tequilaOS/platform_bionic
2
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
platform_bionic/libc/Android.mk

1519 lines
52 KiB
Makefile
Raw Normal View History

AArch64: Add initial support for AArch64 This is the first patch out of a series of patches that add support for AArch64, the new 64bit execution state of the ARMv8 Architecture. The patches add support for LP64 programming model. The patch adds: * "arch-aarch64" to the architecture directories. * "arch-aarch64/include" - headers used by libc * "arch-aarch64/bionic": - crtbegin, crtend support; - aarch64 specific syscall stubs; - setjmp, clone, vfork assembly files. Change-Id: If72b859f81928d03ad05d4ccfcb54c2f5dbf99a5 Signed-off-by: Serban Constantinescu <serban.constantinescu@arm.com>
2013-09-26 12:37:10 +02:00
LOCAL_PATH := $(call my-dir)
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
Fix bionic_coverage for kati. Kati won't override something set with :=, only ?=. Note that the actually reveals that our coverage build has been broken for... some time. The issue is that we have static libraries built with clang and others built with gcc linked into the same .so. If we use both libclang_rt.profile and libgcov together, we get duplicate symbol errors. If we only use one or the other, we don't have all the symbols we need. This will be fixed once chh's TLS stuff makes it into our clang and we can start building libc_thread_atexit_impl with clang. Change-Id: I45d0f2e558a881ad8be71088102320724ced8fe5
2015-10-08 23:04:53 +02:00
bionic_coverage ?= false
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
Add dependencies on included makefiles. Bug: 11050594 Merge from internal master. (cherry-picked from f389284e86bbcbdb257d7388a17fde007113b0d6) Change-Id: I2b3e38329a09d26c16870906f9ed1257e2a9dbc8
2013-10-03 22:26:22 +02:00
# Make everything depend on any changes to included makefiles.
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
libc_common_additional_dependencies := $(LOCAL_PATH)/Android.mk
# Load config for TARGET_ARCH
my_2nd_arch_prefix :=
include $(LOCAL_PATH)/arch-$(TARGET_ARCH)/$(TARGET_ARCH).mk
libc_common_additional_dependencies += \
$(LOCAL_PATH)/arch-$(TARGET_ARCH)/$(TARGET_ARCH).mk
ifdef TARGET_2ND_ARCH
# Load config for TARGET_2ND_ARCH
my_2nd_arch_prefix := $(TARGET_2ND_ARCH_VAR_PREFIX)
include $(LOCAL_PATH)/arch-$(TARGET_2ND_ARCH)/$(TARGET_2ND_ARCH).mk
my_2nd_arch_prefix :=
libc_common_additional_dependencies += \
$(LOCAL_PATH)/arch-$(TARGET_2ND_ARCH)/$(TARGET_2ND_ARCH).mk
endif
# crt obj files
# ========================================================
# crtbrand.c needs <stdint.h> and a #define for the platform SDK version.
libc_crt_target_cflags := \
-I$(LOCAL_PATH)/include \
-DPLATFORM_SDK_VERSION=$(PLATFORM_SDK_VERSION) \
my_2nd_arch_prefix :=
include $(LOCAL_PATH)/crt.mk
ifdef TARGET_2ND_ARCH
my_2nd_arch_prefix := $(TARGET_2ND_ARCH_VAR_PREFIX)
include $(LOCAL_PATH)/crt.mk
my_2nd_arch_prefix :=
endif
Add dependencies on included makefiles. Bug: 11050594 Merge from internal master. (cherry-picked from f389284e86bbcbdb257d7388a17fde007113b0d6) Change-Id: I2b3e38329a09d26c16870906f9ed1257e2a9dbc8
2013-10-03 22:26:22 +02:00
libc: Cleanup the libc makefile. Removes a lot of redundant code, and reorganizes the file. We first setup all the comomn variables we may need, and then define the actual libraries that need to be generated. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
# Define the common source files for all the libc instances
# =========================================================
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
libc_common_src_files := \
Fix missing args. Change-Id: I89a366db984ea6b2553896a4351e59bb91f93040
2014-02-13 01:40:55 +01:00
bionic/bindresvport.c \
bionic/ether_aton.c \
bionic/ether_ntoa.c \
bionic/fts.c \
bionic/getpriority.c \
bionic/initgroups.c \
bionic/isatty.c \
bionic/memmem.c \
bionic/pututline.c \
bionic/sched_cpualloc.c \
bionic/sched_cpucount.c \
bionic/sigblock.c \
bionic/siginterrupt.c \
bionic/sigsetmask.c \
bionic/system_properties_compat.c \
Avoid pathological behavior in OpenBSD's fread. Bug: https://code.google.com/p/android/issues/detail?id=81155 Bug: 18556607 Change-Id: Idc60976b79610e2202cc42dc393dcb4ca6c42e05
2014-12-02 01:13:30 +01:00
stdio/fread.c \
Fix stdio read after EOF behavior. Bug: https://code.google.com/p/android/issues/detail?id=184847 Change-Id: Ia20ce94007c2a09649f0763b1dc7ba959f2f618d
2015-10-10 02:57:26 +02:00
stdio/refill.c \
Clean up sys_signame and sys_siglist a little. We don't need quite so much duplication because we already have a way to get the signal number from its name, and that already copes with the fact that the mips/mips64 numbers are different from everyone else's. Also remove sys_signame from LP64. glibc doesn't have this BSD-ism. Change-Id: I6dc411a3d73589383c85d3b07d9d648311492a10
2014-02-12 04:57:06 +01:00
stdio/snprintf.c\
stdio/sprintf.c \
Move stdio implementation details around a little. Change-Id: I24594426d5479bdd55cbef0ab1b7d76c249dbd0c
2016-01-20 01:28:15 +01:00
stdio/stdio.cpp \
Add <stdio_ext.h> for elfutils. Bug: 17139679 Change-Id: I1605ac382dbb6f23b2d874dbb9769f3cde4a6a99
2014-08-20 02:00:33 +02:00
stdio/stdio_ext.cpp \
Unregister pthread_atfork handlers on dlclose() Bug: http://b/20339788 Change-Id: I874c87faa377645fa9e0752f4fc166d81fd9ef7e
2014-11-21 05:47:02 +01:00
stdlib/atexit.c \
Call __cxa_thread_finalize for the main thread. Bug: http://b/20231984 Bug: http://b/16696563 Change-Id: I71cfddd0d404d1d4a593ec8d3bca9741de8cb90f
2015-04-16 00:31:51 +02:00
stdlib/exit.c \
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
Optimize __memset_chk, __memcpy_chk. DO NOT MERGE. This change creates assembler versions of __memcpy_chk/__memset_chk that is implemented in the memcpy/memset assembler code. This change avoids an extra call to memcpy/memset, instead allowing a simple fall through to occur from the chk code into the body of the real implementation. Testing: - Ran the libc_test on __memcpy_chk/__memset_chk on all nexus devices. - Wrote a small test executable that has three calls to __memcpy_chk and three calls to __memset_chk. First call dest_len is length + 1. Second call dest_len is length. Third call dest_len is length - 1. Verified that the first two calls pass, and the third fails. Examined the logcat output on all nexus devices to verify that the fortify error message was sent properly. - I benchmarked the new __memcpy_chk and __memset_chk on all systems. For __memcpy_chk and large copies, the savings is relatively small (about 1%). For small copies, the savings is large on cortex-a15/krait devices (between 5% to 30%). For cortex-a9 and small copies, the speed up is present, but relatively small (about 3% to 5%). For __memset_chk and large copies, the savings is also small (about 1%). However, all processors show larger speed-ups on small copies (about 30% to 100%). Bug: 9293744 Merge from internal master. (cherry-picked from 7c860db0747f6276a6e43984d43f8fa5181ea936) Change-Id: I916ad305e4001269460ca6ebd38aaa0be8ac7f52
2013-08-01 22:13:33 +02:00
# Fortify implementations of libc functions.
libc_common_src_files += \
FORTIFY_SOURCE: Add __FD_* checks Add FORTIFY_SOURCE checks for the following macros: * FD_CLR * FD_ISSET * FD_SET Bug: 11047121 Change-Id: I3c5952136aec9eff3288b91b1318677ff971525c
2013-10-03 01:11:30 +02:00
bionic/__FD_chk.cpp \
Optimize __memset_chk, __memcpy_chk. DO NOT MERGE. This change creates assembler versions of __memcpy_chk/__memset_chk that is implemented in the memcpy/memset assembler code. This change avoids an extra call to memcpy/memset, instead allowing a simple fall through to occur from the chk code into the body of the real implementation. Testing: - Ran the libc_test on __memcpy_chk/__memset_chk on all nexus devices. - Wrote a small test executable that has three calls to __memcpy_chk and three calls to __memset_chk. First call dest_len is length + 1. Second call dest_len is length. Third call dest_len is length - 1. Verified that the first two calls pass, and the third fails. Examined the logcat output on all nexus devices to verify that the fortify error message was sent properly. - I benchmarked the new __memcpy_chk and __memset_chk on all systems. For __memcpy_chk and large copies, the savings is relatively small (about 1%). For small copies, the savings is large on cortex-a15/krait devices (between 5% to 30%). For cortex-a9 and small copies, the speed up is present, but relatively small (about 3% to 5%). For __memset_chk and large copies, the savings is also small (about 1%). However, all processors show larger speed-ups on small copies (about 30% to 100%). Bug: 9293744 Merge from internal master. (cherry-picked from 7c860db0747f6276a6e43984d43f8fa5181ea936) Change-Id: I916ad305e4001269460ca6ebd38aaa0be8ac7f52
2013-08-01 22:13:33 +02:00
bionic/__fgets_chk.cpp \
add fortified implementations of fread/fwrite A __size_mul_overflow utility is used to take advantage of the checked overflow intrinsics in Clang and GCC (>= 5). The fallback for older compilers is the optimized but less than ideal overflow checking pattern used in OpenBSD. Change-Id: Ibb0d4fd9b5acb67983e6a9f46844c2fd444f7e69
2015-07-18 19:55:51 +02:00
bionic/__fread_chk.cpp \
bionic/__fwrite_chk.cpp \
add a fortified implementation of getcwd Change-Id: Ice3e6d3e1ff07788305dc85f8ee4059baad5fac4
2015-05-20 21:31:26 +02:00
bionic/__getcwd_chk.cpp \
add fortified memchr/memrchr implementations Change-Id: I38c473cc0c608b08f971409a95eb8b853cb2ba1c
2015-04-18 00:16:57 +02:00
bionic/__memchr_chk.cpp \
Optimize __memset_chk, __memcpy_chk. DO NOT MERGE. This change creates assembler versions of __memcpy_chk/__memset_chk that is implemented in the memcpy/memset assembler code. This change avoids an extra call to memcpy/memset, instead allowing a simple fall through to occur from the chk code into the body of the real implementation. Testing: - Ran the libc_test on __memcpy_chk/__memset_chk on all nexus devices. - Wrote a small test executable that has three calls to __memcpy_chk and three calls to __memset_chk. First call dest_len is length + 1. Second call dest_len is length. Third call dest_len is length - 1. Verified that the first two calls pass, and the third fails. Examined the logcat output on all nexus devices to verify that the fortify error message was sent properly. - I benchmarked the new __memcpy_chk and __memset_chk on all systems. For __memcpy_chk and large copies, the savings is relatively small (about 1%). For small copies, the savings is large on cortex-a15/krait devices (between 5% to 30%). For cortex-a9 and small copies, the speed up is present, but relatively small (about 3% to 5%). For __memset_chk and large copies, the savings is also small (about 1%). However, all processors show larger speed-ups on small copies (about 30% to 100%). Bug: 9293744 Merge from internal master. (cherry-picked from 7c860db0747f6276a6e43984d43f8fa5181ea936) Change-Id: I916ad305e4001269460ca6ebd38aaa0be8ac7f52
2013-08-01 22:13:33 +02:00
bionic/__memmove_chk.cpp \
add fortified memchr/memrchr implementations Change-Id: I38c473cc0c608b08f971409a95eb8b853cb2ba1c
2015-04-18 00:16:57 +02:00
bionic/__memrchr_chk.cpp \
Fortify poll and ppoll. And remove the test for FD_ZERO fortification, which never made much sense anyway. Change-Id: Id1009c5298d461fa4722189e8ecaf22f0c529536
2015-02-02 18:15:19 +01:00
bionic/__poll_chk.cpp \
add fortified implementations of pread/pread64 Change-Id: Iec39c3917e0bc94371bd81541619392f5abe29b9
2015-04-16 15:07:45 +02:00
bionic/__pread64_chk.cpp \
bionic/__pread_chk.cpp \
add fortified implementations of write/pwrite{,64} These are just based on the read/pread{,64} implementations with the function calls and error messages adjusted as appropriate. The only difference is that the buffer parameters are const. Change-Id: Ida1597a903807f583f230d74bcedffdb7b24fcf6
2015-07-21 03:37:29 +02:00
bionic/__pwrite64_chk.cpp \
bionic/__pwrite_chk.cpp \
FORTIFY_SOURCE: fortify read() Change-Id: Ic7de163fe121db13e00560adb257331bc709814d
2013-10-10 05:16:34 +02:00
bionic/__read_chk.cpp \
add fortified readlink/readlinkat implementations Change-Id: Ia4b1824d20cad3a072b9162047492dade8576779
2015-04-17 17:26:36 +02:00
bionic/__readlink_chk.cpp \
bionic/__readlinkat_chk.cpp \
libc: fortify recvfrom() Fortify calls to recv() and recvfrom(). We use __bos0 to match glibc's behavior, and because I haven't tested using __bos. Change-Id: Iad6ae96551a89af17a9c347b80cdefcf2020c505
2013-09-25 01:32:07 +02:00
bionic/__recvfrom_chk.cpp \
Add stpcpy/stpncpy. Add tests for the above. Add the fortify implementations of __stpcpy_chk and __stpncpy_chk. Modify the strncpy test to cover more cases and use this template for stpncpy. Add all of the fortify test cases. Bug: 13746695 Change-Id: I8c0f0d4991a878b8e8734fff12c8b73b07fdd344
2014-04-04 23:38:18 +02:00
bionic/__stpcpy_chk.cpp \
bionic/__stpncpy_chk.cpp \
Optimize __memset_chk, __memcpy_chk. DO NOT MERGE. This change creates assembler versions of __memcpy_chk/__memset_chk that is implemented in the memcpy/memset assembler code. This change avoids an extra call to memcpy/memset, instead allowing a simple fall through to occur from the chk code into the body of the real implementation. Testing: - Ran the libc_test on __memcpy_chk/__memset_chk on all nexus devices. - Wrote a small test executable that has three calls to __memcpy_chk and three calls to __memset_chk. First call dest_len is length + 1. Second call dest_len is length. Third call dest_len is length - 1. Verified that the first two calls pass, and the third fails. Examined the logcat output on all nexus devices to verify that the fortify error message was sent properly. - I benchmarked the new __memcpy_chk and __memset_chk on all systems. For __memcpy_chk and large copies, the savings is relatively small (about 1%). For small copies, the savings is large on cortex-a15/krait devices (between 5% to 30%). For cortex-a9 and small copies, the speed up is present, but relatively small (about 3% to 5%). For __memset_chk and large copies, the savings is also small (about 1%). However, all processors show larger speed-ups on small copies (about 30% to 100%). Bug: 9293744 Merge from internal master. (cherry-picked from 7c860db0747f6276a6e43984d43f8fa5181ea936) Change-Id: I916ad305e4001269460ca6ebd38aaa0be8ac7f52
2013-08-01 22:13:33 +02:00
bionic/__strchr_chk.cpp \
bionic/__strlcat_chk.cpp \
bionic/__strlcpy_chk.cpp \
bionic/__strlen_chk.cpp \
bionic/__strncat_chk.cpp \
bionic/__strncpy_chk.cpp \
bionic/__strrchr_chk.cpp \
bionic/__umask_chk.cpp \
bionic/__vsnprintf_chk.cpp \
bionic/__vsprintf_chk.cpp \
add fortified implementations of write/pwrite{,64} These are just based on the read/pread{,64} implementations with the function calls and error messages adjusted as appropriate. The only difference is that the buffer parameters are const. Change-Id: Ida1597a903807f583f230d74bcedffdb7b24fcf6
2015-07-21 03:37:29 +02:00
bionic/__write_chk.cpp
Optimize __memset_chk, __memcpy_chk. DO NOT MERGE. This change creates assembler versions of __memcpy_chk/__memset_chk that is implemented in the memcpy/memset assembler code. This change avoids an extra call to memcpy/memset, instead allowing a simple fall through to occur from the chk code into the body of the real implementation. Testing: - Ran the libc_test on __memcpy_chk/__memset_chk on all nexus devices. - Wrote a small test executable that has three calls to __memcpy_chk and three calls to __memset_chk. First call dest_len is length + 1. Second call dest_len is length. Third call dest_len is length - 1. Verified that the first two calls pass, and the third fails. Examined the logcat output on all nexus devices to verify that the fortify error message was sent properly. - I benchmarked the new __memcpy_chk and __memset_chk on all systems. For __memcpy_chk and large copies, the savings is relatively small (about 1%). For small copies, the savings is large on cortex-a15/krait devices (between 5% to 30%). For cortex-a9 and small copies, the speed up is present, but relatively small (about 3% to 5%). For __memset_chk and large copies, the savings is also small (about 1%). However, all processors show larger speed-ups on small copies (about 30% to 100%). Bug: 9293744 Merge from internal master. (cherry-picked from 7c860db0747f6276a6e43984d43f8fa5181ea936) Change-Id: I916ad305e4001269460ca6ebd38aaa0be8ac7f52
2013-08-01 22:13:33 +02:00
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
libc_bionic_ndk_src_files := \
Clean up abort. * A dlmalloc usage error shouldn't call abort(3) because we want to cause a SIGSEGV by writing the address dlmalloc didn't like to an address the kernel won't like, so that debuggerd will dump the memory around the address that upset dlmalloc. * Switch to the simpler FreeBSD/NetBSD style of registering stdio cleanup. Hopefully this will let us simplify more of the stdio implementation. * Clear the stdio cleanup handler before we abort because of a dlmalloc corruption error. This fixes the reported bug, where we'd hang inside dlmalloc because the stdio cleanup reentered dlmalloc. Bug: 9301265 Change-Id: Ief31b389455d6876e5a68f0f5429567d37277dbc
2013-06-12 23:05:46 +02:00
bionic/abort.cpp \
Mark sockets on accept(). (cherry picked from commit 58b1f3f6a30a660ad81637c2b50382c3d279243b) Change-Id: I5d09be413cf720fbed905f96313b007997ada76c
2014-05-14 00:40:26 +02:00
bionic/accept.cpp \
Mark sockets on creation (socket()) and accept4(). Remove the separate syscall for accept() and implement it as accept4(..., 0). Change-Id: Ib0b8f5d7c5013b91eae6bbc3847852eb355c7714
2014-05-19 22:39:57 +02:00
bionic/accept4.cpp \
Move away from the __ARCH_WANT_SYSCALL_NO_AT system calls. Modern architectures only get the *at(2) system calls. For example, aarch64 doesn't have open(2), and expects userspace to use openat(2) instead. Change-Id: I87b4ed79790cb8a80844f5544ac1a13fda26c7b5
2013-10-22 22:28:46 +02:00
bionic/access.cpp \
Fix inet_addr/inet_aton/inet_network. Rewrite inet_addr and inet_network in terms of inet_aton, and reimplement that to include all the missing error checks. Bug: http://b/24754503 Change-Id: I5dfa971c87201968985a0894df419f0fbf54768a
2015-10-09 03:04:49 +02:00
bionic/arpa_inet.cpp \
Clean up the implementation of the <dirent.h> functions. Change-Id: I3c647cc9588525afc41fee90ee468d58cd13503a
2012-10-26 05:55:23 +02:00
bionic/assert.cpp \
Ensure we always have symbols for atof, strtof, strtold. We'll need a better implementation of strtold for LP64, but all our long double functions are currently broken for LP64 anyway so this isn't a regression. Change-Id: I2bdebac11245d31521d5fa09a16331c03dc4339c
2014-03-13 00:12:57 +01:00
bionic/atof.cpp \
Implement if_nameindex(3)/if_freenameindex(3). This is just a subset of the recently-implemented getifaddrs(3), though if we want to handle interfaces (such as "rmnet_*") that don't have an address, we need to either expose ifaddrs_storage and keep track of which interfaces we've already seen (which is pretty messy), or refactor the netlink code so we can reuse it and just extract the information we need for if_nameindex(3). This patch goes the latter route. Also clean up if_nametoindex(3) and if_indextoname(3). Change-Id: I5ffc5df0bab62286cdda2e7af06f032c767119a8
2016-01-16 06:02:56 +01:00
bionic/bionic_netlink.cpp \
Added a bionic systrace class and tracing to pthread_mutex.cpp. bionic_systrace.h contains an implementation of tracing that can be used with systrace.py and its associated viewer. pthread_mutex now uses this tracing to track pthread_mutex contention, which can be enabled by using the "bionic" command line option to systrace. Bug: 15116468 Change-Id: I30ed5b377c91ca4c36568a0e647ddf95d4e4a61a
2014-07-22 00:38:06 +02:00
bionic/bionic_systrace.cpp \
Remove dependencies on obsolete __ARCH_WANT_SYSCALL_DEPRECATED system calls. (aarch64 kernels don't have these system calls.) Change-Id: I6f64075aa412f71520f2df71c3d69b647f91c1ca
2013-10-25 00:15:14 +02:00
bionic/bionic_time_conversions.cpp \
Remove bogus extra alignment from sbrk. Bug: https://code.google.com/p/android/issues/detail?id=37349 Change-Id: I970c7b6be7bb7fbe6bbbe2c332f05816aeb0e09f
2013-02-06 01:10:59 +01:00
bionic/brk.cpp \
Adds functionality specified by uchar.h mbrtoc32 and c32rtomb get their implementations from mbrtowc and wcrtomb. The wc functions now simply call the c32 functions. Bug: 14646575 Change-Id: I49d4b95fed0f9d790260c996c4d0f8bfd1686324
2014-06-02 20:33:04 +02:00
bionic/c16rtomb.cpp \
bionic/c32rtomb.cpp \
Move away from the __ARCH_WANT_SYSCALL_NO_AT system calls. Modern architectures only get the *at(2) system calls. For example, aarch64 doesn't have open(2), and expects userspace to use openat(2) instead. Change-Id: I87b4ed79790cb8a80844f5544ac1a13fda26c7b5
2013-10-22 22:28:46 +02:00
bionic/chmod.cpp \
bionic/chown.cpp \
Switch to the upstream OpenBSD getenv/putenv/setenv implementation. This fixes all the bugs found by the new tests. Change-Id: Id5a5f9f39a0620208bafa053f871a044725b4795
2014-04-23 02:41:00 +02:00
bionic/clearenv.cpp \
Reimplement clock(3) and switch to OpenBSD time(3). The new implementation is a better approximation to the processor time used by the process because it's actually based on resource usage rather than just elapsed wall clock time. Change-Id: I9e13b69c1d3048cadf0eb9dec1e3ebc78225596a
2014-03-11 21:37:11 +01:00
bionic/clock.cpp \
support _POSIX_CPUTIME Bug: 18490039 Change-Id: I01fa83b48e1b908de1f227b6e4f98e280bcd50ee
2014-11-26 23:04:26 +01:00
bionic/clock_getcpuclockid.cpp \
Add clock_settime and clock_nanosleep. Add the missing prototypes, fix the existing prototypes to use clockid_t rather than int, fix clock_nanosleep's failure behavior, and add simple tests. Bug: 17644443 Bug: https://code.google.com/p/android/issues/detail?id=77372 Change-Id: I03fba369939403918abcabae9551a7123953d780 Signed-off-by: Haruki Hasegawa <h6a.h4i.0@gmail.com>
2014-10-12 17:50:47 +02:00
bionic/clock_nanosleep.cpp \
Clean up forking and cloning. The kernel now maintains the pthread_internal_t::tid field for us, and __clone was only used in one place so let's inline it so we don't have to leave such a dangerous function lying around. Also rename files to match their content and remove some useless #includes. Change-Id: I24299fb4a940e394de75f864ee36fdabbd9438f9
2013-11-19 22:31:58 +01:00
bionic/clone.cpp \
Simplify close(2) EINTR handling. This doesn't affect code like Chrome that correctly ignores EINTR on close, makes code that tries TEMP_FAILURE_RETRY work (where before it might have closed a different fd and appeared to succeed, or had a bogus EBADF), and makes "goto fail" code work (instead of mistakenly assuming that EINTR means that the close failed). Who loses? Anyone actively trying to detect that they caught a signal while in close(2). I don't think those people exist, and I think they have better alternatives available. Bug: https://code.google.com/p/chromium/issues/detail?id=269623 Bug: http://b/20501816 Change-Id: I11e2f66532fe5d1b0082b2433212e24bdda8219b
2015-04-23 06:40:38 +02:00
bionic/close.cpp \
Move __cmsg_nxthdr to its proper file. Change-Id: I095fb6bc5318f2d256baa97658bdfe679d235aec
2014-07-10 13:47:13 +02:00
bionic/__cmsg_nxthdr.cpp \
Introduce netd_client, a dynamic library that talks to netd. The library exists outside bionic. It is dynamically loaded, to replace selected standard socket syscalls with versions that talk to netd. Change connect() to use the library if available. (cherry picked from commit 3a6b627a14df8111b03e452f2df4b5f4938e0e49) Change-Id: Ib6198e19dbc306521a26fcecfdf6e8424d163fc9
2014-05-12 20:19:16 +02:00
bionic/connect.cpp \
Revert "Revert "Add locale aware APIs."" This reverts commit 063e20c26943ec82ef1d53a544545e79054e93d3. Change-Id: Ib8c9004efefe75a5346b3af50dfe37952d91eb21
2014-07-10 00:51:34 +02:00
bionic/ctype.cpp \
libc: Fix alphasort() signature (and implementation). The declaration for alphasort() in <dirent.h> used the deprecated: int alphasort(const void*, const void*); while both Posix and GLibc use instead: int alphasort(const struct dirent** a, const struct dirent** b); See: http://pubs.opengroup.org/onlinepubs/9699919799/functions/alphasort.html This patch does the following: - Update the declaration to match Posix/GLibc - Get rid of the upstream BSD code which isn't compatible with the new signature. - Implement a new trivial alphasort() with the right signature, and ensure that it uses strcoll() instead of strcmp(). - Remove Bionic-specific #ifdef .. #else .. #endif block in dirent_test.cpp which uses alphasort(). Even through strcoll() currently uses strcmp(), this does the right thing in the case where we decide to update strcoll() to properly implement locale-specific ordered comparison. Change-Id: I4fd45604d8a940aaf2eb0ecd7d73e2f11c9bca96
2012-10-29 15:32:54 +01:00
bionic/dirent.cpp \
Remove dependencies on obsolete __ARCH_WANT_SYSCALL_NO_FLAGS syscalls. (aarch64 kernels only have the newer system calls.) Also expose the new functionality that's exposed by glibc in our header files. Change-Id: I45d2d168a03f88723d1f7fbf634701006a4843c5
2013-10-23 18:48:29 +02:00
bionic/dup2.cpp \
bionic/epoll_create.cpp \
Remove dependencies on obsolete __ARCH_WANT_SYSCALL_DEPRECATED system calls. (aarch64 kernels don't have these system calls.) Change-Id: I6f64075aa412f71520f2df71c3d69b647f91c1ca
2013-10-25 00:15:14 +02:00
bionic/epoll_pwait.cpp \
Clean up sys_signame and sys_siglist a little. We don't need quite so much duplication because we already have a way to get the signal number from its name, and that already copes with the fact that the mips/mips64 numbers are different from everyone else's. Also remove sys_signame from LP64. glibc doesn't have this BSD-ism. Change-Id: I6dc411a3d73589383c85d3b07d9d648311492a10
2014-02-12 04:57:06 +01:00
bionic/epoll_wait.cpp \
Fix __errno for LP64 and clean up __get_tls. If __get_tls has the right type, a lot of confusing casting can disappear. It was probably a mistake that __get_tls was exposed as a function for mips and x86 (but not arm), so let's (a) ensure that the __get_tls function always matches the macro, (b) that we have the function for arm too, and (c) that we don't have the function for any 64-bit architecture. Change-Id: Ie9cb989b66e2006524ad7733eb6e1a65055463be
2013-10-09 03:50:24 +02:00
bionic/__errno.cpp \
Add <error.h>. Bug: https://code.google.com/p/android/issues/detail?id=79170 Change-Id: Id91765fac45124545e2674a5b2c814707c1a448b
2015-01-25 03:36:29 +01:00
bionic/error.cpp \
Prevent name conflict for eventfd.cpp and eventfd.s when building libc.a - eventfd.cpp and eventfd.s will output to the same file when building libc.a out/target/product/*/obj/STATIC_LIBRARIES/libc_intermediates/WHOLE/libc_common_objs/eventfd.o - And then `eventfd` will undefined when statically linked to libc. Also add a unit test. (cherry-pick of 8baa929d5d3bcf63381cf78ba76168c80c303f5e.) Change-Id: Icd0eb0f4ce0511fb9ec00a504d491afd47d744d3
2013-04-03 05:29:40 +02:00
bionic/eventfd_read.cpp \
bionic/eventfd_write.cpp \
Fix "faccessat ignores flags" The kernel system call faccessat() does not have any flags arguments, so passing flags to the kernel is currently ignored. Fix the kernel system call so that no flags argument is passed in. Ensure that we don't support AT_SYMLINK_NOFOLLOW. This non-POSIX (http://pubs.opengroup.org/onlinepubs/9699919799/functions/access.html) flag is a glibc extension, and has non-intuitive, error prone behavior. For example, consider the following code: symlink("foo.is.dangling", "foo"); if (faccessat(AT_FDCWD, "foo", R_OK, AT_SYMLINK_NOFOLLOW) == 0) { int fd = openat(AT_FDCWD, "foo", O_RDONLY | O_NOFOLLOW); } The faccessat() call in glibc will return true, but an attempt to open the dangling symlink will end up failing. GLIBC documents this as returning the access mode of the symlink itself, which will always return true for any symlink on Linux. Some further discussions of this are at: * http://lists.landley.net/pipermail/toybox-landley.net/2014-September/003617.html * http://permalink.gmane.org/gmane.linux.lib.musl.general/6952 AT_SYMLINK_NOFOLLOW seems broken by design. I suspect this is why this function was never added to POSIX. (note that "access" is pretty much broken by design too, since it introduces a race condition between check and action). We shouldn't support this until it's clearly documented by POSIX or we can have it produce intuitive results. Don't support AT_EACCESS for now. Implementing it is complicated, and pretty much useless on Android, since we don't have setuid binaries. See http://git.musl-libc.org/cgit/musl/commit/?id=0a05eace163cee9b08571d2ff9d90f5e82d9c228 for how an implementation might look. Bug: 18867827 Change-Id: I25b86c5020f3152ffa3ac3047f6c4152908d0e04
2015-02-24 22:40:43 +01:00
bionic/faccessat.cpp \
Add fchmodat(AT_SYMLINK_NOFOLLOW) and fchmod O_PATH support Many libc functions have an option to not follow symbolic links. This is useful to avoid security sensitive code from inadvertantly following attacker supplied symlinks and taking inappropriate action on files it shouldn't. For example, open() has O_NOFOLLOW, chown() has lchown(), stat() has lstat(), etc. There is no such equivalent function for chmod(), such as lchmod(). To address this, POSIX introduced fchmodat(AT_SYMLINK_NOFOLLOW), which is intended to provide a way to perform a chmod operation which doesn't follow symlinks. Currently, the Linux kernel doesn't implement AT_SYMLINK_NOFOLLOW. In GLIBC, attempting to use the AT_SYMLINK_NOFOLLOW flag causes fchmodat to return ENOTSUP. Details are in "man fchmodat". Bionic currently differs from GLIBC in that AT_SYMLINK_NOFOLLOW is silently ignored and treated as if the flag wasn't present. This patch provides a userspace implementation of AT_SYMLINK_NOFOLLOW for bionic. Using open(O_PATH | O_NOFOLLOW), we can provide a way to atomically change the permissions on files without worrying about race conditions. As part of this change, we add support for fchmod on O_PATH file descriptors, because it's relatively straight forward and could be useful in the future. The basic idea behind this implementation comes from https://sourceware.org/bugzilla/show_bug.cgi?id=14578 , specifically comment #10. Change-Id: I1eba0cdb2c509d9193ceecf28f13118188a3cfa7
2015-02-01 04:57:46 +01:00
bionic/fchmod.cpp \
bionic/fchmodat.cpp \
Let the compiler worry about implementing ffs(3). It does at least as good a job as our old hand-written assembly anyway. Change-Id: If7c4a1ac508bace0b71ee7b67808caa6eabf11d2
2013-10-25 01:29:40 +02:00
bionic/ffs.cpp \
libc: Add O_PATH support for fgetxattr / fsetxattr Support O_PATH file descriptors when handling fgetxattr and fsetxattr. This avoids requiring file read access to pull extended attributes. This is needed to support O_PATH file descriptors when calling SELinux's fgetfilecon() call. In particular, this allows the querying and setting of SELinux file context by using something like the following code: int dirfd = open("/path/to/dir", O_DIRECTORY); int fd = openat(dirfd, "file", O_PATH | O_NOFOLLOW); char *context; fgetfilecon(fd, &context); This change was motivated by a comment in https://android-review.googlesource.com/#/c/152680/1/toys/posix/ls.c Change-Id: Ic0cdf9f9dd0e35a63b44a4c4a08400020041eddf
2015-05-31 22:43:13 +02:00
bionic/fgetxattr.cpp \
Add O_PATH support for flistxattr() A continuation of commit 2825f10b7f61558c264231a536cf3affc0d84204. Add O_PATH compatibility support for flistxattr(). This allows a process to list out all the extended attributes associated with O_PATH file descriptors. Change-Id: Ie2285ac7ad2e4eac427ddba6c2d182d41b130f75
2015-06-06 20:23:26 +02:00
bionic/flistxattr.cpp \
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
bionic/flockfile.cpp \
Reimplement isinf/isnan/fpclassify. Also move isinf and isnan into libc like everyone else. Also move fpclassify to libc like the BSDs (but unlike glibc). We need this to be able to upgrade our float/double/long double parsing to gdtoa. Also add some missing aliases. We now have all of: isnan, __isnan, isnanf, __isnanf, isnanl, __isnanl, isinf, __isinf, isinff, __isinff, isinfl, __isinfl, __fpclassify, __fpclassifyd, __fpclassifyf, __fpclassifyl. Bug: 13469877 Change-Id: I407ffbac06c765a6c5fffda8106c37d7db04f27d
2014-04-12 02:02:20 +02:00
bionic/fpclassify.cpp \
libc: Add O_PATH support for fgetxattr / fsetxattr Support O_PATH file descriptors when handling fgetxattr and fsetxattr. This avoids requiring file read access to pull extended attributes. This is needed to support O_PATH file descriptors when calling SELinux's fgetfilecon() call. In particular, this allows the querying and setting of SELinux file context by using something like the following code: int dirfd = open("/path/to/dir", O_DIRECTORY); int fd = openat(dirfd, "file", O_PATH | O_NOFOLLOW); char *context; fgetfilecon(fd, &context); This change was motivated by a comment in https://android-review.googlesource.com/#/c/152680/1/toys/posix/ls.c Change-Id: Ic0cdf9f9dd0e35a63b44a4c4a08400020041eddf
2015-05-31 22:43:13 +02:00
bionic/fsetxattr.cpp \
Fix error handling for negative size in ftruncate. Bug: 21309901 Change-Id: I54692ab8105dd09db6af7a2c0894a17bdd118aa0
2015-05-20 03:17:31 +02:00
bionic/ftruncate.cpp \
Add futimens. Bug: 10239370 Change-Id: I518340084103dc339ef8a065d4837d6258a1381d
2013-08-09 02:13:33 +02:00
bionic/futimens.cpp \
Clean up the implementation of the <dirent.h> functions. Change-Id: I3c647cc9588525afc41fee90ee468d58cd13503a
2012-10-26 05:55:23 +02:00
bionic/getcwd.cpp \
gethostname.c: change to report ENAMETOOLONG error when buflen is less change to behaviour the same as glibc for the check about buflen Change-Id: I98265a8fe441df6fed2527686f89b087364ca53d Signed-off-by: Yongqin Liu <yongqin.liu@linaro.org>
2014-10-30 09:34:55 +01:00
bionic/gethostname.cpp \
Take ownership of various simple syscall wrappers. None of these trivial functions is something we're going to get from BSD. Change-Id: Iee1d1281b73db67da5ec303da7a49748121464bf
2014-03-11 01:17:01 +01:00
bionic/getpgrp.cpp \
Cache getpid. In practice, with this implementation we never need to make a system call. We get the main thread's tid (which is the same as our pid) back from the set_tid_address system call we have to make during initialization. A new pthread will have the same pid as its parent, and a fork child's main (and only) thread will have a pid equal to its tid, which we get for free from the kernel before clone returns. The only time we'd actually have to make a getpid system call now is if we take a signal during fork and the signal handler calls getpid. (That, or we call getpid in the dynamic linker while it's still dealing with its own relocations and hasn't even set up the main thread yet.) Bug: 15387103 Change-Id: I6d4718ed0a5c912fc75b5f738c49a023dbed5189
2014-06-20 01:39:01 +02:00
bionic/getpid.cpp \
Avoid a system call in 'gettid'. System calls can be pretty slow. This is mako, which has one of our lowest latencies: iterations ns/op BM_unistd_getpid 10000000 209 BM_unistd_gettid 200000000 8 Bug: 15297299 (kernel panic from too many gettid calls) Bug: 15315766 (excessive gettid overhead in liblogd) Change-Id: I49656c0fc5b5d092390264a59e4f2c0d8a8b1aeb
2014-05-31 04:00:03 +02:00
bionic/gettid.cpp \
Implement the GNU basename(3) in addition to the POSIX one. Code like perf(1) needs this. Bug: 11860789 Change-Id: I907eb448052a7b165e4012d74303330d32328cb2
2014-08-19 23:30:30 +02:00
bionic/__gnu_basename.cpp \
Revert "Revert "Implement getifaddrs(3)/freeifaddrs(3)."" This reverts commit 76814a8250fd5e1502909ef5e10e4d166cc96c0e. This differs from the original in fixing the GCC -Werror build: bionic/libc/bionic/ifaddrs.cpp: In function 'void __handle_netlink_response(ifaddrs**, nlmsghdr*)': bionic/libc/bionic/ifaddrs.cpp:113:62: error: use of old-style cast [-Werror=old-style-cast] ifinfomsg* ifi = reinterpret_cast<ifinfomsg*>(NLMSG_DATA(hdr)); This appears to be a GCC bug; the GCC command-line correctly uses -isystem, and manually adding #pragma GCC system_header doesn't help. So just turn the warning off for GCC for now. We won't need to worry about building with GCC soon anyway. Bug: http://b/26238832 Change-Id: I01615bd335edf11baf487b1c83a9157cd780f4a1
2016-01-04 21:38:05 +01:00
bionic/ifaddrs.cpp \
Remove dependencies on obsolete __ARCH_WANT_SYSCALL_NO_FLAGS syscalls. (aarch64 kernels only have the newer system calls.) Also expose the new functionality that's exposed by glibc in our header files. Change-Id: I45d2d168a03f88723d1f7fbf634701006a4843c5
2013-10-23 18:48:29 +02:00
bionic/inotify_init.cpp \
fix the mremap signature The mremap definition was incorrect (unsigned long instead of int) and it was missing the optional new_address parameter. Change-Id: Ib9d0675aaa098c21617cedc9b2b8cf267be3aec4
2015-11-03 11:14:08 +01:00
bionic/ioctl.cpp \
Move away from the __ARCH_WANT_SYSCALL_NO_AT system calls. Modern architectures only get the *at(2) system calls. For example, aarch64 doesn't have open(2), and expects userspace to use openat(2) instead. Change-Id: I87b4ed79790cb8a80844f5544ac1a13fda26c7b5
2013-10-22 22:28:46 +02:00
bionic/lchown.cpp \
Added mkstemp64 Bug: 13076637 Change-Id: I41bf28ab3e6c7325470781e9323eeec023483df5
2014-02-21 16:11:03 +01:00
bionic/lfs64_support.cpp \
Make SIGRTMIN hide the real-time signals we use internally. __SIGRTMIN will continue to tell the truth. This matches glibc's behavior (as evidenced by the fact that we don't need a special case in the strsignal test now). Change-Id: I1abe1681d516577afa8cd39c837ef12467f68dd2
2014-04-30 18:45:40 +02:00
bionic/__libc_current_sigrtmax.cpp \
bionic/__libc_current_sigrtmin.cpp \
Revert "fix incompatibility with -fstack-protector-strong" Reports of devices not booting with a build range including this patch. Reverting while I investigate. This reverts commit 96dd2b6e85e03ca9ab2600c2031490c22d5626ad. Bug: 23470962 Change-Id: I056f1d8668f52d31412868975f2143249fc7e8f0
2015-08-24 08:58:39 +02:00
bionic/libc_init_common.cpp \
Clean up internal libc logging. We only need one logging API, and I prefer the one that does no allocation and is thus safe to use in any context. Also use O_CLOEXEC when opening the /dev/log files. Move everything logging-related into one header file. Change-Id: Ic1e3ea8e9b910dc29df351bff6c0aa4db26fbb58
2013-03-15 23:30:25 +01:00
bionic/libc_logging.cpp \
Clean up the <libgen.h> implementation a little, bring in tests. Most of these tests were in system/extras, but I've added more to cover other cases explicitly mentioned by POSIX. Change-Id: I5e8d77e4179028d77306935cceadbb505515dcde
2012-10-29 22:27:10 +01:00
bionic/libgen.cpp \
Move away from the __ARCH_WANT_SYSCALL_NO_AT system calls. Modern architectures only get the *at(2) system calls. For example, aarch64 doesn't have open(2), and expects userspace to use openat(2) instead. Change-Id: I87b4ed79790cb8a80844f5544ac1a13fda26c7b5
2013-10-22 22:28:46 +02:00
bionic/link.cpp \
Clean up localeconv(3). The OpenBSD doesn't support C99, and the extent to which we support locales is trivial, so just do it ourselves. Change-Id: If0a06e627ecc593f7b8ea3e9389365782e49b00e
2014-04-08 23:34:12 +02:00
bionic/locale.cpp \
Implement POSIX lockf. This has been requested a few times over the years. This is basically a very late rebase of https://android-review.googlesource.com/45470 which was abandoned years ago. One addition is that this version has _FILE_OFFSET_BITS=64 support. POSIX puts this in <unistd.h>. glibc also has it in <fcntl.h>. Bug: http://b/13077650 Change-Id: I5862b1dc326e326c01ad92438ecc1578d19ba739
2016-01-26 03:06:24 +01:00
bionic/lockf.cpp \
Move away from the __ARCH_WANT_SYSCALL_NO_AT system calls. Modern architectures only get the *at(2) system calls. For example, aarch64 doesn't have open(2), and expects userspace to use openat(2) instead. Change-Id: I87b4ed79790cb8a80844f5544ac1a13fda26c7b5
2013-10-22 22:28:46 +02:00
bionic/lstat.cpp \
Implement malloc_info(3). Expose jemalloc stats through the malloc_info(3) interface. Bug: 16874689 Change-Id: I4358ac283002e60ff161107028d1a3fb1e9afb0a
2014-08-20 18:16:57 +02:00
bionic/malloc_info.cpp \
Adds functionality specified by uchar.h mbrtoc32 and c32rtomb get their implementations from mbrtowc and wcrtomb. The wc functions now simply call the c32 functions. Bug: 14646575 Change-Id: I49d4b95fed0f9d790260c996c4d0f8bfd1686324
2014-06-02 20:33:04 +02:00
bionic/mbrtoc16.cpp \
bionic/mbrtoc32.cpp \
bionic/mbstate.cpp \
Add GNU extensions mempcpy and wmemcpy. Used by elfutils. On the bright side, they stopped using __mempcpy. Bug: 18374026 Change-Id: Id29bbe6ef1c5ed5a171bb6c32182f129d8332abb
2015-02-19 06:29:13 +01:00
bionic/mempcpy.cpp \
Move away from the __ARCH_WANT_SYSCALL_NO_AT system calls. Modern architectures only get the *at(2) system calls. For example, aarch64 doesn't have open(2), and expects userspace to use openat(2) instead. Change-Id: I87b4ed79790cb8a80844f5544ac1a13fda26c7b5
2013-10-22 22:28:46 +02:00
bionic/mkdir.cpp \
Make sure we have a mkfifo symbol. Bug: https://code.google.com/p/android/issues/detail?id=58888 Change-Id: Ic0a883a5f30beb82cb7be3c4e81b6d693d5fbb4d
2013-10-22 19:54:11 +02:00
bionic/mkfifo.cpp \
Move away from the __ARCH_WANT_SYSCALL_NO_AT system calls. Modern architectures only get the *at(2) system calls. For example, aarch64 doesn't have open(2), and expects userspace to use openat(2) instead. Change-Id: I87b4ed79790cb8a80844f5544ac1a13fda26c7b5
2013-10-22 22:28:46 +02:00
bionic/mknod.cpp \
Add getmntent_r to our collection of <mntent.h> compatibility stubs. This helps build 'external/flo' out of the box. Bug: 14841211 Change-Id: I30dde77239cceaf1f5743163744eb3604d27a266
2014-05-13 19:44:07 +02:00
bionic/mntent.cpp \
fix the mremap signature The mremap definition was incorrect (unsigned long instead of int) and it was missing the optional new_address parameter. Change-Id: Ib9d0675aaa098c21617cedc9b2b8cf267be3aec4
2015-11-03 11:14:08 +01:00
bionic/mremap.cpp \
Fix build (take two). Make sure __netdClientDispatch is defined in the same set of libraries that refer to it (e.g.: with connect.cpp). Change-Id: I86d7bf2df5bde09f75a35b204eac0e1361747e22
2014-05-14 02:24:03 +02:00
bionic/NetdClientDispatch.cpp \
Implement if_nameindex(3)/if_freenameindex(3). This is just a subset of the recently-implemented getifaddrs(3), though if we want to handle interfaces (such as "rmnet_*") that don't have an address, we need to either expose ifaddrs_storage and keep track of which interfaces we've already seen (which is pretty messy), or refactor the netlink code so we can reuse it and just extract the information we need for if_nameindex(3). This patch goes the latter route. Also clean up if_nametoindex(3) and if_indextoname(3). Change-Id: I5ffc5df0bab62286cdda2e7af06f032c767119a8
2016-01-16 06:02:56 +01:00
bionic/net_if.cpp \
Move away from the __ARCH_WANT_SYSCALL_NO_AT system calls. Modern architectures only get the *at(2) system calls. For example, aarch64 doesn't have open(2), and expects userspace to use openat(2) instead. Change-Id: I87b4ed79790cb8a80844f5544ac1a13fda26c7b5
2013-10-22 22:28:46 +02:00
bionic/open.cpp \
Clean up fpathconf(3)/pathconf(3). fpathconf(3) and pathconf(3) can share code. There's no such header file as <pathconf.h>. glibc/POSIX and BSD disagree about where the _POSIX_* definitions should go. Change-Id: I4a67f1595c9f5fbb26700a131178eedebd6bf712
2014-09-22 23:49:07 +02:00
bionic/pathconf.cpp \
Remove dependencies on obsolete __ARCH_WANT_SYSCALL_DEPRECATED system calls. (aarch64 kernels don't have these system calls.) Change-Id: I6f64075aa412f71520f2df71c3d69b647f91c1ca
2013-10-25 00:15:14 +02:00
bionic/pause.cpp \
Remove dependencies on obsolete __ARCH_WANT_SYSCALL_NO_FLAGS syscalls. (aarch64 kernels only have the newer system calls.) Also expose the new functionality that's exposed by glibc in our header files. Change-Id: I45d2d168a03f88723d1f7fbf634701006a4843c5
2013-10-23 18:48:29 +02:00
bionic/pipe.cpp \
Remove dependencies on obsolete __ARCH_WANT_SYSCALL_DEPRECATED system calls. (aarch64 kernels don't have these system calls.) Change-Id: I6f64075aa412f71520f2df71c3d69b647f91c1ca
2013-10-25 00:15:14 +02:00
bionic/poll.cpp \
Add posix_fadvise(3). (cherry-pick of 00008263782e484020420c606f7d145fe7d0a4d8.) Bug: 12449798 Change-Id: I07cbf3f670a0d1304b68148325a774f266b5c433
2014-09-11 02:39:00 +02:00
bionic/posix_fadvise.cpp \
Add fallocate/fallocate64/posix_fallocate/posix_fallocate64. Bug: 5287571 Bug: 12612860 Change-Id: I4501b9c6cdf9a830336ce0b3afc4ea716b6a0f6f
2014-02-04 01:20:46 +01:00
bionic/posix_fallocate.cpp \
implement posix_madvise Bug: 18472477 Change-Id: I8183de6c281acf69ed5f7f88351b056b9827b162
2014-11-26 05:17:27 +01:00
bionic/posix_madvise.cpp \
Rewrite the POSIX timer functions. This is a much simpler implementation that lets the kernel do as much as possible. Co-authored-by: Jörgen Strand <jorgen.strand@sonymobile.com> Co-authored-by: Snild Dolkow <snild.dolkow@sonymobile.com> Change-Id: Iad19f155de977667aea09410266d54e63e8a26bf
2014-03-05 00:58:02 +01:00
bionic/posix_timers.cpp \
<pthread.h> fixes and pthread cleanup. <pthread.h> was missing nonnull attributes, noreturn on pthread_exit, and had incorrect cv qualifiers for several standard functions. I've also marked the non-standard stuff (where I count glibc rather than POSIX as "standard") so we can revisit this cruft for LP64 and try to ensure we're compatible with glibc. I've also broken out the pthread_cond* functions into a new file. I've made the remaining pthread files (plus ptrace) part of the bionic code and fixed all the warnings. I've added a few more smoke tests for chunks of untested pthread functionality. We no longer need the libc_static_common_src_files hack for any of the pthread implementation because we long since stripped out the rest of the armv5 support, and this hack was just to ensure that __get_tls in libc.a went via the kernel if necessary. This patch also finishes the job of breaking up the pthread.c monolith, and adds a handful of new tests. Change-Id: Idc0ae7f5d8aa65989598acd4c01a874fe21582c7
2013-10-30 22:40:09 +01:00
bionic/ptrace.cpp \
Fix linkage of grantpt(3). Also clean up the implementation of all the pty functions, add tests, and fix the stub implementations of ttyname(3) and ttyname_r(3). Bug: https://code.google.com/p/android/issues/detail?id=58888 Change-Id: I0fb36438cd1abf8d4e87c29415f03db9ba13c3c2
2014-07-26 02:24:00 +02:00
bionic/pty.cpp \
Use pthread_kill() in raise() raise() should use pthread_kill() in a pthreads environment. For bionic this means it should always be used. Change-Id: Ic679272b664d2b8a7068b628fb83a9f7395c441f
2012-12-08 03:41:10 +01:00
bionic/raise.cpp \
Implement rand/srand in terms of random/srandom. Code developed for glibc or older versions of bionic might expect more randomness than the BSD implementation provides. Bug: 15829381 Change-Id: Ia5a908a816e0a5f0639f514107a6384a51ec157e
2014-07-14 20:57:03 +02:00
bionic/rand.cpp \
Move away from the __ARCH_WANT_SYSCALL_NO_AT system calls. Modern architectures only get the *at(2) system calls. For example, aarch64 doesn't have open(2), and expects userspace to use openat(2) instead. Change-Id: I87b4ed79790cb8a80844f5544ac1a13fda26c7b5
2013-10-22 22:28:46 +02:00
bionic/readlink.cpp \
Clean up various warnings in bionic. Change-Id: Ic57541d0a567fd4ae79f0ad59b2ffde1130eb7d2
2014-02-17 21:33:16 +01:00
bionic/reboot.cpp \
Take ownership of various simple syscall wrappers. None of these trivial functions is something we're going to get from BSD. Change-Id: Iee1d1281b73db67da5ec303da7a49748121464bf
2014-03-11 01:17:01 +01:00
bionic/recv.cpp \
Move away from the __ARCH_WANT_SYSCALL_NO_AT system calls. Modern architectures only get the *at(2) system calls. For example, aarch64 doesn't have open(2), and expects userspace to use openat(2) instead. Change-Id: I87b4ed79790cb8a80844f5544ac1a13fda26c7b5
2013-10-22 22:28:46 +02:00
bionic/rename.cpp \
bionic/rmdir.cpp \
Reimplement scandir(3). The old scandir implementation didn't take into account the varying size of directory entries, and didn't correctly clean up on its error exits. Bug: 7339844 Change-Id: Ib40e3564709752241a3119a496cbb2192e3f9abe
2013-02-25 22:14:31 +01:00
bionic/scandir.cpp \
Add missing include and function prototype for sched_getaffinity. Change-Id: I649633c8d074def5d05bd0c8b92fb680d9d11d51
2013-04-09 18:57:42 +02:00
bionic/sched_getaffinity.cpp \
Clean up <sched.h>. This patch switches to using the uapi constants. It also adds the missing setns system call, fixes sched_getcpu's error behavior, and fixes the gensyscalls script now ARM is uapi-only too. Change-Id: I8e16b1693d6d32cd9b8499e46b5d8b0a50bc4f1d
2014-01-02 21:05:50 +01:00
bionic/sched_getcpu.cpp \
Add semaphore tests, fix sem_destroy. Bug: https://code.google.com/p/android/issues/detail?id=76088 Change-Id: I4a0561b23e90312384d40a1c804ca64ee98f4066
2014-09-19 01:11:59 +02:00
bionic/semaphore.cpp \
Take ownership of various simple syscall wrappers. None of these trivial functions is something we're going to get from BSD. Change-Id: Iee1d1281b73db67da5ec303da7a49748121464bf
2014-03-11 01:17:01 +01:00
bionic/send.cpp \
Clean up the cpuacct cruft. Change-Id: I6ed63af8dfc2368e211420389fa8af4d5dc0908f
2013-10-10 02:35:36 +02:00
bionic/setegid.cpp \
Reduce the exposure of the __set_errno implementation detail. Change-Id: I395e1b46a9491e34fc53e71853e932ea90b3d1cc
2012-11-30 20:58:57 +01:00
bionic/__set_errno.cpp \
Clean up the cpuacct cruft. Change-Id: I6ed63af8dfc2368e211420389fa8af4d5dc0908f
2013-10-10 02:35:36 +02:00
bionic/seteuid.cpp \
Ensure we always have symbols for atof, strtof, strtold. We'll need a better implementation of strtold for LP64, but all our long double functions are currently broken for LP64 anyway so this isn't a regression. Change-Id: I2bdebac11245d31521d5fa09a16331c03dc4339c
2014-03-13 00:12:57 +01:00
bionic/setpgrp.cpp \
Fix sigaction(3) for 64-bit. Also clean up <signal.h> and revert the hacks that were necessary for 64-bit in linker/debugger.cpp until now. Change-Id: I3b0554ca8a49ee1c97cda086ce2c1954ebc11892
2013-10-17 07:27:54 +02:00
bionic/sigaction.cpp \
bionic/sigaddset.cpp \
bionic/sigdelset.cpp \
bionic/sigemptyset.cpp \
bionic/sigfillset.cpp \
bionic/sigismember.cpp \
bionic/signal.cpp \
Clean up sys_signame and sys_siglist a little. We don't need quite so much duplication because we already have a way to get the signal number from its name, and that already copes with the fact that the mips/mips64 numbers are different from everyone else's. Also remove sys_signame from LP64. glibc doesn't have this BSD-ism. Change-Id: I6dc411a3d73589383c85d3b07d9d648311492a10
2014-02-12 04:57:06 +01:00
bionic/signalfd.cpp \
Switch sigpending over to rt_sigpending. Change-Id: I7b28984796b5fb343cfbcc47e0afc3a84293d417
2013-10-16 21:53:58 +02:00
bionic/sigpending.cpp \
Clean up the sigprocmask/pthread_sigmask implementation. Let's have both use rt_sigprocmask, like in glibc. The 64-bit ABIs can share the same code as the 32-bit ABIs. Also, let's test the return side of these calls, not just the setting. Bug: 11069919 Change-Id: I11da99f85b5b481870943c520d05ec929b15eddb
2013-10-15 20:23:57 +02:00
bionic/sigprocmask.cpp \
support _POSIX_REALTIME_SIGNALS Bug: 18489947 Change-Id: I2e834d68bc10ca5fc7ebde047b517a3074179475
2014-12-02 02:41:04 +01:00
bionic/sigqueue.cpp \
Fix sigsuspend to use rt_sigsuspend on all platforms. Change-Id: I981c1a66d35480d4457a0a08a1b042dac94daa5b
2013-10-16 03:01:56 +02:00
bionic/sigsuspend.cpp \
support _POSIX_REALTIME_SIGNALS Bug: 18489947 Change-Id: I2e834d68bc10ca5fc7ebde047b517a3074179475
2014-12-02 02:41:04 +01:00
bionic/sigtimedwait.cpp \
Only have one copy of the kernel_sigset_t hack, and add more tests. Change-Id: I377522fcba6fb4b5fd2754ab15b091014bd7c16f
2013-01-10 23:42:14 +01:00
bionic/sigwait.cpp \
support _POSIX_REALTIME_SIGNALS Bug: 18489947 Change-Id: I2e834d68bc10ca5fc7ebde047b517a3074179475
2014-12-02 02:41:04 +01:00
bionic/sigwaitinfo.cpp \
Mark sockets on creation (socket()) and accept4(). Remove the separate syscall for accept() and implement it as accept4(..., 0). Change-Id: Ib0b8f5d7c5013b91eae6bbc3847852eb355c7714
2014-05-19 22:39:57 +02:00
bionic/socket.cpp \
Move away from the __ARCH_WANT_SYSCALL_NO_AT system calls. Modern architectures only get the *at(2) system calls. For example, aarch64 doesn't have open(2), and expects userspace to use openat(2) instead. Change-Id: I87b4ed79790cb8a80844f5544ac1a13fda26c7b5
2013-10-22 22:28:46 +02:00
bionic/stat.cpp \
Add <sys/statvfs.h>. Bug: 2512019 Change-Id: I6e7fd3fa281977cc4bc270481a95416b5b2dc351
2013-07-09 22:25:03 +02:00
bionic/statvfs.cpp \
Add strchrnul. Bug: http://b/18374026 Change-Id: Iea923309c090a51a2d41c5a83320ab3789f40f1c
2015-08-26 18:59:29 +02:00
bionic/strchrnul.cpp \
Clean up the implementation of the <dirent.h> functions. Change-Id: I3c647cc9588525afc41fee90ee468d58cd13503a
2012-10-26 05:55:23 +02:00
bionic/strerror.cpp \
bionic/strerror_r.cpp \
bionic/strsignal.cpp \
Ensure we always have symbols for atof, strtof, strtold. We'll need a better implementation of strtold for LP64, but all our long double functions are currently broken for LP64 anyway so this isn't a regression. Change-Id: I2bdebac11245d31521d5fa09a16331c03dc4339c
2014-03-13 00:12:57 +01:00
bionic/strtold.cpp \
Clean up the implementation of the <dirent.h> functions. Change-Id: I3c647cc9588525afc41fee90ee468d58cd13503a
2012-10-26 05:55:23 +02:00
bionic/stubs.cpp \
Move away from the __ARCH_WANT_SYSCALL_NO_AT system calls. Modern architectures only get the *at(2) system calls. For example, aarch64 doesn't have open(2), and expects userspace to use openat(2) instead. Change-Id: I87b4ed79790cb8a80844f5544ac1a13fda26c7b5
2013-10-22 22:28:46 +02:00
bionic/symlink.cpp \
Add functions in sysinfo.h. Bug: 17703902 Change-Id: I699d12d76a8fc483c919be730c2ac63641b71a1d
2014-11-15 00:51:58 +01:00
bionic/sysinfo.cpp \
Rewrite syslog(3) to use Android logging. Since we don't have syslogd on Android and you can't run one on a non-rooted device, it's more useful if syslog output just goes to the regular Android logging system. Bug: 14292866 Change-Id: Icee7f088b97f88ccbdaf471b98cbac7f19f9210a
2014-07-22 01:35:24 +02:00
bionic/syslog.cpp \
Clean up sys_signame and sys_siglist a little. We don't need quite so much duplication because we already have a way to get the signal number from its name, and that already copes with the fact that the mips/mips64 numbers are different from everyone else's. Also remove sys_signame from LP64. glibc doesn't have this BSD-ism. Change-Id: I6dc411a3d73589383c85d3b07d9d648311492a10
2014-02-12 04:57:06 +01:00
bionic/sys_siglist.c \
bionic/sys_signame.c \
Ensure we always have symbols for atof, strtof, strtold. We'll need a better implementation of strtold for LP64, but all our long double functions are currently broken for LP64 anyway so this isn't a regression. Change-Id: I2bdebac11245d31521d5fa09a16331c03dc4339c
2014-03-13 00:12:57 +01:00
bionic/system_properties.cpp \
Clean up the implementation of the <dirent.h> functions. Change-Id: I3c647cc9588525afc41fee90ee468d58cd13503a
2012-10-26 05:55:23 +02:00
bionic/tdestroy.cpp \
Clean up our termios implementation. It's safe to fix our constant definitions because we know we never had symbols before, so can't be passing the bad old constants to the new functions, or the correct new constants to the old inlines. Change-Id: I858fc680df39bdd3ba471e867833bdfa71f6224e
2014-03-12 00:06:23 +01:00
bionic/termios.cpp \
Sync to current upstream arc4random. This is actually revision 1.33, which is no longer the latest, but it's as close to head as we can currently reasonably get. I've also switched to the OpenBSD getentropy_linux.c implementation of getentropy, lightly modified to try to report an error on failure. Bug: 14499627 Change-Id: Ia7c561184b1f366c9bf66f248aa60f0d53535fcb
2014-06-21 07:49:20 +02:00
bionic/thread_private.cpp \
Clean up the implementation of the <dirent.h> functions. Change-Id: I3c647cc9588525afc41fee90ee468d58cd13503a
2012-10-26 05:55:23 +02:00
bionic/tmpfile.cpp \
Take ownership of various simple syscall wrappers. None of these trivial functions is something we're going to get from BSD. Change-Id: Iee1d1281b73db67da5ec303da7a49748121464bf
2014-03-11 01:17:01 +01:00
bionic/umount.cpp \
Move away from the __ARCH_WANT_SYSCALL_NO_AT system calls. Modern architectures only get the *at(2) system calls. For example, aarch64 doesn't have open(2), and expects userspace to use openat(2) instead. Change-Id: I87b4ed79790cb8a80844f5544ac1a13fda26c7b5
2013-10-22 22:28:46 +02:00
bionic/unlink.cpp \
bionic/utimes.cpp \
Expose wait4 as wait4 rather than __wait4. This helps strace(1) compile with one fewer hack. Change-Id: I5296d0cfec5546709cda990abd705ad33d7c4626
2013-03-21 23:43:53 +01:00
bionic/wait.cpp \
Clean up the implementation of the <dirent.h> functions. Change-Id: I3c647cc9588525afc41fee90ee468d58cd13503a
2012-10-26 05:55:23 +02:00
bionic/wchar.cpp \
Switch to a working UTF-8 mb/wc implementation. Although glibc gets by with an 8-byte mbstate_t, OpenBSD uses 12 bytes (of the 128 bytes it reserves!). We can actually implement UTF-8 encoding/decoding with a 0-byte mbstate_t which means we can make things work on LP32 too, as long as we accept the limitation that the caller needs to present us with a complete sequence before we'll process it. Our behavior is fine when going from characters to bytes; we just update the source wchar_t** to say how far through the input we got. I'll come back and use the 4 bytes we do have to cope with byte sequences split across multiple input buffers. The fact that we don't support UTF-8 sequences longer than 4 bytes plus the fact that the first byte of a UTF-8 sequence encodes the length means we shouldn't need the other fields OpenBSD used (at the cost of some recomputation in cases where a sequence is split across buffers). This patch also makes the minimal changes necessary to setlocale(3) to make us behave like glibc when an app requests UTF-8. (The difference being that our "C" locale is the same as our "C.UTF-8" locale.) Change-Id: Ied327a8c4643744b3611bf6bb005a9b389ba4c2f
2014-05-01 07:03:12 +02:00
bionic/wctype.cpp \
Add GNU extensions mempcpy and wmemcpy. Used by elfutils. On the bright side, they stopped using __mempcpy. Bug: 18374026 Change-Id: Id29bbe6ef1c5ed5a171bb6c32182f129d8332abb
2015-02-19 06:29:13 +01:00
bionic/wmempcpy.cpp \
Move setlocale(3) and the wchar stubs over to .cpp. Also separate out the C++ files so we can use -Werror on them. I'd rather wait for LOCAL_CPPFLAGS to be in AOSP, but this also lets us see which files still need to be sorted into one bucket or the other. Change-Id: I6acc1f7c043935c70a3b089f705d218b9aaaba0a
2012-10-23 02:05:27 +02:00
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
libc_bionic_src_files :=
Enable using clang to build __cxa_thread_atexit_impl. Remove previous workaround as we no longer use __thread in __cxa_thread_atexit_impl.cpp. Change-Id: Ic1062995db488859b341acdda0b5f6635e10d7e8
2015-11-30 23:07:58 +01:00
# The following implementations depend on pthread data, so we can't include
# them in libc_ndk.a.
libc_bionic_src_files += \
bionic/__cxa_thread_atexit_impl.cpp \
bionic/fork.cpp \
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
# The data that backs getauxval is initialized in the libc init functions which
# are invoked by the linker. If this file is included in libc_ndk.a, only one of
# the copies of the global data will be initialized, resulting in nullptr
# dereferences.
libc_bionic_src_files += bionic/getauxval.cpp
Implement setjmp cookies on ARM. Reuse the top bits of _JB_SIGFLAG field previously used to store a boolean to store a cookie that's validated by [sig]longjmp to make it harder to use as a ROP gadget. Additionally, encrypt saved registers with the cookie so that an attacker can't modify a register's value to a specific value without knowing the cookie. Bug: http://b/23942752 Change-Id: Id0eb8d06916e89d5d776bfcaa9458f8826717ba3
2015-09-11 00:40:24 +02:00
# These four require getauxval, which isn't available on older platforms.
Update NDK compatlib for 4.1.2. The replacement new failures present in newer versions are present here as well, with the following new issues: XPASS std/localization/locale.categories/category.numeric/locale.nm.put/facet.num.put.members/put_long_double.pass.cpp This is from the -NaN formatting fix in bionic. We previously had this wrong, and the upstream test is also wrong. There's currently an XFAIL for Android in this test because I haven't fixed the upstream test yet. After that is done, I'll need to teach the test runner how to XFAIL older Android versions... FAIL std/localization/locale.categories/category.ctype/facet.ctype.special/facet.ctype.char.dtor/dtor.pass.cpp dtor.pass.cpp:39: int main(): assertion "globalMemCounter.checkDeleteArrayCalledEq(1)" failed Haven't investigated this one yet. http://b/19412688 Note that this also needs the libgcc link ordering to be fixed in the build system, as we'll otherwise depend on libgcc symbols from libc that may or may not have been there. The build fix can't be submitted because the proper link order causes the libgcc unwinder to be used instead of the EHABI one: http://b/18471342 Bug: 18471532 Change-Id: Icf560485a9b8f5ebbe01e4458703e62ec94df5e1
2015-02-13 01:35:09 +01:00
libc_bionic_src_files += bionic/getentropy_linux.c
libc_bionic_src_files += bionic/sysconf.cpp
libc_bionic_src_files += bionic/vdso.cpp
Implement setjmp cookies on ARM. Reuse the top bits of _JB_SIGFLAG field previously used to store a boolean to store a cookie that's validated by [sig]longjmp to make it harder to use as a ROP gadget. Additionally, encrypt saved registers with the cookie so that an attacker can't modify a register's value to a specific value without knowing the cookie. Bug: http://b/23942752 Change-Id: Id0eb8d06916e89d5d776bfcaa9458f8826717ba3
2015-09-11 00:40:24 +02:00
libc_bionic_src_files += bionic/setjmp_cookie.cpp
Update NDK compatlib for 4.1.2. The replacement new failures present in newer versions are present here as well, with the following new issues: XPASS std/localization/locale.categories/category.numeric/locale.nm.put/facet.num.put.members/put_long_double.pass.cpp This is from the -NaN formatting fix in bionic. We previously had this wrong, and the upstream test is also wrong. There's currently an XFAIL for Android in this test because I haven't fixed the upstream test yet. After that is done, I'll need to teach the test runner how to XFAIL older Android versions... FAIL std/localization/locale.categories/category.ctype/facet.ctype.special/facet.ctype.char.dtor/dtor.pass.cpp dtor.pass.cpp:39: int main(): assertion "globalMemCounter.checkDeleteArrayCalledEq(1)" failed Haven't investigated this one yet. http://b/19412688 Note that this also needs the libgcc link ordering to be fixed in the build system, as we'll otherwise depend on libgcc symbols from libc that may or may not have been there. The build fix can't be submitted because the proper link order causes the libgcc unwinder to be used instead of the EHABI one: http://b/18471342 Bug: 18471532 Change-Id: Icf560485a9b8f5ebbe01e4458703e62ec94df5e1
2015-02-13 01:35:09 +01:00
Switch to LOCAL_SRC_FILES_EXCLUDE This moves the generic arm/arm64/x86 settings into the main makefiles and makes the rest of them derivatives. This better aligns with how soong handles arch/cpu variants. Also updates the Android.bp to make it consistent with the make versions. Change-Id: I5a0275d992bc657459eb6fe1697ad2336731d122
2015-10-15 23:49:45 +02:00
libc_bionic_src_files += \
bionic/__memcpy_chk.cpp \
bionic/__memset_chk.cpp \
bionic/__strcat_chk.cpp \
bionic/__strcpy_chk.cpp \
bionic/strchr.cpp \
bionic/strnlen.c \
bionic/strrchr.cpp \
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
libc_cxa_src_files := \
bionic/__cxa_guard.cpp \
bionic/__cxa_pure_virtual.cpp \
bionic/new.cpp \
Move realpath.c to upstream-freebsd. This is actually a slightly newer upstream version than the one I originally pulled. Hopefully now it's in upstream-freebsd it will be easier to track upstream, though I still need to sit down and write the necessary scripts at some point. Bug: 5110679 Change-Id: I87e563f0f95aa8e68b45578e2a8f448bbf827a33
2013-03-02 01:59:46 +01:00
libc_upstream_freebsd_src_files := \
Switch to upstream FreeBSD ldexp.c. (Yes, this should be in libm, but it isn't. For once, this mistake predates bionic!) Change-Id: I1340407acbcdc0151dee969a38ba490bdd5e6767
2014-03-11 00:23:09 +01:00
upstream-freebsd/lib/libc/gen/ldexp.c \
Switch to upstream sleep(3) and usleep(3). Also fix the signature of usleep, and the definition of useconds_t which should be unsigned, as the 'u' in its name implies. This patch also cleans up the existing FreeBSD hacks by moving the libm stuff from <sys/cdefs.h> to a libm-private header, and adding comments about the hacks we use to build FreeBSD source. Change-Id: Ibe5067a380502df94a0a3a7901969b35411085b6
2013-11-21 01:09:06 +01:00
upstream-freebsd/lib/libc/gen/sleep.c \
upstream-freebsd/lib/libc/gen/usleep.c \
Switch to current upstream getopt_long. Change-Id: I4c646dcb8be9e88dd54d069a03bbc5fbfd92de03
2013-06-25 23:48:10 +02:00
upstream-freebsd/lib/libc/stdlib/getopt_long.c \
Switch to current FreeBSD qsort. Change-Id: Ic46cd0b663dc5fa78c99dd38db0bfe849a25e789
2013-04-12 03:08:34 +02:00
upstream-freebsd/lib/libc/stdlib/qsort.c \
Adds quick_exit(3) and at_quick_exit(3) from freebsd Change-Id: I4fe88abd8f7b8aa45e58aeb2529d59a8d555d338
2014-04-30 02:49:06 +02:00
upstream-freebsd/lib/libc/stdlib/quick_exit.c \
Move realpath.c to upstream-freebsd. This is actually a slightly newer upstream version than the one I originally pulled. Hopefully now it's in upstream-freebsd it will be easier to track upstream, though I still need to sit down and write the necessary scripts at some point. Bug: 5110679 Change-Id: I87e563f0f95aa8e68b45578e2a8f448bbf827a33
2013-03-02 01:59:46 +01:00
upstream-freebsd/lib/libc/stdlib/realpath.c \
Switch to upstream-freebsd for the unmolested wchar code. Change-Id: I87b4d76ff8da04109ae53638eec4f11629798960
2013-03-02 03:35:56 +01:00
upstream-freebsd/lib/libc/string/wcpcpy.c \
upstream-freebsd/lib/libc/string/wcpncpy.c \
upstream-freebsd/lib/libc/string/wcscasecmp.c \
Switch to LOCAL_SRC_FILES_EXCLUDE This moves the generic arm/arm64/x86 settings into the main makefiles and makes the rest of them derivatives. This better aligns with how soong handles arch/cpu variants. Also updates the Android.bp to make it consistent with the make versions. Change-Id: I5a0275d992bc657459eb6fe1697ad2336731d122
2015-10-15 23:49:45 +02:00
upstream-freebsd/lib/libc/string/wcscat.c \
upstream-freebsd/lib/libc/string/wcschr.c \
upstream-freebsd/lib/libc/string/wcscmp.c \
upstream-freebsd/lib/libc/string/wcscpy.c \
Switch to upstream-freebsd for the unmolested wchar code. Change-Id: I87b4d76ff8da04109ae53638eec4f11629798960
2013-03-02 03:35:56 +01:00
upstream-freebsd/lib/libc/string/wcscspn.c \
upstream-freebsd/lib/libc/string/wcsdup.c \
upstream-freebsd/lib/libc/string/wcslcat.c \
Switch to LOCAL_SRC_FILES_EXCLUDE This moves the generic arm/arm64/x86 settings into the main makefiles and makes the rest of them derivatives. This better aligns with how soong handles arch/cpu variants. Also updates the Android.bp to make it consistent with the make versions. Change-Id: I5a0275d992bc657459eb6fe1697ad2336731d122
2015-10-15 23:49:45 +02:00
upstream-freebsd/lib/libc/string/wcslen.c \
Switch to upstream-freebsd for the unmolested wchar code. Change-Id: I87b4d76ff8da04109ae53638eec4f11629798960
2013-03-02 03:35:56 +01:00
upstream-freebsd/lib/libc/string/wcsncasecmp.c \
upstream-freebsd/lib/libc/string/wcsncat.c \
upstream-freebsd/lib/libc/string/wcsncmp.c \
upstream-freebsd/lib/libc/string/wcsncpy.c \
upstream-freebsd/lib/libc/string/wcsnlen.c \
upstream-freebsd/lib/libc/string/wcspbrk.c \
Switch to LOCAL_SRC_FILES_EXCLUDE This moves the generic arm/arm64/x86 settings into the main makefiles and makes the rest of them derivatives. This better aligns with how soong handles arch/cpu variants. Also updates the Android.bp to make it consistent with the make versions. Change-Id: I5a0275d992bc657459eb6fe1697ad2336731d122
2015-10-15 23:49:45 +02:00
upstream-freebsd/lib/libc/string/wcsrchr.c \
Switch to upstream-freebsd for the unmolested wchar code. Change-Id: I87b4d76ff8da04109ae53638eec4f11629798960
2013-03-02 03:35:56 +01:00
upstream-freebsd/lib/libc/string/wcsspn.c \
upstream-freebsd/lib/libc/string/wcstok.c \
upstream-freebsd/lib/libc/string/wmemchr.c \
Switch to LOCAL_SRC_FILES_EXCLUDE This moves the generic arm/arm64/x86 settings into the main makefiles and makes the rest of them derivatives. This better aligns with how soong handles arch/cpu variants. Also updates the Android.bp to make it consistent with the make versions. Change-Id: I5a0275d992bc657459eb6fe1697ad2336731d122
2015-10-15 23:49:45 +02:00
upstream-freebsd/lib/libc/string/wmemcmp.c \
upstream-freebsd/lib/libc/string/wmemmove.c \
Switch to upstream-freebsd for the unmolested wchar code. Change-Id: I87b4d76ff8da04109ae53638eec4f11629798960
2013-03-02 03:35:56 +01:00
upstream-freebsd/lib/libc/string/wmemset.c \
Move realpath.c to upstream-freebsd. This is actually a slightly newer upstream version than the one I originally pulled. Hopefully now it's in upstream-freebsd it will be easier to track upstream, though I still need to sit down and write the necessary scripts at some point. Bug: 5110679 Change-Id: I87e563f0f95aa8e68b45578e2a8f448bbf827a33
2013-03-02 01:59:46 +01:00
Revert "Revert "Switch to NetBSD's strxfrm(3)."" This reverts commit 8793e7c7d21a0434d1b5e63364c88b2b125a3d29, and fixes the build by building upstream NetBSD source as a separate library that's then swallowed whole into libc_common. Change-Id: I6c9317d8c48b5ccaf85a7b185bc07fb31176ff97
2012-08-10 20:39:58 +02:00
libc_upstream_netbsd_src_files := \
Get the full set of PRNG functions in <stdlib.h> Bug: https://code.google.com/p/android/issues/detail?id=58888 Change-Id: I435250bdae302e8bd7e29977d0fde7b9afbfca5e
2014-06-12 20:48:04 +02:00
upstream-netbsd/common/lib/libc/stdlib/random.c \
Move the upstream-netbsd libc files into the correct directory. I screwed up when I originally imported these files; they're in lib/libc/ in the upstream tree; there is no top-level libc/ (though there is a top-level common/, so those files stay where they are). Change-Id: I7c5e2224a4441ab0e33616a855a8c6aacfeac46f
2014-02-20 23:15:09 +01:00
upstream-netbsd/lib/libc/gen/ftw.c \
upstream-netbsd/lib/libc/gen/nftw.c \
upstream-netbsd/lib/libc/gen/nice.c \
upstream-netbsd/lib/libc/gen/popen.c \
upstream-netbsd/lib/libc/gen/psignal.c \
upstream-netbsd/lib/libc/gen/utime.c \
Switch to NetBSD utmp.c. Change-Id: Ibe94888aa48b5b28fea97fd5719a1ed7a23ddeb3
2014-03-11 00:32:35 +01:00
upstream-netbsd/lib/libc/gen/utmp.c \
Switch to upstream NetBSD nsap_addr.c. These symbols should be public (and Firefox uses them), and we'd also probably rather have the upstream thread-safe implementation. Bug: https://bugzilla.mozilla.org/show_bug.cgi?id=1030899 Change-Id: I2a5888fbb3198546848398f576fd2195ff3fe00c
2014-06-30 21:03:43 +02:00
upstream-netbsd/lib/libc/inet/nsap_addr.c \
Move the upstream-netbsd libc files into the correct directory. I screwed up when I originally imported these files; they're in lib/libc/ in the upstream tree; there is no top-level libc/ (though there is a top-level common/, so those files stay where they are). Change-Id: I7c5e2224a4441ab0e33616a855a8c6aacfeac46f
2014-02-20 23:15:09 +01:00
upstream-netbsd/lib/libc/regex/regcomp.c \
upstream-netbsd/lib/libc/regex/regerror.c \
upstream-netbsd/lib/libc/regex/regexec.c \
upstream-netbsd/lib/libc/regex/regfree.c \
upstream-netbsd/lib/libc/stdlib/bsearch.c \
upstream-netbsd/lib/libc/stdlib/div.c \
upstream-netbsd/lib/libc/stdlib/drand48.c \
upstream-netbsd/lib/libc/stdlib/erand48.c \
upstream-netbsd/lib/libc/stdlib/jrand48.c \
Add POSIX lcong48. Change-Id: I821046816661d77275cb02c8c99d526bb41afb9c
2014-11-06 21:40:08 +01:00
upstream-netbsd/lib/libc/stdlib/lcong48.c \
Move the upstream-netbsd libc files into the correct directory. I screwed up when I originally imported these files; they're in lib/libc/ in the upstream tree; there is no top-level libc/ (though there is a top-level common/, so those files stay where they are). Change-Id: I7c5e2224a4441ab0e33616a855a8c6aacfeac46f
2014-02-20 23:15:09 +01:00
upstream-netbsd/lib/libc/stdlib/ldiv.c \
upstream-netbsd/lib/libc/stdlib/lldiv.c \
upstream-netbsd/lib/libc/stdlib/lrand48.c \
upstream-netbsd/lib/libc/stdlib/mrand48.c \
upstream-netbsd/lib/libc/stdlib/nrand48.c \
upstream-netbsd/lib/libc/stdlib/_rand48.c \
Get the full set of PRNG functions in <stdlib.h> Bug: https://code.google.com/p/android/issues/detail?id=58888 Change-Id: I435250bdae302e8bd7e29977d0fde7b9afbfca5e
2014-06-12 20:48:04 +02:00
upstream-netbsd/lib/libc/stdlib/rand_r.c \
Sync with upstream NetBSD lib/libc/regex. Bug: http://b/22850181 Change-Id: I11a51a2031e68a953ccd5691da98c699c7d01904
2015-09-15 03:45:52 +02:00
upstream-netbsd/lib/libc/stdlib/reallocarr.c \
Move the upstream-netbsd libc files into the correct directory. I screwed up when I originally imported these files; they're in lib/libc/ in the upstream tree; there is no top-level libc/ (though there is a top-level common/, so those files stay where they are). Change-Id: I7c5e2224a4441ab0e33616a855a8c6aacfeac46f
2014-02-20 23:15:09 +01:00
upstream-netbsd/lib/libc/stdlib/seed48.c \
upstream-netbsd/lib/libc/stdlib/srand48.c \
upstream-netbsd/lib/libc/string/memccpy.c \
upstream-netbsd/lib/libc/string/strcasestr.c \
upstream-netbsd/lib/libc/string/strcoll.c \
upstream-netbsd/lib/libc/string/strxfrm.c \
Revert "Revert "Switch to NetBSD's strxfrm(3)."" This reverts commit 8793e7c7d21a0434d1b5e63364c88b2b125a3d29, and fixes the build by building upstream NetBSD source as a separate library that's then swallowed whole into libc_common. Change-Id: I6c9317d8c48b5ccaf85a7b185bc07fb31176ff97
2012-08-10 20:39:58 +02:00
Switch to gdtoa. This gives us a real strtold for LP64 and fixes various LP64 bugs. Bug: 13563801 Change-Id: I277858d718ee746e136b6b6308a495ba50dfa488
2014-04-11 02:48:14 +02:00
libc_upstream_openbsd_gdtoa_src_files := \
Clean up and document our hacks for building BSD source. Change-Id: Ic591e22fa5b363bb68376b9f25814c0e5bd83fbf
2014-04-22 02:13:46 +02:00
upstream-openbsd/android/gdtoa_support.cpp \
Switch to gdtoa. This gives us a real strtold for LP64 and fixes various LP64 bugs. Bug: 13563801 Change-Id: I277858d718ee746e136b6b6308a495ba50dfa488
2014-04-11 02:48:14 +02:00
upstream-openbsd/lib/libc/gdtoa/dmisc.c \
upstream-openbsd/lib/libc/gdtoa/dtoa.c \
upstream-openbsd/lib/libc/gdtoa/gdtoa.c \
upstream-openbsd/lib/libc/gdtoa/gethex.c \
upstream-openbsd/lib/libc/gdtoa/gmisc.c \
upstream-openbsd/lib/libc/gdtoa/hd_init.c \
upstream-openbsd/lib/libc/gdtoa/hdtoa.c \
upstream-openbsd/lib/libc/gdtoa/hexnan.c \
upstream-openbsd/lib/libc/gdtoa/ldtoa.c \
upstream-openbsd/lib/libc/gdtoa/misc.c \
upstream-openbsd/lib/libc/gdtoa/smisc.c \
upstream-openbsd/lib/libc/gdtoa/strtod.c \
upstream-openbsd/lib/libc/gdtoa/strtodg.c \
upstream-openbsd/lib/libc/gdtoa/strtof.c \
upstream-openbsd/lib/libc/gdtoa/strtord.c \
upstream-openbsd/lib/libc/gdtoa/sum.c \
upstream-openbsd/lib/libc/gdtoa/ulp.c \
libc_upstream_openbsd_gdtoa_src_files_32 := \
$(libc_upstream_openbsd_gdtoa_src_files) \
libc_upstream_openbsd_gdtoa_src_files_64 := \
$(libc_upstream_openbsd_gdtoa_src_files) \
upstream-openbsd/lib/libc/gdtoa/strtorQ.c \
Implement setjmp cookies on ARM. Reuse the top bits of _JB_SIGFLAG field previously used to store a boolean to store a cookie that's validated by [sig]longjmp to make it harder to use as a ROP gadget. Additionally, encrypt saved registers with the cookie so that an attacker can't modify a register's value to a specific value without knowing the cookie. Bug: http://b/23942752 Change-Id: Id0eb8d06916e89d5d776bfcaa9458f8826717ba3
2015-09-11 00:40:24 +02:00
# These two depend on getentropy_linux.c, which isn't in libc_ndk.a.
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
libc_upstream_openbsd_src_files := \
Use upstream OpenBSD's arc4random. The getentropy_linux.c is lightly modified to build on Android, but we're now completely in sync with upstream OpenBSD's arc4random implementation. Change-Id: If32229fc28aba908035fb38703190d41ddcabc95
2014-07-19 00:57:41 +02:00
upstream-openbsd/lib/libc/crypt/arc4random.c \
upstream-openbsd/lib/libc/crypt/arc4random_uniform.c \
Update NDK compatlib for 4.1.2. The replacement new failures present in newer versions are present here as well, with the following new issues: XPASS std/localization/locale.categories/category.numeric/locale.nm.put/facet.num.put.members/put_long_double.pass.cpp This is from the -NaN formatting fix in bionic. We previously had this wrong, and the upstream test is also wrong. There's currently an XFAIL for Android in this test because I haven't fixed the upstream test yet. After that is done, I'll need to teach the test runner how to XFAIL older Android versions... FAIL std/localization/locale.categories/category.ctype/facet.ctype.special/facet.ctype.char.dtor/dtor.pass.cpp dtor.pass.cpp:39: int main(): assertion "globalMemCounter.checkDeleteArrayCalledEq(1)" failed Haven't investigated this one yet. http://b/19412688 Note that this also needs the libgcc link ordering to be fixed in the build system, as we'll otherwise depend on libgcc symbols from libc that may or may not have been there. The build fix can't be submitted because the proper link order causes the libgcc unwinder to be used instead of the EHABI one: http://b/18471342 Bug: 18471532 Change-Id: Icf560485a9b8f5ebbe01e4458703e62ec94df5e1
2015-02-13 01:35:09 +01:00
Switch to LOCAL_SRC_FILES_EXCLUDE This moves the generic arm/arm64/x86 settings into the main makefiles and makes the rest of them derivatives. This better aligns with how soong handles arch/cpu variants. Also updates the Android.bp to make it consistent with the make versions. Change-Id: I5a0275d992bc657459eb6fe1697ad2336731d122
2015-10-15 23:49:45 +02:00
libc_upstream_openbsd_src_files += \
upstream-openbsd/lib/libc/string/memchr.c \
upstream-openbsd/lib/libc/string/memmove.c \
upstream-openbsd/lib/libc/string/memrchr.c \
upstream-openbsd/lib/libc/string/stpcpy.c \
upstream-openbsd/lib/libc/string/stpncpy.c \
upstream-openbsd/lib/libc/string/strcat.c \
upstream-openbsd/lib/libc/string/strcpy.c \
upstream-openbsd/lib/libc/string/strlcat.c \
upstream-openbsd/lib/libc/string/strlcpy.c \
upstream-openbsd/lib/libc/string/strncat.c \
upstream-openbsd/lib/libc/string/strncmp.c \
upstream-openbsd/lib/libc/string/strncpy.c \
Update NDK compatlib for 4.1.2. The replacement new failures present in newer versions are present here as well, with the following new issues: XPASS std/localization/locale.categories/category.numeric/locale.nm.put/facet.num.put.members/put_long_double.pass.cpp This is from the -NaN formatting fix in bionic. We previously had this wrong, and the upstream test is also wrong. There's currently an XFAIL for Android in this test because I haven't fixed the upstream test yet. After that is done, I'll need to teach the test runner how to XFAIL older Android versions... FAIL std/localization/locale.categories/category.ctype/facet.ctype.special/facet.ctype.char.dtor/dtor.pass.cpp dtor.pass.cpp:39: int main(): assertion "globalMemCounter.checkDeleteArrayCalledEq(1)" failed Haven't investigated this one yet. http://b/19412688 Note that this also needs the libgcc link ordering to be fixed in the build system, as we'll otherwise depend on libgcc symbols from libc that may or may not have been there. The build fix can't be submitted because the proper link order causes the libgcc unwinder to be used instead of the EHABI one: http://b/18471342 Bug: 18471532 Change-Id: Icf560485a9b8f5ebbe01e4458703e62ec94df5e1
2015-02-13 01:35:09 +01:00
libc_upstream_openbsd_ndk_src_files := \
upstream-openbsd/lib/libc/compat-43/killpg.c \
Switch to upstream alarm(3). The only way the setitimer call can fail is if the unsigned number of seconds is too large to fit in the kernel's signed number of seconds. If you schedule a 68-year alarm, glibc will fail by returning 0 and BSD will fail by returning -1. Change-Id: Ic3721b01428f5402d99f31fd7f2ba2cc58805607
2014-03-03 23:38:20 +01:00
upstream-openbsd/lib/libc/gen/alarm.c \
Upgrade our <ctype.h> implementation to OpenBSD head. Adding the perfunctory <ctype.h> tests showed that we'd accidentally dropped several symbols. This puts everything back in its proper place and switches us to upstream head at the same time. Change-Id: Ib527ad280c9baded81e667fa598698526d93e66f
2014-04-18 19:29:16 +02:00
upstream-openbsd/lib/libc/gen/ctype_.c \
Switch to OpenBSD daemon(3). Change-Id: I1fd0be09fdb24aa6f1d945410eba5987f8a949b4
2014-08-27 01:25:19 +02:00
upstream-openbsd/lib/libc/gen/daemon.c \
Switch to OpenBSD <err.h> implementation. Change-Id: Ia950d88871a30f68e74f9ac7dbd87788e128e02f
2014-08-27 02:00:37 +02:00
upstream-openbsd/lib/libc/gen/err.c \
upstream-openbsd/lib/libc/gen/errx.c \
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
upstream-openbsd/lib/libc/gen/exec.c \
upstream-openbsd/lib/libc/gen/fnmatch.c \
Add getprogname/setprogname for BSD compatibility. This is one less change we have to make when porting BSD code. Bug: https://code.google.com/p/android/issues/detail?id=34898 Change-Id: If9b1a8d16996c7a19abcce8d3a456afc3e105a41
2014-03-01 01:23:27 +01:00
upstream-openbsd/lib/libc/gen/ftok.c \
upstream-openbsd/lib/libc/gen/getprogname.c \
Upgrade our <ctype.h> implementation to OpenBSD head. Adding the perfunctory <ctype.h> tests showed that we'd accidentally dropped several symbols. This puts everything back in its proper place and switches us to upstream head at the same time. Change-Id: Ib527ad280c9baded81e667fa598698526d93e66f
2014-04-18 19:29:16 +02:00
upstream-openbsd/lib/libc/gen/isctype.c \
Add getprogname/setprogname for BSD compatibility. This is one less change we have to make when porting BSD code. Bug: https://code.google.com/p/android/issues/detail?id=34898 Change-Id: If9b1a8d16996c7a19abcce8d3a456afc3e105a41
2014-03-01 01:23:27 +01:00
upstream-openbsd/lib/libc/gen/setprogname.c \
Reimplement clock(3) and switch to OpenBSD time(3). The new implementation is a better approximation to the processor time used by the process because it's actually based on resource usage rather than just elapsed wall clock time. Change-Id: I9e13b69c1d3048cadf0eb9dec1e3ebc78225596a
2014-03-11 21:37:11 +01:00
upstream-openbsd/lib/libc/gen/time.c \
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
upstream-openbsd/lib/libc/gen/tolower_.c \
Add getprogname/setprogname for BSD compatibility. This is one less change we have to make when porting BSD code. Bug: https://code.google.com/p/android/issues/detail?id=34898 Change-Id: If9b1a8d16996c7a19abcce8d3a456afc3e105a41
2014-03-01 01:23:27 +01:00
upstream-openbsd/lib/libc/gen/toupper_.c \
Switch to OpenBSD <err.h> implementation. Change-Id: Ia950d88871a30f68e74f9ac7dbd87788e128e02f
2014-08-27 02:00:37 +02:00
upstream-openbsd/lib/libc/gen/verr.c \
upstream-openbsd/lib/libc/gen/verrx.c \
upstream-openbsd/lib/libc/gen/vwarn.c \
upstream-openbsd/lib/libc/gen/vwarnx.c \
upstream-openbsd/lib/libc/gen/warn.c \
upstream-openbsd/lib/libc/gen/warnx.c \
Switch to the OpenBSD wcsto* functions. This replaces a partial set of non-functional functions with a complete set of functions, all of which actually work. This requires us to implement mbsnrtowcs and wcsnrtombs which completes the set of what we need for libc++. The mbsnrtowcs is basically a copy & paste of wcsnrtombs, but I'm going to go straight to looking at using the OpenBSD UTF-8 implementation rather than keep polishing our home-grown turd. (This patch also opportunistically switches us over to upstream btowc, mbrlen, and wctob, since they're all trivially expressed in terms of other functions.) Change-Id: I0f81443840de0f1aa73b96f0b51988976793a323
2014-04-29 23:46:56 +02:00
upstream-openbsd/lib/libc/locale/btowc.c \
upstream-openbsd/lib/libc/locale/mbrlen.c \
upstream-openbsd/lib/libc/locale/mbstowcs.c \
Add mbtowc and fix mbrtowc. Change-Id: I48786cd82587e61188d40f6fd6e11ac05e857ae9
2014-04-29 02:51:13 +02:00
upstream-openbsd/lib/libc/locale/mbtowc.c \
Switch to OpenBSD wcscoll. Change-Id: I1050ef728e18124656e76f32b42ac9a4688f252d
2014-03-11 00:11:59 +01:00
upstream-openbsd/lib/libc/locale/wcscoll.c \
Switch to the OpenBSD wcsto* functions. This replaces a partial set of non-functional functions with a complete set of functions, all of which actually work. This requires us to implement mbsnrtowcs and wcsnrtombs which completes the set of what we need for libc++. The mbsnrtowcs is basically a copy & paste of wcsnrtombs, but I'm going to go straight to looking at using the OpenBSD UTF-8 implementation rather than keep polishing our home-grown turd. (This patch also opportunistically switches us over to upstream btowc, mbrlen, and wctob, since they're all trivially expressed in terms of other functions.) Change-Id: I0f81443840de0f1aa73b96f0b51988976793a323
2014-04-29 23:46:56 +02:00
upstream-openbsd/lib/libc/locale/wcstod.c \
upstream-openbsd/lib/libc/locale/wcstof.c \
Switch to the OpenBSD implementations of the wide scanf functions. This also gets us the C99 wcstoimax and wcstoumax, and a working fgetwc and ungetwc, all of which are needed in the implementation. This also brings several other files closer to upstream. Change-Id: I23b025a8237a6dbb9aa50d2a96765ea729a85579
2014-04-30 01:28:56 +02:00
upstream-openbsd/lib/libc/locale/wcstoimax.c \
Switch to the OpenBSD wcsto* functions. This replaces a partial set of non-functional functions with a complete set of functions, all of which actually work. This requires us to implement mbsnrtowcs and wcsnrtombs which completes the set of what we need for libc++. The mbsnrtowcs is basically a copy & paste of wcsnrtombs, but I'm going to go straight to looking at using the OpenBSD UTF-8 implementation rather than keep polishing our home-grown turd. (This patch also opportunistically switches us over to upstream btowc, mbrlen, and wctob, since they're all trivially expressed in terms of other functions.) Change-Id: I0f81443840de0f1aa73b96f0b51988976793a323
2014-04-29 23:46:56 +02:00
upstream-openbsd/lib/libc/locale/wcstol.c \
upstream-openbsd/lib/libc/locale/wcstold.c \
upstream-openbsd/lib/libc/locale/wcstoll.c \
upstream-openbsd/lib/libc/locale/wcstombs.c \
upstream-openbsd/lib/libc/locale/wcstoul.c \
upstream-openbsd/lib/libc/locale/wcstoull.c \
Switch to the OpenBSD implementations of the wide scanf functions. This also gets us the C99 wcstoimax and wcstoumax, and a working fgetwc and ungetwc, all of which are needed in the implementation. This also brings several other files closer to upstream. Change-Id: I23b025a8237a6dbb9aa50d2a96765ea729a85579
2014-04-30 01:28:56 +02:00
upstream-openbsd/lib/libc/locale/wcstoumax.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/locale/wcsxfrm.c \
Switch to the OpenBSD wcsto* functions. This replaces a partial set of non-functional functions with a complete set of functions, all of which actually work. This requires us to implement mbsnrtowcs and wcsnrtombs which completes the set of what we need for libc++. The mbsnrtowcs is basically a copy & paste of wcsnrtombs, but I'm going to go straight to looking at using the OpenBSD UTF-8 implementation rather than keep polishing our home-grown turd. (This patch also opportunistically switches us over to upstream btowc, mbrlen, and wctob, since they're all trivially expressed in terms of other functions.) Change-Id: I0f81443840de0f1aa73b96f0b51988976793a323
2014-04-29 23:46:56 +02:00
upstream-openbsd/lib/libc/locale/wctob.c \
Switch to a working UTF-8 mb/wc implementation. Although glibc gets by with an 8-byte mbstate_t, OpenBSD uses 12 bytes (of the 128 bytes it reserves!). We can actually implement UTF-8 encoding/decoding with a 0-byte mbstate_t which means we can make things work on LP32 too, as long as we accept the limitation that the caller needs to present us with a complete sequence before we'll process it. Our behavior is fine when going from characters to bytes; we just update the source wchar_t** to say how far through the input we got. I'll come back and use the 4 bytes we do have to cope with byte sequences split across multiple input buffers. The fact that we don't support UTF-8 sequences longer than 4 bytes plus the fact that the first byte of a UTF-8 sequence encodes the length means we shouldn't need the other fields OpenBSD used (at the cost of some recomputation in cases where a sequence is split across buffers). This patch also makes the minimal changes necessary to setlocale(3) to make us behave like glibc when an app requests UTF-8. (The difference being that our "C" locale is the same as our "C.UTF-8" locale.) Change-Id: Ied327a8c4643744b3611bf6bb005a9b389ba4c2f
2014-05-01 07:03:12 +02:00
upstream-openbsd/lib/libc/locale/wctomb.c \
Flesh out <arpa/inet.h>. Use the upstream OpenBSD implementations of these functions. Also ensure we have symbols for htonl, htons, ntohl, and ntohs. gtest doesn't like us using the macro versions in ASSERT_EQ. Bug: 14840760 Change-Id: I68720e9aca14838df457d2bb27b999d5818ac2b5
2014-05-14 01:05:51 +02:00
upstream-openbsd/lib/libc/net/htonl.c \
upstream-openbsd/lib/libc/net/htons.c \
upstream-openbsd/lib/libc/net/inet_lnaof.c \
upstream-openbsd/lib/libc/net/inet_makeaddr.c \
upstream-openbsd/lib/libc/net/inet_netof.c \
upstream-openbsd/lib/libc/net/inet_ntoa.c \
Switch back to OpenBSD inet_ntop now they're FORTIFY clean. Change-Id: I04798a4966c352a8a12feebdff9646e3bad5933f
2014-05-20 00:55:29 +02:00
upstream-openbsd/lib/libc/net/inet_ntop.c \
Flesh out <arpa/inet.h>. Use the upstream OpenBSD implementations of these functions. Also ensure we have symbols for htonl, htons, ntohl, and ntohs. gtest doesn't like us using the macro versions in ASSERT_EQ. Bug: 14840760 Change-Id: I68720e9aca14838df457d2bb27b999d5818ac2b5
2014-05-14 01:05:51 +02:00
upstream-openbsd/lib/libc/net/inet_pton.c \
upstream-openbsd/lib/libc/net/ntohl.c \
upstream-openbsd/lib/libc/net/ntohs.c \
Sync with upstream OpenBSD. Also some minor build cleanup. Change-Id: Ibd20602d7ea45685f0c47fceb00ffd6c7ae35954
2016-01-16 01:30:18 +01:00
upstream-openbsd/lib/libc/net/res_random.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/asprintf.c \
upstream-openbsd/lib/libc/stdio/clrerr.c \
Switch to POSIX dprintf/vdprintf. Bug: 11156955 Change-Id: I734bd02db514367ab119a48304aae9767958e367
2014-05-22 10:24:30 +02:00
upstream-openbsd/lib/libc/stdio/dprintf.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/feof.c \
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
upstream-openbsd/lib/libc/stdio/ferror.c \
upstream-openbsd/lib/libc/stdio/fflush.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/fgetc.c \
upstream-openbsd/lib/libc/stdio/fgetln.c \
upstream-openbsd/lib/libc/stdio/fgets.c \
Switch to the OpenBSD implementations of the wide scanf functions. This also gets us the C99 wcstoimax and wcstoumax, and a working fgetwc and ungetwc, all of which are needed in the implementation. This also brings several other files closer to upstream. Change-Id: I23b025a8237a6dbb9aa50d2a96765ea729a85579
2014-04-30 01:28:56 +02:00
upstream-openbsd/lib/libc/stdio/fgetwc.c \
Switch to OpenBSD stdio wide get/put functions. Change-Id: I71f8769cdea874e55d397ca7682d9d4e659d3dcb
2014-04-30 02:08:03 +02:00
upstream-openbsd/lib/libc/stdio/fgetws.c \
Switch to OpenBSD flags.c. Change-Id: I0a35e5bd9f8edba27e0c73e5c8150636346d6a81
2014-09-24 05:02:42 +02:00
upstream-openbsd/lib/libc/stdio/flags.c \
Add POSIX-2008 fmemopen, open_memstream, and open_wmemstream. Bug: 17164505 Change-Id: I59e28a08ff8b6ab632230b11a5807cfd5278aeb5
2014-08-21 01:10:49 +02:00
upstream-openbsd/lib/libc/stdio/fmemopen.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/fprintf.c \
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
upstream-openbsd/lib/libc/stdio/fpurge.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/fputc.c \
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
upstream-openbsd/lib/libc/stdio/fputs.c \
Switch to OpenBSD stdio wide get/put functions. Change-Id: I71f8769cdea874e55d397ca7682d9d4e659d3dcb
2014-04-30 02:08:03 +02:00
upstream-openbsd/lib/libc/stdio/fputwc.c \
upstream-openbsd/lib/libc/stdio/fputws.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/fscanf.c \
Switch to upstream OpenBSD fread/fvwrite. Change-Id: I3dc11083693bc8d99edc0cbcc6f70dc9e5dc6565
2014-04-18 22:13:04 +02:00
upstream-openbsd/lib/libc/stdio/fvwrite.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/fwalk.c \
Switch to OpenBSD stdio wide get/put functions. Change-Id: I71f8769cdea874e55d397ca7682d9d4e659d3dcb
2014-04-30 02:08:03 +02:00
upstream-openbsd/lib/libc/stdio/fwide.c \
Switch to OpenBSD stdio wide printf functions. Change-Id: Icf4f8685d021ec6b7482ca1cc021ce8184098e4a
2014-04-30 02:39:29 +02:00
upstream-openbsd/lib/libc/stdio/fwprintf.c \
Switch to current upstream OpenBSD fwrite.c. Change-Id: Ife527aafc1e5438f477d711902efe6e6f59f3f8e
2014-05-03 03:18:46 +02:00
upstream-openbsd/lib/libc/stdio/fwrite.c \
Switch to the OpenBSD implementations of the wide scanf functions. This also gets us the C99 wcstoimax and wcstoumax, and a working fgetwc and ungetwc, all of which are needed in the implementation. This also brings several other files closer to upstream. Change-Id: I23b025a8237a6dbb9aa50d2a96765ea729a85579
2014-04-30 01:28:56 +02:00
upstream-openbsd/lib/libc/stdio/fwscanf.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/getc.c \
upstream-openbsd/lib/libc/stdio/getchar.c \
upstream-openbsd/lib/libc/stdio/getdelim.c \
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
upstream-openbsd/lib/libc/stdio/getline.c \
upstream-openbsd/lib/libc/stdio/gets.c \
Switch to OpenBSD stdio wide get/put functions. Change-Id: I71f8769cdea874e55d397ca7682d9d4e659d3dcb
2014-04-30 02:08:03 +02:00
upstream-openbsd/lib/libc/stdio/getwc.c \
upstream-openbsd/lib/libc/stdio/getwchar.c \
Register _cleanup function with atexit * Register cleanup function with atexit instead of calling it explicitly on exit() * abort() no longer calls _cleanup: Flushing stdio buffers on abort is no longer required by POSIX. * dlmalloc no longer need to reset cleanup (see above) * Upstream findfp.c makebuf.c setvbuf.cexit.c to openbsd versions. Bug: 14415367 Change-Id: I277058852485a9d3dbb13e5c232db5f9948d78ac
2014-05-15 08:11:05 +02:00
upstream-openbsd/lib/libc/stdio/makebuf.c \
Switch to upstream OpenBSD mktemp.c. Almost all of our stdio is actually OpenBSD, so although this isn't really a core part of stdio (it doesn't touch struct FILE, for example) it probably makes sense for it to come from the same upstream. My actual motivation though is that it's the only FreeBSD file we have compiler warnings from. This patch moves us over to -Werror by default, with only the DNS code having -Wno-error. Change-Id: Id244a5b445cba41b0a1ca30298ca7b1ed177810c
2014-07-02 20:22:26 +02:00
upstream-openbsd/lib/libc/stdio/mktemp.c \
Add POSIX-2008 fmemopen, open_memstream, and open_wmemstream. Bug: 17164505 Change-Id: I59e28a08ff8b6ab632230b11a5807cfd5278aeb5
2014-08-21 01:10:49 +02:00
upstream-openbsd/lib/libc/stdio/open_memstream.c \
upstream-openbsd/lib/libc/stdio/open_wmemstream.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/perror.c \
upstream-openbsd/lib/libc/stdio/printf.c \
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
upstream-openbsd/lib/libc/stdio/putc.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/putchar.c \
upstream-openbsd/lib/libc/stdio/puts.c \
Switch to OpenBSD stdio wide get/put functions. Change-Id: I71f8769cdea874e55d397ca7682d9d4e659d3dcb
2014-04-30 02:08:03 +02:00
upstream-openbsd/lib/libc/stdio/putwc.c \
upstream-openbsd/lib/libc/stdio/putwchar.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/remove.c \
upstream-openbsd/lib/libc/stdio/rewind.c \
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
upstream-openbsd/lib/libc/stdio/rget.c \
upstream-openbsd/lib/libc/stdio/scanf.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/setbuf.c \
upstream-openbsd/lib/libc/stdio/setbuffer.c \
Register _cleanup function with atexit * Register cleanup function with atexit instead of calling it explicitly on exit() * abort() no longer calls _cleanup: Flushing stdio buffers on abort is no longer required by POSIX. * dlmalloc no longer need to reset cleanup (see above) * Upstream findfp.c makebuf.c setvbuf.cexit.c to openbsd versions. Bug: 14415367 Change-Id: I277058852485a9d3dbb13e5c232db5f9948d78ac
2014-05-15 08:11:05 +02:00
upstream-openbsd/lib/libc/stdio/setvbuf.c \
Upgrade to current upstream scanf implementation. Also add a basic test. Change-Id: Icc0e68a5716b9579244f6eb8bac1ab5a24eda85a
2014-03-13 01:10:41 +01:00
upstream-openbsd/lib/libc/stdio/sscanf.c \
Switch to OpenBSD stdio wide printf functions. Change-Id: Icf4f8685d021ec6b7482ca1cc021ce8184098e4a
2014-04-30 02:39:29 +02:00
upstream-openbsd/lib/libc/stdio/swprintf.c \
Switch to the OpenBSD implementations of the wide scanf functions. This also gets us the C99 wcstoimax and wcstoumax, and a working fgetwc and ungetwc, all of which are needed in the implementation. This also brings several other files closer to upstream. Change-Id: I23b025a8237a6dbb9aa50d2a96765ea729a85579
2014-04-30 01:28:56 +02:00
upstream-openbsd/lib/libc/stdio/swscanf.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/tempnam.c \
upstream-openbsd/lib/libc/stdio/tmpnam.c \
upstream-openbsd/lib/libc/stdio/ungetc.c \
Switch to the OpenBSD implementations of the wide scanf functions. This also gets us the C99 wcstoimax and wcstoumax, and a working fgetwc and ungetwc, all of which are needed in the implementation. This also brings several other files closer to upstream. Change-Id: I23b025a8237a6dbb9aa50d2a96765ea729a85579
2014-04-30 01:28:56 +02:00
upstream-openbsd/lib/libc/stdio/ungetwc.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/vasprintf.c \
Switch to POSIX dprintf/vdprintf. Bug: 11156955 Change-Id: I734bd02db514367ab119a48304aae9767958e367
2014-05-22 10:24:30 +02:00
upstream-openbsd/lib/libc/stdio/vdprintf.c \
Upgrade to current vfprintf. This gets us various bug fixes and missing C99 functionality. Bug: https://code.google.com/p/android/issues/detail?id=64886 Change-Id: Ie9f8ac569e9b5eec1e4a1faacfe2c21662eaf895
2014-04-18 02:30:03 +02:00
upstream-openbsd/lib/libc/stdio/vfprintf.c \
Upgrade to current upstream scanf implementation. Also add a basic test. Change-Id: Icc0e68a5716b9579244f6eb8bac1ab5a24eda85a
2014-03-13 01:10:41 +01:00
upstream-openbsd/lib/libc/stdio/vfscanf.c \
Switch to OpenBSD stdio wide printf functions. Change-Id: Icf4f8685d021ec6b7482ca1cc021ce8184098e4a
2014-04-30 02:39:29 +02:00
upstream-openbsd/lib/libc/stdio/vfwprintf.c \
Switch to the OpenBSD implementations of the wide scanf functions. This also gets us the C99 wcstoimax and wcstoumax, and a working fgetwc and ungetwc, all of which are needed in the implementation. This also brings several other files closer to upstream. Change-Id: I23b025a8237a6dbb9aa50d2a96765ea729a85579
2014-04-30 01:28:56 +02:00
upstream-openbsd/lib/libc/stdio/vfwscanf.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/vprintf.c \
upstream-openbsd/lib/libc/stdio/vscanf.c \
upstream-openbsd/lib/libc/stdio/vsnprintf.c \
upstream-openbsd/lib/libc/stdio/vsprintf.c \
Upgrade to current upstream scanf implementation. Also add a basic test. Change-Id: Icc0e68a5716b9579244f6eb8bac1ab5a24eda85a
2014-03-13 01:10:41 +01:00
upstream-openbsd/lib/libc/stdio/vsscanf.c \
Switch to OpenBSD stdio wide printf functions. Change-Id: Icf4f8685d021ec6b7482ca1cc021ce8184098e4a
2014-04-30 02:39:29 +02:00
upstream-openbsd/lib/libc/stdio/vswprintf.c \
Switch to the OpenBSD implementations of the wide scanf functions. This also gets us the C99 wcstoimax and wcstoumax, and a working fgetwc and ungetwc, all of which are needed in the implementation. This also brings several other files closer to upstream. Change-Id: I23b025a8237a6dbb9aa50d2a96765ea729a85579
2014-04-30 01:28:56 +02:00
upstream-openbsd/lib/libc/stdio/vswscanf.c \
Switch to OpenBSD stdio wide printf functions. Change-Id: Icf4f8685d021ec6b7482ca1cc021ce8184098e4a
2014-04-30 02:39:29 +02:00
upstream-openbsd/lib/libc/stdio/vwprintf.c \
Switch to the OpenBSD implementations of the wide scanf functions. This also gets us the C99 wcstoimax and wcstoumax, and a working fgetwc and ungetwc, all of which are needed in the implementation. This also brings several other files closer to upstream. Change-Id: I23b025a8237a6dbb9aa50d2a96765ea729a85579
2014-04-30 01:28:56 +02:00
upstream-openbsd/lib/libc/stdio/vwscanf.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdio/wbuf.c \
Switch to OpenBSD stdio wide printf functions. Change-Id: Icf4f8685d021ec6b7482ca1cc021ce8184098e4a
2014-04-30 02:39:29 +02:00
upstream-openbsd/lib/libc/stdio/wprintf.c \
Switch to the OpenBSD implementations of the wide scanf functions. This also gets us the C99 wcstoimax and wcstoumax, and a working fgetwc and ungetwc, all of which are needed in the implementation. This also brings several other files closer to upstream. Change-Id: I23b025a8237a6dbb9aa50d2a96765ea729a85579
2014-04-30 01:28:56 +02:00
upstream-openbsd/lib/libc/stdio/wscanf.c \
Switch to current upstream OpenBSD wsetup.c. Change-Id: I2c1123f3e1d3c4af7fd7bf354e763934a39b78c0
2014-05-03 03:29:25 +02:00
upstream-openbsd/lib/libc/stdio/wsetup.c \
Move several FreeBSD functions to OpenBSD. There's no difference between the two BSDs for these functions, so let's go with the majority. Change-Id: Ic296f23153a48bed5ad152f5b31989c4e00e60bd
2015-09-15 04:01:58 +02:00
upstream-openbsd/lib/libc/stdlib/abs.c \
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
upstream-openbsd/lib/libc/stdlib/atoi.c \
upstream-openbsd/lib/libc/stdlib/atol.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdlib/atoll.c \
Switch to the upstream OpenBSD getenv/putenv/setenv implementation. This fixes all the bugs found by the new tests. Change-Id: Id5a5f9f39a0620208bafa053f871a044725b4795
2014-04-23 02:41:00 +02:00
upstream-openbsd/lib/libc/stdlib/getenv.c \
Switch to OpenBSD insque/remque/killpg. Change-Id: Ie02290ad3187b1c1596dd776fd1c8a743a55f7ef
2014-09-23 23:06:56 +02:00
upstream-openbsd/lib/libc/stdlib/insque.c \
Move several FreeBSD functions to OpenBSD. There's no difference between the two BSDs for these functions, so let's go with the majority. Change-Id: Ic296f23153a48bed5ad152f5b31989c4e00e60bd
2015-09-15 04:01:58 +02:00
upstream-openbsd/lib/libc/stdlib/imaxabs.c \
upstream-openbsd/lib/libc/stdlib/imaxdiv.c \
upstream-openbsd/lib/libc/stdlib/labs.c \
upstream-openbsd/lib/libc/stdlib/llabs.c \
Implement twalk(3), add unit tests. I've also added insque(3) and remque(3) (from NetBSD because the OpenBSD ones are currently broken for non-circular lists). I've not added the three hash table functions that should be in this header because they operate on a single global hash table and thus aren't likely to be useful. Bug: https://code.google.com/p/android/issues/detail?id=73719 Change-Id: I97397a7b921e2e860fd9c8032cafd9097380498a
2014-07-24 01:02:26 +02:00
upstream-openbsd/lib/libc/stdlib/lsearch.c \
Sync with upstream OpenBSD fts.c. Change-Id: Ia7e26b603f6461095b5a8d7e8a6cdf4f8279fa84
2014-11-22 04:25:27 +01:00
upstream-openbsd/lib/libc/stdlib/reallocarray.c \
Switch to OpenBSD insque/remque/killpg. Change-Id: Ie02290ad3187b1c1596dd776fd1c8a743a55f7ef
2014-09-23 23:06:56 +02:00
upstream-openbsd/lib/libc/stdlib/remque.c \
Switch to the upstream OpenBSD getenv/putenv/setenv implementation. This fixes all the bugs found by the new tests. Change-Id: Id5a5f9f39a0620208bafa053f871a044725b4795
2014-04-23 02:41:00 +02:00
upstream-openbsd/lib/libc/stdlib/setenv.c \
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
upstream-openbsd/lib/libc/stdlib/strtoimax.c \
upstream-openbsd/lib/libc/stdlib/strtol.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/stdlib/strtoll.c \
upstream-openbsd/lib/libc/stdlib/strtoul.c \
upstream-openbsd/lib/libc/stdlib/strtoull.c \
upstream-openbsd/lib/libc/stdlib/strtoumax.c \
Switch to current OpenBSD system.c. Change-Id: Ie2df60f0d55dca69149ffe064f10183411407987
2014-02-26 00:19:26 +01:00
upstream-openbsd/lib/libc/stdlib/system.c \
Implement twalk(3), add unit tests. I've also added insque(3) and remque(3) (from NetBSD because the OpenBSD ones are currently broken for non-circular lists). I've not added the three hash table functions that should be in this header because they operate on a single global hash table and thus aren't likely to be useful. Bug: https://code.google.com/p/android/issues/detail?id=73719 Change-Id: I97397a7b921e2e860fd9c8032cafd9097380498a
2014-07-24 01:02:26 +02:00
upstream-openbsd/lib/libc/stdlib/tfind.c \
upstream-openbsd/lib/libc/stdlib/tsearch.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/string/strcasecmp.c \
upstream-openbsd/lib/libc/string/strcspn.c \
upstream-openbsd/lib/libc/string/strdup.c \
upstream-openbsd/lib/libc/string/strndup.c \
upstream-openbsd/lib/libc/string/strpbrk.c \
upstream-openbsd/lib/libc/string/strsep.c \
upstream-openbsd/lib/libc/string/strspn.c \
upstream-openbsd/lib/libc/string/strstr.c \
upstream-openbsd/lib/libc/string/strtok.c \
Clean up <stdlib.h> slightly. Interestingly, this mostly involves cleaning up our implementation of various <string.h> functions. Change-Id: Ifaef49b5cb997134f7bc0cc31bdac844bdb9e089
2015-01-26 22:34:58 +01:00
upstream-openbsd/lib/libc/string/wmemcpy.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/string/wcslcpy.c \
Replace our broken wcswcs with the working upstream one. Change-Id: I2952684df5674d10f0564d92c2cd42597725c0e3
2014-04-29 01:28:51 +02:00
upstream-openbsd/lib/libc/string/wcsstr.c \
More stdio cleanup. Change-Id: Idc909cd3dc7b072f1edd2ae4980932d6550e8568
2014-03-11 23:05:50 +01:00
upstream-openbsd/lib/libc/string/wcswidth.c \
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
libc_pthread_src_files := \
bionic/pthread_atfork.cpp \
bionic/pthread_attr.cpp \
Implement pthread barrier. Bug: 24341262 Change-Id: I5472549e5d7545c1c3f0bef78235f545557b9630
2015-11-06 07:06:09 +01:00
bionic/pthread_barrier.cpp \
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
bionic/pthread_cond.cpp \
bionic/pthread_create.cpp \
bionic/pthread_detach.cpp \
bionic/pthread_equal.cpp \
bionic/pthread_exit.cpp \
bionic/pthread_getcpuclockid.cpp \
bionic/pthread_getschedparam.cpp \
bionic/pthread_gettid_np.cpp \
Let g_thread_list_lock only protect g_thread_list. As glibc/netbsd don't protect access to thread struct members by a global lock, we don't want to do it either. This change reduces the responsibility of g_thread_list_lock to only protect g_thread_list. Bug: 19636317 Change-Id: I897890710653dac165d8fa4452c7ecf74abdbf2b
2015-03-19 22:19:19 +01:00
bionic/pthread_internal.cpp \
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
bionic/pthread_join.cpp \
bionic/pthread_key.cpp \
bionic/pthread_kill.cpp \
bionic/pthread_mutex.cpp \
bionic/pthread_once.cpp \
bionic/pthread_rwlock.cpp \
bionic/pthread_self.cpp \
bionic/pthread_setname_np.cpp \
bionic/pthread_setschedparam.cpp \
bionic/pthread_sigmask.cpp \
Implement pthread spin. In order to run tsan unit tests, we need to support pthread spin APIs. Bug: 18623621 Bug: 25392375 Change-Id: Icbb4a74e72e467824b3715982a01600031868e29
2015-11-18 01:03:18 +01:00
bionic/pthread_spinlock.cpp \
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
libc_arch_static_src_files := \
bionic/dl_iterate_phdr_static.cpp \
Break bionic implementations into arch versions. DO NOT MERGE Move arch specific code for arm, mips, x86 into separate makefiles. In addition, add different arm cpu versions of memcpy/memset. Bug: 8005082 (cherry picked from commit acdde8c1cf8e8beed98c052757d96695b820b50c) Change-Id: I0108d432af9f6283ae99adfc92a3399e5ab3e31d
2013-02-26 10:30:00 +01:00
Fix a few libc makefile mishaps. These are the problems fixed in this CL: - The ndk_cruft.cpp file was not compiled for 64 bit. - Some files where not compiled for libc_bionic.a due to a missing patch-up-arch-specific-flags call. - The libc_bionic_ndk.a patch up call was using bionic src, not the bionic_ndk_src_files. - Move mmap.cpp to the ndk src list where it was before because of the above errors. Bug: 21640784 Change-Id: I31e97bc8bf347c8e584cc17aa4b984a7d122b298
2015-06-05 04:24:07 +02:00
# Various kinds of cruft.
Share LP32 makefile settings between arches Add <var>_32 to patch-up-arch-specific-flags, and move the LP32 cruft varaibles from the 32-bit arch specific makefiles into the top level Android.mk. Change-Id: Id3fcf6805d4af048c2524c94b1295416ebe7d057
2015-02-05 02:35:49 +01:00
# ========================================================
Fix a few libc makefile mishaps. These are the problems fixed in this CL: - The ndk_cruft.cpp file was not compiled for 64 bit. - Some files where not compiled for libc_bionic.a due to a missing patch-up-arch-specific-flags call. - The libc_bionic_ndk.a patch up call was using bionic src, not the bionic_ndk_src_files. - Move mmap.cpp to the ndk src list where it was before because of the above errors. Bug: 21640784 Change-Id: I31e97bc8bf347c8e584cc17aa4b984a7d122b298
2015-06-05 04:24:07 +02:00
libc_common_src_files += \
bionic/ndk_cruft.cpp \
libc_bionic_ndk_src_files_32 += \
Share LP32 makefile settings between arches Add <var>_32 to patch-up-arch-specific-flags, and move the LP32 cruft varaibles from the 32-bit arch specific makefiles into the top level Android.mk. Change-Id: Id3fcf6805d4af048c2524c94b1295416ebe7d057
2015-02-05 02:35:49 +01:00
bionic/mmap.cpp \
libc_common_src_files_32 += \
bionic/legacy_32_bit_support.cpp \
bionic/time64.c \
libc_netbsd_src_files_32 += \
upstream-netbsd/common/lib/libc/hash/sha1/sha1.c \
libc_openbsd_src_files_32 += \
upstream-openbsd/lib/libc/stdio/putw.c \
libc: Cleanup the libc makefile. Removes a lot of redundant code, and reorganizes the file. We first setup all the comomn variables we may need, and then define the actual libraries that need to be generated. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
# Define some common cflags
# ========================================================
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
libc_common_cflags := \
Give the timer_create SIGEV_THREAD helper threads sensible names. Bug: 6609676 Change-Id: I286b197c75beee4d9930b0973f2d7dd47c14e91c
2012-06-06 19:32:56 +02:00
-D_LIBC=1 \
Turn on -Wunused and fix the mistakes it uncovers. Change-Id: I023d2d8b547fbc21d4124bb7510d42b06a0dc501
2014-06-04 00:22:34 +02:00
-Wall -Wextra -Wunused \
Don't allow int<->pointer conversions. Normally we don't have -Werror for upstream code, but for those warnings that probably point to 32-bit assumptions about pointers, we want those warnings to always be errors. Change-Id: Ibece9caf09b2f7989ca600ef448d07868669a8fb
2013-10-09 02:04:33 +02:00
Enable clang compilation with libc but not linker. Clang is still disabled for x86 and x86_64 long double code, for x86_64 special assembly instruction, and the linker module. BUG: 17163651 BUG: 17302991 BUG: 17403674 Change-Id: I43c5f5f0ddc0c2a31918f00b57150dadd8f3f26b
2014-09-05 00:19:52 +02:00
use_clang := $(USE_CLANG_PLATFORM_BUILD)
Use gcc to build and test x86_64 long double code. Clang/llvm x86_64 configuration of long double is still incompatible with gcc. https://llvm.org/bugs/show_bug.cgi?id=23897 BUG: 21899249 Change-Id: I47fd6d0ce2daa84ae0035e7885a4d90e4c3056f7
2015-06-23 23:18:14 +02:00
# Clang/llvm has incompatible long double (fp128) for x86_64.
# https://llvm.org/bugs/show_bug.cgi?id=23897
ifeq ($(TARGET_ARCH),x86_64)
use_clang := false
endif
Disable clang for mips/mips64 libc. * Many processes, including adbd, failed to start in mips/mips64 emulator when libc.so was compiled by clang. BUG: 25291096 Change-Id: If3434ebdca4a3a6bf6102b120ee838a7ab66cd74
2015-11-20 18:55:35 +01:00
# b/25291096, Clang/llvm compiled libc.so for mips/mips64 failed to boot.
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),mips mips64))
use_clang := false
endif
Enable clang compilation with libc but not linker. Clang is still disabled for x86 and x86_64 long double code, for x86_64 special assembly instruction, and the linker module. BUG: 17163651 BUG: 17302991 BUG: 17403674 Change-Id: I43c5f5f0ddc0c2a31918f00b57150dadd8f3f26b
2014-09-05 00:19:52 +02:00
ifeq ($(use_clang),)
Switch libc default to clang. * Default to clang when USE_CLANG_PLATFORM_BUILD is not set and the target has no clang bug. BUG: 26102335 Change-Id: Ied6c9dc5593bfbadbb8d8b38e66ea237d649bae5
2016-01-19 19:59:34 +01:00
use_clang := true
Enable clang compilation with libc but not linker. Clang is still disabled for x86 and x86_64 long double code, for x86_64 special assembly instruction, and the linker module. BUG: 17163651 BUG: 17302991 BUG: 17403674 Change-Id: I43c5f5f0ddc0c2a31918f00b57150dadd8f3f26b
2014-09-05 00:19:52 +02:00
endif
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
Don't allow int<->pointer conversions. Normally we don't have -Werror for upstream code, but for those warnings that probably point to 32-bit assumptions about pointers, we want those warnings to always be errors. Change-Id: Ibece9caf09b2f7989ca600ef448d07868669a8fb
2013-10-09 02:04:33 +02:00
# Try to catch typical 32-bit assumptions that break with 64-bit pointers.
libc_common_cflags += \
-Werror=pointer-to-int-cast \
-Werror=int-to-pointer-cast \
-Werror=type-limits \
Switch to upstream OpenBSD mktemp.c. Almost all of our stdio is actually OpenBSD, so although this isn't really a core part of stdio (it doesn't touch struct FILE, for example) it probably makes sense for it to come from the same upstream. My actual motivation though is that it's the only FreeBSD file we have compiler warnings from. This patch moves us over to -Werror by default, with only the DNS code having -Wno-error. Change-Id: Id244a5b445cba41b0a1ca30298ca7b1ed177810c
2014-07-02 20:22:26 +02:00
-Werror \
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
libc: Replace a reference to TARGET_BUILD_TYPE with DEBUG_BIONIC_LIBC We're soon going to stop using TARGET_BUILD_TYPE==debug to debug native modules. Signed-off-by: Dave Bort <dbort@android.com>
2009-04-24 00:50:03 +02:00
ifeq ($(strip $(DEBUG_BIONIC_LIBC)),true)
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
libc_common_cflags += -DDEBUG
endif
Remove dlmalloc. Bug: http://b/17207577 Change-Id: Ie009badca6deb1f91b27a4340b70cdd6bedff893
2016-01-07 02:57:44 +01:00
libc_malloc_src := bionic/jemalloc_wrapper.cpp
libc_common_c_includes += external/jemalloc/include
Support for jemalloc to replace dlmalloc. To use jemalloc, add MALLOC_IMPL = jemalloc in a board config file and you get the new version automatically. Update the pthread_create_key tests since jemalloc uses a few keys. Add a new test to verify memalign works as expected. Bug: 981363 Change-Id: I16eb152b291a95bd2499e90492fc6b4bd7053836
2014-05-08 20:14:03 +02:00
Use gnu++11 and gnu99 explicitly for C++/C files. Clang and gcc default to different standards, so we should be explicit about the versions we want to compile for. Change-Id: I65495a2392dd29f36373b94c616c2506173e6033
2013-10-01 02:10:53 +02:00
# Define some common conlyflags
libc_common_conlyflags := \
-std=gnu99
# Define some common cppflags
libc_common_cppflags := \
libc: Cleanup the libc makefile. Removes a lot of redundant code, and reorganizes the file. We first setup all the comomn variables we may need, and then define the actual libraries that need to be generated. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
# Define some common includes
# ========================================================
Make jemalloc the default choice. Change-Id: I5ff0cf0c396b7ae6278e6fcd5a0d50f8b966fb54
2014-07-11 03:53:41 +02:00
libc_common_c_includes += \
Fix missing args. Change-Id: I89a366db984ea6b2553896a4351e59bb91f93040
2014-02-13 01:40:55 +01:00
$(LOCAL_PATH)/stdio \
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
# ========================================================
Share LP32 makefile settings between arches Add <var>_32 to patch-up-arch-specific-flags, and move the LP32 cruft varaibles from the 32-bit arch specific makefiles into the top level Android.mk. Change-Id: Id3fcf6805d4af048c2524c94b1295416ebe7d057
2015-02-05 02:35:49 +01:00
# Add in the arch or 32-bit specific flags
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
# Must be called with $(eval).
# $(1): the LOCAL_ variable name
# $(2): the bionic variable name to pull in
define patch-up-arch-specific-flags
$(1)_$(TARGET_ARCH) += $($(2)_$(TARGET_ARCH))
Share LP32 makefile settings between arches Add <var>_32 to patch-up-arch-specific-flags, and move the LP32 cruft varaibles from the 32-bit arch specific makefiles into the top level Android.mk. Change-Id: Id3fcf6805d4af048c2524c94b1295416ebe7d057
2015-02-05 02:35:49 +01:00
$(1)_32 += $($(2)_32)
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
ifdef TARGET_2ND_ARCH
$(1)_$(TARGET_2ND_ARCH) += $($(2)_$(TARGET_2ND_ARCH))
MIPS support to the libc Makefile Change-Id: Ia1f3f545ca61a5abb235e99046cb3cdaf24d4205
2012-08-07 21:23:24 +02:00
endif
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
endef
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
Don't use -fstack-protector on ssp.c libc's stack protector initialization routine (__guard_setup) is in bionic/ssp.c. This code deliberately modifies the stack canary. This code should never be compiled with -fstack-protector-all otherwise it will crash (mismatched canary value). Force bionic/ssp.c to be compiled with -fno-stack-protector Change-Id: Ib95a5736e4bafe1a460d6b4e522ca660b417d8d6
2012-07-10 19:46:19 +02:00
# ========================================================
Rename the stack protector intermediate library. libbionic_ssp already confused at least one person, and characters in filenames are cheap, so let's just call this library what it is. Change-Id: I69ab950bf52fa4d267a6891efb49b5e177efc0c4
2014-03-11 00:15:11 +01:00
# libc_stack_protector.a - stack protector code
Don't use -fstack-protector on ssp.c libc's stack protector initialization routine (__guard_setup) is in bionic/ssp.c. This code deliberately modifies the stack canary. This code should never be compiled with -fstack-protector-all otherwise it will crash (mismatched canary value). Force bionic/ssp.c to be compiled with -fno-stack-protector Change-Id: Ib95a5736e4bafe1a460d6b4e522ca660b417d8d6
2012-07-10 19:46:19 +02:00
# ========================================================
#
Defend against -fstack-protector in libc startup. Exactly which functions get a stack protector is up to the compiler, so let's separate the code that sets up the environment stack protection requires and explicitly build it with -fno-stack-protector. Bug: http://b/26276517 Change-Id: I8719e23ead1f1e81715c32c1335da868f68369b5
2016-01-07 04:51:43 +01:00
# Code that implements the stack protector (or that runs
# before TLS has been set up) needs to be compiled with
# -fno-stack-protector, since it accesses the stack canary
# TLS slot.
Don't use -fstack-protector on ssp.c libc's stack protector initialization routine (__guard_setup) is in bionic/ssp.c. This code deliberately modifies the stack canary. This code should never be compiled with -fstack-protector-all otherwise it will crash (mismatched canary value). Force bionic/ssp.c to be compiled with -fno-stack-protector Change-Id: Ib95a5736e4bafe1a460d6b4e522ca660b417d8d6
2012-07-10 19:46:19 +02:00
include $(CLEAR_VARS)
Defend against -fstack-protector in libc startup. Exactly which functions get a stack protector is up to the compiler, so let's separate the code that sets up the environment stack protection requires and explicitly build it with -fno-stack-protector. Bug: http://b/26276517 Change-Id: I8719e23ead1f1e81715c32c1335da868f68369b5
2016-01-07 04:51:43 +01:00
LOCAL_SRC_FILES := \
bionic/__libc_init_main_thread.cpp \
bionic/__stack_chk_fail.cpp \
LOCAL_SRC_FILES_arm64 := arch-arm64/bionic/__set_tls.c
Fix -fstack-protector-strong for x86. We need to ensure %gs:20 is set up early enough for -fstack-protector-strong on x86, and that __set_tls doesn't get stack protector checks because it's a prerequisite for them. x86 devices/emulators won't boot without this. Bug: http://b/26073874 Change-Id: Icf0d34294648cc0c8cb406a3617befe0d45c525a
2015-12-08 21:43:13 +01:00
LOCAL_SRC_FILES_x86 := arch-x86/bionic/__set_tls.c
Defend against -fstack-protector in libc startup. Exactly which functions get a stack protector is up to the compiler, so let's separate the code that sets up the environment stack protection requires and explicitly build it with -fno-stack-protector. Bug: http://b/26276517 Change-Id: I8719e23ead1f1e81715c32c1335da868f68369b5
2016-01-07 04:51:43 +01:00
LOCAL_SRC_FILES_x86_64 := arch-x86_64/bionic/__set_tls.c
Switch to upstream OpenBSD mktemp.c. Almost all of our stdio is actually OpenBSD, so although this isn't really a core part of stdio (it doesn't touch struct FILE, for example) it probably makes sense for it to come from the same upstream. My actual motivation though is that it's the only FreeBSD file we have compiler warnings from. This patch moves us over to -Werror by default, with only the DNS code having -Wno-error. Change-Id: Id244a5b445cba41b0a1ca30298ca7b1ed177810c
2014-07-02 20:22:26 +02:00
LOCAL_CFLAGS := $(libc_common_cflags) -fno-stack-protector
Use gnu++11 and gnu99 explicitly for C++/C files. Clang and gcc default to different standards, so we should be explicit about the versions we want to compile for. Change-Id: I65495a2392dd29f36373b94c616c2506173e6033
2013-10-01 02:10:53 +02:00
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
Don't use -fstack-protector on ssp.c libc's stack protector initialization routine (__guard_setup) is in bionic/ssp.c. This code deliberately modifies the stack canary. This code should never be compiled with -fstack-protector-all otherwise it will crash (mismatched canary value). Force bionic/ssp.c to be compiled with -fno-stack-protector Change-Id: Ib95a5736e4bafe1a460d6b4e522ca660b417d8d6
2012-07-10 19:46:19 +02:00
LOCAL_C_INCLUDES := $(libc_common_c_includes)
Rename the stack protector intermediate library. libbionic_ssp already confused at least one person, and characters in filenames are cheap, so let's just call this library what it is. Change-Id: I69ab950bf52fa4d267a6891efb49b5e177efc0c4
2014-03-11 00:15:11 +01:00
LOCAL_MODULE := libc_stack_protector
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_CLANG := $(use_clang)
Add dependencies on included makefiles. Bug: 11050594 Merge from internal master. (cherry-picked from f389284e86bbcbdb257d7388a17fde007113b0d6) Change-Id: I2b3e38329a09d26c16870906f9ed1257e2a9dbc8
2013-10-03 22:26:22 +02:00
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Don't use -fstack-protector on ssp.c libc's stack protector initialization routine (__guard_setup) is in bionic/ssp.c. This code deliberately modifies the stack canary. This code should never be compiled with -fstack-protector-all otherwise it will crash (mismatched canary value). Force bionic/ssp.c to be compiled with -fno-stack-protector Change-Id: Ib95a5736e4bafe1a460d6b4e522ca660b417d8d6
2012-07-10 19:46:19 +02:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Don't use -fstack-protector on ssp.c libc's stack protector initialization routine (__guard_setup) is in bionic/ssp.c. This code deliberately modifies the stack canary. This code should never be compiled with -fstack-protector-all otherwise it will crash (mismatched canary value). Force bionic/ssp.c to be compiled with -fno-stack-protector Change-Id: Ib95a5736e4bafe1a460d6b4e522ca660b417d8d6
2012-07-10 19:46:19 +02:00
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
Don't use -fstack-protector on ssp.c libc's stack protector initialization routine (__guard_setup) is in bionic/ssp.c. This code deliberately modifies the stack canary. This code should never be compiled with -fstack-protector-all otherwise it will crash (mismatched canary value). Force bionic/ssp.c to be compiled with -fno-stack-protector Change-Id: Ib95a5736e4bafe1a460d6b4e522ca660b417d8d6
2012-07-10 19:46:19 +02:00
include $(BUILD_STATIC_LIBRARY)
Defend against -fstack-protector in libc startup. Exactly which functions get a stack protector is up to the compiler, so let's separate the code that sets up the environment stack protection requires and explicitly build it with -fno-stack-protector. Bug: http://b/26276517 Change-Id: I8719e23ead1f1e81715c32c1335da868f68369b5
2016-01-07 04:51:43 +01:00
# libc_init_static.cpp also needs to be built without stack protector,
# because it's responsible for setting up TLS for static executables.
# This isn't the case for dynamic executables because the dynamic linker
# has already set up the main thread's TLS.
include $(CLEAR_VARS)
LOCAL_SRC_FILES := bionic/libc_init_static.cpp
LOCAL_CFLAGS := $(libc_common_cflags) -fno-stack-protector
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
LOCAL_C_INCLUDES := $(libc_common_c_includes)
LOCAL_MODULE := libc_init_static
LOCAL_CLANG := $(use_clang)
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
LOCAL_CXX_STL := none
LOCAL_SYSTEM_SHARED_LIBRARIES :=
LOCAL_SANITIZE := never
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
include $(BUILD_STATIC_LIBRARY)
Upgrade to tzcode2013d. Well, kinda... localtime.c still contains a bunch of Android-specific hacks, as does strftime.c. But the other files are now exactly the same as upstream. This catches up with several years of bug fixes, and fixes most of the compiler warnings that were in this code. (Just two remain.) Bug: 1744909 Change-Id: I2ddfecb6fd408c847397c17afb0fff859e27feef
2013-07-13 02:31:11 +02:00
# ========================================================
# libc_tzcode.a - upstream 'tzcode' code
# ========================================================
include $(CLEAR_VARS)
Don't explicitly list the tzcode files. Change-Id: I3ac19899523ac1f5fd2fdd6c942aefce75a964a4
2014-02-20 20:28:20 +01:00
LOCAL_SRC_FILES := $(call all-c-files-under,tzcode)
Use the OpenBSD wcsftime. Change-Id: I81929355d245ba1e58b4a464ca6cf45915e0238e
2014-05-03 00:53:03 +02:00
# tzcode doesn't include wcsftime, so we use the OpenBSD one.
LOCAL_SRC_FILES += upstream-openbsd/lib/libc/time/wcsftime.c
Don't use so much stack in tzcode. Bug: 14313703 Bug: https://code.google.com/p/android/issues/detail?id=61130 Change-Id: Id9b240fce20d4a6f2660792070e02b2b5f287655
2014-04-26 01:55:04 +02:00
Build tzcode with hidden visibility. Bug: 11156955 Change-Id: Ib98d837b56cbbdfd01687cb3054fe3103eec0da9
2014-06-19 23:49:30 +02:00
LOCAL_CFLAGS := $(libc_common_cflags) \
-fvisibility=hidden \
Sync with upstream tzcode (2015g). This is quite a large patch because we haven't updated for some time, but the good news is that upstream is now thread-safe so a lot of our changes go away in this update and the remaining diff is a lot smaller. (Note that our whitespace still doesn't match upstream. I use diff -wub to compare. Upstream doesn't even really have a consistent style. New code seems to be two spaces, old code tabs.) From the intervening changelogs (eliding the changes that only affected the tools, which we don't use): 2014a: An uninitialized-storage bug in 'localtime' has been fixed. (Thanks to Logan Chien.) 2014b: 'zic' and 'localtime' no longer reject locations needing four transitions per year for the forseeable future. (Thanks to Andrew Main (Zefram).) 2014c: <None> 2014d: <None> 2014e: <None> 2014f: 'localtime', 'mktime', etc. now use much less stack space if ALL_STATE is defined. (Thanks to Elliott Hughes for reporting the problem.) Some lint has been removed when using GCC_DEBUG_FLAGS with GCC 4.9.0. 2014g: Unless NETBSD_INSPIRED is defined to 0, the tz library now supplies functions for creating and using objects that represent time zones. The new functions are tzalloc, tzfree, localtime_rz, mktime_z, and (if STD_INSPIRED is also defined) posix2time_z and time2posix_z. They are intended for performance: for example, localtime_rz (unlike localtime_r) is trivially thread-safe without locking. (Thanks to Christos Zoulas for proposing NetBSD-inspired functions, and to Alan Barrett and Jonathan Lennox for helping to debug the change.) If THREAD_SAFE is defined to 1, the tz library is now thread-safe. Although not needed for tz's own applications, which are single-threaded, this supports POSIX better if the tz library is used in multithreaded apps. Some crashes have been fixed when zdump or the tz library is given invalid or outlandish input. The tz library no longer mishandles leap seconds on platforms with unsigned time_t in time zones that lack ordinary transitions after 1970. The tz code now attempts to infer TM_GMTOFF and TM_ZONE if not already defined, to make it easier to configure on common platforms. Define NO_TM_GMTOFF and NO_TM_ZONE to suppress this. Unless the new macro UNINIT_TRAP is defined to 1, the tz code now assumes that reading uninitialized memory yields garbage values but does not cause other problems such as traps. If TM_GMTOFF is defined and UNINIT_TRAP is 0, mktime is now more likely to guess right for ambiguous time stamps near transitions where tm_isdst does not change. If HAVE_STRFTIME_L is defined to 1, the tz library now defines strftime_l for compatibility with recent versions of POSIX. Only the C locale is supported, though. HAVE_STRFTIME_L defaults to 1 on recent POSIX versions, and to 0 otherwise. tzselect -c now uses a hybrid distance measure that works better in Africa. (Thanks to Alan Barrett for noting the problem.) The C source code now ports to NetBSD when GCC_DEBUG_FLAGS is used, or when time_tz is defined. When HAVE_UTMPX_H is set the 'date' command now builds on systems whose <utmpx.h> file does not define WTMPX_FILE, and when setting the date it updates the wtmpx file if _PATH_WTMPX is defined. This affects GNU/Linux and similar systems. For easier maintenance later, some C code has been simplified, some lint has been removed, and the code has been tweaked so that plain 'make' is more likely to work. The C type 'bool' is now used for boolean values, instead of 'int'. The long-obsolete LOCALE_HOME code has been removed. The long-obsolete 'gtime' function has been removed. 2014h: The tz library's localtime and mktime functions now set tzname to a value appropriate for the requested time stamp, and zdump now uses this on platforms not defining TM_ZONE, fixing a 2014g regression. (Thanks to Tim Parenti for reporting the problem.) The tz library no longer sets tzname if localtime or mktime fails. An access to uninitalized data has been fixed. (Thanks to Jörg Richter for reporting the problem.) When THREAD_SAFE is defined, the code ports to the C11 memory model. A memory leak has been fixed if ALL_STATE and THREAD_SAFE are defined and two threads race to initialize data used by gmtime-like functions. (Thanks to Andy Heninger for reporting the problems.) 2014i: The time-related library functions now set errno on failure, and some crashes in the new tzalloc-related library functions have been fixed. (Thanks to Christos Zoulas for reporting most of these problems and for suggesting fixes.) If USG_COMPAT is defined and the requested time stamp is standard time, the tz library's localtime and mktime functions now set the extern variable timezone to a value appropriate for that time stamp; and similarly for ALTZONE, daylight saving time, and the altzone variable. This change is a companion to the tzname change in 2014h, and is designed to make timezone and altzone more compatible with tzname. The tz library's functions now set errno to EOVERFLOW if they fail because the result cannot be represented. ctime and ctime_r now return NULL and set errno when a time stamp is out of range, rather than having undefined behavior. Some bugs associated with the new 2014g functions have been fixed. This includes a bug that largely incapacitated the new functions time2posix_z and posix2time_z. (Thanks to Christos Zoulas.) It also includes some uses of uninitialized variables after tzalloc. The new code uses the standard type 'ssize_t', which the Makefile now gives porting advice about. 2014j: <None> 2015a: tzalloc now scrubs time zone abbreviations compatibly with the way that tzset always has, by replacing invalid bytes with '_' and by shortening too-long abbreviations. 2015b: Fix integer overflow bug in reference 'mktime' implementation. (Problem reported by Jörg Richter.) Allow -Dtime_tz=time_t compilations, and allow -Dtime_tz=... libraries to be used in the same executable as standard-library time_t functions. (Problems reported by Bradley White.) 2015c: <None> 2015d: <None> 2015e: <None> 2015f: <None> 2015g: localtime no longer mishandles America/Anchorage after 2037. (Thanks to Bradley White for reporting the bug.) On hosts with signed 32-bit time_t, localtime no longer mishandles Pacific/Fiji after 2038-01-16 14:00 UTC. The localtime module allows the variables 'timezone', 'daylight', and 'altzone' to be in common storage shared with other modules, and declares them in case the system <time.h> does not. (Problems reported by Kees Dekker.) On platforms with tm_zone, strftime.c now assumes it is not NULL. This simplifies the code and is consistent with zdump.c. (Problem reported by Christos Zoulas.) Change-Id: I9eb0a8323cb8bd9968fcfe612dc14f45aa3b59d2
2015-10-08 02:13:40 +02:00
-Wno-unused-parameter \
Build tzcode with hidden visibility. Bug: 11156955 Change-Id: Ib98d837b56cbbdfd01687cb3054fe3103eec0da9
2014-06-19 23:49:30 +02:00
Don't use so much stack in tzcode. Bug: 14313703 Bug: https://code.google.com/p/android/issues/detail?id=61130 Change-Id: Id9b240fce20d4a6f2660792070e02b2b5f287655
2014-04-26 01:55:04 +02:00
# Don't use ridiculous amounts of stack.
LOCAL_CFLAGS += -DALL_STATE
# Include tzsetwall, timelocal, timegm, time2posix, and posix2time.
LOCAL_CFLAGS += -DSTD_INSPIRED
Fix strftime if tm_zone is null. Upstream tzcode said "On platforms with tm_zone, strftime.c now assumes it is not NULL". Which is fine for any struct tm generated by tzcode, but not necessarily true of a struct tm constructed by arbitrary code. In particular, Netflix on Nexus Player was failing to start because they format "%Z" with a struct tm whose tm_zone is null (the other fields are valid, but, yeah, that's probably not intentional). glibc takes a null tm_zone to mean "the current time zone", so let's do that too. (Historically Android would use the empty string, and POSIX doesn't clarify which of this is the appropriate behavior when tm_zone is null.) Bug: http://b/25170306 Change-Id: Idbf68bfe90d143aca7dada8607742905188b1d33
2015-11-13 01:51:31 +01:00
# Obviously, we want to be thread-safe.
Sync with upstream tzcode (2015g). This is quite a large patch because we haven't updated for some time, but the good news is that upstream is now thread-safe so a lot of our changes go away in this update and the remaining diff is a lot smaller. (Note that our whitespace still doesn't match upstream. I use diff -wub to compare. Upstream doesn't even really have a consistent style. New code seems to be two spaces, old code tabs.) From the intervening changelogs (eliding the changes that only affected the tools, which we don't use): 2014a: An uninitialized-storage bug in 'localtime' has been fixed. (Thanks to Logan Chien.) 2014b: 'zic' and 'localtime' no longer reject locations needing four transitions per year for the forseeable future. (Thanks to Andrew Main (Zefram).) 2014c: <None> 2014d: <None> 2014e: <None> 2014f: 'localtime', 'mktime', etc. now use much less stack space if ALL_STATE is defined. (Thanks to Elliott Hughes for reporting the problem.) Some lint has been removed when using GCC_DEBUG_FLAGS with GCC 4.9.0. 2014g: Unless NETBSD_INSPIRED is defined to 0, the tz library now supplies functions for creating and using objects that represent time zones. The new functions are tzalloc, tzfree, localtime_rz, mktime_z, and (if STD_INSPIRED is also defined) posix2time_z and time2posix_z. They are intended for performance: for example, localtime_rz (unlike localtime_r) is trivially thread-safe without locking. (Thanks to Christos Zoulas for proposing NetBSD-inspired functions, and to Alan Barrett and Jonathan Lennox for helping to debug the change.) If THREAD_SAFE is defined to 1, the tz library is now thread-safe. Although not needed for tz's own applications, which are single-threaded, this supports POSIX better if the tz library is used in multithreaded apps. Some crashes have been fixed when zdump or the tz library is given invalid or outlandish input. The tz library no longer mishandles leap seconds on platforms with unsigned time_t in time zones that lack ordinary transitions after 1970. The tz code now attempts to infer TM_GMTOFF and TM_ZONE if not already defined, to make it easier to configure on common platforms. Define NO_TM_GMTOFF and NO_TM_ZONE to suppress this. Unless the new macro UNINIT_TRAP is defined to 1, the tz code now assumes that reading uninitialized memory yields garbage values but does not cause other problems such as traps. If TM_GMTOFF is defined and UNINIT_TRAP is 0, mktime is now more likely to guess right for ambiguous time stamps near transitions where tm_isdst does not change. If HAVE_STRFTIME_L is defined to 1, the tz library now defines strftime_l for compatibility with recent versions of POSIX. Only the C locale is supported, though. HAVE_STRFTIME_L defaults to 1 on recent POSIX versions, and to 0 otherwise. tzselect -c now uses a hybrid distance measure that works better in Africa. (Thanks to Alan Barrett for noting the problem.) The C source code now ports to NetBSD when GCC_DEBUG_FLAGS is used, or when time_tz is defined. When HAVE_UTMPX_H is set the 'date' command now builds on systems whose <utmpx.h> file does not define WTMPX_FILE, and when setting the date it updates the wtmpx file if _PATH_WTMPX is defined. This affects GNU/Linux and similar systems. For easier maintenance later, some C code has been simplified, some lint has been removed, and the code has been tweaked so that plain 'make' is more likely to work. The C type 'bool' is now used for boolean values, instead of 'int'. The long-obsolete LOCALE_HOME code has been removed. The long-obsolete 'gtime' function has been removed. 2014h: The tz library's localtime and mktime functions now set tzname to a value appropriate for the requested time stamp, and zdump now uses this on platforms not defining TM_ZONE, fixing a 2014g regression. (Thanks to Tim Parenti for reporting the problem.) The tz library no longer sets tzname if localtime or mktime fails. An access to uninitalized data has been fixed. (Thanks to Jörg Richter for reporting the problem.) When THREAD_SAFE is defined, the code ports to the C11 memory model. A memory leak has been fixed if ALL_STATE and THREAD_SAFE are defined and two threads race to initialize data used by gmtime-like functions. (Thanks to Andy Heninger for reporting the problems.) 2014i: The time-related library functions now set errno on failure, and some crashes in the new tzalloc-related library functions have been fixed. (Thanks to Christos Zoulas for reporting most of these problems and for suggesting fixes.) If USG_COMPAT is defined and the requested time stamp is standard time, the tz library's localtime and mktime functions now set the extern variable timezone to a value appropriate for that time stamp; and similarly for ALTZONE, daylight saving time, and the altzone variable. This change is a companion to the tzname change in 2014h, and is designed to make timezone and altzone more compatible with tzname. The tz library's functions now set errno to EOVERFLOW if they fail because the result cannot be represented. ctime and ctime_r now return NULL and set errno when a time stamp is out of range, rather than having undefined behavior. Some bugs associated with the new 2014g functions have been fixed. This includes a bug that largely incapacitated the new functions time2posix_z and posix2time_z. (Thanks to Christos Zoulas.) It also includes some uses of uninitialized variables after tzalloc. The new code uses the standard type 'ssize_t', which the Makefile now gives porting advice about. 2014j: <None> 2015a: tzalloc now scrubs time zone abbreviations compatibly with the way that tzset always has, by replacing invalid bytes with '_' and by shortening too-long abbreviations. 2015b: Fix integer overflow bug in reference 'mktime' implementation. (Problem reported by Jörg Richter.) Allow -Dtime_tz=time_t compilations, and allow -Dtime_tz=... libraries to be used in the same executable as standard-library time_t functions. (Problems reported by Bradley White.) 2015c: <None> 2015d: <None> 2015e: <None> 2015f: <None> 2015g: localtime no longer mishandles America/Anchorage after 2037. (Thanks to Bradley White for reporting the bug.) On hosts with signed 32-bit time_t, localtime no longer mishandles Pacific/Fiji after 2038-01-16 14:00 UTC. The localtime module allows the variables 'timezone', 'daylight', and 'altzone' to be in common storage shared with other modules, and declares them in case the system <time.h> does not. (Problems reported by Kees Dekker.) On platforms with tm_zone, strftime.c now assumes it is not NULL. This simplifies the code and is consistent with zdump.c. (Problem reported by Christos Zoulas.) Change-Id: I9eb0a8323cb8bd9968fcfe612dc14f45aa3b59d2
2015-10-08 02:13:40 +02:00
LOCAL_CFLAGS += -DTHREAD_SAFE
Don't use so much stack in tzcode. Bug: 14313703 Bug: https://code.google.com/p/android/issues/detail?id=61130 Change-Id: Id9b240fce20d4a6f2660792070e02b2b5f287655
2014-04-26 01:55:04 +02:00
# The name of the tm_gmtoff field in our struct tm.
LOCAL_CFLAGS += -DTM_GMTOFF=tm_gmtoff
# Where we store our tzdata.
LOCAL_CFLAGS += -DTZDIR=\"/system/usr/share/zoneinfo\"
# Include timezone and daylight globals.
LOCAL_CFLAGS += -DUSG_COMPAT=1
Fix strftime if tm_zone is null. Upstream tzcode said "On platforms with tm_zone, strftime.c now assumes it is not NULL". Which is fine for any struct tm generated by tzcode, but not necessarily true of a struct tm constructed by arbitrary code. In particular, Netflix on Nexus Player was failing to start because they format "%Z" with a struct tm whose tm_zone is null (the other fields are valid, but, yeah, that's probably not intentional). glibc takes a null tm_zone to mean "the current time zone", so let's do that too. (Historically Android would use the empty string, and POSIX doesn't clarify which of this is the appropriate behavior when tm_zone is null.) Bug: http://b/25170306 Change-Id: Idbf68bfe90d143aca7dada8607742905188b1d33
2015-11-13 01:51:31 +01:00
# Use the empty string (instead of " ") as the timezone abbreviation fallback.
LOCAL_CFLAGS += -DWILDABBR=\"\"
Fix <math.h> to quieten most of our warnings. I've reported the wcsftime bug upstream, but we really just want to use -D to ensure the buggy code isn't built. (I've also brought our strftime a bit closer to upstream now we have the right define.) I don't think upstream is likely to fix all their sign-compare and uninitialized warnings, so let's just silence them. As for libm, again upstream isn't likely to fix all their warnings, and silencing those made the ones that were our fault stand out. I've fixed our <math.h> to fix the warnings caused by our lack of definitions for the non-imprecise long-double functions. I checked the C99 standard, and all these functions are there. Change-Id: Iee8e1182c1db375058fb2c451eceb212bab47a37
2014-05-06 05:28:28 +02:00
LOCAL_CFLAGS += -DNO_RUN_TIME_WARNINGS_ABOUT_YEAR_2000_PROBLEMS_THANK_YOU
Build tzcode with hidden visibility. Bug: 11156955 Change-Id: Ib98d837b56cbbdfd01687cb3054fe3103eec0da9
2014-06-19 23:49:30 +02:00
LOCAL_CFLAGS += -Dlint
Don't use so much stack in tzcode. Bug: 14313703 Bug: https://code.google.com/p/android/issues/detail?id=61130 Change-Id: Id9b240fce20d4a6f2660792070e02b2b5f287655
2014-04-26 01:55:04 +02:00
Use gnu++11 and gnu99 explicitly for C++/C files. Clang and gcc default to different standards, so we should be explicit about the versions we want to compile for. Change-Id: I65495a2392dd29f36373b94c616c2506173e6033
2013-10-01 02:10:53 +02:00
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
Use the OpenBSD wcsftime. Change-Id: I81929355d245ba1e58b4a464ca6cf45915e0238e
2014-05-03 00:53:03 +02:00
LOCAL_C_INCLUDES := $(libc_common_c_includes) $(LOCAL_PATH)/tzcode/
Upgrade to tzcode2013d. Well, kinda... localtime.c still contains a bunch of Android-specific hacks, as does strftime.c. But the other files are now exactly the same as upstream. This catches up with several years of bug fixes, and fixes most of the compiler warnings that were in this code. (Just two remain.) Bug: 1744909 Change-Id: I2ddfecb6fd408c847397c17afb0fff859e27feef
2013-07-13 02:31:11 +02:00
LOCAL_MODULE := libc_tzcode
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_CLANG := $(use_clang)
Add dependencies on included makefiles. Bug: 11050594 Merge from internal master. (cherry-picked from f389284e86bbcbdb257d7388a17fde007113b0d6) Change-Id: I2b3e38329a09d26c16870906f9ed1257e2a9dbc8
2013-10-03 22:26:22 +02:00
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Upgrade to tzcode2013d. Well, kinda... localtime.c still contains a bunch of Android-specific hacks, as does strftime.c. But the other files are now exactly the same as upstream. This catches up with several years of bug fixes, and fixes most of the compiler warnings that were in this code. (Just two remain.) Bug: 1744909 Change-Id: I2ddfecb6fd408c847397c17afb0fff859e27feef
2013-07-13 02:31:11 +02:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Upgrade to tzcode2013d. Well, kinda... localtime.c still contains a bunch of Android-specific hacks, as does strftime.c. But the other files are now exactly the same as upstream. This catches up with several years of bug fixes, and fixes most of the compiler warnings that were in this code. (Just two remain.) Bug: 1744909 Change-Id: I2ddfecb6fd408c847397c17afb0fff859e27feef
2013-07-13 02:31:11 +02:00
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
Upgrade to tzcode2013d. Well, kinda... localtime.c still contains a bunch of Android-specific hacks, as does strftime.c. But the other files are now exactly the same as upstream. This catches up with several years of bug fixes, and fixes most of the compiler warnings that were in this code. (Just two remain.) Bug: 1744909 Change-Id: I2ddfecb6fd408c847397c17afb0fff859e27feef
2013-07-13 02:31:11 +02:00
include $(BUILD_STATIC_LIBRARY)
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
# ========================================================
# libc_dns.a - modified NetBSD DNS code
# ========================================================
include $(CLEAR_VARS)
Use default hidden visibility to build libc_dns. Bug: 11156955 Change-Id: Ia443705f5fbee0681039d4480abc89850968f475
2014-05-16 21:04:10 +02:00
LOCAL_SRC_FILES := \
$(call all-c-files-under,dns) \
upstream-netbsd/lib/libc/isc/ev_streams.c \
upstream-netbsd/lib/libc/isc/ev_timers.c \
Move mtctxres.c to libc_dns.a. Has the effect of making ___mtctxres hidden. Bug: 17007799 Change-Id: I5aa5f49344ad5ecb33f48737430561b329bcbb0d
2014-08-15 00:43:25 +02:00
upstream-netbsd/lib/libc/resolv/mtctxres.c \
Use default hidden visibility to build libc_dns. Bug: 11156955 Change-Id: Ia443705f5fbee0681039d4480abc89850968f475
2014-05-16 21:04:10 +02:00
Switch to OpenBSD res_random. Change-Id: Ia971d647832893e9bb4601697132a105524d2f96
2014-07-22 00:31:01 +02:00
LOCAL_CFLAGS += \
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
$(libc_common_cflags) \
Clean up reentrancy cruft. The DNS copy of reentrant.h was unused, so remove it. The strtod implementation can use the upstream-netbsd reentrant.h and get a little closer to what was then upstream. (It's since been replaced by gdtoa, and we'll have to follow at some point, but for now this doesn't make anything any worse.) ANDROID_CHANGES is (now) only used in the DNS code, so push the -D down. The <locale.h> change prevents an LP32 hack from leaking into LP64. Change-Id: Idf30b98a59d7ca8f7c6cd6d07020b512057911ef
2014-03-14 00:17:43 +01:00
-DANDROID_CHANGES \
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
-DINET6 \
Use default hidden visibility to build libc_dns. Bug: 11156955 Change-Id: Ia443705f5fbee0681039d4480abc89850968f475
2014-05-16 21:04:10 +02:00
-fvisibility=hidden \
Switch to upstream OpenBSD mktemp.c. Almost all of our stdio is actually OpenBSD, so although this isn't really a core part of stdio (it doesn't touch struct FILE, for example) it probably makes sense for it to come from the same upstream. My actual motivation though is that it's the only FreeBSD file we have compiler warnings from. This patch moves us over to -Werror by default, with only the DNS code having -Wno-error. Change-Id: Id244a5b445cba41b0a1ca30298ca7b1ed177810c
2014-07-02 20:22:26 +02:00
-Wno-unused-parameter \
Use default hidden visibility to build libc_dns. Bug: 11156955 Change-Id: Ia443705f5fbee0681039d4480abc89850968f475
2014-05-16 21:04:10 +02:00
-include netbsd-compat.h \
Fix the newly-introduced warning and turn on -Werror. I cleaned up most of our warnings last week but forgot to turn on -Werror, so of course we're getting new warnings already. I've left -Werror commented out in those places where we still have warnings to deal with before we can turn on -Werror. Change-Id: Ia58ff8b8c1ada4bf81eec6f19ec1d34e133cf4b1
2014-05-17 01:50:34 +02:00
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
Use LOCAL_C_INCLUDES instead of LOCAL_CFLAGS for include dirs Change-Id: If61d6faae5d91c3bbe198e733e5922cd877ba353
2015-02-06 23:16:17 +01:00
LOCAL_C_INCLUDES := $(libc_common_c_includes) \
$(LOCAL_PATH)/dns/include \
$(LOCAL_PATH)/private \
$(LOCAL_PATH)/upstream-netbsd/lib/libc/include \
$(LOCAL_PATH)/upstream-netbsd/android/include \
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
LOCAL_MODULE := libc_dns
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_CLANG := $(use_clang)
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
include $(BUILD_STATIC_LIBRARY)
Move realpath.c to upstream-freebsd. This is actually a slightly newer upstream version than the one I originally pulled. Hopefully now it's in upstream-freebsd it will be easier to track upstream, though I still need to sit down and write the necessary scripts at some point. Bug: 5110679 Change-Id: I87e563f0f95aa8e68b45578e2a8f448bbf827a33
2013-03-02 01:59:46 +01:00
# ========================================================
# libc_freebsd.a - upstream FreeBSD C library code
# ========================================================
#
# These files are built with the freebsd-compat.h header file
# automatically included.
include $(CLEAR_VARS)
LOCAL_SRC_FILES := $(libc_upstream_freebsd_src_files)
LOCAL_CFLAGS := \
$(libc_common_cflags) \
Fix <math.h> to quieten most of our warnings. I've reported the wcsftime bug upstream, but we really just want to use -D to ensure the buggy code isn't built. (I've also brought our strftime a bit closer to upstream now we have the right define.) I don't think upstream is likely to fix all their sign-compare and uninitialized warnings, so let's just silence them. As for libm, again upstream isn't likely to fix all their warnings, and silencing those made the ones that were our fault stand out. I've fixed our <math.h> to fix the warnings caused by our lack of definitions for the non-imprecise long-double functions. I checked the C99 standard, and all these functions are there. Change-Id: Iee8e1182c1db375058fb2c451eceb212bab47a37
2014-05-06 05:28:28 +02:00
-Wno-sign-compare -Wno-uninitialized \
Fix the newly-introduced warning and turn on -Werror. I cleaned up most of our warnings last week but forgot to turn on -Werror, so of course we're getting new warnings already. I've left -Werror commented out in those places where we still have warnings to deal with before we can turn on -Werror. Change-Id: Ia58ff8b8c1ada4bf81eec6f19ec1d34e133cf4b1
2014-05-17 01:50:34 +02:00
-include freebsd-compat.h \
Use gnu++11 and gnu99 explicitly for C++/C files. Clang and gcc default to different standards, so we should be explicit about the versions we want to compile for. Change-Id: I65495a2392dd29f36373b94c616c2506173e6033
2013-10-01 02:10:53 +02:00
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
Use LOCAL_C_INCLUDES instead of LOCAL_CFLAGS for include dirs Change-Id: If61d6faae5d91c3bbe198e733e5922cd877ba353
2015-02-06 23:16:17 +01:00
LOCAL_C_INCLUDES := $(libc_common_c_includes) \
$(LOCAL_PATH)/upstream-freebsd/android/include \
$(LOCAL_PATH)/upstream-freebsd/lib/libc/include \
Move realpath.c to upstream-freebsd. This is actually a slightly newer upstream version than the one I originally pulled. Hopefully now it's in upstream-freebsd it will be easier to track upstream, though I still need to sit down and write the necessary scripts at some point. Bug: 5110679 Change-Id: I87e563f0f95aa8e68b45578e2a8f448bbf827a33
2013-03-02 01:59:46 +01:00
LOCAL_MODULE := libc_freebsd
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_CLANG := $(use_clang)
Add dependencies on included makefiles. Bug: 11050594 Merge from internal master. (cherry-picked from f389284e86bbcbdb257d7388a17fde007113b0d6) Change-Id: I2b3e38329a09d26c16870906f9ed1257e2a9dbc8
2013-10-03 22:26:22 +02:00
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Move realpath.c to upstream-freebsd. This is actually a slightly newer upstream version than the one I originally pulled. Hopefully now it's in upstream-freebsd it will be easier to track upstream, though I still need to sit down and write the necessary scripts at some point. Bug: 5110679 Change-Id: I87e563f0f95aa8e68b45578e2a8f448bbf827a33
2013-03-02 01:59:46 +01:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Move realpath.c to upstream-freebsd. This is actually a slightly newer upstream version than the one I originally pulled. Hopefully now it's in upstream-freebsd it will be easier to track upstream, though I still need to sit down and write the necessary scripts at some point. Bug: 5110679 Change-Id: I87e563f0f95aa8e68b45578e2a8f448bbf827a33
2013-03-02 01:59:46 +01:00
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
Switch to LOCAL_SRC_FILES_EXCLUDE This moves the generic arm/arm64/x86 settings into the main makefiles and makes the rest of them derivatives. This better aligns with how soong handles arch/cpu variants. Also updates the Android.bp to make it consistent with the make versions. Change-Id: I5a0275d992bc657459eb6fe1697ad2336731d122
2015-10-15 23:49:45 +02:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_SRC_FILES_EXCLUDE,libc_freebsd_src_files_exclude))
Move realpath.c to upstream-freebsd. This is actually a slightly newer upstream version than the one I originally pulled. Hopefully now it's in upstream-freebsd it will be easier to track upstream, though I still need to sit down and write the necessary scripts at some point. Bug: 5110679 Change-Id: I87e563f0f95aa8e68b45578e2a8f448bbf827a33
2013-03-02 01:59:46 +01:00
include $(BUILD_STATIC_LIBRARY)
Revert "Revert "Switch to NetBSD's strxfrm(3)."" This reverts commit 8793e7c7d21a0434d1b5e63364c88b2b125a3d29, and fixes the build by building upstream NetBSD source as a separate library that's then swallowed whole into libc_common. Change-Id: I6c9317d8c48b5ccaf85a7b185bc07fb31176ff97
2012-08-10 20:39:58 +02:00
# ========================================================
# libc_netbsd.a - upstream NetBSD C library code
# ========================================================
#
# These files are built with the netbsd-compat.h header file
# automatically included.
include $(CLEAR_VARS)
LOCAL_SRC_FILES := $(libc_upstream_netbsd_src_files)
LOCAL_CFLAGS := \
$(libc_common_cflags) \
Fix <math.h> to quieten most of our warnings. I've reported the wcsftime bug upstream, but we really just want to use -D to ensure the buggy code isn't built. (I've also brought our strftime a bit closer to upstream now we have the right define.) I don't think upstream is likely to fix all their sign-compare and uninitialized warnings, so let's just silence them. As for libm, again upstream isn't likely to fix all their warnings, and silencing those made the ones that were our fault stand out. I've fixed our <math.h> to fix the warnings caused by our lack of definitions for the non-imprecise long-double functions. I checked the C99 standard, and all these functions are there. Change-Id: Iee8e1182c1db375058fb2c451eceb212bab47a37
2014-05-06 05:28:28 +02:00
-Wno-sign-compare -Wno-uninitialized \
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
-DPOSIX_MISTAKE \
Fix the newly-introduced warning and turn on -Werror. I cleaned up most of our warnings last week but forgot to turn on -Werror, so of course we're getting new warnings already. I've left -Werror commented out in those places where we still have warnings to deal with before we can turn on -Werror. Change-Id: Ia58ff8b8c1ada4bf81eec6f19ec1d34e133cf4b1
2014-05-17 01:50:34 +02:00
-include netbsd-compat.h \
Use gnu++11 and gnu99 explicitly for C++/C files. Clang and gcc default to different standards, so we should be explicit about the versions we want to compile for. Change-Id: I65495a2392dd29f36373b94c616c2506173e6033
2013-10-01 02:10:53 +02:00
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
Use LOCAL_C_INCLUDES instead of LOCAL_CFLAGS for include dirs Change-Id: If61d6faae5d91c3bbe198e733e5922cd877ba353
2015-02-06 23:16:17 +01:00
LOCAL_C_INCLUDES := $(libc_common_c_includes) \
$(LOCAL_PATH)/upstream-netbsd/android/include \
$(LOCAL_PATH)/upstream-netbsd/lib/libc/include \
Revert "Revert "Switch to NetBSD's strxfrm(3)."" This reverts commit 8793e7c7d21a0434d1b5e63364c88b2b125a3d29, and fixes the build by building upstream NetBSD source as a separate library that's then swallowed whole into libc_common. Change-Id: I6c9317d8c48b5ccaf85a7b185bc07fb31176ff97
2012-08-10 20:39:58 +02:00
LOCAL_MODULE := libc_netbsd
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_CLANG := $(use_clang)
Add dependencies on included makefiles. Bug: 11050594 Merge from internal master. (cherry-picked from f389284e86bbcbdb257d7388a17fde007113b0d6) Change-Id: I2b3e38329a09d26c16870906f9ed1257e2a9dbc8
2013-10-03 22:26:22 +02:00
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Revert "Revert "Switch to NetBSD's strxfrm(3)."" This reverts commit 8793e7c7d21a0434d1b5e63364c88b2b125a3d29, and fixes the build by building upstream NetBSD source as a separate library that's then swallowed whole into libc_common. Change-Id: I6c9317d8c48b5ccaf85a7b185bc07fb31176ff97
2012-08-10 20:39:58 +02:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Revert "Revert "Switch to NetBSD's strxfrm(3)."" This reverts commit 8793e7c7d21a0434d1b5e63364c88b2b125a3d29, and fixes the build by building upstream NetBSD source as a separate library that's then swallowed whole into libc_common. Change-Id: I6c9317d8c48b5ccaf85a7b185bc07fb31176ff97
2012-08-10 20:39:58 +02:00
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
Remove SHA1 from LP64 in favor of libcrypto. Keep the symbols around on LP32 for binary compatibility. Bug: 11156955 Change-Id: I379a7e0fa3092e9a70daeafcbcb2aacfc314031a
2014-06-17 01:52:27 +02:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_SRC_FILES,libc_netbsd_src_files))
Revert "Revert "Switch to NetBSD's strxfrm(3)."" This reverts commit 8793e7c7d21a0434d1b5e63364c88b2b125a3d29, and fixes the build by building upstream NetBSD source as a separate library that's then swallowed whole into libc_common. Change-Id: I6c9317d8c48b5ccaf85a7b185bc07fb31176ff97
2012-08-10 20:39:58 +02:00
include $(BUILD_STATIC_LIBRARY)
Update NDK compatlib for 4.1.2. The replacement new failures present in newer versions are present here as well, with the following new issues: XPASS std/localization/locale.categories/category.numeric/locale.nm.put/facet.num.put.members/put_long_double.pass.cpp This is from the -NaN formatting fix in bionic. We previously had this wrong, and the upstream test is also wrong. There's currently an XFAIL for Android in this test because I haven't fixed the upstream test yet. After that is done, I'll need to teach the test runner how to XFAIL older Android versions... FAIL std/localization/locale.categories/category.ctype/facet.ctype.special/facet.ctype.char.dtor/dtor.pass.cpp dtor.pass.cpp:39: int main(): assertion "globalMemCounter.checkDeleteArrayCalledEq(1)" failed Haven't investigated this one yet. http://b/19412688 Note that this also needs the libgcc link ordering to be fixed in the build system, as we'll otherwise depend on libgcc symbols from libc that may or may not have been there. The build fix can't be submitted because the proper link order causes the libgcc unwinder to be used instead of the EHABI one: http://b/18471342 Bug: 18471532 Change-Id: Icf560485a9b8f5ebbe01e4458703e62ec94df5e1
2015-02-13 01:35:09 +01:00
# ========================================================
# libc_openbsd_ndk.a - upstream OpenBSD C library code
# that can be safely included in the libc_ndk.a (doesn't
# contain any troublesome global data or constructors).
# ========================================================
#
# These files are built with the openbsd-compat.h header file
# automatically included.
include $(CLEAR_VARS)
LOCAL_SRC_FILES := $(libc_upstream_openbsd_ndk_src_files)
Allow building libc long double code with clang/llvm. This requires fix of clang's Android x86 long double size and format. That bug has been fixed in https://android-review.git.corp.google.com/#/c/146254/ Change-Id: I182c6c493085212f88c694356659f72227c8b8c7
2015-03-16 21:49:40 +01:00
LOCAL_CLANG := $(use_clang)
Update NDK compatlib for 4.1.2. The replacement new failures present in newer versions are present here as well, with the following new issues: XPASS std/localization/locale.categories/category.numeric/locale.nm.put/facet.num.put.members/put_long_double.pass.cpp This is from the -NaN formatting fix in bionic. We previously had this wrong, and the upstream test is also wrong. There's currently an XFAIL for Android in this test because I haven't fixed the upstream test yet. After that is done, I'll need to teach the test runner how to XFAIL older Android versions... FAIL std/localization/locale.categories/category.ctype/facet.ctype.special/facet.ctype.char.dtor/dtor.pass.cpp dtor.pass.cpp:39: int main(): assertion "globalMemCounter.checkDeleteArrayCalledEq(1)" failed Haven't investigated this one yet. http://b/19412688 Note that this also needs the libgcc link ordering to be fixed in the build system, as we'll otherwise depend on libgcc symbols from libc that may or may not have been there. The build fix can't be submitted because the proper link order causes the libgcc unwinder to be used instead of the EHABI one: http://b/18471342 Bug: 18471532 Change-Id: Icf560485a9b8f5ebbe01e4458703e62ec94df5e1
2015-02-13 01:35:09 +01:00
LOCAL_CFLAGS := \
$(libc_common_cflags) \
-Wno-sign-compare \
-Wno-uninitialized \
-Wno-unused-parameter \
-include openbsd-compat.h \
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
LOCAL_C_INCLUDES := $(libc_common_c_includes) \
$(LOCAL_PATH)/private \
$(LOCAL_PATH)/upstream-openbsd/android/include \
$(LOCAL_PATH)/upstream-openbsd/lib/libc/include \
$(LOCAL_PATH)/upstream-openbsd/lib/libc/gdtoa/ \
LOCAL_MODULE := libc_openbsd_ndk
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
LOCAL_CXX_STL := none
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Update NDK compatlib for 4.1.2. The replacement new failures present in newer versions are present here as well, with the following new issues: XPASS std/localization/locale.categories/category.numeric/locale.nm.put/facet.num.put.members/put_long_double.pass.cpp This is from the -NaN formatting fix in bionic. We previously had this wrong, and the upstream test is also wrong. There's currently an XFAIL for Android in this test because I haven't fixed the upstream test yet. After that is done, I'll need to teach the test runner how to XFAIL older Android versions... FAIL std/localization/locale.categories/category.ctype/facet.ctype.special/facet.ctype.char.dtor/dtor.pass.cpp dtor.pass.cpp:39: int main(): assertion "globalMemCounter.checkDeleteArrayCalledEq(1)" failed Haven't investigated this one yet. http://b/19412688 Note that this also needs the libgcc link ordering to be fixed in the build system, as we'll otherwise depend on libgcc symbols from libc that may or may not have been there. The build fix can't be submitted because the proper link order causes the libgcc unwinder to be used instead of the EHABI one: http://b/18471342 Bug: 18471532 Change-Id: Icf560485a9b8f5ebbe01e4458703e62ec94df5e1
2015-02-13 01:35:09 +01:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
include $(BUILD_STATIC_LIBRARY)
Clean up our OpenBSD usage. Also undo some of the mess where we have OpenBSD <stdio.h> but a mix of different BSD's implementations. In this first pass, I've only moved easy OpenBSD stuff. Change-Id: Iae67b02cde6dba9d8d06fedeb53efbfdac0a8cf6
2014-02-21 03:04:58 +01:00
# ========================================================
# libc_openbsd.a - upstream OpenBSD C library code
# ========================================================
#
# These files are built with the openbsd-compat.h header file
# automatically included.
include $(CLEAR_VARS)
More OpenBSD cleanup (primarily string). This patch removes the string/ and wchar/ directories. Change-Id: Ia489904bc67047e4bc79acb1f3eec21aa3fe5f0d
2014-02-25 03:00:43 +01:00
LOCAL_SRC_FILES := $(libc_upstream_openbsd_src_files)
Allow building libc long double code with clang/llvm. This requires fix of clang's Android x86 long double size and format. That bug has been fixed in https://android-review.git.corp.google.com/#/c/146254/ Change-Id: I182c6c493085212f88c694356659f72227c8b8c7
2015-03-16 21:49:40 +01:00
LOCAL_CLANG := $(use_clang)
Enable clang compilation with libc but not linker. Clang is still disabled for x86 and x86_64 long double code, for x86_64 special assembly instruction, and the linker module. BUG: 17163651 BUG: 17302991 BUG: 17403674 Change-Id: I43c5f5f0ddc0c2a31918f00b57150dadd8f3f26b
2014-09-05 00:19:52 +02:00
Clean up our OpenBSD usage. Also undo some of the mess where we have OpenBSD <stdio.h> but a mix of different BSD's implementations. In this first pass, I've only moved easy OpenBSD stuff. Change-Id: Iae67b02cde6dba9d8d06fedeb53efbfdac0a8cf6
2014-02-21 03:04:58 +01:00
LOCAL_CFLAGS := \
$(libc_common_cflags) \
Sync with upstream findfp.c. Change-Id: Ib298f46de0c4d0fd4ce0a9b456c163e8b8b06342
2014-09-24 19:27:09 +02:00
-Wno-sign-compare \
-Wno-uninitialized \
-Wno-unused-parameter \
Fix the newly-introduced warning and turn on -Werror. I cleaned up most of our warnings last week but forgot to turn on -Werror, so of course we're getting new warnings already. I've left -Werror commented out in those places where we still have warnings to deal with before we can turn on -Werror. Change-Id: Ia58ff8b8c1ada4bf81eec6f19ec1d34e133cf4b1
2014-05-17 01:50:34 +02:00
-include openbsd-compat.h \
Clean up our OpenBSD usage. Also undo some of the mess where we have OpenBSD <stdio.h> but a mix of different BSD's implementations. In this first pass, I've only moved easy OpenBSD stuff. Change-Id: Iae67b02cde6dba9d8d06fedeb53efbfdac0a8cf6
2014-02-21 03:04:58 +01:00
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
Use LOCAL_C_INCLUDES instead of LOCAL_CFLAGS for include dirs Change-Id: If61d6faae5d91c3bbe198e733e5922cd877ba353
2015-02-06 23:16:17 +01:00
LOCAL_C_INCLUDES := $(libc_common_c_includes) \
$(LOCAL_PATH)/private \
$(LOCAL_PATH)/upstream-openbsd/android/include \
$(LOCAL_PATH)/upstream-openbsd/lib/libc/include \
$(LOCAL_PATH)/upstream-openbsd/lib/libc/gdtoa/ \
Clean up our OpenBSD usage. Also undo some of the mess where we have OpenBSD <stdio.h> but a mix of different BSD's implementations. In this first pass, I've only moved easy OpenBSD stuff. Change-Id: Iae67b02cde6dba9d8d06fedeb53efbfdac0a8cf6
2014-02-21 03:04:58 +01:00
LOCAL_MODULE := libc_openbsd
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Clean up our OpenBSD usage. Also undo some of the mess where we have OpenBSD <stdio.h> but a mix of different BSD's implementations. In this first pass, I've only moved easy OpenBSD stuff. Change-Id: Iae67b02cde6dba9d8d06fedeb53efbfdac0a8cf6
2014-02-21 03:04:58 +01:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Clean up our OpenBSD usage. Also undo some of the mess where we have OpenBSD <stdio.h> but a mix of different BSD's implementations. In this first pass, I've only moved easy OpenBSD stuff. Change-Id: Iae67b02cde6dba9d8d06fedeb53efbfdac0a8cf6
2014-02-21 03:04:58 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
Re-add putw for LP32 This was missed when switching to LOCAL_SRC_FILES_EXCLUDE Change-Id: I6ea23c9eb31abe11e0ec4abfc2ee2f2a43c76ce2
2015-12-03 02:26:15 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_SRC_FILES,libc_openbsd_src_files))
Switch to LOCAL_SRC_FILES_EXCLUDE This moves the generic arm/arm64/x86 settings into the main makefiles and makes the rest of them derivatives. This better aligns with how soong handles arch/cpu variants. Also updates the Android.bp to make it consistent with the make versions. Change-Id: I5a0275d992bc657459eb6fe1697ad2336731d122
2015-10-15 23:49:45 +02:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_SRC_FILES_EXCLUDE,libc_openbsd_src_files_exclude))
Clean up our OpenBSD usage. Also undo some of the mess where we have OpenBSD <stdio.h> but a mix of different BSD's implementations. In this first pass, I've only moved easy OpenBSD stuff. Change-Id: Iae67b02cde6dba9d8d06fedeb53efbfdac0a8cf6
2014-02-21 03:04:58 +01:00
include $(BUILD_STATIC_LIBRARY)
Switch to gdtoa. This gives us a real strtold for LP64 and fixes various LP64 bugs. Bug: 13563801 Change-Id: I277858d718ee746e136b6b6308a495ba50dfa488
2014-04-11 02:48:14 +02:00
# ========================================================
# libc_gdtoa.a - upstream OpenBSD C library gdtoa code
# ========================================================
#
# These files are built with the openbsd-compat.h header file
# automatically included.
include $(CLEAR_VARS)
LOCAL_SRC_FILES_32 := $(libc_upstream_openbsd_gdtoa_src_files_32)
LOCAL_SRC_FILES_64 := $(libc_upstream_openbsd_gdtoa_src_files_64)
Allow building libc long double code with clang/llvm. This requires fix of clang's Android x86 long double size and format. That bug has been fixed in https://android-review.git.corp.google.com/#/c/146254/ Change-Id: I182c6c493085212f88c694356659f72227c8b8c7
2015-03-16 21:49:40 +01:00
LOCAL_CLANG := $(use_clang)
Enable clang compilation with libc but not linker. Clang is still disabled for x86 and x86_64 long double code, for x86_64 special assembly instruction, and the linker module. BUG: 17163651 BUG: 17302991 BUG: 17403674 Change-Id: I43c5f5f0ddc0c2a31918f00b57150dadd8f3f26b
2014-09-05 00:19:52 +02:00
Switch to gdtoa. This gives us a real strtold for LP64 and fixes various LP64 bugs. Bug: 13563801 Change-Id: I277858d718ee746e136b6b6308a495ba50dfa488
2014-04-11 02:48:14 +02:00
LOCAL_CFLAGS := \
$(libc_common_cflags) \
Fix <math.h> to quieten most of our warnings. I've reported the wcsftime bug upstream, but we really just want to use -D to ensure the buggy code isn't built. (I've also brought our strftime a bit closer to upstream now we have the right define.) I don't think upstream is likely to fix all their sign-compare and uninitialized warnings, so let's just silence them. As for libm, again upstream isn't likely to fix all their warnings, and silencing those made the ones that were our fault stand out. I've fixed our <math.h> to fix the warnings caused by our lack of definitions for the non-imprecise long-double functions. I checked the C99 standard, and all these functions are there. Change-Id: Iee8e1182c1db375058fb2c451eceb212bab47a37
2014-05-06 05:28:28 +02:00
-Wno-sign-compare -Wno-uninitialized \
Use -fvisibility=hidden to build libc_gdtoa. Bug: 12177745 Change-Id: Ia4ac71e1a3a237e764c809cc591622f8952ed642
2014-04-22 21:32:49 +02:00
-fvisibility=hidden \
Clean up and document our hacks for building BSD source. Change-Id: Ic591e22fa5b363bb68376b9f25814c0e5bd83fbf
2014-04-22 02:13:46 +02:00
-include openbsd-compat.h \
Switch to gdtoa. This gives us a real strtold for LP64 and fixes various LP64 bugs. Bug: 13563801 Change-Id: I277858d718ee746e136b6b6308a495ba50dfa488
2014-04-11 02:48:14 +02:00
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
Use LOCAL_C_INCLUDES instead of LOCAL_CFLAGS for include dirs Change-Id: If61d6faae5d91c3bbe198e733e5922cd877ba353
2015-02-06 23:16:17 +01:00
LOCAL_C_INCLUDES := $(libc_common_c_includes) \
$(LOCAL_PATH)/private \
$(LOCAL_PATH)/upstream-openbsd/android/include \
$(LOCAL_PATH)/upstream-openbsd/lib/libc/include \
Switch to gdtoa. This gives us a real strtold for LP64 and fixes various LP64 bugs. Bug: 13563801 Change-Id: I277858d718ee746e136b6b6308a495ba50dfa488
2014-04-11 02:48:14 +02:00
LOCAL_MODULE := libc_gdtoa
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Switch to gdtoa. This gives us a real strtold for LP64 and fixes various LP64 bugs. Bug: 13563801 Change-Id: I277858d718ee746e136b6b6308a495ba50dfa488
2014-04-11 02:48:14 +02:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Switch to gdtoa. This gives us a real strtold for LP64 and fixes various LP64 bugs. Bug: 13563801 Change-Id: I277858d718ee746e136b6b6308a495ba50dfa488
2014-04-11 02:48:14 +02:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
include $(BUILD_STATIC_LIBRARY)
Move setlocale(3) and the wchar stubs over to .cpp. Also separate out the C++ files so we can use -Werror on them. I'd rather wait for LOCAL_CPPFLAGS to be in AOSP, but this also lets us see which files still need to be sorted into one bucket or the other. Change-Id: I6acc1f7c043935c70a3b089f705d218b9aaaba0a
2012-10-23 02:05:27 +02:00
# ========================================================
# libc_bionic.a - home-grown C library code
# ========================================================
Fix compilation of crtbegin.c and pthread_debug.cpp. Because there was no default := for the aarch64 libc_crt_target_cflags, the += was causing libc_crt_target_cflags to be recursively-defined variable, which meant that when we were compiling crtbegin.c LOCAL_PATH would be bionic/tests/ and we'd have -Ibionic/tests/include/ and find none of our include files. Also fix linking of pthread_debug.cpp, at least in the disabled mode. The enabled mode was already broken for all architectures, and continues to be broken after this change. It's been broken for long enough that we might want to just remove it... (aarch64 is using the FSF linker where arm uses the gold linker.) Change-Id: I7db2e386694f6933db043138e6e97e5ae54d4174
2013-12-18 05:47:06 +01:00
Move setlocale(3) and the wchar stubs over to .cpp. Also separate out the C++ files so we can use -Werror on them. I'd rather wait for LOCAL_CPPFLAGS to be in AOSP, but this also lets us see which files still need to be sorted into one bucket or the other. Change-Id: I6acc1f7c043935c70a3b089f705d218b9aaaba0a
2012-10-23 02:05:27 +02:00
include $(CLEAR_VARS)
LOCAL_SRC_FILES := $(libc_bionic_src_files)
Reduce stack usage of tmpfile(3). Also ensure that none of our home-grown code uses more than 2KiB per frame. Change-Id: I8987a17d72f4b7f082bb7fa25e137c8433664c14
2014-05-13 19:14:22 +02:00
LOCAL_CFLAGS := $(libc_common_cflags) \
-Wframe-larger-than=2048 \
Use gnu++11 and gnu99 explicitly for C++/C files. Clang and gcc default to different standards, so we should be explicit about the versions we want to compile for. Change-Id: I65495a2392dd29f36373b94c616c2506173e6033
2013-10-01 02:10:53 +02:00
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
Turn on -Wold-style-cast and fix the errors. A couple of dodgy cases where we cast away const, but otherwise pretty boring. Change-Id: Ibc39ebd525377792b5911464be842121c20f03b9
2015-01-22 01:19:07 +01:00
LOCAL_CPPFLAGS := $(libc_common_cppflags) -Wold-style-cast
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_C_INCLUDES := $(libc_common_c_includes) bionic/libstdc++/include
Move setlocale(3) and the wchar stubs over to .cpp. Also separate out the C++ files so we can use -Werror on them. I'd rather wait for LOCAL_CPPFLAGS to be in AOSP, but this also lets us see which files still need to be sorted into one bucket or the other. Change-Id: I6acc1f7c043935c70a3b089f705d218b9aaaba0a
2012-10-23 02:05:27 +02:00
LOCAL_MODULE := libc_bionic
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_CLANG := $(use_clang)
Add dependencies on included makefiles. Bug: 11050594 Merge from internal master. (cherry-picked from f389284e86bbcbdb257d7388a17fde007113b0d6) Change-Id: I2b3e38329a09d26c16870906f9ed1257e2a9dbc8
2013-10-03 22:26:22 +02:00
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Move setlocale(3) and the wchar stubs over to .cpp. Also separate out the C++ files so we can use -Werror on them. I'd rather wait for LOCAL_CPPFLAGS to be in AOSP, but this also lets us see which files still need to be sorted into one bucket or the other. Change-Id: I6acc1f7c043935c70a3b089f705d218b9aaaba0a
2012-10-23 02:05:27 +02:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Move setlocale(3) and the wchar stubs over to .cpp. Also separate out the C++ files so we can use -Werror on them. I'd rather wait for LOCAL_CPPFLAGS to be in AOSP, but this also lets us see which files still need to be sorted into one bucket or the other. Change-Id: I6acc1f7c043935c70a3b089f705d218b9aaaba0a
2012-10-23 02:05:27 +02:00
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
Fix a few libc makefile mishaps. These are the problems fixed in this CL: - The ndk_cruft.cpp file was not compiled for 64 bit. - Some files where not compiled for libc_bionic.a due to a missing patch-up-arch-specific-flags call. - The libc_bionic_ndk.a patch up call was using bionic src, not the bionic_ndk_src_files. - Move mmap.cpp to the ndk src list where it was before because of the above errors. Bug: 21640784 Change-Id: I31e97bc8bf347c8e584cc17aa4b984a7d122b298
2015-06-05 04:24:07 +02:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_SRC_FILES,libc_bionic_src_files))
Switch to LOCAL_SRC_FILES_EXCLUDE This moves the generic arm/arm64/x86 settings into the main makefiles and makes the rest of them derivatives. This better aligns with how soong handles arch/cpu variants. Also updates the Android.bp to make it consistent with the make versions. Change-Id: I5a0275d992bc657459eb6fe1697ad2336731d122
2015-10-15 23:49:45 +02:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_SRC_FILES_EXCLUDE,libc_bionic_src_files_exclude))
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
include $(BUILD_STATIC_LIBRARY)
# ========================================================
# libc_bionic_ndk.a - The portions of libc_bionic that can
# be safely used in libc_ndk.a (no troublesome global data
# or constructors).
# ========================================================
include $(CLEAR_VARS)
LOCAL_SRC_FILES := $(libc_bionic_ndk_src_files)
LOCAL_CFLAGS := $(libc_common_cflags) \
-Wframe-larger-than=2048 \
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags) -Wold-style-cast
LOCAL_C_INCLUDES := $(libc_common_c_includes) bionic/libstdc++/include
LOCAL_MODULE := libc_bionic_ndk
LOCAL_CLANG := $(use_clang)
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
LOCAL_CXX_STL := none
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
Fix a few libc makefile mishaps. These are the problems fixed in this CL: - The ndk_cruft.cpp file was not compiled for 64 bit. - Some files where not compiled for libc_bionic.a due to a missing patch-up-arch-specific-flags call. - The libc_bionic_ndk.a patch up call was using bionic src, not the bionic_ndk_src_files. - Move mmap.cpp to the ndk src list where it was before because of the above errors. Bug: 21640784 Change-Id: I31e97bc8bf347c8e584cc17aa4b984a7d122b298
2015-06-05 04:24:07 +02:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_SRC_FILES,libc_bionic_ndk_src_files))
Move setlocale(3) and the wchar stubs over to .cpp. Also separate out the C++ files so we can use -Werror on them. I'd rather wait for LOCAL_CPPFLAGS to be in AOSP, but this also lets us see which files still need to be sorted into one bucket or the other. Change-Id: I6acc1f7c043935c70a3b089f705d218b9aaaba0a
2012-10-23 02:05:27 +02:00
include $(BUILD_STATIC_LIBRARY)
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
# ========================================================
# libc_pthread.a - pthreads parts that previously lived in
# libc_bionic.a. Relocated to their own library because
# they can't be included in libc_ndk.a (as they layout of
# pthread_t has changed over the years and has ABI
# compatibility issues).
# ========================================================
include $(CLEAR_VARS)
LOCAL_SRC_FILES := $(libc_pthread_src_files)
LOCAL_CFLAGS := $(libc_common_cflags) \
-Wframe-larger-than=2048 \
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags) -Wold-style-cast
LOCAL_C_INCLUDES := $(libc_common_c_includes)
LOCAL_MODULE := libc_pthread
LOCAL_CLANG := $(use_clang)
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
LOCAL_CXX_STL := none
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
include $(BUILD_STATIC_LIBRARY)
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
# ========================================================
# libc_cxa.a - Things traditionally in libstdc++
# ========================================================
include $(CLEAR_VARS)
LOCAL_SRC_FILES := $(libc_cxa_src_files)
LOCAL_CFLAGS := $(libc_common_cflags) \
-fvisibility=hidden \
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_C_INCLUDES := $(libc_common_c_includes) bionic/libstdc++/include
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
LOCAL_MODULE := libc_cxa
LOCAL_CLANG := true # GCC refuses to hide new/delete
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Fix coverage build after adding clang coverage. Static libraries are painful. Details are in the comment. I'll try to get prebuilts in to the tree so I can fix this on Monday. Until then, this isn't actually a regression because we've not had coverage available for this library until now anyway. Bug: 17574078 Change-Id: I7505c8a94007203e15a6cf192caa06004849d7d9
2015-01-25 23:27:51 +01:00
# b/17574078: Need to disable coverage until we have a prebuilt libprofile_rt.
# Since this is a static library built with clang, it needs to link
# libprofile_rt when it is linked into the final binary. Since the final binary
# is built with GCC, it won't link libprofile_rt. We can't very easily just add
# libprofile_rt to all link lines the way we've done for libgcov because
# libprofile_rt isn't prebuilt, and it would be tricky to write a rule that
# would make sure libprofile_rt is built.
LOCAL_NATIVE_COVERAGE := false
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
include $(BUILD_STATIC_LIBRARY)
Build syscall stubs in their own library. This lets us lose the auto-generated makefiles. Change-Id: I2de0c71b3b9c08f9cce8f4ff7fd7254dda008c86
2014-02-11 22:32:02 +01:00
# ========================================================
# libc_syscalls.a
# ========================================================
include $(CLEAR_VARS)
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
LOCAL_SRC_FILES_$(TARGET_ARCH) := $(call all-S-files-under,arch-$(TARGET_ARCH)/syscalls)
ifdef TARGET_2ND_ARCH
LOCAL_SRC_FILES_$(TARGET_2ND_ARCH) := $(call all-S-files-under,arch-$(TARGET_2ND_ARCH)/syscalls)
endif
Build syscall stubs in their own library. This lets us lose the auto-generated makefiles. Change-Id: I2de0c71b3b9c08f9cce8f4ff7fd7254dda008c86
2014-02-11 22:32:02 +01:00
LOCAL_MODULE := libc_syscalls
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_CLANG := $(use_clang)
Build syscall stubs in their own library. This lets us lose the auto-generated makefiles. Change-Id: I2de0c71b3b9c08f9cce8f4ff7fd7254dda008c86
2014-02-11 22:32:02 +01:00
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Build syscall stubs in their own library. This lets us lose the auto-generated makefiles. Change-Id: I2de0c71b3b9c08f9cce8f4ff7fd7254dda008c86
2014-02-11 22:32:02 +01:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Build syscall stubs in their own library. This lets us lose the auto-generated makefiles. Change-Id: I2de0c71b3b9c08f9cce8f4ff7fd7254dda008c86
2014-02-11 22:32:02 +01:00
include $(BUILD_STATIC_LIBRARY)
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
# ========================================================
# libc_aeabi.a
# This is an LP32 ARM-only library that needs to be built with -fno-builtin
# to avoid infinite recursion. For the other architectures we just build an
# empty library to keep this makefile simple.
# ========================================================
include $(CLEAR_VARS)
LOCAL_SRC_FILES_arm := arch-arm/bionic/__aeabi.c
LOCAL_MODULE := libc_aeabi
LOCAL_CLANG := $(use_clang)
LOCAL_CFLAGS := $(libc_common_cflags) -fno-builtin
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
include $(BUILD_STATIC_LIBRARY)
Add a basic NDK compatibility library. We know we can safely statically link `libm`, since it doesn't have any dependencies on the OS or the layout of a data type that has changed between releases (like `pthread_t`). We can safely statically link `libc_syscalls` because the user can check for and handle `ENOSYS`. Update `ndk_missing_symbols.py` to account for symbols that are in the compatibility library. Improve `symbols.py` to be able to pull symbols from a static library. Change-Id: Ifb0ede1e8b4a8f0f33865d9fed72fb8b4d443fbc
2015-01-22 02:50:29 +01:00
# ========================================================
# libc_ndk.a
# Compatibility library for the NDK. This library contains
# all the parts of libc that are safe to statically link.
# We can't safely statically link things that can only run
# on a certain version of the OS. Examples include
# anything that talks to netd (a large portion of the DNS
# code) and anything that is dependent on the layout of a
# data structure that has changed across releases (such as
# pthread_t).
# ========================================================
include $(CLEAR_VARS)
LOCAL_MODULE := libc_ndk
Parameterize use of clang in libc_ndk. We still have issues with clang coverage in static libraries, so we need to make sure we follow suit with the rest of libc for now. Bug: 17574078 Change-Id: I2ab58a84b1caa0d8d08415d240c35adec5b1e150
2015-02-19 02:37:52 +01:00
LOCAL_CLANG := $(use_clang)
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
LOCAL_ASFLAGS := $(LOCAL_CFLAGS)
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CFLAGS := $(libc_common_cflags) -fvisibility=hidden -O0
LOCAL_CPPFLAGS := $(libc_common_cppflags)
LOCAL_C_INCLUDES := $(libc_common_c_includes)
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
LOCAL_SYSTEM_SHARED_LIBRARIES :=
LOCAL_SRC_FILES := \
$(libc_common_src_files) \
$(libc_arch_dynamic_src_files) \
$(libc_ndk_stub_src_files) \
Malloc debug rewrite. The major components of the rewrite: - Completely remove the qemu shared library code. Nobody was using it and it appears to have broken at some point. - Adds the ability to enable/disable different options independently. - Adds a new option that can enable the backtrace on alloc/free when a process gets a specific signal. - Adds a new way to enable malloc debug. If a special property is set, and the process has an environment variable set, then debug malloc will be enabled. This allows something that might be a derivative of app_process to be started with an environment variable being enabled. - get_malloc_leak_info() used to return one element for each pointer that had the exact same backtrace. The new version returns information for every one of the pointers with same backtrace. It turns out ddms already automatically coalesces these, so the old method simply hid the fact that there where multiple pointers with the same amount of backtrace. - Moved all of the malloc debug specific code into the library. Nothing related to the malloc debug data structures remains in libc. - Removed the calls to the debug malloc cleanup routine. Instead, I added an atexit call with the debug malloc cleanup routine. This gets around most problems related to the timing of doing the cleanup. The new properties and environment variables: libc.debug.malloc.options Set by option name (such as "backtrace"). Setting this to a bad value will cause a usage statement to be printed to the log. libc.debug.malloc.program Same as before. If this is set, then only the program named will be launched with malloc debug enabled. This is not a complete match, but if any part of the property is in the program name, malloc debug is enabled. libc.debug.malloc.env_enabled If set, then malloc debug is only enabled if the running process has the environment variable LIBC_DEBUG_MALLOC_ENABLE set. Bug: 19145921 Change-Id: I7b0e58cc85cc6d4118173fe1f8627a391b64c0d7
2015-11-17 02:30:32 +01:00
bionic/malloc_common.cpp \
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
LOCAL_SRC_FILES_arm += \
arch-common/bionic/crtbegin_so.c \
arch-arm/bionic/atexit_legacy.c \
arch-common/bionic/crtend_so.S \
Add a basic NDK compatibility library. We know we can safely statically link `libm`, since it doesn't have any dependencies on the OS or the layout of a data type that has changed between releases (like `pthread_t`). We can safely statically link `libc_syscalls` because the user can check for and handle `ENOSYS`. Update `ndk_missing_symbols.py` to account for symbols that are in the compatibility library. Improve `symbols.py` to be able to pull symbols from a static library. Change-Id: Ifb0ede1e8b4a8f0f33865d9fed72fb8b4d443fbc
2015-01-22 02:50:29 +01:00
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
LOCAL_CFLAGS := $(libc_common_cflags) \
-DLIBC_STATIC \
LOCAL_WHOLE_STATIC_LIBRARIES := \
libc_bionic_ndk \
libc_cxa \
libc_freebsd \
libc_gdtoa \
libc_malloc \
libc_netbsd \
Update NDK compatlib for 4.1.2. The replacement new failures present in newer versions are present here as well, with the following new issues: XPASS std/localization/locale.categories/category.numeric/locale.nm.put/facet.num.put.members/put_long_double.pass.cpp This is from the -NaN formatting fix in bionic. We previously had this wrong, and the upstream test is also wrong. There's currently an XFAIL for Android in this test because I haven't fixed the upstream test yet. After that is done, I'll need to teach the test runner how to XFAIL older Android versions... FAIL std/localization/locale.categories/category.ctype/facet.ctype.special/facet.ctype.char.dtor/dtor.pass.cpp dtor.pass.cpp:39: int main(): assertion "globalMemCounter.checkDeleteArrayCalledEq(1)" failed Haven't investigated this one yet. http://b/19412688 Note that this also needs the libgcc link ordering to be fixed in the build system, as we'll otherwise depend on libgcc symbols from libc that may or may not have been there. The build fix can't be submitted because the proper link order causes the libgcc unwinder to be used instead of the EHABI one: http://b/18471342 Bug: 18471532 Change-Id: Icf560485a9b8f5ebbe01e4458703e62ec94df5e1
2015-02-13 01:35:09 +01:00
libc_openbsd_ndk \
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
libc_stack_protector \
libc_syscalls \
libc_tzcode \
libm \
Remove dlmalloc. Bug: http://b/17207577 Change-Id: Ie009badca6deb1f91b27a4340b70cdd6bedff893
2016-01-07 02:57:44 +01:00
libjemalloc \
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
LOCAL_WHOLE_STATIC_LIBRARIES_arm := libc_aeabi
LOCAL_CXX_STL := none
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
$(eval $(call patch-up-arch-specific-flags,LOCAL_SRC_FILES,libc_common_src_files))
$(eval $(call patch-up-arch-specific-flags,LOCAL_SRC_FILES,libc_arch_dynamic_src_files))
$(eval $(call patch-up-arch-specific-flags,LOCAL_ASFLAGS,LOCAL_CFLAGS))
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Add a basic NDK compatibility library. We know we can safely statically link `libm`, since it doesn't have any dependencies on the OS or the layout of a data type that has changed between releases (like `pthread_t`). We can safely statically link `libc_syscalls` because the user can check for and handle `ENOSYS`. Update `ndk_missing_symbols.py` to account for symbols that are in the compatibility library. Improve `symbols.py` to be able to pull symbols from a static library. Change-Id: Ifb0ede1e8b4a8f0f33865d9fed72fb8b4d443fbc
2015-01-22 02:50:29 +01:00
include $(BUILD_STATIC_LIBRARY)
libc: Cleanup the libc makefile. Removes a lot of redundant code, and reorganizes the file. We first setup all the comomn variables we may need, and then define the actual libraries that need to be generated. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
# ========================================================
# libc_common.a
# ========================================================
Revert "Revert "Switch to NetBSD's strxfrm(3)."" This reverts commit 8793e7c7d21a0434d1b5e63364c88b2b125a3d29, and fixes the build by building upstream NetBSD source as a separate library that's then swallowed whole into libc_common. Change-Id: I6c9317d8c48b5ccaf85a7b185bc07fb31176ff97
2012-08-10 20:39:58 +02:00
libc: Cleanup the libc makefile. Removes a lot of redundant code, and reorganizes the file. We first setup all the comomn variables we may need, and then define the actual libraries that need to be generated. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
include $(CLEAR_VARS)
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
libc: Cleanup the libc makefile. Removes a lot of redundant code, and reorganizes the file. We first setup all the comomn variables we may need, and then define the actual libraries that need to be generated. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
LOCAL_SRC_FILES := $(libc_common_src_files)
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
LOCAL_CFLAGS := $(libc_common_cflags)
Use gnu++11 and gnu99 explicitly for C++/C files. Clang and gcc default to different standards, so we should be explicit about the versions we want to compile for. Change-Id: I65495a2392dd29f36373b94c616c2506173e6033
2013-10-01 02:10:53 +02:00
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
libc: Cleanup the libc makefile. Removes a lot of redundant code, and reorganizes the file. We first setup all the comomn variables we may need, and then define the actual libraries that need to be generated. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
LOCAL_C_INCLUDES := $(libc_common_c_includes)
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
LOCAL_MODULE := libc_common
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_CLANG := $(use_clang)
Add dependencies on included makefiles. Bug: 11050594 Merge from internal master. (cherry-picked from f389284e86bbcbdb257d7388a17fde007113b0d6) Change-Id: I2b3e38329a09d26c16870906f9ed1257e2a9dbc8
2013-10-03 22:26:22 +02:00
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Upgrade to tzcode2013d. Well, kinda... localtime.c still contains a bunch of Android-specific hacks, as does strftime.c. But the other files are now exactly the same as upstream. This catches up with several years of bug fixes, and fixes most of the compiler warnings that were in this code. (Just two remain.) Bug: 1744909 Change-Id: I2ddfecb6fd408c847397c17afb0fff859e27feef
2013-07-13 02:31:11 +02:00
LOCAL_WHOLE_STATIC_LIBRARIES := \
libc_bionic \
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
libc_bionic_ndk \
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
libc_cxa \
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
libc_dns \
Upgrade to tzcode2013d. Well, kinda... localtime.c still contains a bunch of Android-specific hacks, as does strftime.c. But the other files are now exactly the same as upstream. This catches up with several years of bug fixes, and fixes most of the compiler warnings that were in this code. (Just two remain.) Bug: 1744909 Change-Id: I2ddfecb6fd408c847397c17afb0fff859e27feef
2013-07-13 02:31:11 +02:00
libc_freebsd \
Switch to gdtoa. This gives us a real strtold for LP64 and fixes various LP64 bugs. Bug: 13563801 Change-Id: I277858d718ee746e136b6b6308a495ba50dfa488
2014-04-11 02:48:14 +02:00
libc_gdtoa \
Hide dlmalloc symbols. We need to leave dlmalloc_trim and dlmalloc_inspect_all exposed for the VM, but if we're seriously looking at other malloc implementations, that's something we're going to have to fix. Bug: 11156955 Change-Id: If85156c280044f1616c09a3c50ba674aaf0e8d3a
2014-06-04 03:47:17 +02:00
libc_malloc \
Upgrade to tzcode2013d. Well, kinda... localtime.c still contains a bunch of Android-specific hacks, as does strftime.c. But the other files are now exactly the same as upstream. This catches up with several years of bug fixes, and fixes most of the compiler warnings that were in this code. (Just two remain.) Bug: 1744909 Change-Id: I2ddfecb6fd408c847397c17afb0fff859e27feef
2013-07-13 02:31:11 +02:00
libc_netbsd \
Clean up our OpenBSD usage. Also undo some of the mess where we have OpenBSD <stdio.h> but a mix of different BSD's implementations. In this first pass, I've only moved easy OpenBSD stuff. Change-Id: Iae67b02cde6dba9d8d06fedeb53efbfdac0a8cf6
2014-02-21 03:04:58 +01:00
libc_openbsd \
Update NDK compatlib for 4.1.2. The replacement new failures present in newer versions are present here as well, with the following new issues: XPASS std/localization/locale.categories/category.numeric/locale.nm.put/facet.num.put.members/put_long_double.pass.cpp This is from the -NaN formatting fix in bionic. We previously had this wrong, and the upstream test is also wrong. There's currently an XFAIL for Android in this test because I haven't fixed the upstream test yet. After that is done, I'll need to teach the test runner how to XFAIL older Android versions... FAIL std/localization/locale.categories/category.ctype/facet.ctype.special/facet.ctype.char.dtor/dtor.pass.cpp dtor.pass.cpp:39: int main(): assertion "globalMemCounter.checkDeleteArrayCalledEq(1)" failed Haven't investigated this one yet. http://b/19412688 Note that this also needs the libgcc link ordering to be fixed in the build system, as we'll otherwise depend on libgcc symbols from libc that may or may not have been there. The build fix can't be submitted because the proper link order causes the libgcc unwinder to be used instead of the EHABI one: http://b/18471342 Bug: 18471532 Change-Id: Icf560485a9b8f5ebbe01e4458703e62ec94df5e1
2015-02-13 01:35:09 +01:00
libc_openbsd_ndk \
Update the NDK compatlib for 4.4.4 and ToT. Now passes all libc++ tests for these targets, with the exception of the usual failing replacement new tests since libc uses new/delete for things. I don't know if we can ever really fix these. Bug: 18471532 Change-Id: Ibc0a15f26b0e4613249b5e15ecf3cf80e523467c
2015-02-10 02:45:22 +01:00
libc_pthread \
Rename the stack protector intermediate library. libbionic_ssp already confused at least one person, and characters in filenames are cheap, so let's just call this library what it is. Change-Id: I69ab950bf52fa4d267a6891efb49b5e177efc0c4
2014-03-11 00:15:11 +01:00
libc_stack_protector \
Build syscall stubs in their own library. This lets us lose the auto-generated makefiles. Change-Id: I2de0c71b3b9c08f9cce8f4ff7fd7254dda008c86
2014-02-11 22:32:02 +01:00
libc_syscalls \
Fix x86_64 build, clean up intermediate libraries. The x86_64 build was failing because clone.S had a call to __thread_entry which was being added to a different intermediate .a on the way to making libc.so, and the linker couldn't guarantee statically that such a relocation would be possible. ld: error: out/target/product/generic_x86_64/obj/STATIC_LIBRARIES/libc_common_intermediates/libc_common.a(clone.o): requires dynamic R_X86_64_PC32 reloc against '__thread_entry' which may overflow at runtime; recompile with -fPIC This patch addresses that by ensuring that the caller and callee end up in the same intermediate .a. While I'm here, I've tried to clean up some of the mess that led to this situation too. In particular, this removes libc/private/ from the default include path (except for the DNS code), and splits out the DNS code into its own library (since it's a weird special case of upstream NetBSD code that's diverged so heavily it's unlikely ever to get back in sync). There's more cleanup of the DNS situation possible, but this is definitely a step in the right direction, and it's more than enough to get x86_64 building cleanly. Change-Id: I00425a7245b7a2573df16cc38798187d0729e7c4
2013-10-10 00:50:50 +02:00
libc_tzcode \
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_WHOLE_STATIC_LIBRARIES_arm := libc_aeabi
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Split out LOCAL_ASFLAGS. Bug: 6967574 Change-Id: I5be50d582c1cb8075039b06292cf0745a996c3c4
2012-08-11 23:23:19 +02:00
# TODO: split out the asflags.
LOCAL_ASFLAGS := $(LOCAL_CFLAGS)
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Split out LOCAL_ASFLAGS. Bug: 6967574 Change-Id: I5be50d582c1cb8075039b06292cf0745a996c3c4
2012-08-11 23:23:19 +02:00
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
$(eval $(call patch-up-arch-specific-flags,LOCAL_SRC_FILES,libc_common_src_files))
$(eval $(call patch-up-arch-specific-flags,LOCAL_ASFLAGS,LOCAL_CFLAGS))
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
include $(BUILD_STATIC_LIBRARY)
libc: Add an intermediate version of the static libc without malloc For now, this will only used by the dynamic loader to ensure that malloc and friends are never called. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
# ========================================================
# libc_nomalloc.a
# ========================================================
#
# This is a version of the static C library that does not
Everyone has a TLS register. Change-Id: Id7cdf67087aa7d5074c9c59b7e595bc391d9f146
2013-02-13 06:56:42 +01:00
# include malloc. It's useful in situations when the user wants
# to provide their own malloc implementation, or wants to
Fix comment typo. Change-Id: I1909f724826b531daf14fec4853e40d578fc0f59
2014-05-16 03:12:11 +02:00
# explicitly disallow the use of malloc, such as in the
# dynamic linker.
libc: Add an intermediate version of the static libc without malloc For now, this will only used by the dynamic loader to ensure that malloc and friends are never called. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
include $(CLEAR_VARS)
LOCAL_SRC_FILES := \
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
$(libc_arch_static_src_files) \
libc: Add an intermediate version of the static libc without malloc For now, this will only used by the dynamic loader to ensure that malloc and friends are never called. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
LOCAL_C_INCLUDES := $(libc_common_c_includes)
Fix __get_tls() in static C library to use kernel helpers. This is needed to fix gdbserver's handling of threaded programs, among other things. Change-Id: I823387c602cef9891532da946a01db14be780ab0
2010-08-27 17:19:19 +02:00
LOCAL_CFLAGS := $(libc_common_cflags) \
Fix the newly-introduced warning and turn on -Werror. I cleaned up most of our warnings last week but forgot to turn on -Werror, so of course we're getting new warnings already. I've left -Werror commented out in those places where we still have warnings to deal with before we can turn on -Werror. Change-Id: Ia58ff8b8c1ada4bf81eec6f19ec1d34e133cf4b1
2014-05-17 01:50:34 +02:00
-DLIBC_STATIC \
Use gnu++11 and gnu99 explicitly for C++/C files. Clang and gcc default to different standards, so we should be explicit about the versions we want to compile for. Change-Id: I65495a2392dd29f36373b94c616c2506173e6033
2013-10-01 02:10:53 +02:00
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
libc: Add an intermediate version of the static libc without malloc For now, this will only used by the dynamic loader to ensure that malloc and friends are never called. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
LOCAL_MODULE := libc_nomalloc
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_CLANG := $(use_clang)
Add dependencies on included makefiles. Bug: 11050594 Merge from internal master. (cherry-picked from f389284e86bbcbdb257d7388a17fde007113b0d6) Change-Id: I2b3e38329a09d26c16870906f9ed1257e2a9dbc8
2013-10-03 22:26:22 +02:00
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Defend against -fstack-protector in libc startup. Exactly which functions get a stack protector is up to the compiler, so let's separate the code that sets up the environment stack protection requires and explicitly build it with -fno-stack-protector. Bug: http://b/26276517 Change-Id: I8719e23ead1f1e81715c32c1335da868f68369b5
2016-01-07 04:51:43 +01:00
LOCAL_WHOLE_STATIC_LIBRARIES := libc_common libc_init_static
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
libc: Add an intermediate version of the static libc without malloc For now, this will only used by the dynamic loader to ensure that malloc and friends are never called. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
libc: Add an intermediate version of the static libc without malloc For now, this will only used by the dynamic loader to ensure that malloc and friends are never called. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
$(eval $(call patch-up-arch-specific-flags,LOCAL_SRC_FILES,libc_arch_static_src_files))
libc: Add an intermediate version of the static libc without malloc For now, this will only used by the dynamic loader to ensure that malloc and friends are never called. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
include $(BUILD_STATIC_LIBRARY)
Hide dlmalloc symbols. We need to leave dlmalloc_trim and dlmalloc_inspect_all exposed for the VM, but if we're seriously looking at other malloc implementations, that's something we're going to have to fix. Bug: 11156955 Change-Id: If85156c280044f1616c09a3c50ba674aaf0e8d3a
2014-06-04 03:47:17 +02:00
# ========================================================
# libc_malloc.a: the _prefixed_ malloc functions (like dlcalloc).
# ========================================================
include $(CLEAR_VARS)
LOCAL_SRC_FILES := $(libc_malloc_src)
LOCAL_CFLAGS := $(libc_common_cflags) \
-fvisibility=hidden \
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
LOCAL_C_INCLUDES := $(libc_common_c_includes)
LOCAL_MODULE := libc_malloc
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_CLANG := $(use_clang)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Hide dlmalloc symbols. We need to leave dlmalloc_trim and dlmalloc_inspect_all exposed for the VM, but if we're seriously looking at other malloc implementations, that's something we're going to have to fix. Bug: 11156955 Change-Id: If85156c280044f1616c09a3c50ba674aaf0e8d3a
2014-06-04 03:47:17 +02:00
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
include $(BUILD_STATIC_LIBRARY)
libc: Cleanup the libc makefile. Removes a lot of redundant code, and reorganizes the file. We first setup all the comomn variables we may need, and then define the actual libraries that need to be generated. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
# ========================================================
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
# libc.a
# ========================================================
include $(CLEAR_VARS)
LOCAL_SRC_FILES := \
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
$(libc_arch_static_src_files) \
Malloc debug rewrite. The major components of the rewrite: - Completely remove the qemu shared library code. Nobody was using it and it appears to have broken at some point. - Adds the ability to enable/disable different options independently. - Adds a new option that can enable the backtrace on alloc/free when a process gets a specific signal. - Adds a new way to enable malloc debug. If a special property is set, and the process has an environment variable set, then debug malloc will be enabled. This allows something that might be a derivative of app_process to be started with an environment variable being enabled. - get_malloc_leak_info() used to return one element for each pointer that had the exact same backtrace. The new version returns information for every one of the pointers with same backtrace. It turns out ddms already automatically coalesces these, so the old method simply hid the fact that there where multiple pointers with the same amount of backtrace. - Moved all of the malloc debug specific code into the library. Nothing related to the malloc debug data structures remains in libc. - Removed the calls to the debug malloc cleanup routine. Instead, I added an atexit call with the debug malloc cleanup routine. This gets around most problems related to the timing of doing the cleanup. The new properties and environment variables: libc.debug.malloc.options Set by option name (such as "backtrace"). Setting this to a bad value will cause a usage statement to be printed to the log. libc.debug.malloc.program Same as before. If this is set, then only the program named will be launched with malloc debug enabled. This is not a complete match, but if any part of the property is in the program name, malloc debug is enabled. libc.debug.malloc.env_enabled If set, then malloc debug is only enabled if the running process has the environment variable LIBC_DEBUG_MALLOC_ENABLE set. Bug: 19145921 Change-Id: I7b0e58cc85cc6d4118173fe1f8627a391b64c0d7
2015-11-17 02:30:32 +01:00
bionic/malloc_common.cpp \
bionic/libc_init_static.cpp \
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
Split libc_debug.so into two .so modules loaded on demand from libc.so This change is intended to eliminate need to replace libc.so with libc_debug.so in order to enablememory allocation debugging. This is also the first step towards implementing extended memoryallocation debugging using emulator's capabilities in monitoring memory access.
2009-11-17 23:13:38 +01:00
LOCAL_CFLAGS := $(libc_common_cflags) \
Fix the newly-introduced warning and turn on -Werror. I cleaned up most of our warnings last week but forgot to turn on -Werror, so of course we're getting new warnings already. I've left -Werror commented out in those places where we still have warnings to deal with before we can turn on -Werror. Change-Id: Ia58ff8b8c1ada4bf81eec6f19ec1d34e133cf4b1
2014-05-17 01:50:34 +02:00
-DLIBC_STATIC \
Use gnu++11 and gnu99 explicitly for C++/C files. Clang and gcc default to different standards, so we should be explicit about the versions we want to compile for. Change-Id: I65495a2392dd29f36373b94c616c2506173e6033
2013-10-01 02:10:53 +02:00
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
LOCAL_C_INCLUDES := $(libc_common_c_includes)
libc: Cleanup the libc makefile. Removes a lot of redundant code, and reorganizes the file. We first setup all the comomn variables we may need, and then define the actual libraries that need to be generated. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
LOCAL_MODULE := libc
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_CLANG := $(use_clang)
Add dependencies on included makefiles. Bug: 11050594 Merge from internal master. (cherry-picked from f389284e86bbcbdb257d7388a17fde007113b0d6) Change-Id: I2b3e38329a09d26c16870906f9ed1257e2a9dbc8
2013-10-03 22:26:22 +02:00
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
Remove dlmalloc. Bug: http://b/17207577 Change-Id: Ie009badca6deb1f91b27a4340b70cdd6bedff893
2016-01-07 02:57:44 +01:00
LOCAL_WHOLE_STATIC_LIBRARIES := libc_common libc_init_static libjemalloc
Remove jemalloc.a from libc_nomalloc.a Change-Id: I86edc1a6cf3a26c46e6daef2c859459c1b0f29af
2015-04-23 21:24:43 +02:00
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
$(eval $(call patch-up-arch-specific-flags,LOCAL_SRC_FILES,libc_arch_static_src_files))
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
include $(BUILD_STATIC_LIBRARY)
libc: Cleanup the libc makefile. Removes a lot of redundant code, and reorganizes the file. We first setup all the comomn variables we may need, and then define the actual libraries that need to be generated. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
# ========================================================
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
# libc.so
# ========================================================
include $(CLEAR_VARS)
Switch to upstream OpenBSD mktemp.c. Almost all of our stdio is actually OpenBSD, so although this isn't really a core part of stdio (it doesn't touch struct FILE, for example) it probably makes sense for it to come from the same upstream. My actual motivation though is that it's the only FreeBSD file we have compiler warnings from. This patch moves us over to -Werror by default, with only the DNS code having -Wno-error. Change-Id: Id244a5b445cba41b0a1ca30298ca7b1ed177810c
2014-07-02 20:22:26 +02:00
LOCAL_CFLAGS := $(libc_common_cflags)
Use gnu++11 and gnu99 explicitly for C++/C files. Clang and gcc default to different standards, so we should be explicit about the versions we want to compile for. Change-Id: I65495a2392dd29f36373b94c616c2506173e6033
2013-10-01 02:10:53 +02:00
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
Generate sysv hash for libc, libstdc++ and libm Bug: 19059885 Change-Id: I35e2a9cd0bb3914d9d0c82d163bfaf0fff844667
2015-03-10 03:38:56 +01:00
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
LOCAL_C_INCLUDES := $(libc_common_c_includes)
LOCAL_SRC_FILES := \
Unregister pthread_atfork handlers on dlclose() Bug: http://b/20339788 Change-Id: I874c87faa377645fa9e0752f4fc166d81fd9ef7e
2014-11-21 05:47:02 +01:00
arch-common/bionic/crtbegin_so.c \
arch-common/bionic/crtbrand.S \
Fix compilation of crtbegin.c and pthread_debug.cpp. Because there was no default := for the aarch64 libc_crt_target_cflags, the += was causing libc_crt_target_cflags to be recursively-defined variable, which meant that when we were compiling crtbegin.c LOCAL_PATH would be bionic/tests/ and we'd have -Ibionic/tests/include/ and find none of our include files. Also fix linking of pthread_debug.cpp, at least in the disabled mode. The enabled mode was already broken for all architectures, and continues to be broken after this change. It's been broken for long enough that we might want to just remove it... (aarch64 is using the FSF linker where arm uses the gold linker.) Change-Id: I7db2e386694f6933db043138e6e97e5ae54d4174
2013-12-18 05:47:06 +01:00
$(libc_arch_dynamic_src_files) \
Malloc debug rewrite. The major components of the rewrite: - Completely remove the qemu shared library code. Nobody was using it and it appears to have broken at some point. - Adds the ability to enable/disable different options independently. - Adds a new option that can enable the backtrace on alloc/free when a process gets a specific signal. - Adds a new way to enable malloc debug. If a special property is set, and the process has an environment variable set, then debug malloc will be enabled. This allows something that might be a derivative of app_process to be started with an environment variable being enabled. - get_malloc_leak_info() used to return one element for each pointer that had the exact same backtrace. The new version returns information for every one of the pointers with same backtrace. It turns out ddms already automatically coalesces these, so the old method simply hid the fact that there where multiple pointers with the same amount of backtrace. - Moved all of the malloc debug specific code into the library. Nothing related to the malloc debug data structures remains in libc. - Removed the calls to the debug malloc cleanup routine. Instead, I added an atexit call with the debug malloc cleanup routine. This gets around most problems related to the timing of doing the cleanup. The new properties and environment variables: libc.debug.malloc.options Set by option name (such as "backtrace"). Setting this to a bad value will cause a usage statement to be printed to the log. libc.debug.malloc.program Same as before. If this is set, then only the program named will be launched with malloc debug enabled. This is not a complete match, but if any part of the property is in the program name, malloc debug is enabled. libc.debug.malloc.env_enabled If set, then malloc debug is only enabled if the running process has the environment variable LIBC_DEBUG_MALLOC_ENABLE set. Bug: 19145921 Change-Id: I7b0e58cc85cc6d4118173fe1f8627a391b64c0d7
2015-11-17 02:30:32 +01:00
bionic/malloc_common.cpp \
Fix compilation of crtbegin.c and pthread_debug.cpp. Because there was no default := for the aarch64 libc_crt_target_cflags, the += was causing libc_crt_target_cflags to be recursively-defined variable, which meant that when we were compiling crtbegin.c LOCAL_PATH would be bionic/tests/ and we'd have -Ibionic/tests/include/ and find none of our include files. Also fix linking of pthread_debug.cpp, at least in the disabled mode. The enabled mode was already broken for all architectures, and continues to be broken after this change. It's been broken for long enough that we might want to just remove it... (aarch64 is using the FSF linker where arm uses the gold linker.) Change-Id: I7db2e386694f6933db043138e6e97e5ae54d4174
2013-12-18 05:47:06 +01:00
bionic/libc_init_dynamic.cpp \
Fix build. Change-Id: I33293d8bc62cbb22e23a704c4242e7e9d3fce7c5
2014-05-14 01:30:12 +02:00
bionic/NetdClient.cpp \
Unregister pthread_atfork handlers on dlclose() Bug: http://b/20339788 Change-Id: I874c87faa377645fa9e0752f4fc166d81fd9ef7e
2014-11-21 05:47:02 +01:00
arch-common/bionic/crtend_so.S \
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
Switch to gdtoa. This gives us a real strtold for LP64 and fixes various LP64 bugs. Bug: 13563801 Change-Id: I277858d718ee746e136b6b6308a495ba50dfa488
2014-04-11 02:48:14 +02:00
LOCAL_MODULE := libc
Enable building libc with clang. Bug: 16218205 Change-Id: I024de2044fe016412c7bead22c264e96be4cb39c
2014-07-11 01:57:27 +02:00
LOCAL_CLANG := $(use_clang)
Make bionic's dependency on the tzdata explicit. Also stop building the obsolete three files, now bionic and libcore both use the new single file. Bug: 7012465 Change-Id: I1b9b49af7382c57b6cb8820c2275e6d4044b2bb6
2012-10-18 20:28:56 +02:00
LOCAL_REQUIRED_MODULES := tzdata
Don't export new/delete from libc shared libraries. This used to be handled by -fvisibility=hidden on libc_cxa, but that was broken by the resolution of https://llvm.org/PR22419 (introduced to Android in today's clang update). Now we just use a version script that prevents these from being re-exported from our shared libraries. Change-Id: Ib290e1d0d7426e09ad17a91178162fff6dbdcfa9
2015-04-02 01:26:54 +02:00
LOCAL_ADDITIONAL_DEPENDENCIES := \
$(libc_common_additional_dependencies) \
Generate per-architecture version scripts. Bug: http://b/24767418 Change-Id: I23e49d8d30c0def3215ffbd99d09f207d9483624
2015-10-28 23:14:13 +01:00
$(LOCAL_PATH)/libc.arm.map \
$(LOCAL_PATH)/libc.arm64.map \
$(LOCAL_PATH)/libc.mips.map \
$(LOCAL_PATH)/libc.mips64.map \
$(LOCAL_PATH)/libc.x86.map \
$(LOCAL_PATH)/libc.x86_64.map \
Generate libc version-script for brillo Brillo doesn't use the ndk cruft, so we need separate set of version scripts. Added new "nobrillo" tag to mark such symbols in *.map.txt files. Bug: http://b/26164862 Change-Id: Iaee1b7119f75b68c2971679fc32817e6df29fd94
2015-12-14 23:07:52 +01:00
$(LOCAL_PATH)/libc.arm.brillo.map \
$(LOCAL_PATH)/libc.mips.brillo.map \
$(LOCAL_PATH)/libc.x86.brillo.map \
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
Keep the symbol table in the shared libc. In order to allow the unwinder code to have meaningful names for libc functions, leave the symbol table. This results in the libc.so getting to be about ~130K larger on all arm platforms and about ~70K larger on mips/x86 platforms. Bug: 12958251 Change-Id: I6b3a97e4824142faf5de46aeabf7c1dfb98a8cc6
2014-03-21 19:11:38 +01:00
# Leave the symbols in the shared library so that stack unwinders can produce
# meaningful name resolution.
LOCAL_STRIP_MODULE := keep_symbols
Do not pack relocations for libc.so Bug: http://b/20645321 Bug: http://b/20655855 (cherry picked from commit 452742d2bf4247106055696c11bc4e59ca630f11) Change-Id: Ic9125cc1bc4c9ba9eb20d030de72e3ce1fb86fa6
2015-04-28 23:52:47 +02:00
# Do not pack libc.so relocations; see http://b/20645321 for details.
LOCAL_PACK_MODULE_RELOCATIONS := false
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
# WARNING: The only library libc.so should depend on is libdl.so! If you add other libraries,
# make sure to add -Wl,--exclude-libs=libgcc.a to the LOCAL_LDFLAGS for those libraries. This
# ensures that symbols that are pulled into those new libraries from libgcc.a are not declared
# external; if that were the case, then libc would not pull those symbols from libgcc.a as it
# should, instead relying on the external symbols from the dependent libraries. That would
Switch to gdtoa. This gives us a real strtold for LP64 and fixes various LP64 bugs. Bug: 13563801 Change-Id: I277858d718ee746e136b6b6308a495ba50dfa488
2014-04-11 02:48:14 +02:00
# create a "cloaked" dependency on libgcc.a in libc though the libraries, which is not what
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
# you wanted!
LOCAL_SHARED_LIBRARIES := libdl
Remove dlmalloc. Bug: http://b/17207577 Change-Id: Ie009badca6deb1f91b27a4340b70cdd6bedff893
2016-01-07 02:57:44 +01:00
LOCAL_WHOLE_STATIC_LIBRARIES := libc_common libjemalloc
Remove jemalloc.a from libc_nomalloc.a Change-Id: I86edc1a6cf3a26c46e6daef2c859459c1b0f29af
2015-04-23 21:24:43 +02:00
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
Fix LDFLAGS for arm and x86 Change-Id: I5366f4499cb4da3a51b8484ab3e4ff785f0888ab
2015-10-29 01:24:43 +01:00
# TODO: This is to work around b/24465209. Remove after root cause is fixed
LOCAL_LDFLAGS_arm := -Wl,--hash-style=both
LOCAL_LDFLAGS_x86 := -Wl,--hash-style=both
Don't export new/delete from libc shared libraries. This used to be handled by -fvisibility=hidden on libc_cxa, but that was broken by the resolution of https://llvm.org/PR22419 (introduced to Android in today's clang update). Now we just use a version script that prevents these from being re-exported from our shared libraries. Change-Id: Ib290e1d0d7426e09ad17a91178162fff6dbdcfa9
2015-04-02 01:26:54 +02:00
# Don't re-export new/delete and friends, even if the compiler really wants to.
Generate libc version-script for brillo Brillo doesn't use the ndk cruft, so we need separate set of version scripts. Added new "nobrillo" tag to mark such symbols in *.map.txt files. Bug: http://b/26164862 Change-Id: Iaee1b7119f75b68c2971679fc32817e6df29fd94
2015-12-14 23:07:52 +01:00
ifdef BRILLO
LOCAL_LDFLAGS_arm += -Wl,--version-script,$(LOCAL_PATH)/libc.arm.brillo.map
LOCAL_LDFLAGS_mips += -Wl,--version-script,$(LOCAL_PATH)/libc.mips.brillo.map
LOCAL_LDFLAGS_x86 += -Wl,--version-script,$(LOCAL_PATH)/libc.x86.brillo.map
else
Fix LDFLAGS for arm and x86 Change-Id: I5366f4499cb4da3a51b8484ab3e4ff785f0888ab
2015-10-29 01:24:43 +01:00
LOCAL_LDFLAGS_arm += -Wl,--version-script,$(LOCAL_PATH)/libc.arm.map
LOCAL_LDFLAGS_mips += -Wl,--version-script,$(LOCAL_PATH)/libc.mips.map
LOCAL_LDFLAGS_x86 += -Wl,--version-script,$(LOCAL_PATH)/libc.x86.map
Revert "Revert "Remove remaining ndk_cruft from LP64 platforms"" This reverts commit 23af25b747b5e7b0ca513a87b32f1f6a3cb99698. Bug: http://b/26164862 Change-Id: I201bf49530f33f8ba6a1a35b79723afb9e84c7a5
2015-12-16 20:02:04 +01:00
endif
Generate libc version-script for brillo Brillo doesn't use the ndk cruft, so we need separate set of version scripts. Added new "nobrillo" tag to mark such symbols in *.map.txt files. Bug: http://b/26164862 Change-Id: Iaee1b7119f75b68c2971679fc32817e6df29fd94
2015-12-14 23:07:52 +01:00
LOCAL_LDFLAGS_arm64 += -Wl,--version-script,$(LOCAL_PATH)/libc.arm64.map
LOCAL_LDFLAGS_mips64 += -Wl,--version-script,$(LOCAL_PATH)/libc.mips64.map
Fix LDFLAGS for arm and x86 Change-Id: I5366f4499cb4da3a51b8484ab3e4ff785f0888ab
2015-10-29 01:24:43 +01:00
LOCAL_LDFLAGS_x86_64 += -Wl,--version-script,$(LOCAL_PATH)/libc.x86_64.map
Don't export new/delete from libc shared libraries. This used to be handled by -fvisibility=hidden on libc_cxa, but that was broken by the resolution of https://llvm.org/PR22419 (introduced to Android in today's clang update). Now we just use a version script that prevents these from being re-exported from our shared libraries. Change-Id: Ib290e1d0d7426e09ad17a91178162fff6dbdcfa9
2015-04-02 01:26:54 +02:00
Revert "Revert "Prevent symbols from libgcc from being reexported."" This reverts commit 4e4e1a5d78ff6dd9394b9b116c012046da409e99. Change-Id: I8e78c3a59598e052e02965dd37b3fba8f567eed9
2014-06-24 06:31:08 +02:00
# We'd really like to do this for all architectures, but since this wasn't done
# before, these symbols must continue to be exported on LP32 for binary
# compatibility.
Revert "Fix volantis boot." Bug: http://b/20065774 This reverts commit 76e1cbca75e9fdaf7a7943f2c58e65433b283f4a.
2015-05-08 00:12:24 +02:00
LOCAL_LDFLAGS_64 := -Wl,--exclude-libs,libgcc.a
Revert "Revert "Prevent symbols from libgcc from being reexported."" This reverts commit 4e4e1a5d78ff6dd9394b9b116c012046da409e99. Change-Id: I8e78c3a59598e052e02965dd37b3fba8f567eed9
2014-06-24 06:31:08 +02:00
Prevent the build system from hiding our ABI. Unfortunately --exclude-libs (now passed globally) clobbers our version script, so we have to prevent the build system from using this flag. Bug: http://b/24166967 Change-Id: I33c766d399c418fdc17983c8c0a56608d463201e
2015-09-23 01:30:31 +02:00
# Unfortunately --exclude-libs clobbers our version script, so we have to
# prevent the build system from using this flag.
LOCAL_NO_EXCLUDE_LIBS := true
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
Fix missing args. Change-Id: I89a366db984ea6b2553896a4351e59bb91f93040
2014-02-13 01:40:55 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_SRC_FILES,libc_arch_dynamic_src_files))
Unregister pthread_atfork handlers on dlclose() Bug: http://b/20339788 Change-Id: I874c87faa377645fa9e0752f4fc166d81fd9ef7e
2014-11-21 05:47:02 +01:00
LOCAL_NO_CRT := true
LOCAL_ASFLAGS += $(libc_crt_target_cflags)
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
# special for arm
LOCAL_CFLAGS_arm += -DCRT_LEGACY_WORKAROUND
LOCAL_SRC_FILES_arm += \
Unregister pthread_atfork handlers on dlclose() Bug: http://b/20339788 Change-Id: I874c87faa377645fa9e0752f4fc166d81fd9ef7e
2014-11-21 05:47:02 +01:00
arch-arm/bionic/atexit_legacy.c
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
include $(BUILD_SHARED_LIBRARY)
libc: Cleanup the libc makefile. Removes a lot of redundant code, and reorganizes the file. We first setup all the comomn variables we may need, and then define the actual libraries that need to be generated. Signed-off-by: Dima Zavin <dima@android.com>
2009-05-27 19:52:37 +02:00
# ========================================================
Malloc debug rewrite. The major components of the rewrite: - Completely remove the qemu shared library code. Nobody was using it and it appears to have broken at some point. - Adds the ability to enable/disable different options independently. - Adds a new option that can enable the backtrace on alloc/free when a process gets a specific signal. - Adds a new way to enable malloc debug. If a special property is set, and the process has an environment variable set, then debug malloc will be enabled. This allows something that might be a derivative of app_process to be started with an environment variable being enabled. - get_malloc_leak_info() used to return one element for each pointer that had the exact same backtrace. The new version returns information for every one of the pointers with same backtrace. It turns out ddms already automatically coalesces these, so the old method simply hid the fact that there where multiple pointers with the same amount of backtrace. - Moved all of the malloc debug specific code into the library. Nothing related to the malloc debug data structures remains in libc. - Removed the calls to the debug malloc cleanup routine. Instead, I added an atexit call with the debug malloc cleanup routine. This gets around most problems related to the timing of doing the cleanup. The new properties and environment variables: libc.debug.malloc.options Set by option name (such as "backtrace"). Setting this to a bad value will cause a usage statement to be printed to the log. libc.debug.malloc.program Same as before. If this is set, then only the program named will be launched with malloc debug enabled. This is not a complete match, but if any part of the property is in the program name, malloc debug is enabled. libc.debug.malloc.env_enabled If set, then malloc debug is only enabled if the running process has the environment variable LIBC_DEBUG_MALLOC_ENABLE set. Bug: 19145921 Change-Id: I7b0e58cc85cc6d4118173fe1f8627a391b64c0d7
2015-11-17 02:30:32 +01:00
# libc_logging.a
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
# ========================================================
include $(CLEAR_VARS)
Remove references to MALLOC_LEAK_CHECK. This define isn't really used any more. Bug: 17378595 Change-Id: I39722b83cd98955a3a2514577de34a239524617f
2014-09-19 21:26:09 +02:00
LOCAL_CFLAGS := $(libc_common_cflags)
Use gnu++11 and gnu99 explicitly for C++/C files. Clang and gcc default to different standards, so we should be explicit about the versions we want to compile for. Change-Id: I65495a2392dd29f36373b94c616c2506173e6033
2013-10-01 02:10:53 +02:00
LOCAL_CONLYFLAGS := $(libc_common_conlyflags)
LOCAL_CPPFLAGS := $(libc_common_cppflags)
Split libc_debug.so into two .so modules loaded on demand from libc.so This change is intended to eliminate need to replace libc.so with libc_debug.so in order to enablememory allocation debugging. This is also the first step towards implementing extended memoryallocation debugging using emulator's capabilities in monitoring memory access.
2009-11-17 23:13:38 +01:00
LOCAL_C_INCLUDES := $(libc_common_c_includes)
LOCAL_SRC_FILES := \
Do not include libc_common in malloc debug code. The inclusion of the static libc_common library in the malloc_debug_XXX.so shared libraries causes constructors to be called twice. This doesn't seem to have caused any issues when setting the libc.debug.malloc property. However, jemalloc crashes because there are two jemalloc implementations, one in the static libc_common library and one in the shared library. Each implementation has created overlapping thread keys that are not the same. The crash comes because one of the jemalloc keys is actually used by the locale setting code. Thus if someone sets the locale, the jemalloc code crashes trying to access the same key. Change-Id: Iaac650a82d69064db148a6333e9403744f68b4a4
2014-07-10 02:16:07 +02:00
bionic/libc_logging.cpp \
Split libc_debug.so into two .so modules loaded on demand from libc.so This change is intended to eliminate need to replace libc.so with libc_debug.so in order to enablememory allocation debugging. This is also the first step towards implementing extended memoryallocation debugging using emulator's capabilities in monitoring memory access.
2009-11-17 23:13:38 +01:00
Malloc debug rewrite. The major components of the rewrite: - Completely remove the qemu shared library code. Nobody was using it and it appears to have broken at some point. - Adds the ability to enable/disable different options independently. - Adds a new option that can enable the backtrace on alloc/free when a process gets a specific signal. - Adds a new way to enable malloc debug. If a special property is set, and the process has an environment variable set, then debug malloc will be enabled. This allows something that might be a derivative of app_process to be started with an environment variable being enabled. - get_malloc_leak_info() used to return one element for each pointer that had the exact same backtrace. The new version returns information for every one of the pointers with same backtrace. It turns out ddms already automatically coalesces these, so the old method simply hid the fact that there where multiple pointers with the same amount of backtrace. - Moved all of the malloc debug specific code into the library. Nothing related to the malloc debug data structures remains in libc. - Removed the calls to the debug malloc cleanup routine. Instead, I added an atexit call with the debug malloc cleanup routine. This gets around most problems related to the timing of doing the cleanup. The new properties and environment variables: libc.debug.malloc.options Set by option name (such as "backtrace"). Setting this to a bad value will cause a usage statement to be printed to the log. libc.debug.malloc.program Same as before. If this is set, then only the program named will be launched with malloc debug enabled. This is not a complete match, but if any part of the property is in the program name, malloc debug is enabled. libc.debug.malloc.env_enabled If set, then malloc debug is only enabled if the running process has the environment variable LIBC_DEBUG_MALLOC_ENABLE set. Bug: 19145921 Change-Id: I7b0e58cc85cc6d4118173fe1f8627a391b64c0d7
2015-11-17 02:30:32 +01:00
LOCAL_MODULE := libc_logging
Split libc_debug.so into two .so modules loaded on demand from libc.so This change is intended to eliminate need to replace libc.so with libc_debug.so in order to enablememory allocation debugging. This is also the first step towards implementing extended memoryallocation debugging using emulator's capabilities in monitoring memory access.
2009-11-17 23:13:38 +01:00
Malloc debug rewrite. The major components of the rewrite: - Completely remove the qemu shared library code. Nobody was using it and it appears to have broken at some point. - Adds the ability to enable/disable different options independently. - Adds a new option that can enable the backtrace on alloc/free when a process gets a specific signal. - Adds a new way to enable malloc debug. If a special property is set, and the process has an environment variable set, then debug malloc will be enabled. This allows something that might be a derivative of app_process to be started with an environment variable being enabled. - get_malloc_leak_info() used to return one element for each pointer that had the exact same backtrace. The new version returns information for every one of the pointers with same backtrace. It turns out ddms already automatically coalesces these, so the old method simply hid the fact that there where multiple pointers with the same amount of backtrace. - Moved all of the malloc debug specific code into the library. Nothing related to the malloc debug data structures remains in libc. - Removed the calls to the debug malloc cleanup routine. Instead, I added an atexit call with the debug malloc cleanup routine. This gets around most problems related to the timing of doing the cleanup. The new properties and environment variables: libc.debug.malloc.options Set by option name (such as "backtrace"). Setting this to a bad value will cause a usage statement to be printed to the log. libc.debug.malloc.program Same as before. If this is set, then only the program named will be launched with malloc debug enabled. This is not a complete match, but if any part of the property is in the program name, malloc debug is enabled. libc.debug.malloc.env_enabled If set, then malloc debug is only enabled if the running process has the environment variable LIBC_DEBUG_MALLOC_ENABLE set. Bug: 19145921 Change-Id: I7b0e58cc85cc6d4118173fe1f8627a391b64c0d7
2015-11-17 02:30:32 +01:00
LOCAL_CLANG := $(use_clang)
LOCAL_ADDITIONAL_DEPENDENCIES := $(libc_common_additional_dependencies)
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
LOCAL_SYSTEM_SHARED_LIBRARIES :=
bionic: remove LOCAL_PRELINK_MODULE Change-Id: Iad65ae18446b20d2aace7b8fe19b1a032b616e9c Signed-off-by: Iliyan Malchev <malchev@google.com>
2011-03-14 22:02:05 +01:00
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
auto import from //depot/cupcake/@135843
2009-03-04 04:28:35 +01:00
Reconfig libc's Android.mk to build for multilib 1. Moved arch-specific setup to their own files: - <arch>/<arch>.mk, arch-specific configs. Variables in those config end with the arch name. - removed the extra complexity introduced by function libc-add-cpu-variant-src, which seems to be not very useful these days. 2. Separated out the crt object files generation rules and set up the rules for both TARGET_ARCH and TARGET_2ND_ARCH. 3. Build all the libraries for both TARGET_ARCH and TARGET_2ND_ARCH, with the arch-specific LOCAL_ variables. Bug: 11654773 Change-Id: I9c2d85db0affa49199d182236d2210060a321421
2014-01-24 00:17:50 +01:00
$(eval $(call patch-up-arch-specific-flags,LOCAL_CFLAGS,libc_common_cflags))
Malloc debug rewrite. The major components of the rewrite: - Completely remove the qemu shared library code. Nobody was using it and it appears to have broken at some point. - Adds the ability to enable/disable different options independently. - Adds a new option that can enable the backtrace on alloc/free when a process gets a specific signal. - Adds a new way to enable malloc debug. If a special property is set, and the process has an environment variable set, then debug malloc will be enabled. This allows something that might be a derivative of app_process to be started with an environment variable being enabled. - get_malloc_leak_info() used to return one element for each pointer that had the exact same backtrace. The new version returns information for every one of the pointers with same backtrace. It turns out ddms already automatically coalesces these, so the old method simply hid the fact that there where multiple pointers with the same amount of backtrace. - Moved all of the malloc debug specific code into the library. Nothing related to the malloc debug data structures remains in libc. - Removed the calls to the debug malloc cleanup routine. Instead, I added an atexit call with the debug malloc cleanup routine. This gets around most problems related to the timing of doing the cleanup. The new properties and environment variables: libc.debug.malloc.options Set by option name (such as "backtrace"). Setting this to a bad value will cause a usage statement to be printed to the log. libc.debug.malloc.program Same as before. If this is set, then only the program named will be launched with malloc debug enabled. This is not a complete match, but if any part of the property is in the program name, malloc debug is enabled. libc.debug.malloc.env_enabled If set, then malloc debug is only enabled if the running process has the environment variable LIBC_DEBUG_MALLOC_ENABLE set. Bug: 19145921 Change-Id: I7b0e58cc85cc6d4118173fe1f8627a391b64c0d7
2015-11-17 02:30:32 +01:00
include $(BUILD_STATIC_LIBRARY)
Split libc_debug.so into two .so modules loaded on demand from libc.so This change is intended to eliminate need to replace libc.so with libc_debug.so in order to enablememory allocation debugging. This is also the first step towards implementing extended memoryallocation debugging using emulator's capabilities in monitoring memory access.
2009-11-17 23:13:38 +01:00
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
# ========================================================
# libstdc++.so
# ========================================================
libstdcxx_common_src_files := \
bionic/__cxa_guard.cpp \
bionic/__cxa_pure_virtual.cpp \
bionic/new.cpp \
include $(CLEAR_VARS)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_C_INCLUDES := $(libc_common_c_includes) bionic/libstdc++/include
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
LOCAL_CFLAGS := $(libc_common_cflags)
Refactoring: C++11 style DISABLE_ bionic marcos Enable the -std=gnu++11 flag for libstdc++ static and dynamic libs. ScopeGuard uses DISABLE_ macros instead of '= delete'; Change-Id: I07e21b306f95fffd49345f7fa136cfdac61e0225
2014-09-09 01:22:22 +02:00
LOCAL_CPPFLAGS := $(libc_common_cppflags)
Generate sysv hash for libc, libstdc++ and libm Bug: 19059885 Change-Id: I35e2a9cd0bb3914d9d0c82d163bfaf0fff844667
2015-03-10 03:38:56 +01:00
Fix bug number for linker related workarounds Bug: 24425865 Change-Id: I2021674a9df0e768f312ca024d906302db1c494f
2015-09-28 19:14:17 +02:00
# TODO: This is to work around b/24465209. Remove after root cause is fixed
Switch libc/libcstdc++/libm to hash-style=both Bug: http://b/19059885 Change-Id: Ie42703d91f291fd689a1cf9c35bcab5c487b9a84
2015-05-08 21:05:47 +02:00
LOCAL_LDFLAGS_arm := -Wl,--hash-style=both
Apply work around b/19059885 to x86 Bug: http://b/19059885 Bug: http://b/21203348 Change-Id: Ic375e9f877d68de8f866d17362879a7dde638465 (cherry picked from commit 69a5fb951d69689dedd83cb033ae3dcd0ef05b65)
2015-05-16 02:53:39 +02:00
LOCAL_LDFLAGS_x86 := -Wl,--hash-style=both
Generate sysv hash for libc, libstdc++ and libm Bug: 19059885 Change-Id: I35e2a9cd0bb3914d9d0c82d163bfaf0fff844667
2015-03-10 03:38:56 +01:00
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
LOCAL_SRC_FILES := $(libstdcxx_common_src_files)
LOCAL_MODULE:= libstdc++
LOCAL_ADDITIONAL_DEPENDENCIES := $(LOCAL_PATH)/Android.mk
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
LOCAL_SYSTEM_SHARED_LIBRARIES := libc
Malloc debug rewrite. The major components of the rewrite: - Completely remove the qemu shared library code. Nobody was using it and it appears to have broken at some point. - Adds the ability to enable/disable different options independently. - Adds a new option that can enable the backtrace on alloc/free when a process gets a specific signal. - Adds a new way to enable malloc debug. If a special property is set, and the process has an environment variable set, then debug malloc will be enabled. This allows something that might be a derivative of app_process to be started with an environment variable being enabled. - get_malloc_leak_info() used to return one element for each pointer that had the exact same backtrace. The new version returns information for every one of the pointers with same backtrace. It turns out ddms already automatically coalesces these, so the old method simply hid the fact that there where multiple pointers with the same amount of backtrace. - Moved all of the malloc debug specific code into the library. Nothing related to the malloc debug data structures remains in libc. - Removed the calls to the debug malloc cleanup routine. Instead, I added an atexit call with the debug malloc cleanup routine. This gets around most problems related to the timing of doing the cleanup. The new properties and environment variables: libc.debug.malloc.options Set by option name (such as "backtrace"). Setting this to a bad value will cause a usage statement to be printed to the log. libc.debug.malloc.program Same as before. If this is set, then only the program named will be launched with malloc debug enabled. This is not a complete match, but if any part of the property is in the program name, malloc debug is enabled. libc.debug.malloc.env_enabled If set, then malloc debug is only enabled if the running process has the environment variable LIBC_DEBUG_MALLOC_ENABLE set. Bug: 19145921 Change-Id: I7b0e58cc85cc6d4118173fe1f8627a391b64c0d7
2015-11-17 02:30:32 +01:00
LOCAL_STATIC_LIBRARIES := libc_logging
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
include $(BUILD_SHARED_LIBRARY)
# ========================================================
# libstdc++.a
# ========================================================
include $(CLEAR_VARS)
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_C_INCLUDES := $(libc_common_c_includes) bionic/libstdc++/include
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
LOCAL_CFLAGS := $(libc_common_cflags)
Refactoring: C++11 style DISABLE_ bionic marcos Enable the -std=gnu++11 flag for libstdc++ static and dynamic libs. ScopeGuard uses DISABLE_ macros instead of '= delete'; Change-Id: I07e21b306f95fffd49345f7fa136cfdac61e0225
2014-09-09 01:22:22 +02:00
LOCAL_CPPFLAGS := $(libc_common_cppflags)
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
LOCAL_SRC_FILES := $(libstdcxx_common_src_files)
LOCAL_MODULE:= libstdc++
LOCAL_ADDITIONAL_DEPENDENCIES := $(LOCAL_PATH)/Android.mk
Exclude libstdc++ from linked libraries. Also explicitly add libstdc++ includes for necessary targets. Change-Id: If712ba0ae7908d8147a69e29da5c453a183d6540
2014-09-16 19:10:25 +02:00
LOCAL_CXX_STL := none
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
LOCAL_SYSTEM_SHARED_LIBRARIES := libc
Use new sanitizer option. Change-Id: I21b04a1f3be3b10ee557bf9e0caa5db638686364
2015-06-16 22:57:54 +02:00
LOCAL_SANITIZE := never
Coverage configs for bionic. Bug: 17553780 Change-Id: Iee7d0c210e2379e71e0c1e4e7058bce23f61e8cd
2014-09-23 19:44:01 +02:00
LOCAL_NATIVE_COVERAGE := $(bionic_coverage)
Hide C++ stuff in libc. Put it back in libstdc++. Bug: 17062445 Change-Id: I027b186719654c2865b08c3fa83f90fa00c1e838
2014-08-15 23:20:04 +02:00
include $(BUILD_STATIC_LIBRARY)
Reference in a new issue Copy permalink