pcc wrongly passes any option beginning with -m to the linker, and
will break at link time if these options were added to CFLAGS. testing
linking lets us catch this at configure time and skip them.

these functions must behave as if they obtain the lock via flockfile
to satisfy POSIX requirements. since another thread can provably hold
the lock when they are called, they must wait to obtain the lock
before they can return, even if the correct return value could be
obtained without locking. in the case of fclose and freopen, failure
to do so could cause correct (albeit obscure) programs to crash or
otherwise misbehave; in the case of feof, ferror, and fwide, failure
to obtain the lock could sometimes return incorrect results. in any
case, having these functions proceed and return while another thread
held the lock was wrong.

1. don't open /dev/null just as a basis to copy flags; use shared
__fmodeflags function to get the right file flags for the mode.

2. handle the case (probably invalid, but whatever) case where the
original stream's file descriptor was closed; previously, the logic
re-closed it.

3. accept the "e" mode flag for close-on-exec; update dup3 to fallback
to using dup2 so we can simply call __dup3 instead of putting fallback
logic in freopen itself.

gcc seems to be generating identical or near-identical code for both
versions, but the newer code is more expressive of what it's doing.

the W* namespace is not reserved, so the nonstandard ones must be
moved under extension features. also WNOHANG and WUNTRACED were
missing.

the behavior of putenv is left undefined if the argument does not
contain an equal sign, but traditional implementations behave this way
and gnulib replaces putenv if it doesn't do this.

this will prevent gnulib from wrapping our strtod to handle this
useless feature.

signal mask was not being restored after fork, but instead blocked again.

this was broken during the early dynamic-linked TLS commits, which
rearranged some of the code for handling new relocation types.

__release_ptc() is only valid in the parent; if it's performed in the
child, the lock will be unlocked early then double-unlocked later,
corrupting the lock state.

with this commit, based on testing with patches to qemu which are not
yet upstream,

this is necessary to allow $CC with arguments in it

since it did not set the return-value register, the caller could
wrongly interpret this as failure.

these macros are supported by more compilers

since we target systems without overcommit, special care should be
taken that system() and popen(), like posix_spawn(), do not fail in
processes whose commit charges are too high to allow ordinary forking.

this in turn requires special precautions to ensure that the parent
process's signal handlers do not end up running in the shared-memory
child, where they could corrupt the state of the parent process.

popen has also been updated to use pipe2, so it does not have a
fd-leak race in multi-threaded programs. since pipe2 is missing on
older kernels, (non-atomic) emulation has been added.

some silly bugs in the old code should be gone too.

only @PLT relocations are considered functions for purposes of
-Bsymbolic-functions, so always use @PLT. it should not hurt in the
static-linked case.

based on strstr. passes gnulib tests and a few quick checks of my own.

despite documentation that makes it sound a lot different, the only
ABI-constraint difference between TLS variants II and I seems to be
that variant II stores the initial TLS segment immediately below the
thread pointer (i.e. the thread pointer points to the end of it) and
variant I stores the initial TLS segment above the thread pointer,
requiring the thread descriptor to be stored below. the actual value
stored in the thread pointer register also tends to have per-arch
random offsets applied to it for silly micro-optimization purposes.

with these changes applied, TLS should be basically working on all
supported archs except microblaze. I'm still working on getting the
necessary information and a working toolchain that can build TLS
binaries for microblaze, but in theory, static-linked programs with
TLS and dynamic-linked programs where only the main executable uses
TLS should already work on microblaze.

alignment constraints have not yet been heavily tested, so it's
possible that this code does not always align TLS segments correctly
on archs that need TLS variant I.

usage of vfork creates a situation where a process of lower privilege
may momentarily have write access to the memory of a process of higher
privilege.

consider the case of a multi-threaded suid program which is calling
posix_spawn in one thread while another thread drops the elevated
privileges then runs untrusted (relative to the elevated privilege)
code as the original invoking user. this untrusted code can then
potentially modify the data the child process will use before calling
exec, for example changing the pathname or arguments that will be
passed to exec.

note that if vfork is implemented as fork, the lock will not be held
until the child execs, but since memory is not shared it does not
matter.

this change brings the behavior in line with the static-linked code,
which seems to be correct.

with this change, pcc-built musl libc.so seems to work correctly. the
problem is that pcc generates GOT lookups for external-linkage symbols
even if they are hidden, rather than using GOT-relative addressing.
the entire reason we're using hidden visibility on the __libc object
is to make it accessible prior to relocations -- not to mention
inexpensive to access. unfortunately, the workaround makes it even
more expensive on pcc.

when the pcc issue is fixed, an appropriate version test should be
added so new pcc can use the much more efficient variant.

this is actually a rather subtle issue: do arrays decay to pointers
when used as inline asm args? gcc says yes, but currently pcc says no.
hopefully this discrepency in pcc will be fixed, but since the
behavior is not clearly defined anywhere I can find, I'm using an
explicit operation to cause the decay to occur.

this makes it so the #undef libc and __libc name are no longer needed,
which were problematic because the "accessor function" mode for
accessing the libc struct could not be used, breaking build on any
compiler without (working) visibility.

this is necessary because posix_spawn calls sigaction after vfork, and
if the thread pointer is not already initialized, initializing it in
the child corrupts the parent process's state.

this doubles the performance of the fastest syscalls on the atom I
tested it on; improvement is reportedly much more dramatic on
worst-case cpus. cannot be used for cancellable syscalls.