a non-deterministic output file.
Test: m nothing && mv out/soong/build.ninja /tmp && m nothing && diff -u out/soong/build.ninja /tmp/build.ninja
Change-Id: I889736715dab491bb7e69f3499ab1a5e2c876171
apexDepsMutator can be called on multiple apex modules in parallel,
and then two goroutines could call BuildForApex on the same module
in parallel, leading to a data race appending to apexVariations.
This also results in random ordering of the entries in
apexVariations.
Hold a mutex around appending to apexVariations, and sort it before
passing it to ctx.CreateVariations.
Fixes: 134425751
Test: m nothing
Change-Id: If5a3b53a778daacb3e26ac05cde872cf8eb980b3
There are several firmware files that are being installed to
system/etc/firmware or vendor/firmware. A new module type is
defined to install firmware files to install in the system
or vendor image.
Fixes: b/133711895
Test: Ran lunch, wrote unit test cases, created a sample Android.bp
with defined prebuilt_firmware Soong and ran mma.
Change-Id: Iaa791cdda4402936037c5a456cc8bf8e6c905b3e
The new package manager behavior requires packages without source code
to have an application element with hasCode attribute set to false in
their manifest. With this change, Soong can now automatically insert one
for codeless apps.
Test: app_test.go, manifest_fixer_test.py
Fixes: 124375490
Change-Id: Ied89a8d07c63805ab910859a4f7c45fc1c60bb73
This reverts commit d71dc6b5e0.
Reason for revert: ag/974065 fixes the problem that caused the revert
Change-Id: I3ba9eecdab34602aaf19fa518520ac932c9b97af
I'm not sure if these sections are used, but by adding them to
libgcc_stripped.a, we can end up with multiple .gnu_debuglink sections in
an executable, where the first two unhelpfully point at "libgcc.a". e.g.:
$ readelf -SW -x.gnu_debuglink out/target/product/walleye/system/bin/bootstrap/linker64
...
Section Headers:
...
[18] .gnu_debuglink PROGBITS 0000000000000000 10f19c 000020 00 0 0 1
...
[22] .gnu_debuglink PROGBITS 0000000000000000 18a2c8 000010 00 0 0 4
...
Hex dump of section '.gnu_debuglink':
0x00000000 6c696267 63632e61 00000000 3d5f65d8 libgcc.a....=_e.
0x00000010 6c696267 63632e61 00000000 3d5f65d8 libgcc.a....=_e.
Hex dump of section '.gnu_debuglink':
0x00000000 6c696e6b 65723634 00000000 f85db7ea linker64.....]..
Bug: none
Test: m libgcc_stripped linker
Test: no .gnu_debuglink section in libgcc_stripped.a
Test: only one .gnu_debuglink section in system/bin/bootstrap/linker64
Change-Id: I0c47a9acd6d52f5cbc5bfa8b67d072ebb075a242
