This allows SourceProvider modules to create rust_library variants so
that generated source can be referenced as an external crate rather than
via an include macro. This is done by including rust_bindgen modules
like any other library, as a dependency in either rlibs, dylibs, or
rustlibs.
This renames the stem and flags properties for rust_bindgen modules to
source_stem and bindgen_flags, respectively. This deconflicts with the
usage in baseCompiler.
This also removes 'subName' from the Module struct and moves it over to
SourceProvider, which was the only user. This allows us to set it in
baseSourceProvider's AndroidMk; setting it in Module's AndroidMk was
causing problems finding NOTICE files for bindgen library variants.
Bug: 159064919
Test: New Soong tests pass.
Test: Local test rust_binary can use rust_bindgen module as a crate.
Change-Id: Ieb2cb614c2dd0b5aa7120541d77f6f822a6a1806
rust_library modules no longer produce "link" variants alongside
"rust_libraries" variants as the former have been moved to rust_ffi
modules. rust_library modules however still create empty link variants
and code paths still assume that modules can support both rust linkage
and cc linkage. This adds unnecessary complexity that no longer serves a
purpose.
This cleans this up by removing the unused "link" variant for
rust_library modules and simplifies code paths that can now assume
modules are either a rust library or a cc library, but not both.
This also fixes a bug where Shared() was returning the wrong value.
Bug: 159064919
Test: cd external/rust/; mma
Test: cd external/crosvm/; mma
Change-Id: I3b5498c80b315c56b621dcb1388022fecc1bfc1b
Previously, the prebuilt mutators were added by the
cc.RegisterRequiredBuildComponentsForTest() function as a convenience
but unfortunately it lead to some of the mutators being in a different
order in the tests than in the normal build.
This change:
* Extracts the RegisterPrebuiltMutators() call from
cc.RegisterRequiredBuildComponentsForTest()
* Makes sure that the prebuilt mutators are registered before the
visibility gatherer and enforcer mutators.
Bug: 162505935
Test: m nothing
Change-Id: I7d959b558200b502f0a5e4653c41ea01414e142a
dependencyTag uses a set of predefined tags to identify different types
of dependencies. There are already multiple bits of metadata stored
in the dependency tag (Library, Shared, ReexportFlags), and supporting
them all requires a combinatorial explosion of predefined tags and
causes issues when using equality comparisons if a new bit of metadata
is added.
Add a new libraryDependencyTag type that will contain the metadata
bits, and replace the quality comparisons with checks on the metadata
bits.
There are 5 TODOs where modifying the checks identified problems with
the existing checks. These were left in place to produce identical
build output and will be fixed separately.
Bug: 162437057
Test: no change to build.ninja or {Android,make_vars,late}-${TARGET_PRODUCT}.mk
Change-Id: I72d4207dcf381c07c92e00e5a03968ebb5ed8d30
rust_bindgen was not hermetic previously as it would pull in host
headers for device targets. This fixes that by using the same flags we
use when compiling with Clang. This also makes sure our rust_bindgen
headers are built as similar as possible to their respective
cc_libraries.
This also pulls in the bionic dependencies as well, which provide the
headers required for device targets.
Bug: 162007475
Test: device rust_bindgen deps file does not reference host headers.
Change-Id: I4efdf333e011a6c6d73a0345e5485823f166d17a
This makes debugging easier when the bindings contain errors.
Bug: 161730753
Test: Generated bindings are formatted.
Change-Id: Ic46f69d8b6d07b4dc6912b7f02d7427ea6c11cde
Modules defined in the srcs property are automatically added as
dependencies with AddDependency(), which will use any variant available.
This can cause incorrect architecture bindings to be silently pulled in,
such as when a host module uses a rust_bindgen module that doesn't
create a host variant.
This moves populating depPaths.SrcDeps over to depsToPaths and adds a
check for SourceProviders to make sure the correct OS and architecture
is being used.
Bug: 161826371
Test: Soong no longer silently pulls in bindings for the wrong
target.
Test: New Soong test to catch this case passes.
Change-Id: I2b3651cf6fc7dabf4081434df1c455e637f5b3a4
clippy was not being provided generated sources previously, so it would
error out on being unable to locate the sources.
Test: Clippy no longer errors out on generated sources.
Bug: 161748694
Change-Id: I85907327ca40c82a9520e0c0dbd643bfba6b7315
Allow rust modules to depend on and use generated source code provided
by SourceProvider modules and genrule modules without resorting to
hardcoded output paths.
All generated sources are now copied to a dependent module's
intermediates directory, then OUT_DIR is set to point to that path when
calling rustc. This matches the common convention used in most rust
crates to include generated source code from the path defined in the
OUT_DIR environment variable.
A couple other small notable changes are included in this CL:
* prebuiltLibraries can no longer include generated source files as they
should be prebuilt.
* srcPathFromModuleSrcs now excludes the main source file from the
second return value so its a list of only the generated sources.
Bug: 159064919
Test: Local example rust_library compiles with rust_bindgen dependency.
Test: Local example rust_library compiles with genrule dependency.
Test: Collision detected when multiple providers produce similar output.
Test: New Soong tests pass.
Change-Id: I59f54a25368c680b9086420c47ec24ab8cd1de6b
Add SourceProvider modules which provides a base interface for more
complex code generation usecases such as bindgen. Also adds the
rust_bindgen module type which calls bindgen to generate Rust FFI
bindings to C.
Bug: 159064919
Test: Local test module generates bindings.
Test: New Soong tests pass.
Change-Id: Ie31467bbbe423497666ad837cf5fe1acd1e76bd8
This behavior is non-hermetic, but matches the behavior of the C++
support. On Linux, using lld works, but on OSX, this fails because lld
does not support Apple's new .tbd files.
Bug: 155302034
Test: On both Linux and Mac hosts:
cd external/rust; mma
Change-Id: I0ad489113d720bdb9c3b7a67cce9d1e72266f428
rustc expects Rust dylibs to have a .dylib extension on darwin.
Bug: 155302034
Test: rust_library_host outputs have correct extension on darwin.
Change-Id: I17b0d7eaf0f42f8186185e3f14c3a8e992c74fde
* "srcs" list contains one main Rust source file,
followed by optional dependent modules.
* A dependent module included in the "srcs" list is
the module name prefixed with ":".
* Add a simple test.
Bug: 160331255
Test: make and manual test build dependencies on genrule modules
Change-Id: I4f079138c2599158810b6412fce81b612a3f64a4
With proper prebuilt modules, we can avoid any rustc implicit sysroot
searching.
Asd a bonus, this should make rust-project.json generation correctly
grab otherwise implicit dependencies.
Prebuilt rlibs may include several dependency rlibs. Without a
link_dirs attribute, every dependency (even if unexported) would need a
separate module.
Previously we were casing out on exact structs, which might be OK when
libraryDecorator and procMacroDecorator were the only possibilities, but
repeating the logic for three types is too much. Using an interface
makes this logic scale better.
Bug: 159591910
Test: cd external/rust; mma; m crosvm.experimental
Change-Id: Ia1124e09f48cd05e39f094bbcb988622ebd2272f
This change makes it possible to use a single module to provide both
dylib and rlib varieties of a library. This allows the use of libstd and
libtest from a rustlibs property, allowing linkage type to change
for different variants.
Bug: 159718669
Test: cd external crates; mma; m crosvm.experimental
Change-Id: I477c4d2faec63703fdc6dd42ba020747d6a50714
Adds the rustlibs dependency type which will automatically select
between rlib and dylib based on the type of the library.
Bug: 143217452
Test: cd external/rust; mma
Change-Id: I97faadae98bf957090a32939cfb2d3a10f74a057
rust_library now produces only rlib and dylib variants.
rust_library_foreign now produces static and shared variants.
This change was made because both are common cases, but all four
together essentially never happens. This allows us to have fewer repeat
module definitions.
Bug: 143217452
Test: cd external/rust; mma
Change-Id: Iaf69e8da38f7c12710331c51464699cf7cbbb656
Previously, we had just called AddProperties inside the factory on
libraryDecorator's properties directly. This inadvertantly missed their
MutatedProperties. Inheriting properties from libraryDecorator rather
than baseCompiler will avoid that type of error in the future.
Test: cd external/rust; mma; m crosvm.experimental
Change-Id: Ic08c410e69a8e272d63f9ffbdbe16ba9c3cc3de1
Add documentation on how lints are defined and used in Android. Merge
the deny_warnings attribute with a new attribute (no_lint) which can be
used to disable the default linting parameters.
Explicitly allow all lints for external/ and prebuilts/, which remove
any warning when building sysroot for the devices.
Test: cd external/rust/crates; mma
Test: add dummy internal Rust module; mma
Change-Id: I62be1c41aeda4068fb9e288038727c1de5ffe547
* Add HostToolPath to use rust_binary_host modules as host tools.
Bug: 157666974
Test: make with local genrule examples
Change-Id: I9648313c0f0695d2e695f18ded4710350b2d6475
Depending on the location of the repository (e.g. external/, vendor/), a
different set of lints will be enabled. Add the clippy property to the
rust_* modules. This property can be used to overwrite the default
behaviour.
Test: m checkbuild
Bug: 157238651
Change-Id: Ife0f723ef4a74abb102597f8486a7b9f30e7d351
The testRustContext function is used to prepare the environment for the
unit tests. As such, it is marked as a Helper function. Remove a
second superfluous call within its definition.
Test: m nothing
Change-Id: I37ea9a584b33498794c1cac828a7557d4bea8d7c
This CL adds "dists" to the base property struct to support multiple
dist file configurations, and generic tag support to dist tagged outputs
of modules.
Fixes: b/152834186
Test: soong tests and `m sdk dist`
Change-Id: I80c86bc9b7b09e671f640a4480c45d438bdd9a2a
Signed-off-by: Jingwen Chen <jingwen@google.com>
The moduleContextImpl structure contains duplicate information on the
contexts. BaseModuleContext, ModuleContext and DepsContext can already
surface which Module is being processed via Module(). Because most of
the Rust-specific methods on the *Context structures simply forward to
the Module methods, expose a RustModule() method.
Test: cd external/rust/crates; mma
Change-Id: Ifee90825d54081fc5e9a8df0b7c4580412e9158c
Make uses NATIVE_COVERAGE to enable gcov coverage and CLANG_COVERAGE
to enable clang coverage. NATIVE_COVERAGE is translated to the Soong
Native_coverage product variable which triggers the native_coverage
clause in Android.bp files. The clause also needs to be triggered
for CLANG_COVERAGE.
Rename the existing Native_coverage product variable to GcovCoverage,
and regenerate Native_coverage when either GcovCoverage or
ClangCoverage are set.
Also remove NativeLineCoverage, it wasn't doing anything differently
than Native_coverage.
Bug: 159059537
Test: m checkbuild
Change-Id: I215124a9b35a2ad50ad562079d392e3d33da11f4
Bug: 144861059
Test: cd external/rust/crates; mma
Test: Manual verification through ldd that libstd is a dependency
Change-Id: I603cf519215317aa8c400cd0f6ebb1b58f5bcf15
strings.TrimPrefix() args are reversed, leading to an incorrect gcov
path prefix for Rust coverage.
Bug: 156482307
Test: built and ran unicode-xid_device_tests_unicode_xid with coverage
Change-Id: I6476cee14093a7e53738019c84dbcf4333b0b9b2
When a lib has sdk_version set, an SDK variant and a platform variant
are created by the sdkMutator. Then by the versionMutator, if the
library had 'stubs.versions' property, one or more versioned variants
and one impl variant are created for each of the two (SDK and platform)
variants. As a concrete example,
cc_library {
name: "foo",
sdk_version: "current",
stubs: { versions: ["1", "2"], },
}
would create 6 variants:
1) (sdk: "", version: "")
2) (sdk: "", version: "1")
3) (sdk: "", version: "2")
4) (sdk: "sdk", version: "")
5) (sdk: "sdk", version: "1")
6) (sdk: "sdk", version: "2")
This is somewhat uncessary because the need for the SDK mutator is to
have the platform variant (sdk:"") of a lib where sdk_version is unset,
which actually makes sens for the impl variant (version:""), but not
the versioned variants (version:"1" or version:"2").
This is not only unncessary, but also causes duplicate module
definitions in the Make side when doing an unbundled build. Specifically,
The #1 and #4 above both are emitted to Make and get the same name
"foo".
To fix the problem and not to create unnecessary variants, the versioned
variants are no longer created for the sdk variant. So, foo now has
the following variants only.
1) (sdk: "", version: "") // not emitted to Make (by versionMutator)
2) (sdk: "", version: "1") // not emitted to Make (by versionMutator)
3) (sdk: "", version: "2") // emitted to Make (by versionMutator)
4) (sdk: "sdk", version: "") // not emitted to Make (by versionMutator)
Bug: 159106705
Test: Add sdk_version:"minimum" to libnativehelper in libnativehelper/Android.bp.
m SOONG_ALLOW_MISSING_DEPENDENCIES=true TARGET_BUILD_UNBUNDLED=true libnativehelper
Change-Id: I6f02f4189e5504286174ccff1642166da82d00c9
This changes the way the output filename is calculated for rust_test
binaries to be more similar to cc_test.
This also removes the option to define multiple test binaries in a
single rust_test module via the TestPerSrc mutator. Now each rust_test
module corresponds to a single test binary.
Bug: 158500462
Test: m -j pin-utils_tests_pin_utils
Test: m -j unicode-xid_device_tests_unicode_xid
Change-Id: I6e0f79dcb4e49fa49d6ebb36abeef67a9eb180a0
Use the -Z profile-emit flag to ensure that rust gcda files have
/proc/self/cwd/ appended to them similar to cc modules. This makes sure
our gcda output is consistent no matter what language the binary is
written in.
Bug: 156482307
Test: gcda outputs to the new path.
Change-Id: Ife4d55f885c7e33dffa66f7436bc2bf8b5916586
Because we are not relying on Cargo.toml for our crate dependencies,
we need to provide a structured file to rust-analyzer which describes
the locations of the crates. Add a generator for that purpose, similarly
to cc/compdb and cc/ccdeps.
Bug: 156395307
Test: SOONG_GEN_RUST_PROJECT=1 m nothing && \
cat ${ANDROID_BUILD_TOP}/out/soong/rust-project.json
Change-Id: I46efe0adeddae281eaf86707504c3aa15b5e80b8
Annotate the test config properties with android:"path", and always
use LOCAL_FULL_TEST_CONFIG to pass the path to Make.
Bug: 157621988
Test: manual
Change-Id: I1c020e642f83d994d63c9e9cb56e686086d5acaa
Rename `CoveragePath` and `CoverageExcludePaths` as
`NativeCoveragePath` and `NativeCoverageExcludePaths` (resp.).
Also rename function `android.CoverageEnabledForPath` as
`android.NativeCoverageEnabledForPath`.
Test: m nothing
Bug: 158212027
Change-Id: Id2c11a638e88088096420b537effa866d7667304
Without specifying -soname, cc binaries are unable to locate Rust
provided shared libraries. This adds the -soname linker flag for shared
libraries.
Bug: 158490355
Test: readelf -d <cc_binary> shows the expected SONAME
Change-Id: I66852a7ce24d5ea5e426f11bc1834fb56a150628
Give prebuilt_etc and sh_binary their own packages and split the
gigantic main Android.bp up to small, per-package ones.
Test: m nothing, TreeHugger
Bug: 156980228
Change-Id: I7b00cd344b9f16861f1ff39edf0029f016b853d0
Darwin builds aren't using the correct suffix for shared libraries or
proc macros (.dylib). This fixes that.
Bug: 155302034
Test: "cd external/rust; mma" no longer indicates wrong extension.
Change-Id: I5093e0bcc78dbb2dd9fcf39069f326101bb8f411
This adds gcov coverage support for Rust device library and binary
modules (including test modules). Support is provided to pass Rust
static library gcno files to CC modules and visa versa.
Additional changes:
* Begin mutator added for Rust modules.
* SuffixInList added to android package.
* CoverageEnabled added to Coverage interface.
* CoverageFiles added to LinkableLibrary interface.
* Fix in coverage mutator for non-CC modules which marked the wrong
variant as the coverage variant.
* Added coverage libraries to the cc.GatherRequiredDepsForTest.
Bug: 146448203
Test: NATIVE_COVERAGE=true COVERAGE_PATHS='*' m -j <rust_module>
Change-Id: If20728bdde42a1dd544a35a40f0d981b80a5835f
CC libraries which depend on Rust libraries get missing dependency
errors when building 32-bit variants dependent on Rust modules which
don't explicitly have "multilib: both" declared.
Because CC libraries use MultilibBoth by default, Rust should do the
same.
This also fixes a bug where the ARM32 toolchain incorrectly embedded
toolchain64Bit instead of toolchain32Bit.
Bug: 154730212
Test: Rust libraries provide both variants by default.
Change-Id: Ia545fe069d3c6b77c3d18f4f10267e2c72ee0bab
The list of required deps for testing is a lot shorter as we no longer
include as many prebuilt libraries and our list of stdlibs is much
shorter. Remove the libraries we no longer need, and add the
non-prebuilt libstd and libtest to the deps generated for every test.
Bug: 153463188
Test: Soong tests pass.
Change-Id: I94f7b514797ab8d6aedb543ea6afdb07611b7293
"version" mutator creates stubs variants for "cc" libraries with
"stubs.versions". These stubs are for APEX-APEX or APEX-Platform
boundaries.
For host/ramdisk/recovery variants, stubs are not necessary.
Exempt-From-Owner-Approval: cp from internal
Bug: 153698496
Test: m
Merged-In: Id576c4318d9d69246a4a7e2fb4145d5fd2ab9416
Change-Id: Id576c4318d9d69246a4a7e2fb4145d5fd2ab9416
(cherry picked from commit c40b5193fe)
With kernel support being removed for execute-only memory layouts,
remove support for XOM until there's kernel support again.
Bug: 153457617
Test: Rust binaries no longer have execute-only text sections.
Change-Id: I19cfb5b347a1f44394a619e17bb0c057c09e76a9
Rust binaries were installing to the wrong paths. This fixes
the install path by only including the archtype subdir when there's a
multilib conflict for that archtype.
Bug: 153423714
Test: Example rust module installs to expected path.
Change-Id: I63ea1ce7dae0d97d65522e0917d916b44808f442
Revert submission 1242911-sdk_version_variant
Reason for revert: b/153394225
Reverted Changes:
Ife99745fb:Use libnativewindow for platform variant of libagq...
I1bae84c43:Use libnativewindow for platform variant of androi...
I6e6021ed3:Use stl to depend on libc++
Ife99745fb:Use libnativewindow for platform variant of libRSS...
I2c9f439b9:Fix static dependency on libprotobuf-cpp-lite-ndk
Iff2aff9cf:Set sdk_version for cc_genrules used by modules wi...
I7d72934aa:Add sdk mutator for native modules
Ief378a007:Use sdk variant of Soong modules when LOCAL_SDK_VE...
Bug: 149591340
Change-Id: I798fa902c779469c6382b6699351e5d12bf14785
Fixes: 153394225
Compiling native modules against the NDK disables platform features
like ASAN. For anything shipped on the system image there is no
reason to compile against the NDK. Add a new mutator to Soong that
creates a platform and an SDK variant for modules that set
sdk_version, and ignore sdk_version for the platform variant. The
SDK variant will be used for embedding in APKs that may be installed
on older platforms. Apexes use their own variants that enforce
backwards compatibility.
Test: sdk_test.go
Test: TestJNIPackaging
Bug: 149591340
Change-Id: I7d72934aaee2e1326cc0ba5f29f51f14feec4521
Native modules within APEX should be linked with proper stub version
according to its min_sdk_version.
For example, when min_sdk_version is set to "29", libfoo in the apex
would be linked to libbar of version 29 from platform, even if it has
a newer version like 30.
Bug: 145796956
Test: m nothing (soong tests)
Change-Id: I4a0b2002587bc24b7deeb5d59b6eeba5e1db5b1f
The latest version of crosvm refers to minijail as a separate crate:
2dae56768c%5E%21/#F1
The current version of minijail in aosp crosvm-master isn't a crate,
but minijail in aosp master is a crate and builds with crosvm.
This allows the latest version of crosvm to build in the Android tree
and use minijail in the same tree.
Change-Id: I2231149f1488727754430183d04072c5cc0f570a
Move the NDK modules defined in app_test.go into cc/testing.go in
preparation for using sdk_version: "current" in more tests.
Test: all soong tests
Change-Id: I80bc31225fb4562fce42781219cefbbd6affae06
When linking native libraries with rustc, be explicit about the
kind of native library being linked. This prevents confusion when
two kinds of one library (e.g. static/dynamic) are available in
the library search paths.
Bug: 147140513
Test: The correct prebuilt is selected when linking native prebuilts.
Change-Id: I37975bcd284e6c33ce3dd45fab8a3b5011b0803b
The APEX dependency is more correctly tracked. Previously, the
dependency was tracked while we gather modules that will be installed to
an APEX. This actually was incorrect because we skipped many dependency
types that we don't need to follow to gather the modules list, such as
the headers dependency.
Now, the dependency is tracked directly when a module is mutated for an
APEX. In other words, if a module is mutated for an apex X, then the
module will appear in the X-deps-into.txt file.
This change also changes the format of the txt file. It now clearly
shows why a module is included in the APEX by showing the list of
modules that depend on the module.
Bug: 146323213
Test: m
Change-Id: I0a70cf9cce56e36565f9d55683fdaace8748a081
If a library happens to contain the extension as a substring in its
name, libNameFromFilePath will return a truncated library name. Change
this calculation to remove the last instance of the extension substring
instead.
Bug: 147140513
Test: Modified rust tests pass.
Change-Id: I0ed91e5f571ed5c4040ee15956a1598846aee43a
Rust has no backwards compatibility concerns, so no need to use
--use-android-relr-tags here.
Bug: http://b/147452927
Test: treehugger
Change-Id: I94de40bfad1ff2dcbc4241b944db90e0891bf9f3
cdylib targets cannot currently dynamically link to Rust crates, so this
flag was not actually doing anything, and in conjunction with -C lto
which we *do* want on for this target, causes a hard error on
rustc-1.40.0.
Bug: 147438124
Test: Build a rust_library module, this implicitly tries to build cdylib
Change-Id: I828b3e6bae7c58f4081db3e73009b443a382a296
If we want to build a component which uses two versions of a crate in
the dependency graph (for example, libc in std and libc from
external/crates), Rust requires at least one of them have a metadata
field to differentiate them.
Since Soong module names should be distinct for a given target, they
should serve as metadata.
Bug: 147397356
Test: Build a rust library with "rlibs: ["liblibc"]" in the blueprint
Change-Id: I5d85ea8b3197972c09de301c9523efaff08adf81
This reverts commit 7cb4d378e7.
Test: m
Test: ALLOW_MISSING_DEPENDENCIES=true DIST_DIR=out/dist ./art/tools/dist_linux_bionic.sh -j80 com.android.art.host
(in the master-art-host branch)
Change-Id: I9beca73aafdf42f03bfa19cf1634b2641dac417b
This reverts commit 956305c61c.
Reason for revert: broke master-art-host branch
Exempt-From-Owner-Approval: reverting a bad change
Change-Id: Id7faed4ee85328c7c65847a3543ea9e67a3d50b3
This is no longer shipped by rustc, and wasn't really intended to be
implicitly available in the first place.
Bug: 147432625
Test: Built a sample rust device binary
Change-Id: I3c33ac3726c788debe14ed39a9ecbae7a7449351
We need to have a way to see the list of modules that directly or
indirectly contribute to an APEX. People find it difficult to determine
whether a module is included in which APEXes because APEX tracks
indirect dependencies as well as direct dependencies. Therefore, just
looking at Android.bp for the APEX itself doesn't give the answer.
This change adds a new make target <apex_name>-deps-info, which
generates out/soong/<apex_name>-deps-info.txt file that shows the
internal and external dependencies of the said APEX.
Here, internal means the dependencies are actually part of the
APEX, while external means the dependencies are still external to the
APEX.
Bug: 146323213
Test: m (apex_test amended)
Change-Id: I33d1ccf5d1ca335d71cd6ced0f5f66b8c3886d13
The cc.GatherRequiredDepsForTest() method returns some default module
definitions that are required when using cc module types like
cc_library. Previously, the registration of the module types and
mutators needed to process those default definitions was duplicated
in the test config initialization.
This change removes that duplicated code and replaces it with calls
to cc.RegisterRequiredBuildComponentsForTest(ctx) which registers all
the required build components.
Test: m checkbuild
Bug: 146540677
Change-Id: I80b6913c5691ff164ce9d308b9e1da24940f2d42
The filesystem object was available through ModuleContext.Fs(), but
gives too much access to the filesystem without enforicing correct
dependencies. In order to support sandboxing the soong_build
process move the filesystem into the Config. The next change will
make it private.
Bug: 146437378
Test: all Soong tests
Change-Id: I5d3ae9108f120fd335b21efd612aefa078378813
The clang-r370808 upgrade contains a change to LLD allow PT_LOAD
segments to reside at non-multiples of the page size in the resulting
object file. https://reviews.llvm.org/rL369344
While this helps reduce the alignment waste and resulting image size, it
has interesting implications for execute only memory (XOM): The runtime
loader will now load code or data from other segments into pages with
different protections than intended.
This would partially defeat execute only (XOM) text sections as the
segment could now overlap with previous and following sections. This
might allow for code or data from the preceding and following sections
(like .eh_frame, and .data.rel.ro) to be executable, and either ends of
.text to be readable.
When the runtime loader (linker[64]) `mmap`s segments from *.so files,
the file offset parameter (see `man 2 mmap`) MUST be a multiple of the
page size. Since the updated LLD can now pack segments in a file (which
helps minimize resulting object file size) (previously, the segment
offsets were page aligned), this has interesting implications.
To appreciate the current bug, consider the following output from
`readelf` before this patch is applied, but after the toolchain upgrade:
```
$ readelf -lSW $OUT/symbols/apex/com.android.runtime/lib64/bionic/libc.so
...
[Nr] Name Type Address Off Size ES Flg Lk Inf Al
...
[13] .eh_frame PROGBITS 000000000002e7c0 02e7c0 013374 00 A 0 0 8
[14] .text PROGBITS 0000000000042b40 041b40 09ecb4 00 AX 0 0 64
[15] .plt PROGBITS 00000000000e1800 0e0800 001f30 00 AX 0 0 16
[16] .data.rel.ro PROGBITS 00000000000e4740 0e2740 005208 00 WA 0 0 32
...
Type Offset VirtAddr PhysAddr FileSiz MemSiz Flg Align
PHDR 0x000040 0x0000000000000040 0x0000000000000040 0x000230 0x000230 R 0x8
LOAD 0x000000 0x0000000000000000 0x0000000000000000 0x041b34 0x041b34 R 0x1000
LOAD 0x041b40 0x0000000000042b40 0x0000000000042b40 0x0a0bf0 0x0a0bf0 E 0x1000
LOAD 0x0e2740 0x00000000000e4740 0x00000000000e4740 0x006720 0x006720 RW 0x1000
...
01 .note.android.ident .note.gnu.build-id .dynsym .gnu.version .gnu.version_d .gnu.version_r .gnu.hash .dynstr .rela.dyn .rela.plt .rodata .eh_frame_hdr .eh_frame
02 .text .plt
03 .data.rel.ro .fini_array .init_array .dynamic .got .got.plt
...
The above output tells us:
1. .text will wind up in the third (02) segment.
2. The third segment will be (LOAD)'ed as (E)xecutable.
3. Because the file (Offset) of the first segment (0x41b40) is NOT a
multiple of the page size, it cannot be passed as the `offset` to
`mmap`. As such it will be rounded down to the first multiple of the
page size, 0x41000.
4. The preceding section (.eh_frame) will be loaded in the preceding
segment (01). It occupies file (Off)set range [(0x2e7c0):0x41b34].
0x41b34 is not explicit in the output, instead you must use the
formula:
Off + Size == End
ie.
0x2e7c0 + 0x13374 == 0x41b34
(This happens to match (FileSiz) of the second segment, which makes
sense as .eh_frame is the final section in the second segment.)
5. mmap'ing file offset 0x41000 when loading the second segment will
include 0x4c0 bytes (0x42000 - 0x41b40) from .text, now mapped as
readable (oops). Suddenly code from .text is now readable (and thus
scannable for gadgets for ROP chains).
6. mmap'ing file offset 0x41000 when loading the third segment will
include 0xb34 bytes (0x41b34 - 0x41000) from .eh_frame, now mapped as
executable (oops). Suddenly data from .eh_frame is now executable
(and thus a potential gadget for ROP chains).
7. mmap'ing file offset 0xe2000 when loading the third segment will
include 0x8CO bytes (0xe3000 - 0xe2740) from .data.rel.ro, now mapped
as executable (oops). Suddenly data from .data.rel.ro is now
executable (and thus a potential gadget for ROP chains).
8. mmap'ing file offset 0xe2000 when loading the fourth segment will
include 0x730 bytes (0xe0800 + 0x1f30 - 0xe2000) from .plt, now
mapped as readable (oops). Suddenly data from .plt is now readable
(and thus scannable for gadgets for ROP chains).
All these oops' could be avoided if the linker placed .text+.plt at page
size aligned file offsets, which is what `-Wl,-z,separate-code` code
does. After this patch, we have:
```
$ readelf -lSW $OUT/symbols/apex/com.android.runtime/lib64/bionic/libc.so
...
Type Offset VirtAddr PhysAddr FileSiz MemSiz Flg Align
PHDR 0x000040 0x0000000000000040 0x0000000000000040 0x000230 0x000230 R 0x8
LOAD 0x000000 0x0000000000000000 0x0000000000000000 0x041b34 0x041b34 R 0x1000
LOAD 0x042000 0x0000000000042000 0x0000000000042000 0x0a0be0 0x0a0be0 E 0x1000
LOAD 0x0e3000 0x00000000000e3000 0x00000000000e3000 0x006720 0x006720 RW 0x1000
```
In the future, we could go back to tightly packing segments in the
binary if the runtime loader was improved to detect the previously
stated problem, and `memset` over the problematic ranges of the freshly
`mmap`ed pages (implying additional startup cost for reduced binary
size). This might save ~6 KB from each native binary, which adds up to
~17 MB for an AOSP image.
Also, prefer
-Wl,--execute-only
rather than
-Wl,-execute-only
Bug: 139945549
Bug: 146144180
Test: readelf -lSW $OUT/symbols/apex/com.android.runtime/lib64/bionic/libc.so
Change-Id: I64527e034ca3c71565ea52ed06f81f75d5216627
Reported-by: Ryan Prichard <rprichard@google.com>
Suggested-by: Fangrui Song <maskray@google.com>
Signed-off-by: Nick Desaulniers <ndesaulniers@google.com>
Use the empty string for the core image variant so that modules
added to imageMutator do not change their build directory.
Bug: 142286466
Test: m checkbuild
Change-Id: Ida4534d9a4d6176236aaa480fed359ce27acfaa1
Merged-In: Ida4534d9a4d6176236aaa480fed359ce27acfaa1
(cherry picked from commit 72d685ee7f45e5393be44ae4159edf083ac918de)
* Now the installation directories match those for C/C++ tests:
* The relative_install_path refers to path under nativetest(64).
* Device test files are installed in data/nativetest(64).
* Automatically generated configuration files and copied test binaries
are still in the "testcases" directory.
* Change host test configuration to run test binary files
in testcases/<mutated_module_name>/<arch_type>/<stem_name>
Bug: 140938178
Test: atest --include-subdirs under external/rust/crates
Change-Id: I4b29afb897f4ba8749e87f79857c5b1a959bb2b0
* Copy current configuration from
rust/config/x86_64_device.go and
cc/config/x86_device.go.
Bug: 145608833
Test: build rust modules for aosp_x86-userdebug
Change-Id: Ie4951dd4703596f8e503d813a6c4ac03c1f20980
Move the ImageMutator to be registered just after the archMutator
in preparation for moving it between osMutator and archMutator.
Requries updating variants in a few tests that now run the
ImageMutator.
Bug: 142286466
Test: no change to build.ninja
Test: all soong tests
Change-Id: Ia9d2a7bc0e225bedec3c9a83ea04f471a931bf47
* The RustBinaryTest runner and device test config template
are derived from GTest. Device tests are pushed to /data/local/tmp.
Bug: 140938178
Test: make unicode-xid projects; run atest
Change-Id: Idb4cab7872b48c6e25bc59b85aa6d9dece4383ec
Avoid having to pass ModuleFactoryAdaptor to every call to
RegisterModuleType in a test by wrapping RegisterModuleType.
Test: all soong tests
Change-Id: If8847d16487de0479cc3020b728256922b3cadba
Rust C libraries cannot currently depend on dylibs. This prevents device
libraries from attempting to link the stdlib dynamically, instead using
rlibs.
Bug: 144861059
Test: Simple Rust library compiles as a shared/static library.
Change-Id: I38e5aae90e50ddcc5fc72d5a8d1820d92955cf48
cc.LinkableInterfaces should be able to request if an empty variant is
required from the LinkageMutator. We were previously assuming that if only
static or shared libraries were being built, then we didn't need the
empty variant. However this should be explicit to handle cases where the
default set of variants being created may not include both static and
shared libraries.
Bug: 144861059
Test: m -j crosvm.experimental
Change-Id: I66724bdc16ff350b06dfa4d049c82f33019979a5
Prepare for making the image mutator available to all modules and
moving it between the os and arch mutators by moving it into the
android package and using an interface implemented by the module
types to control it.
Bug: 142286466
Test: No unexpected changes to out/soong/build.ninja
Change-Id: I0dcc9c7b5ec80edffade340c367f6ae4da34151b
* Rust test modules should have NATIVE_TESTS class,
not EXECUTABLES, to work with atest --host.
Bug: 140938178
Test: make rust projects; run atest --host .
Change-Id: Ie9e237ee8c17b6e2d41f08a77c2b79d462a2c193
* Rename testBinaryDecorator to testDecorator
* Add TestProperties
* Add install function for testDecorator to install config files
* Add tradefed.AutoGenRustHostTestConfig
* Depend on new build/make/core/rust_host_test_config_template.xml
and new tradefed.testtype.rust.RustBinaryHostTest class
* Add autogenTemplateWithName in tradefed/autogen.go
to generate config files with customized(mutated) executable name.
* Make rust_test module names more robust and easy to use.
* Use crate name instead of source file name as the Stem
for single source file modules, to match original user
specified output file name in Cargo.toml.
* Do not set up test module SubName when Stem is empty
or when the module name already contains Stem suffix.
That happens when TestPerSrcMutator is disabled or when
there is only one source file with renamed output file name.
* In TEST_MAPPING, references to mutated rust_test modules should be
(1) <module_name> for single source file modules without mutation, or
(2) <module_name>_<crate_name> for single source file modules, or
(3) <module_name>_<source_file_base_name> for multi-file modules.
Bug: 140938178
Test: mm in rust projects, check output test file names
Change-Id: Ifdbfa14d5eed4f10b4fb983f82c93bbb9be3f899
Make sure LinkabeInterfaces can export include dirs to cc modules. This
fixes the Rust implementation and makes sure these are pulled in for cc
modules.
Bug: 144052093
Test: cc module depending on a rust library includes dirs correctly.
Change-Id: I6b5d5e6ab6afb507178b4f2cbdc45f01031dbfe4
Now that we have sysroots built in Soong, we can enable these devices
and they just work.
Bug: 141251907
Bug: 141381044
Change-Id: I562256fee372d331db883f7c1b46405945295c24
In order to ensure we are using current platform Bionic for any platform
Rust binaries, we need to build the sysroot in Soong. This will also
enable us too hook the "test" crate if necessary.
While both a dynamic and static sysroot are available, on device only a
dynamic sysroot will be injected. On host, we continue using the sysroot
used to build the compiler as before.
Bug: 139486496
Change-Id: I127377e5b056610ceb5015a34d266250320fbc31
Break up some of the longer lines in the Rust source and refactor the
TransformSrcto* functions to take fewer arguments.
Bug: N/A
Test: m -j crosvm.experimental
Change-Id: Ia0deb8bfe3e1bfd07ef35633e9a7e6f5e799b3a3
rustc expects libraries and proc_macro filenames to conform to
a particular format, alphanumeric with underscores and lib${crate_name}.*.
Enforce this with a check when getStem() is called.
This makes the crate_name property required for proc_macros and
libraries. This also removes the notion of a default crate name derived
from the module name. It's not needed for binaries, so this won't impact
them.
Bug: 143579265
Test: m -j crosvm.experimental
Change-Id: I2770cf7d02dd4291c3d240d58d242b940098dcee
* Rust tests are like binary files compiled with --test.
New test.go follows binary.go code patterns and reuses
some code in binary.go.
* Generate one test per source file as testPerSrc in cc/test.go.
The "all tests" variation feature of cc/test.go is not copied yet.
Fix some Stem and SubName settings to make testPerSrc work.
* Move cc.CheckDuplicate to android.CheckDuplicate,
which is now shared by cc and rust.
* Refactor tests in binary_test.go and add new test_test.go.
Bug: 140938178
Test: mm in rust projects, added rust_test and rust_test_host
Change-Id: Ia6fec8b4cf2572fd352ab1938a1f3c7b5cca2212
Adds the ability for rust modules to be compiled as C libraries, and
allows cc modules to depend on these rust-generated modules. This also
means that soong-rust should not have any dependencies on soong-cc aside
from what's required for testing.
There's a couple small fixes included as well:
- A bug in libNameFromFilePath that caused issues when library's had
"lib" in their name.
- VariantName is removed from rust library MutatedProperties since this
was unused.
Bug: 140726209
Test: Soong tests pass.
Test: Example cc_binary can include a rust shared library as a dep.
Test: m crosvm.experimental
Change-Id: Ia7deed1345d2423001089014cc65ce7934123da4
Adds an interface, CcLinkableInterface, for cc linkable dependencies
which come from other toolchains such as Rust.
Bug: 140726209
Test: Soong tests pass, rust modules still compile.
Change-Id: I7378a46fad94fd0b735746aaf4e265fd2c2c04d8
Support for building Rust modules on darwin hosts.
Bug: 140640858
Test: m libremain works on darwin
Change-Id: Ieb1ff9167de34cffbebebab31fa48da07081c8a6
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
Add a ToMakePath() method that returns a new path that points out
out/ instead of out/soong/, and replace the
"$(OUT_DIR)/" + path.RelPathString()
pattern with
path.ToMakePath().String()
Bug: 141877526
Test: m checkbuild
Change-Id: I391b9f2ed78c83a58d905d48355ce9b01d610d16
Create a new type InstallPath that is similar to OutputPath to
differentiate intermediates output paths from installed output
paths.
RelPathString is a poorly defined, undocumented function that is
primarily used to get an install path relative to out/soong to
generate an equivalent install path for Make relative to $(OUT_DIR).
Move it to InstallPath for now, and fix the one remaining user on
OutputPath.
Add a method to create an NDK install path so that ndk_sysroot.go
doesn't have to do it manually with PathForOutput.
Bug: 141877526
Test: m checkbuild
Change-Id: I83c5a0bd1fd6c3dba8d3b6d20d039f64f353ddd5
* Edition and Deny_warnings should not be set when
constructing a BaseCompilerProperties, or the
initialized values will reject values inherited from rust_defaults.
* Use getEdition and getDenyWarnings to retrieve those properties
with defaults from config.
Bug: 141699953
Test: mm in rust projects
Change-Id: Id1ae357caeaf656cd33732bf4e54920e206f4ead
We need to pass --remap-path-prefix to Rust builds to ensure that they
are reproducible across different paths. This adds that for both host
and device builds.
Test: Verify host and device builds are reproducible with different
directories.
Change-Id: I68ba31537b2332fd05613aa0fb00a1502d78e410
* "-D warnings" means "deny all warnings" and make them errors.
* Modules with warnings should fix all warnings or use
deny_warnings: false
Bug: 141699953
Test: mm in projects with Rust modules
Change-Id: I6310dee8e34b7780937e8fc1834016a04a943a2f
Ensure that devices can include proc_macros, which are host-only
and may include host-only dependencies.
Bug: 141491501
Test: Soong tests pass.
Test: Test fails as expected when removing CL 1126496
Change-Id: I3ae7ab80283cd1fd4b800a533cb3205b3c108d45
Currently proc_macros don't pull in the correct arch dependencies when
the host arch differs from the target arch.
This fixes how proc_macro dependencies are handled by defining them as
always being host-only and including them as dependencies for
device-modules by using AddFarVariationDependencies.
Bug: 141491501
Test: Example device rust module builds with proc_macro dependency.
Change-Id: Ic037dc406ce90526f8b68c92fffc0d93a498a4ff
Add libtest to our list of stdlib libraries, and remove some others that
are only used when generating proc_macros (which aren't dynamic).
Bug: 141207434
Test: m -j crosvm.experimental
Change-Id: I39cb030940adf1993e861d0142378eeea8b1ff5e
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
Splits out the Rust paths whitelist into a separate file
under config/ so that OWNERS can be defined for it.
Bug: 141207129
Test: m -j crosvm.experimental
Change-Id: I5effa6783e5c47560b4b1eae12ad0eb9e9ba96fe
The Target checks against the os type and arch type occur before we
determine a dependency is a rust.Module / cc.Module. This meant
rust_defaults would fail this check, preventing them from being used.
Bug: 140963307
Test: make a rust_binary_host using rust_defaults.
Change-Id: Ibf43635f14ad367b4ce8016a2196f8c645b36bfe
Both Rust and cc use this function, so move it over to android
package's util.go and export it.
Bug: 140734195
Test: m -j
Change-Id: Ibe8b7a94592e402468a027ad6027b187f29c8e07
The Rust toolchain only supports Linux hosts, so skip Rust tests
that depend on host toolchains when building on other platforms.
Test: Tests still run on Linux hosts
Bug: 140435149
Change-Id: I298b52589ab5c6a7ec3acc4db1111596cd995d76
Adds support to Soong for building rust modules. This currently only
supports x86_64 device and x86 linux host targets. The functionality
is sufficient to build crosvm.
Bug: 136189233
Test: Test module builds.
Test: crosvm builds.
Change-Id: I6ea04615834a6d673578ab10ea1a2eb04259fe09
