The executable can be inside a zip file using the same syntax used for
shared objects: path.zip!/libentry.so.
The linker currently requires an absolute path. This restriction could be
loosened, but it didn't seem important? If it allowed non-absolute paths,
we'd need to decide how to handle:
- foo/bar (relative to CWD?)
- foo (search PATH / LD_LIBRARY_PATH, or also relative to CWD?)
- foo.zip!/bar (normalize_path() requires an absolute path)
The linker adjusts the argc/argv passed to main() and to constructor
functions to hide the initial linker argument, but doesn't adjust the auxv
vector or files like /proc/self/{exe,cmdline,auxv,stat}. Those files will
report that the kernel loaded the linker as an executable.
I think the linker_logger.cpp change guarding against (g_argv == NULL)
isn't actually necessary, but it seemed like a good idea given that I'm
delaying initialization of g_argv until after C++ constructors have run.
Bug: http://b/112050209
Test: bionic unit tests
Change-Id: I846faf98b16fd34218946f6167e8b451897debe5
The tls_nodelete state should apply to load_group not
isolated soinfo. This actually also means that multiple
soinfos may have tls_counter on their dso_handles.
This change replaces TLS_NODELETE flag with secondary counter.
Note that access to the secondary counter (located inside soinfo)
is pretty expensive because it requires soinfo lookup by dso_handle
whereas dso_handle counter is much faster. This is why it is updated
only when dso_handle counter starts or hits 0.
Bug: http://b/80278285
Test: bionic-unit-tests --gtest_filter=dl*
Change-Id: I535583f6714e45fa2a7eaf7bb3126da20ee7cba9
* Since Android does not have __tls_get_addr yet,
-fno-emulated-tls is only a test of compiler.
Lld is correct to reject this at link time.
But if the purpose of this test is to check at run-time,
allow_undefined-symbols is the trick to get the linked library.
Bug: http://b/74361956
Test: build and compare linked file
Change-Id: Ibcdc8b50aeeaec9d7d308b66df2a6997ce3c1a4f
This commit allows users to create a link without soname filters between
two linker namespaces.
The motivation is to establish one-way shared library isolation. For
example, assume that there are two linker namespaces `default` and
`vndk`. We would like to limit the shared libraries that can be used by
the `default` namespace. In the meanwhile, we would like to allow the
`vndk` namespace to use shared libs from the `default` namespace if the
soname cannot be find in the search path or loaded sonames of the `vndk`
namespace.
shared_libs = %VNDK_CORE_LIBRARIES%
shared_libs += %VNDK_SAMEPROCESS_LIBRARIES%
vndk <-------------------------------------------- default
\_______________________________________________/^
allow_all_shared_libs = true
android_link_namespaces_all_libs() is added to libdl, but it is
versioned as LIBC_PRIVATE. android_link_namespaces_all_libs() is only
for unit tests.
Bug: 69824336
Test: adb shell /data/nativetest/linker-unit-tests/linker-unit-tests32
Test: adb shell /data/nativetest64/linker-unit-tests/linker-unit-tests64
Test: adb shell /data/nativetest/bionic-unit-tests/bionic-unit-tests
Test: adb shell /data/nativetest64/bionic-unit-tests/bionic-unit-tests
Test: Update /system/etc/ld.config*.txt and check whether the vndk
linker namespace of the vendor process can access the shared libs from
the default linker namespace.
Change-Id: I2879f0c5f5af60c7e56f8f743ebd2872e552286b
Introduce new flag to mark soinfo as TLS_NODELETE when
there are thread_local dtors associated with dso_handle
belonging to it.
Test: bionic-unit-tests --gtest_filter=dl*
Test: bionic-unit-tests-glibc --gtest_filter=dl*
Bug: https://github.com/android-ndk/ndk/issues/360
Change-Id: I724ef89fc899788f95c47e6372c38b3313f18fed
This change addresses multiple problems introduced by
02586a2a34
1. In the case of unsuccessful dlopen the failure guard is triggered
for two namespaces which leads to double unload.
2. In the case where load_tasks includes libraries from 3 and more
namespaces it results in incorrect linking of libraries shared between
second and third/forth and so on namespaces.
The root cause of these problems was recursive call to find_libraries.
It does not do what it is expected to do. It does not form new load_tasks
list and immediately jumps to linking local_group. Not only this skips
reference counting it also will include unlinked but accessible library
from third (and fourth and fifth) namespaces in invalid local group. The
best case scenario here is that for 3 or more namesapces this will
fail to link. The worse case scenario it will link the library
incorrectly with will lead to very hard to catch bugs.
This change removes recursive call and replaces it with explicit list of
local_groups which should be linked. It also revisits the way we do
reference counting - with this change the reference counts are updated after
after libraries are successfully loaded.
Also update soinfo_free to abort in case when linker tries to free same
soinfo for the second time - this makes linker behavior less undefined.
Test: bionic-unit-tests
Bug: http://b/69787209
Change-Id: Iea25ced181a98c6503cce6e2b832c91d697342d5
This reverts commit 58554ccb8a.
causes /vendor/bin/qseecomd to hit the new abort:
[ 8.983301] c5 603 DEBUG: Abort message: 'soinfo=0x7147894cd0 is not in soinfo_list (double unload?)'
Bug: http://b/69909887
Bug: http://b/69787209
Change-Id: Ied38f797e0a071a1acc5ed41adf1b45e855143c7
In the case of unsuccessful dlopen the failure guard is triggered
for two namespaces which leads to double unload.
Also update soinfo_free to abort in case when linker tries to free same
soinfo for the second time - this makes linker behavior less undefined.
Test: bionic-unit-tests
Bug: http://b/69787209
Change-Id: I886787ee021b050667f967bce7aa2708390886ea
there's no use case for prepending/appending to bool, and string
properties within module struct. Declearing "*bool" and "*string" almost
cover everything user need.
Test: m -j checkbuild
Bug: b/68853585
Change-Id: I8342d4a43fb81f2842532ddb38fba414767ac82a
Added a test to make sure linker produces correct error message
when user attempts to load a library in a linked namespace and fails.
Bug: http://b/67866190
Bug: http://b/64950640
Bug: http://b/64888291
Test: bionic-unit-test --gtest_filter=dlext*
Change-Id: I5b5c2070d1388eff123118350b2b5c8fc7571a29
libdl is part of system_shared_libs now. -ldl -lpthread -lm are now defaults
for host_ldlibs on Linux and Darwin. -lrt is a default for host_ldlibs on
Linux.
Test: m host
Test: mmma bionic
Change-Id: I966e2f88c24fba5e412bee6b6382045a2026a8e4
With ld.config.txt, we now have multiple built-in namespaces other than
the default namespace. Libs (and their dependents) listed in LD_PRELOAD
must be visible to those additional namespaces as well.
This also adds a debugging only feature: path to the linker config file
can be customized via LD_CONFIG_FILE environment variable. This works
only for debuggable builds.
Bug: 38114603
Bug: 62815515
Test: 1. ./external/compiler-rt/lib/asan/scripts/asan_device_setup --lib
prebuilts/clang/host/linux-x86/clang-stable/lib64/clang/5.0/lib/linux
2. enable talkback shortcut
3. in the home screen, hold vol-up/down together
4. device does not reboots and talkback shortcut is toggled
Test: bionic-unit-tests and linker-unit-tests successful
Change-Id: I9a03591053f4a9caea82f0dcb23e7a3d324bb9bd
Used by CFI, so broke cfi_test#early_init@x86, but I've added a specific
test for this (and a similar test for getauxval from preinit, which this
patch does not fix).
Bug: http://b/35885875
Test: ran tests
Change-Id: I43885bedfb88c0a26b4474bd3c27a87dec7bbc97
In the case when there are multiple dependencies on
the same library in the local_group the unload may
in some situations (covered now by tests) result
calling d-tors for some libraries prematurely.
In order to have correct call order loader checks if this
is last dependency in local group before adding it to BFS
queue.
Bug: http://b/35201832
Test: bionic-unit-tests --gtest_filter=dl*:Dl*
Test: bionic-unit-tests-glibc --gtest_filter=dl*
Change-Id: I4c6955b9032acc7147a51d9f09b61d9e0818700c
Also add another test for dlsym(RTLD_NEXT, ..)
Bug: http://b/33106624
Test: run bionic-unit-tests-glibc and bionic-unit-tests
Change-Id: I340165d66bf2360b0e3273d3561a08cb5e7bd025
Host builds were compiling without -Wl,--no-undefined. Add missing
host ldlibs so that -Wl,--no-undefined can be added again.
Test: m -j host
Bug: 32305815
Change-Id: I456098775ead2bddae3e61c7055b2bde3f6b38ff
Fix rpath value for libraries after build system update
Bug: http://b/31221587
Test: run bionic-unit-tests --gtest_filter=dl*:Dl*:*atexit*
Change-Id: Iaed834bef830fd0b252b5de82e27634bafc92d8b
This shows up when you're doing make HOST_SANITIZE=address
Test: mmma bionic && out/host/linux-x86/bin/bionic_tests_zipalign (compare before and after)
Change-Id: Ia94790496327f3818d5fdb7b5ad07e14e60bcae1
Move libraries used in bionic-unit-tests out of system partition to
/data/nativetests../bionic-loader-test-libs
Bug: http://b/22182538
Test: build bionic and run bionic-unit-tests
Change-Id: I170177bef782839d0b4970ae4418bf54d0a77836
The compile-time tests and a few custom libraries for dynamic linker
testing are still compiled in make.
Also converts the make rules to run tests on the host to shell scripts
in tests/run-on-host.sh and benchmarks/run-on-host.sh
Change-Id: I6f174b3a69d58c4ed74d29f4e79332d483681534
