Home
last modified time | relevance | path

Searched refs:as (Results 1 – 25 of 27) sorted by relevance

12

/bionic/docs/
Ddefines.md10 is typically a good choice when you use libc API that's only in bionic, such as
18 useful as you might think, and one of the other choices on this page is usually
31 One thing to note (if your code may also be built as part of the OS itself) is
33 obvious "next" API level such as 19. Once the API level has been decided, the
39 choice when you use Linux-specific API, such as a Linux-specific system call or
46 If your code can be built either as part of an app _or_ as part of the OS
48 This is typically a good choice when your code uses non-NDK API if it's built as
Dnative_allocator.md92 of allocation routines such as what happens when a non-power of two alignment
204 results, but, as mentioned before, these microbenchmark numbers should
205 not be used as absolutes for determining if an allocator is worth using.
253 in a real world allocation scenario. SQL operations were chosen as a
267 These numbers should be as performant as the current allocator.
279 nearly as performant as the current allocator.
319 on a thread as if it was rerunning the real trace. The only issue is that
344 checked in as zip files.
358 they should not be used as absolute value for comparison against the
359 current allocator. But, they should be in the same range as the current
[all …]
Dfdsan.md8as *use-after-close* and *double-close*. These errors are direct analogues of the memory allocatio…
43 …how most memory allocations can have their ownership handled by types such as `std::unique_ptr`, a…
49 ### Enabling fdsan (as a user)
57 - Same as warn-once, except without disabling after the first warning.
119 When running the program, the threads' executions will be interleaved as follows:
186 …d for closing `victim`'s fd. If we store `bystander`'s fd in a `unique_fd` as well, we should get …
227 In a real application, things are probably not going to be as detectable or reproducible as our toy…
229 ### Enabling fdsan (as a C++ library implementer)
Dstatus.md54 * Full C11 `<threads.h>` (available as inlines for older API levels).
76 * Using `%n` with the printf family is now reported as a FORTIFY failure.
107 runtime and reported as a FORTIFY failure.
109 reported as a FORTIFY failure.
136 reported as a FORTIFY failure. Most commonly this is a result of confusing
284 Note that this doesn't change behavior as much as you might think: the
311 inherently non-portable as-is.) The best alternative on Android is to
332 overrun is detected, the program is safely aborted as in this
D32-bit-abi.md13 to make `off64_t` and corresponding functions such as `ftruncate64` available.
24 and replaces each `off_t` function with its `off64_t` counterpart, such as
91 The header `<time64.h>` and type `time64_t` exist as a workaround,
109 mutexes for tids that don't fit in 16 bits. This typically manifests as
Dlibc_assembler.md118 NOTE: It is assumed that these commands are executed in adb as the shell user
120 machine, you might need to escape the special shell characters such as **$**.
130 When verifying an assembler routine that operates on buffer data (such as
163 to verify unwind cases, such as around branches, where unwind information
Delf-tls.md122 variables that are both part of the same module as the reference. Instead of a call to
225 As long as a shared object is one of the initially-loaded modules, a better option is to use
404 > ... All OSes use basically the same algorithm (a per-module lookup table) as detailed in Ulrich
407 > expose the structure of the DTV via metadata exported as symbols from the .so itself, designed
558 Maybe Bionic should do the same as musl? Perhaps musl's robustness argument holds for Bionic,
808 (`0x23581321` for arm32 and `0x23581321345589` for arm64). As long as it finds its constant at a
825 AOSP hikey960 build, only `/system/bin/netd` has a TLS segment, and it's only 32 bytes. As long as
854 overall ELF TLS design and ABI details for x86 and x86-64 (as well as several other architectures
/bionic/libdl/
DNOTICE11 and distribution as defined by Sections 1 through 9 of this document.
37 Object form, made available under the License, as indicated by a
44 represent, as a whole, an original work of authorship. For the purposes
61 designated in writing by the copyright owner as "Not a Contribution."
77 (except as stated in this section) patent license to make, have made,
88 as of the date such litigation is filed.
107 (d) If the Work includes a "NOTICE" text file as part of its
113 as part of the Derivative Works; within the Source form or
120 or as an addendum to the NOTICE text from the Work, provided
122 as modifying the License.
[all …]
DAndroid.bp63 // for x86, exclude libgcc_eh.a for the same reasons as above
155 // for x86, exclude libgcc_eh.a for the same reasons as above
/bionic/libc/async_safe/
DREADME.md9 costly to connect a socket for each thread as some processes, such as system_server, have over 100
/bionic/
DREADME.md6 # Using bionic as an app developer
50 The `tests/` directory contains unit tests. Roughly arranged as one file per
154 that would make more sense as the place to add the wrapper.
200 If you make a change that is likely to have a wide effect on the tree (such as a
205 either, as `make tests` covers a few additional modules, but generally speaking
232 Most of the unit tests are executed by CTS. By default, CTS runs as
233 a non-root user, so the unit tests must also pass when not run as root.
234 Some tests cannot do any useful work unless run as root. In this case,
249 $ adb unroot # Because real CTS doesn't run as root.
264 You can supply gtest flags as extra arguments to this script.
[all …]
Dandroid-changes-for-ndk-developers.md12 any architecture, as we will be doing basic inspection only. On Linux
23 a warning, as appropriate.
32 as toasts. Experience has shown that many developers don’t habitually
91 to dlopen(3) (as opposed to being referenced by DT_NEEDED entries).
134 your APK. Just use `System.loadLibrary("foo")` exactly as normal but set
139 and reported to the user as space taken up by your app). Any .so file
163 private C/C++ symbols. Private symbols aren't tested as part of the
167 as many developers found when Android 6.0 Marshmallow switched from
239 relocations. That is, the code must be loaded as is and must not be
243 hand-written assembler. This is not common. Use the scanelf tool as
[all …]
/bionic/libc/malloc_debug/
DREADME_marshmallow_and_earlier.md17 run as root on the device.
78 A 32 byte buffer is placed before the returned allocation (known as
81 as a rear guard). This buffer is filled with the pattern (0xbb).
100 are printed to the log as leaks. This isn't very useful since it tends
DREADME_api.md38 *backtrace\_size* as returned by the original call to
42 backtrace and size as this allocation. On Android Nougat, this value was
DREADME.md118 Android devices). The format of this dumped data is the same format as
466 generated by am dumpheap -n or, as of P, by the signal or on exit.
529 NUM\_ALLOCATIONS value should be treated as always 1 no matter what is
636 It is possible to use the backtrace\_enable\_on\_signal option as well,
/bionic/benchmarks/linker_relocation/
DAndroid.bp35 // library in the "data" property, but that doesn't work, because each library is listed as having a
36 // "link:shared" variant (as opposed to "link:static"), whereas the cc_benchmark has no
/bionic/benchmarks/
DREADME.md34 the file will be found as relative to the current directory. If the option specifies the full path
35 to an XML file such as `/data/nativetest/suites/example.xml`, it will be used as-is.
/bionic/linker/
DAndroid.bp9 // embedded as the entry point, and the linker is embedded as ELF sections in
386 // for x86, exclude libgcc_eh.a for the same reasons as above
/bionic/benchmarks/spawn/
DAndroid.bp69 // Install these binaries in the same directory as the main benchmark binary.
/bionic/libc/dns/nameser/
Dns_name.c755 ns_name_eq(ns_nname_ct a, size_t as, ns_nname_ct b, size_t bs) { in ns_name_eq() argument
756 ns_nname_ct ae = a + as, be = b + bs; in ns_name_eq()
/bionic/libc/kernel/
DREADME.md73 to determine which directory to use as the destination directory.
/bionic/libc/malloc_hooks/
DREADME.md59 There is no hook for malloc\_usable\_size as of now.
/bionic/tests/
DAndroid.bp483 // library as an emutls symbol. The -fno-emulated-tls flag above configures
513 // library as an emutls symbol. The -fno-emulated-tls flag above configures
/bionic/libc/
DSYSCALLS.TXT15 # function, which is not the same as the standard C exit() function which calls it)
263 # sockets for x86. These are done as an "indexed" call to socketcall syscall.
DAndroid.bp236 // Use the empty string (instead of " ") as the timezone abbreviation
1191 // they can't be included in libc_ndk.a (as the layout of
1320 // data structure that has changed across releases (such as
1621 // then libc would not pull those symbols from libgcc.a as it should, instead relying
1819 // (such as libunwind_llvm) depend on the headers. Setting sdk_version to 1

12