* Allow sanitization of libc (excluding existing global sanitizers)
and disallow sanitization of linker. The latter has not been
necessary before because HWASan is the first sanitizer to support
static binaries (with the exception of CFI, which is not used
globally).
* Static binary startup: initialize HWASan shadow very early so that
almost entire libc can be sanitized. The rest of initialization is
done in a global constructor; until that is done sanitized code can
run but can't report errors (will simply crash with SIGTRAP).
* Switch malloc_common from je_* to __sanitizer_*.
* Call hwasan functions when entering and leaving threads. We can not
intercept pthread_create when libc depends on libclang_rt.hwasan.
An alternative to this would be a callback interface like requested
here:
https://sourceware.org/glibc/wiki/ThreadPropertiesAPI
All of the above is behind a compile-time check
__has_feature(hwaddress_sanitizer). This means that HWASan actually
requires libc to be instrumented, and would not work otherwise. It's
an implementation choice that greatly reduces complexity of the tool.
Instrumented libc also guarantees that hwasan is present and
initialized in every process, which allows piecemeal sanitization
(i.e. library w/o main executable, or even individual static
libraries), unlike ASan.
Change-Id: If44c46b79b15049d1745ba46ec910ae4f355d19c
libc had some -Wimplicit-fallthrough warnings. They all seem to be
benign. We're trying to enable this flag globally, so we need to
annotate these breaks here.
Bug: 112564944
Test: Builds
Change-Id: I5afae694cc4cf26ad1a61e2c8ae91f00cda7c733
This runs through the trace of the allocations in a sql benchmark app
executed in the benchmark thread.
Add one benchmark with decay time set to 0 and another with decay time
set to 1.
Include a script that can generate a header file that can be used to
regenerate the data.
Bug: 112317428
Test: Builds, ran unit tests, ran benchmarks.
Change-Id: I62e287cc06b74b74bcc5a4bbee71b0fac0a196fd
1. fix all_benchmarks. Some new benchmarks were added.
2. move some benchmark suites to the right place.
3. add unit test for https://android-review.googlesource.com/c/platform/bionic/+/724785
Test: compile and run bionic-benchmarks-tests on device.
Change-Id: I2b686b51910a422b664f9ea968199f37944039c0
This patch adds a new strcmp/strcmp benchmark with different set of
inputs. It cover all pre-defined range values (small, medium, large)
for aligned inputs and also small range for unaligned ones.
Test: Ran new string test suite with a single bionic iteration.
Change-Id: Ice78d74c3583a33158eaba3309c10df54044aa28
This patch adds a new memcmp benchmark with different set of inputs.
It covers all pre-defined range values (small, medium, large) for
aligned inputs and also medium range for unaligned ones.
Test: Ran new string test suite with a single bionic iteration.
Change-Id: Icdb95220a62d46bd71fa6ffbbf4567a9a05b7685
This patch add an option to use the pre-define set of ranges along
with macros AT_ONEBUF_MANUAL_ALIGN_* and AT_TWOBUF_MANUAL_ALIGN1_*.
The size argument can be either a number or a string representing
the sets of values:
* SMALL for values between 1 and 256.
* MEDIUM for values between 512 and 128KB.
* LARGE for values between 256KB and 2048KB.
Test: Ran new string test suite with a single bionic iteration.
Change-Id: Ieda81ee9a5019991b0b2f97d4ca3a237127c5848
This change allows the use of read(2) and getdents(2) so that UBSan can
work correctly on x86_64.
Bug: 111999822
Test: CtsWrapWrapDebugTestCases
Change-Id: I44aac17fffdffc5c3f5d87fb3d876af28c8044b2
This patch ues exp, exp2, log, log2, and pow from arm optimized
routines. For pow on x86_64, although slight slower it simplifies
the code required on both bionic and arm-optimized-routines (so
there is no need to select and export the symbol based on
architecture).
Performance-wise the improvements are:
x86_64 throughput latency
exp 1.16x 1.16x
log 1.08x 0.95x
exp2 1.27x 1.55x
log2 1.40x 1.43x
pow 0.77x 0.89x *
* I tried to check if AVX2/FMA but without success.
aarch64 throughput latency
exp 2.33x 2.16x
exp2 1.99x 1.50x
log 1.79x 1.43x
log2 2.15x 1.80x
pow 3.81x 3.07x
Test: ran bionic tests on static mode.
Change-Id: Ib16bf3280c5329fd257a3b3f0b6c4f2f3cb34deb
This is currently a non-functional change, because base_timeout_msec is
never set to a non-default value (other than in unit tests).
Test: netd_unit_test --gtest_filter='ResolverTest.*'
Change-Id: I21989bab5cc9f55ee1392e30965eb405d720a28e
EXPECT_DEATH requires special handling to work with fdsan, because of
our fork detection. Instead of duplicating the EXPECT_FDSAN_DEATH macro
in the libbase tests as well, move the unique_fd fdsan tests here.
Test: bionic_unit_tests
Change-Id: Ia7b65d4560601d5a78d143aced887a6773b401c0
As for exp/exp2 benchmark, this patch add two benchmark for pow:
one which measures thoughput and one which measures latency.
The input data is the same as powf.
Test: ran 32-bit and 64-bit x86 tests on host
Change-Id: I04335fac9e76fb3f39935323dacf6b7be6a6f917
Similar to exp/exp2, this patch add two benchmarks for log and log2:
one which measures thoughput and one which measures latency.
The input data is based on logf/log2f one (reduced trace based on 2.8
billion samples extracted from specpu2017 521.wrf_r benchmark).
Test: ran 32-bit and 64-bit x86 tests on host
Change-Id: I51a9e2960e45d5619585e685eaa7634cc7be901b
This patch add two benchmarks for both exp and exp2, one which measures
thoughput and one which measures latency. The latency benchmark works by
creating a dependency on the previous iteration and with a zero multiply
to avoid changing the input value.
The input is based on expf/exp2f benchmark (reduced trace based on 2.4
billion samples extracted from specpu2017 521.wrf_r benchmark).
Test: ran 32-bit and 64-bit x86 tests on host
The primary motivation is speeding up tests for non-responsive
nameservers. It's also base infrastructure for future work to compute
retry timeouts dynamically.
Using default timeouts (RES_TIMEOUT = 5 seconds):
ResolverTest.GetAddrInfoV6_nonresponsive (45141 ms)
With base_timeout_msec=100:
ResolverTest.GetAddrInfoV6_nonresponsive (1264 ms)
Test: executed unsubmitted test from aosp/713993
Change-Id: Id902089ca69ca8d28032180ba51e8937262ef490
Let the debuggerd handler get the address of the fdsan table so that
crash_dump can parse it and print it out in tombstones.
Test: debuggerd `pidof adbd`
Change-Id: I2297200cae52d0ff24479dba55d1ab4938583501
Make it easier for other code to parse the fdsan table themselves by
making it so that they can include the bionic_fdsan.h header to get a
struct definition with the layout without bringing in other
dependencies.
Test: treehugger
Change-Id: I3583ef113991234aa83cd193a4eb139ad06737c7
This change avoids having to run the genseccomp.py script every time a
policy file is edited, and instead generates these files at
compile-time.
Bug: None
Test: m
Test: find out/soong/ -name x86_64_global_policy.cpp # Shows files
Test: generated policies are equivalent to original policies
Change-Id: I12461fe0c5fb02c008c1b2503fbb994b8aa2f56b
