Commit graph

14 commits

Author SHA1 Message Date
Jooyung Han
33eb615eb0 ndkstubgen: use llndk=<version> for new llndk stub
We want LLNDK symbols to be explicitly marked with llndk tag to
handle LLNDK freezing which happens before SDK freezing. If symbols
need to be frozen as LLNDK, those symbols must be marked explicitly with
correct vFRC version.

In the following example,

LIBFOO { # introduced=35
  foo;
  bar;
  bar; # llndk=202404
  baz; # llndk=202404
  qux; # llndk=202505
};

NDK libfoo will have foo and bar while LLNDK libfoo stub will have bar
and baz for 202404.

Bug: 329012338
Test: test_ndkstubgen test_symbolfile
Change-Id: I384f589b240fa047e8871964bf9550f426024dfc
2024-03-14 06:06:26 +09:00
Steven Moreland
8d5b9d01e9 validate tags for typos
it turns out there were typos which are now fixed :D

Bug: 317293882
Test: build tree
Change-Id: I91dc08f00e8dd2e9dcb9101307d9b04bdcadea11
2024-01-03 21:48:03 +00:00
Elliott Hughes
8cddc00b02 Add riscv64.
Otherwise we hit an unknown --arch error in ndkstubgen as soon as we try
to build an NDK library (such as libc/libdl/libm, which are the very
first libraries we need to build).

Test: treehugger
Change-Id: Id633248d1fa80eeddfd234301355931bb1309dc3
2022-10-10 15:33:50 +00:00
Jiyong Park
4ecbdb6dfd Add --no-ndk to ndkstubgen
By default, ndkstubgen does not omit NDK symbols as long as they
satisfy the API version and the CPU architecture requirements. This
change adds a new flag --no-ndk to ndkstubgen which is used to
override the default behavior. If the flag is set, NDK symbols are
omitted leaving only the annotated symbols (e.g. #apex, #systemapi,
etc.).

This will be used for stub libraries that are not part of NDK. So far,
symbols in such libraries haven't needed to be annotated as #apex, and
that has caused a confusion that those symbols belong to NDK. The
follow-up change will ensure that those symbols are always annoated as
either #apex or #systemapi so that their roles are clearly visible.

Bug: 184712170
Test: atest test_ndkstubgen
Test: atest test_symbolfile
Change-Id: Ic8d2c7d0b32bdef79f7563621035e60f406e4131
2022-10-05 09:06:49 +09:00
Jiyong Park
85cc35a143 Distinguish # apex and # systemapi symbols
Previously, the symbol tag `# apex` was treated the same as `#
systemapi`. With this CL, they have different meanings.

`# systemapi`: APIs that are defined in the platform (the non-updatable
part), and are exposed to unbundled system components like APEX

`# apex`: APIs that are defined in the APEX and are exposed to the
platform or other APEXes

Bug: 239274367
Test: m
Change-Id: I0484ea349656dbbd337e5fe3a5970f0ad275b807
2022-07-19 13:47:19 +09:00
Jiyong Park
3f9c41d2f5 Refactor symbolfile script
Introduce the class `Filter` which encapsulates conditions
(architecture, api level, whether llndk is included, etc.) so that we
don't need to touch many places everytime a new condition is added.

In addition, refactor test_symbolfile to reduce duplications

This CL doesn't add a new functionality.

Bug: 239274367
Test: run test_ndkstubgen and test_symbolfile
Change-Id: I188e482492d39ec18134ecc5c908f0d288a754a9
2022-07-19 13:47:15 +09:00
Spandan Das
3f5659f268 Revert "Revert "Apply pylint to cc/symbolfile""
This reverts commit ea28f4cae7.

Reason for revert: Postsubmit failure was unrelated, and was fixed by this revert CL aosp/1802327

Change-Id: I2693b17c2261cfc3dce0ed69baeb2d1f45f784f8
2021-08-19 19:31:54 +00:00
Junwei Wang
ea28f4cae7 Revert "Apply pylint to cc/symbolfile"
This reverts commit 73de70264a.

Reason for revert: Broken build 7651218 on aosp-master on aosp_x86_64-userdebug -- b/197190129

Change-Id: I18387e5d2765fecb292127d09e3bbe7cf19b5efd
2021-08-19 08:10:05 +00:00
Spandan Das
73de70264a Apply pylint to cc/symbolfile
PEP8 recommends using the not operator over comparision to empty string

Test: pytest build/soong/cc/symbolfile/test_symbolfile.py
Test: pylint --rcfile tools/repohooks/tools/pylintrc
build/soong/cc/symbolfile
Bug: 195738175

Change-Id: I4ed2bc0680beccc19a63d8b3f383983d4dff1baf
2021-08-16 18:59:57 +00:00
Dan Albert
56f52de8bd Add systemapi as an APEX synonym for stub maps.
The apex tag is currently used for platform-to-APEX, APEX-to-APEX, and
APEX-to-platform, and it's difficult to tell when reviewing the map
files. Add a synonym so authors can be explicit about platform-to-APEX
since those have different stability rules.

Test: pytest
Test: mypy
Test: pylint
Test: treehugger
Bug: None
Change-Id: I0aee679a05b1b2d3749307434c19486bf4c7fe52
2021-06-10 16:38:19 -07:00
Dan Albert
ead21552ee Refactor tag handling code in stub generator.
Test: pytest
Test: pylint
Test: mypy
Test: treehugger
Bug: None
Change-Id: I7fb380a3d879b4abcdadfe4aafff0ca4098d2e07
2021-06-10 16:38:19 -07:00
Bob Badour
02040de891 Add LOCAL_LICENSE_KINDS to build/soong
Added SPDX-license-identifier-Apache-2.0 to:
  Android.bp
  android/Android.bp
  android/soongconfig/Android.bp
  androidmk/Android.bp
  apex/Android.bp
  bazel/Android.bp
  bp2build/Android.bp
  bpf/Android.bp
  bpfix/Android.bp
  cc/Android.bp
  cc/config/Android.bp
  cc/libbuildversion/Android.bp
  cc/libbuildversion/tests/Android.bp
  cc/ndk_api_coverage_parser/Android.bp
  cc/ndkstubgen/Android.bp
  cc/symbolfile/Android.bp
  cmd/dep_fixer/Android.bp
  cmd/diff_target_files/Android.bp
  cmd/extract_apks/Android.bp
  cmd/extract_jar_packages/Android.bp
  cmd/extract_linker/Android.bp
  cmd/fileslist/Android.bp
  cmd/host_bionic_inject/Android.bp
  cmd/javac_wrapper/Android.bp
  cmd/merge_zips/Android.bp
  cmd/multiproduct_kati/Android.bp
  cmd/path_interposer/Android.bp
  cmd/pom2bp/Android.bp
  cmd/pom2mk/Android.bp
  cmd/sbox/Android.bp
  cmd/soong_build/Android.bp
  cmd/soong_env/Android.bp
  cmd/soong_ui/Android.bp
  cmd/zip2zip/Android.bp
  cmd/zipsync/Android.bp
  cuj/Android.bp
  dexpreopt/Android.bp
  dexpreopt/dexpreopt_gen/Android.bp
  env/Android.bp
  etc/Android.bp
  filesystem/Android.bp
  finder/Android.bp
  finder/cmd/Android.bp
  genrule/Android.bp
  jar/Android.bp
  java/Android.bp
  java/config/Android.bp
  kernel/Android.bp
  linkerconfig/Android.bp
  linkerconfig/proto/Android.bp
  makedeps/Android.bp
  partner/Android.bp
  phony/Android.bp
  python/Android.bp
  python/tests/Android.bp
  remoteexec/Android.bp
  rust/Android.bp
  rust/config/Android.bp
  scripts/Android.bp
  sdk/Android.bp
  sh/Android.bp
  shared/Android.bp
  symbol_inject/Android.bp
  symbol_inject/cmd/Android.bp
  sysprop/Android.bp
  tradefed/Android.bp
  ui/build/Android.bp
  ui/logger/Android.bp
  ui/metrics/Android.bp
  ui/metrics/proc/Android.bp
  ui/status/Android.bp
  ui/terminal/Android.bp
  ui/tracer/Android.bp
  xml/Android.bp
  zip/Android.bp
  zip/cmd/Android.bp

Added SPDX-license-identifier-Apache-2.0 SPDX-license-identifier-BSD to:
  finder/fs/Android.bp
  third_party/zip/Android.bp

Bug: 68860345
Bug: 151177513
Bug: 151953481

Test: m all

Exempt-From-Owner-Approval: janitorial work

Change-Id: Ia47ca14f16b8c9f84f9d533a07e5b00e2c04e8d4
2021-02-06 04:23:21 +00:00
Dan Albert
af7b36dea5 Add type information to symbolfile and ndkstubgen.
Test: mypy symbolfile
Test: pytest
Bug: None
Change-Id: I6b1045d315e5a10e699d31de9fafc084d82768b2
2020-09-24 14:03:31 -07:00
Dan Albert
06f58afd81 Get NDK python script tests running.
Imports weren't working in tests because the package had been created.
The Python "binaries" built by Soong don't seem to take their own
pkg_path into account, so I split the separate pieces of code here out
into their own packages.

Note that the ndk_api_coverage_parser tests do not actually pass
before or after this change (seems like it might be a
non-deterministic ordering issue in the attributes of the generated
output?), but they can at least be run now.

Test: pytest ndkstubgen
Test: pytest symbolfile
Test: pytest ndk_api_coverage_parser
Test: out/host/linux-x86/nativetest64/test_ndkstubgen/test_ndkstubgen
Test: out/host/linux-x86/nativetest64/test_symbolfile/test_symbolfile
Test: out/host/linux-x86/nativetest64/test_ndk_api_coverage_parser/test_ndk_api_coverage_parser
Bug: None
Change-Id: I2ac22f7ced7566e4808070f2f72fd04355846e0b
2020-07-16 13:23:29 -07:00