Currently, we're building against the native host version of libc++ and
other headers. Instead, use prebuilts of ckati/makeparallel built with
Soong against the included libc++ and sysroot. Building ckati /
makeparallel doesn't work well in unbundled branches that don't have all
of the libc++ dependencies to do successful host builds.
KATI is default variable defined when running under Kati, don't use it
for other purposes. Use CKATI instead.
Bug: 27533820
Change-Id: I82d66f5e34e3341a3cee4c5c83cd1d9a54d057ff
Build binaries usable on older machines even if older SDKs are not
installed. Older SDKs can no longer be installed on newer Xcode
versions.
Change-Id: I0c9f2534466a127a19107820879c2856bfac0076
We've been including the system libc++ headers even if we're building
against our version of libc++. Stop doing that, and only add the headers
to our path if we're using the system libraries.
If nothing is specified, on recent OSX versions, libc++ is the default
c++ library instead of libstdc++. We've been explicitly including the
libc++ headers on all versions, but that breaks old versions. Force us
over to libc++, since the system libstdc++ does not support C++11, and
libc++ is still supported on our oldest version (10.8).
Change-Id: I1fccee8da0f425e10ccc9d3247ed40664eb6ada0
/Developer was removed in Xcode 4.3. Our minimum SDK version is 10.8,
which implies at least Xcode 4.4, and our documentation requires
Xcode 4.5.2.
Change-Id: Ib343df6ded6e98222d8ee2e542e1f3fadd2b1397
Bug: http://b/27482631
This fixes an issue where krait compiles were not passing the proper
flags to the assembler when -no-integrated-as is used. This was
preventing ART from switching to clang in some cases.
Change-Id: I24966bf1ff84305df3586ab01d1aae7d5922aa11
When a static library uses LOCAL_EXPORT_C_INCLUDE_DIRS, then is included
entirely in another library with LOCAL_WHOLE_STATIC_LIBRARIES, all the
symbols are reexported, so reexport the headers too.
Change-Id: I92cf17894fc991a5b5ecb59ca5e095e407c98de4
1. Combined ~10 calls to the make build system to only one.
We added a phony target "dump-many-vars" to the build system to dump
"<var>=<value>" pairs. We then store the pairs as shell variables.
With this cache get_build_var/get_abs_build_var can just return
the shell variables instead of querying the build system.
2. Prune .git when we search for AndroidProduct.mks.
In internal source tree lunch time was reduced from ~15s to ~1.5s.
Bug: 27429759
Change-Id: I24e88598f6fab598ef26518885fd5e86e71a772d
- For host Java libraries, java-lib-files should return javalib.jar.
- Host dalvik Java libraries are special: factor out
host-dex-java-lib-files.
- Be explict that jack-lib-deps equals jack-lib-files.
Bug: 27451686
Change-Id: I9235384354e119ef7ebbf29b7e525d1ceea242e0
(cherry-pick from commit 063d0455e0)
This will allow us to convert icu4j-icudata / icu4j-icutzdata
targets to "host" targets rather than "target" targets.
Right now they are forcing Jack off just so they can use
javac + jarjar (even though the resulting file is used on
host).
Bug: 27218410
(cherry picked from commit 30b21ad579)
Change-Id: I4f16896606bcfdd00164078eeb142d05b29bb832
Also specify "-d keepdepfile" to ninja so .P files won't be
deleted by ninja.
-include for .s files are removed because GCC doesn't
generate .d files for .s files.
Bug: 26839129
Change-Id: If00e93c7a33449ec314a5cdba438475a32979f4e
Similar to how we clean the generated source dir if .aidl becomes
obsolete.
Bug: 26688474
Change-Id: I66c5b6d6513259b5f4f4bac84f384f4bf953d82a
(cherry-pick from commit d5ffec9383)
So that we can disable only one in multilib modules.
Bug: 27442756
Change-Id: I4ca379fac997f9165c47cb93d34bf1f483f5a241
(cherry-pick from commit b6eb8ddbde)
am: a65b01fc09
* commit 'a65b01fc09b47c83e59dc4757c76eafba1e15cbc':
Remove LOCAL_ACP_UNAVAILABLE
Remove unused dependencies on $(ACP)
Replace some uses of acp with cp
Now that copy-file-to-target doesn't use acp, nothing in the acp build
path uses acp, so we don't need to special case it to prevent loops.
Change-Id: I12810c1b064d0c03135a80077a76bc4c9cc18b24
ACP was originally created to overcome differences with cp between
Linux, Darwin, and Windows. We've since dropped Windows as a build host,
and don't use features like '-u' anymore.
For most of our current usecases, 'acp' is identical to 'cp' except that
it splits 'cp -p' into 'acp -p' for mode and ownership, and 'acp -t' for
timestamps. On Linux, this could be specified using
'--preserve=mode,ownership', but Darwin doesn't have this.
Since we're removing the destination file before copying, 'cp' already
preserves the mode (modulated by the umask). So the only extra thing
that gets preserved with 'acp -p' is ownership, which we should not care
about in the build system. (In many cases we shouldn't be preserving
mode either, so that readonly source trees can actually be marked
readonly, but that will be a future change)
Change-Id: Ied96fdc303ac5c774347c07363daec8b6dfb22e4
Run jack with no outputs as a quick check for compilation errors and use
a timestamp to store that check was made.
Bug: 19069325
(cherry picked from commit 43084d9f49)
Change-Id: I9b84b503b28cfdfa245f91da0061ee3a79386b28
If there's a symlink as the destination to one of these macros,
currently we'll write to the destination of that symlink instead of
overwriting the symlink. We've run into this a few times when a module
is added to replace a symlink that used to exist via
LOCAL_POST_INSTALL_CMD. These have required manual discovery, and
additions to CleanSpec.mk files:
http://android-review.googlesource.com/143334
Use `rm -f` for single-file targets to remove the destination before
copying. On Linux, `cp --remove-destination` can work, but is not
supported by Darwin or acp.
There may still be problems with dependencies when symlinks are
involved, since ninja will use the destination of the symlink to check
whether it is up to date. But at least with this change, if any
dependency gets regenerated, we'll properly reset the file.
Change-Id: I6d3ac0bd9ced5e21a0ff9dad0eaff012a7bc9c75
When switch build types or products, it's no longer necessary to special
case these projects that change their command line based on the build
type or product. Ninja keeps track of the command line last used to
create a file, and will mark it as dirty if the new command line is
different.
Change-Id: I905ff9599eae2952bddc05e7328f77f0849be20a
Places whitelisted brillo tests and the whitelist itself
in a zip when run.
BUG: 27385399
Change-Id: I93c2ea8cc521292a6de811bb47bc87a727edd21f
TEST: manual make dist brillo_tests, confirmed desired files were in zip.
When the build type changes (for example, from "shamu-userdebug"
to "shamu-user"), the build system doesn't delete all files
and start over. Rather, build artifacts from the old build type
are reused for the new build type.
This is problematic for the recovery SELinux policy, which differs
between build types. Reusing a userdebug policy on a user build
is inappropriate and could lead to security bugs.
Force the deletion of the recovery SELinux policy when changing
build types, so it can be properly regenerated. This is consistent
with how we treat the normal SELinux policy (see commit
a8b3d54101).
Change-Id: I4ebafe3712dc121644828f6538865061aad58cc0
And everything special-cased on that. Add a warning if USE_NINJA is
set to let users know that it no longer changes anything.
Change-Id: Ib8739151fe26ea6bf8f76b7ac2b8f4097dab0b47
Modify the compiler flags for Jack and javac.
This has the following effects:
1) Generally, some of the type inference rules changed.
2) javac: bytecode is generated with the v52 major version (not v51)
3) jack: Java 8 language features are supported.
The javac / dx toolchain does not support Java 8 language features.
Bug: 26753820
(cherry picked from commit fda1ace26116a6677cc77c92c24e5259817fb86e)
Change-Id: I07769de473775d95b13feb38c0eb37086eb120f7
There are some code policies we want to enforce more strictly, but
it's hard to do so for third party code because we have to either
carry the diff burden or upstream the patch, and in the latter case
the turnaround time for fixes can be problematic, and sometimes
upstream won't accept changes (sometimes people just need to win the
obfuscated C contest).
We define ANDROID_STRICT for any code that we expect to be able to
make these policy fixes as we change policies.
Change-Id: I15faf62cec1932dd859a082f66942b2606d0ff45
It's a valid situation for all three of LOCAL_INIT_RC, LOCAL_INIT_RC_32,
and LOCAL_INIT_RC_64 to be used.
Bug: 26773181
Change-Id: If9661f93b1823279075fc3d55195f7a939e01b6f
Sami Tolvanen