(or loading a dso linked to libthr.so into process which was not
linked against threading library).
- Remove libthr interposers of the libc functions, including
__error(). Instead, functions calls are indirected through the
interposing table, similar to how pthread stubs in libc are already
done. Libc by default points either to syscall trampolines or to
existing libc implementations. On libthr load, libthr rewrites the
pointers to the cancellable implementations already in libthr. The
interposition table is separate from pthreads stubs indirection
table to not pull pthreads stubs into static binaries.
- Postpone the malloc(3) internal mutexes initialization until libthr
is loaded. This avoids recursion between calloc(3) and static
pthread_mutex_t initialization.
- Reinstall signal handlers with wrapper on libthr load. The
_rtld_is_dlopened(3) is used to avoid useless calls to sigaction(2)
when libthr is statically referenced from the main binary.
In the process, fix openat(2), swapcontext(2) and setcontext(2)
interposing. The libc symbols were exported at different versions
than libthr interposers. Export both libc and libthr versions from
libc now, with default set to the higher version from libthr.
Remove unused and disconnected swapcontext(3) userspace implementation
from libc/gen.
No objections from: deischen
Tested by: pho, antoine (exp-run) (previous versions)
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Some files lack required #include <sys/stat.h>. The #ifdef is per ngie's
request; the includes are clearly necessary for struct stat.
The faccessat test fails because it tries to use AT_SYMLINK_NOFOLLOW with
faccessat(), which is not specified by POSIX.1-2008.
Differential Revision: https://reviews.freebsd.org/D1411
Reviewed by: ngie
In general 64-bit ELF notes use 4-byte padding, not 8, despite what is
claimed in various specs.
Upstream elftoolchain ticket 472
https://sourceforge.net/p/elftoolchain/tickets/472/
Sponsored by: The FreeBSD Foundation
To be able to info pages consider installing texinfo from ports print/texinfo or
via pkg: pkg install texinfo
Differential Revision: https://reviews.freebsd.org/D1409
Reviewed by: emaste, imp (previous version)
Relnotes: yes
Fix transformation of add with pc argument to adr for non-immediate
arguments.
This fixes an "Unimplemented" error when assembling certain ARM add
instructions with pc-relative arguments.
Reported by: sbruno
PR: 196412, 196423
job
The h_raw application doesn't do proper bounds checking without the option
being supplied via the build, which means that it doesn't throw signals and
fail as expected
PR: 196430
X-MFC with: r276479
Rename some internal templates to avoid conflict with complier
intrinsics. __is_constructible --> __libcpp_is_constructible,
__is_nothrow_constructible --> __libcpp_is_nothrow_constructible, and
__is_nothrow_assignable --> __libcpp_is_nothrow_assignable. No
functionality change.
Pull in r206805 from upstream libc++ trunk (by Marshall Clow):
Use compiler intrinsic __is_constructible if available
This should fix building parts of world with -std=c++11 enabled.
Reported by: Oliver Hartmann <ohartman@zedat.fu-berlin.de>
MFC after: 1 week
PR20228: don't retain a pointer to a vector element after the
container has been resized.
This fixes a possible crash when compiling certain parts of libc++'s
type_traits header.
The initial fix in r276374 is valid only for 64-bit objects. Revert it
and return an error in CHECK_EHDR if the multiplication would overflow.
The original buffer overflow issue was found with the security/afl
fuzzer and has upstream elftoolchain ticket 462. The 32-bit object issue
with r276374 found by antoine@ during an i386 exp-run.
Sponsored by: The FreeBSD Foundation
This fixes two strip(1) issues found during ports exp-run and adds a
string hash implementation which significantly speeds up certain
operations on objects with large numbers of symbols.
This also improves libdwarf handling for stripped objects with
.eh_frame or .debug_frame (but not other debug) sections.
PR: 196107
Sponsored by: The FreeBSD Foundation
Previously a corrupt ELF file could read beyond the end of e_rawfile.
Upstream elftoolchain ticket 462. Found via the security/afl fuzzer.
Sponsored by: The FreeBSD Foundation
Track r270303:
This provides a minor cleanup in elfdump; there are otherwise no
consumers in the tree. Old SUN documentation can be found for either
variant, but GNU binutils switched to DT_FEATURE around 2000.
Sponsored by: The FreeBSD Foundation
This includes a number of libdwarf improvements (particularly DWARF4
related) and updates to elftoolchain tools such as strip(1). It also
includes a large number of miscellaneous fixes (memory leaks, sign and
cast warnings, integer overflow and underflow, etc.).
This is a merge of r276167,276170-276172 from the
projects/elftoolchain-update-r3130 branch.
Sponsored by: The FreeBSD Foundation
PowerPC: CTR shouldn't fire if a TLS call is in the loop
Determining the address of a TLS variable results in a function call in
certain TLS models. This means that a simple ICmpInst might actually
result in invalidating the CTR register.
In such cases, do not attempt to rely on the CTR register for loop
optimization purposes.
This fixes PR22034.
Differential Revision: http://reviews.llvm.org/D6786
This fixes a "Invalid PPC CTR loop" error when compiling parts of libc
for PowerPC-32.
[PowerPC] Replace foul hackery with real calls to __tls_get_addr
My original support for the general dynamic and local dynamic TLS
models contained some fairly obtuse hacks to generate calls to
__tls_get_addr when lowering a TargetGlobalAddress. Rather than
generating real calls, special GET_TLS_ADDR nodes were used to wrap
the calls and only reveal them at assembly time. I attempted to
provide correct parameter and return values by chaining CopyToReg and
CopyFromReg nodes onto the GET_TLS_ADDR nodes, but this was also not
fully correct. Problems were seen with two back-to-back stores to TLS
variables, where the call sequences ended up overlapping with unhappy
results. Additionally, since these weren't real calls, the proper
register side effects of a call were not recorded, so clobbered values
were kept live across the calls.
The proper thing to do is to lower these into calls in the first
place. This is relatively straightforward; see the changes to
PPCTargetLowering::LowerGlobalTLSAddress() in PPCISelLowering.cpp.
The changes here are standard call lowering, except that we need to
track the fact that these calls will require a relocation. This is
done by adding a machine operand flag of MO_TLSLD or MO_TLSGD to the
TargetGlobalAddress operand that appears earlier in the sequence.
The calls to LowerCallTo() eventually find their way to
LowerCall_64SVR4() or LowerCall_32SVR4(), which call FinishCall(),
which calls PrepareCall(). In PrepareCall(), we detect the calls to
__tls_get_addr and immediately snag the TargetGlobalTLSAddress with
the annotated relocation information. This becomes an extra operand
on the call following the callee, which is expected for nodes of type
tlscall. We change the call opcode to CALL_TLS for this case. Back
in FinishCall(), we change it again to CALL_NOP_TLS for 64-bit only,
since we require a TOC-restore nop following the call for the 64-bit
ABIs.
During selection, patterns in PPCInstrInfo.td and PPCInstr64Bit.td
convert the CALL_TLS nodes into BL_TLS nodes, and convert the
CALL_NOP_TLS nodes into BL8_NOP_TLS nodes. This replaces the code
removed from PPCAsmPrinter.cpp, as the BL_TLS or BL8_NOP_TLS
nodes can now be emitted normally using their patterns and the
associated printTLSCall print method.
Finally, as a result of these changes, all references to get-tls-addr
in its various guises are no longer used, so they have been removed.
There are existing TLS tests to verify the changes haven't messed
anything up). I've added one new test that verifies that the problem
with the original code has been fixed.
This fixes a fatal "Bad machine code" error when compiling parts of
libgomp for 32-bit PowerPC.
