Rich Felker [Fri, 19 Oct 2012 00:26:41 +0000 (20:26 -0400)]
avoid raising spurious division-by-zero exception in printf
Rich Felker [Fri, 19 Oct 2012 00:19:53 +0000 (20:19 -0400)]
floating point environment/exceptions support for mips
Rich Felker [Thu, 18 Oct 2012 20:41:27 +0000 (16:41 -0400)]
fix parent-memory-clobber in posix_spawn (environ)
Rich Felker [Thu, 18 Oct 2012 19:58:23 +0000 (15:58 -0400)]
overhaul system() and popen() to use vfork; fix various related bugs
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.
Rich Felker [Thu, 18 Oct 2012 04:09:36 +0000 (00:09 -0400)]
fix (hopefully; untested) completely broken/incomplete microblaze sigsetjmp
Rich Felker [Thu, 18 Oct 2012 03:45:21 +0000 (23:45 -0400)]
fix microblaze asm relocations for shared libc
only @PLT relocations are considered functions for purposes of
-Bsymbolic-functions, so always use @PLT. it should not hurt in the
static-linked case.
Rich Felker [Wed, 17 Oct 2012 06:47:11 +0000 (02:47 -0400)]
assert() is supposed to have type void
Rich Felker [Tue, 16 Oct 2012 03:02:57 +0000 (23:02 -0400)]
add memmem function (gnu extension)
based on strstr. passes gnulib tests and a few quick checks of my own.
Rich Felker [Tue, 16 Oct 2012 01:01:48 +0000 (21:01 -0400)]
microblaze TLS relocation support, completely untested
Rich Felker [Mon, 15 Oct 2012 22:51:53 +0000 (18:51 -0400)]
add support for TLS variant I, presently needed for arm and mips
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.
Rich Felker [Mon, 15 Oct 2012 15:42:46 +0000 (11:42 -0400)]
block uid/gid changes during posix_spawn
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.
Rich Felker [Mon, 15 Oct 2012 01:10:44 +0000 (21:10 -0400)]
fix overlap of thread stacks with thread tls segments
Rich Felker [Sun, 14 Oct 2012 23:56:50 +0000 (19:56 -0400)]
fix main program TLS alignment for dynamic-linked programs
this change brings the behavior in line with the static-linked code,
which seems to be correct.
Rich Felker [Sun, 14 Oct 2012 03:53:18 +0000 (23:53 -0400)]
workaround broken hidden-visibility handling in pcc
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.
Rich Felker [Sun, 14 Oct 2012 03:46:51 +0000 (23:46 -0400)]
ensure pointer decay in inline-asm arg for i386 syscall6
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.
Rich Felker [Sun, 14 Oct 2012 03:25:20 +0000 (23:25 -0400)]
fix namespace clash (libc) in dynlink.c
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.
Rich Felker [Sun, 14 Oct 2012 03:23:29 +0000 (23:23 -0400)]
remove dead code from dynamic linker
Rich Felker [Fri, 12 Oct 2012 03:09:17 +0000 (23:09 -0400)]
comment possibly-confusing i386 vsyscall asm
Rich Felker [Fri, 12 Oct 2012 02:51:05 +0000 (22:51 -0400)]
avoid the thread-ptr-init behavior of sigaction when not installing handler
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.
Rich Felker [Fri, 12 Oct 2012 02:47:07 +0000 (22:47 -0400)]
i386 vsyscall support (vdso-provided sysenter/syscall instruction based)
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.
Rich Felker [Mon, 8 Oct 2012 15:20:31 +0000 (11:20 -0400)]
ensure that buffer for decoding auxv at startup is initially zero
Rich Felker [Mon, 8 Oct 2012 01:43:46 +0000 (21:43 -0400)]
clean up and refactor program initialization
the code in __libc_start_main is now responsible for parsing auxv,
rather than duplicating the parsing all over the place. this should
shave off a few cycles and some code size. __init_libc is left as an
external-linkage function despite the fact that it could be static, to
prevent it from being inlined and permanently wasting stack space when
main is called.
a few other minor changes are included, like eliminating per-thread
ssp canaries (they were likely broken when combined with certain
dlopen usages, and completely unnecessary) and some other unnecessary
checks. since this code gets linked into every program, it should be
as small and simple as possible.
Rich Felker [Mon, 8 Oct 2012 00:19:11 +0000 (20:19 -0400)]
fix breakage due to initializing thread pointer when loading libs
at initial program load, all libraries must be loaded before the
thread pointer can be setup, since the TP-relative addresses of all
initial TLS objects must be constant.
Rich Felker [Sun, 7 Oct 2012 03:57:51 +0000 (23:57 -0400)]
make new TLS setup block even implementation-internals signals
this is needed to ensure async-cancel-safety, i.e. to make it safe to
access TLS objects when async cancellation is enabled. otherwise, if
cancellation were acter upon after the atomic fetch/add but before the
thread saved the obtained memory, another access to the same TLS in
the cancellation handler could end up performing the atomic fetch/add
again, consuming more memory than is actually available and
overflowing into other objects on the heap.
Rich Felker [Sat, 6 Oct 2012 20:56:35 +0000 (16:56 -0400)]
don't crash if TLS library is loaded into process with no thread pointer
Rich Felker [Sat, 6 Oct 2012 20:51:03 +0000 (16:51 -0400)]
fix buggy TLS size/alignment computations in static-linked TLS
Rich Felker [Sat, 6 Oct 2012 05:36:11 +0000 (01:36 -0400)]
fix symbol acceptance/rejection rules for TLS
symbol value of 0 is not "undefined" for TLS; it's the address of the
first symbol in the TLS segment. however, non-definition TLS
references also have values of 0, so check the section.
hopefully the new logic is more clear, too.
Rich Felker [Sat, 6 Oct 2012 05:22:51 +0000 (01:22 -0400)]
TLS fixes, mainly alignment handling
compute offsets from the thread pointer statically when loading the
library, rather than repeating the logic on each thread creation. not
only is the latter less efficient at runtime; it also fails to provide
solid guarantees that the offsets will remain the same when the
initial alignment of memory is different. the new alignment handling
is both more rigorous and simpler.
the old code was also clobbering TLS bss with random image data in
some cases due to using tls_size (size of TLS segment) instead of
tls_len (length of the TLS data image).
Rich Felker [Fri, 5 Oct 2012 17:09:09 +0000 (13:09 -0400)]
fix/improve shared library ctor/dtor handling, allow recursive dlopen
some libraries call dlopen from their constructors, resulting in
recursive calls to dlopen. previously, this resulted in deadlock. I'm
now unlocking the dlopen lock before running constructors (this is
especially important since the lock also blocked pthread_create and
was being held while application code runs!) and using a separate
recursive mutex protecting the ctor/dtor state instead.
in order to prevent the same ctor from being called more than once, a
module is considered "constructed" just before the ctor runs.
also, switch from using atexit to register each dtor to using a single
atexit call to register the dynamic linker's dtor processing as just
one handler. this is necessary because atexit performs allocation and
may fail, but the library has already been loaded and cannot be
backed-out at the time dtor registration is performed. this change
also ensures that all dtors run after all atexit functions, rather
than in mixed order.
Rich Felker [Fri, 5 Oct 2012 16:09:54 +0000 (12:09 -0400)]
small dynamic linker module search fix
libraries loaded more than once by pathname should not get shortnames
that would cause them to later be used to satisfy non-pathname load
requests.
Rich Felker [Fri, 5 Oct 2012 15:51:50 +0000 (11:51 -0400)]
support for TLS in dynamic-loaded (dlopen) modules
unlike other implementations, this one reserves memory for new TLS in
all pre-existing threads at dlopen-time, and dlopen will fail with no
resources consumed and no new libraries loaded if memory is not
available. memory is not immediately distributed to running threads;
that would be too complex and too costly. instead, assurances are made
that threads needing the new TLS can obtain it in an async-signal-safe
way from a buffer belonging to the dynamic linker/new module (via
atomic fetch-and-add based allocator).
I've re-appropriated the lock that was previously used for __synccall
(synchronizing set*id() syscalls between threads) as a general
pthread_create lock. it's a "backwards" rwlock where the "read"
operation is safe atomic modification of the live thread count, which
multiple threads can perform at the same time, and the "write"
operation is making sure the count does not increase during an
operation that depends on it remaining bounded (__synccall or dlopen).
in static-linked programs that don't use __synccall, this lock is a
no-op and has no cost.
Rich Felker [Fri, 5 Oct 2012 05:15:25 +0000 (01:15 -0400)]
fix race condition in dlopen
orig_tail was being saved before the lock was obtained, allowing
dlopen failure to roll-back other dlopens that had succeeded.
Rich Felker [Fri, 5 Oct 2012 05:00:40 +0000 (01:00 -0400)]
fix incorrect TLS reloc macro names in x86_64 reloc.h
Rich Felker [Fri, 5 Oct 2012 02:48:33 +0000 (22:48 -0400)]
dynamic-linked TLS support for everything but dlopen'd libs
currently, only i386 is tested. x86_64 and arm should probably work.
the necessary relocation types for mips and microblaze have not been
added because I don't understand how they're supposed to work, and I'm
not even sure if it's defined yet on microblaze. I may be able to
reverse engineer the requirements out of gcc/binutils output.
Rich Felker [Fri, 5 Oct 2012 01:08:53 +0000 (21:08 -0400)]
remove freeing of dynamic linker data when dlopen/dlsym are not used
this was an optimization to save/recover a minimal amount of extra
memory for use by malloc, that's becoming increasingly costly to keep
around. freeing this data:
1. breaks debugging with gdb (it can't find library symbols)
2. breaks thread-local storage in shared libraries
it would be possible to disable freeing when TLS is used, but in
addition to the above breakages, tracking whether dlopen/dlsym is used
adds a cost to every symbol lookup, possibly making program startup
slower for large programs. combined with the complexity, it's not
worth it. we already save/recover plenty of memory in the dynamic
linker with reclaim_gaps.
Rich Felker [Fri, 5 Oct 2012 01:01:56 +0000 (21:01 -0400)]
beginnings of full TLS support in shared libraries
this code will not work yet because the necessary relocations are not
supported, and cannot be supported without some internal changes to
how relocation processing works (coming soon).
Rich Felker [Fri, 5 Oct 2012 00:04:13 +0000 (20:04 -0400)]
partial TLS support for dynamic-linked programs
only TLS in the main program is supported so far; TLS defined in
shared libraries will not work yet.
Rich Felker [Thu, 4 Oct 2012 20:35:46 +0000 (16:35 -0400)]
TLS (GNU/C11 thread-local storage) support for static-linked programs
the design for TLS in dynamic-linked programs is mostly complete too,
but I have not yet implemented it. cost is nonzero but still low for
programs which do not use TLS and/or do not use threads (a few hundred
bytes of new code, plus dependency on memcpy). i believe it can be
made smaller at some point by merging __init_tls and __init_security
into __libc_start_main and avoiding duplicate auxv-parsing code.
at the same time, I've also slightly changed the logic pthread_create
uses to allocate guard pages to ensure that guard pages are not
counted towards commit charge.
Rich Felker [Wed, 3 Oct 2012 15:49:58 +0000 (11:49 -0400)]
tell the assembler to mark all files as not requiring executable stack
for some reason this option is undocumented. not sure when it was
added, so I'm using a configure test. gcc was already setting the mark
correctly for C files, but assembler source files would need ugly
.note boilerplate in every single file to achieve this without the
option to the assembler.
blame whoever thought it would be a good idea to make the stack
executable by default rather than doing it the other way around...
Rich Felker [Mon, 1 Oct 2012 00:00:38 +0000 (20:00 -0400)]
add getopt reset support
based on proposed patches by Daniel Cegiełka, with minor changes:
- use a weak symbol for optreset so it doesn't clash with namespace
- also reset optpos (position in multi-option arg like -lR)
- also make getopt_long support reset
Rich Felker [Sun, 30 Sep 2012 23:44:45 +0000 (19:44 -0400)]
protect sem_open against cancellation
also fix one minor bug: failure to free the early-reserved slot when
the semaphore later found to already be mapped.
Rich Felker [Sun, 30 Sep 2012 23:35:40 +0000 (19:35 -0400)]
overhaul sem_open
this function was overly complicated and not even obviously correct.
avoid using openat/linkat just like in shm_open, and instead expand
pathname using code shared with shm_open. remove bogus (and dangerous,
with priorities) use of spinlocks.
this commit also heavily streamlines the code and ensures there are no
failure cases that can happen after a new semaphore has been created
in the filesystem, since that case is unreportable.
Rich Felker [Sun, 30 Sep 2012 21:53:54 +0000 (17:53 -0400)]
clean up, bugfixes, and general improvement for shm_open/shm_unlink
1. don't make non-cloexec file descriptors
2. cancellation safety (cleanup handlers were missing, now unneeded)
3. share name validation/mapping code between open/unlink functions
4. avoid wasteful/slow syscalls
Rich Felker [Sun, 30 Sep 2012 05:52:17 +0000 (01:52 -0400)]
define some _POSIX_* macros that were omitted; required for XSI conformance
Rich Felker [Sat, 29 Sep 2012 22:14:46 +0000 (18:14 -0400)]
more close-on-exec fixes, mostly using new "e" flag to fopen
Rich Felker [Sat, 29 Sep 2012 22:09:34 +0000 (18:09 -0400)]
add 'e' modifier (close-on-exec) to fopen and fdopen
this feature will be in the next version of POSIX, and can be used
internally immediately. there are many internal uses of fopen where
close-on-exec is needed to fix bugs.
Rich Felker [Sat, 29 Sep 2012 21:59:50 +0000 (17:59 -0400)]
fix some more O_CLOEXEC/SOCK_CLOEXEC issues
Rich Felker [Sat, 29 Sep 2012 21:44:33 +0000 (17:44 -0400)]
always expose accept4
it will be in the next version of POSIX
Rich Felker [Sat, 29 Sep 2012 21:42:21 +0000 (17:42 -0400)]
always expose dup3 and pipe2
they will be in the next version of POSIX
Rich Felker [Sat, 29 Sep 2012 21:40:42 +0000 (17:40 -0400)]
move accept4, dup3, and pipe2 to non-linux-specific locations
these interfaces have been adopted by the Austin Group for inclusion
in the next version of POSIX.
Rich Felker [Sat, 29 Sep 2012 21:36:27 +0000 (17:36 -0400)]
emulate SOCK_CLOEXEC and SOCK_NONBLOCK for old (pre-2.6.27) kernels
also update syslog to use SOCK_CLOEXEC rather than separate fcntl
step, to make it safe in multithreaded programs that run external
programs.
emulation is not atomic; it could be made atomic by holding a lock on
forking during the operation, but this seems like overkill. my goal is
not to achieve perfect behavior on old kernels (which have plenty of
other imperfect behavior already) but to avoid catastrophic breakage
in (1) syslog, which would give no output on old kernels with the
change to use SOCK_CLOEXEC, and (2) programs built on a new kernel
where configure scripts detected a working SOCK_CLOEXEC, which later
get run on older kernels (they may otherwise fail to work completely).
Rich Felker [Sat, 29 Sep 2012 20:49:32 +0000 (16:49 -0400)]
sem_open should make process-shared semaphores
this did not matter because we don't yet treat process-shared special.
when private futex support is added, however, it will matter.
Rich Felker [Sat, 29 Sep 2012 20:48:52 +0000 (16:48 -0400)]
use O_CLOEXEC to open semaphore files in sem_open
Rich Felker [Sat, 29 Sep 2012 05:14:07 +0000 (01:14 -0400)]
fix some indention-with-spaces that crept in
Rich Felker [Sat, 29 Sep 2012 05:05:31 +0000 (01:05 -0400)]
microblaze port
based on initial work by rdp, with heavy modifications. some features
including threads are untested because qemu app-level emulation seems
to be broken and I do not have a proper system image for testing.
Rich Felker [Thu, 27 Sep 2012 22:56:10 +0000 (18:56 -0400)]
fix arm clone syscall bug (no effect unless app uses clone)
the code to exit the new thread/process after the start function
returns was mixed up in its syscall convention.
Rich Felker [Thu, 27 Sep 2012 21:19:09 +0000 (17:19 -0400)]
optimize strchrnul/strcspn not to scan string twice on no-match
when strchr fails, and important piece of information already
computed, the string length, is thrown away. have strchrnul (with
namespace protection) be the underlying function so this information
can be kept, and let strchr be a wrapper for it. this also allows
strcspn to be considerably faster in the case where the match set has
a single element that's not matched.
Rich Felker [Thu, 27 Sep 2012 20:56:33 +0000 (16:56 -0400)]
slightly cleaner strlen, also seems to compile to better code
testing with gcc 4.6.3 on x86, -Os, the old version does a duplicate
null byte check after the first loop. this is purely the compiler
being stupid, but the old code was also stupid and unintuitive in how
it expressed the check.
Rich Felker [Wed, 26 Sep 2012 04:56:07 +0000 (00:56 -0400)]
fix dirname to handle input of form "foo/" correctly
also optimized a bit.
Rich Felker [Tue, 25 Sep 2012 02:39:08 +0000 (22:39 -0400)]
fix handling of EINTR during close()
austin group interpretation for defect #529
(http://austingroupbugs.net/view.php?id=529) tightens the
requirements on close such that, if it returns with EINTR, the file
descriptor must not be closed. the linux kernel developers vehemently
disagree with this, and will not change it. we catch and remap EINTR
to EINPROGRESS, which the standard allows close() to return when the
operation was not finished but the file descriptor has been closed.
Rich Felker [Sat, 22 Sep 2012 20:19:06 +0000 (16:19 -0400)]
fix getaddrinfo to accept port 0 (zero)
new behavior can be summarized as:
inputs that parse completely as a decimal number are treated as one,
and rejected only if the result is out of 16-bit range.
inputs that do not parse as a decimal number (where strtoul leaves
anything left over in the input) are searched in /etc/services.
Rich Felker [Sat, 22 Sep 2012 20:08:58 +0000 (16:08 -0400)]
fix remaining IPC_64 issue (shmctl)
also cleanup cruft related to the issue
Rich Felker [Sat, 22 Sep 2012 12:04:17 +0000 (08:04 -0400)]
fix IPC_64 in msgctl too
Rich Felker [Sat, 22 Sep 2012 12:02:42 +0000 (08:02 -0400)]
fix broken semctl on systems that don't use IPC_64 flag
not tested on mips and arm; they may still be broken. x86_64 should be
ok now.
Rich Felker [Fri, 21 Sep 2012 17:47:26 +0000 (13:47 -0400)]
offer REALGCC variable to configure musl-gcc wrapper at runtime
this is useful when the underlying gcc is already a wrapper, which is
the case at least on some uclibc-based system images. it's also useful
for running an older/newer/nondefault version of gcc.
Rich Felker [Fri, 21 Sep 2012 08:05:01 +0000 (04:05 -0400)]
LFS64 alias for prlimit
issue reported/requested by Justin Cormack
Rich Felker [Mon, 17 Sep 2012 02:26:23 +0000 (22:26 -0400)]
add clock_adjtime, remap_file_pages, and syncfs syscall wrappers
patch by Justin Cormack, with slight modification
Rich Felker [Sun, 16 Sep 2012 04:10:08 +0000 (00:10 -0400)]
release notes for 0.9.6
Rich Felker [Sun, 16 Sep 2012 03:45:41 +0000 (23:45 -0400)]
add O_EXEC open mode
the linux O_PATH mode provides the necessary semantics for both the
O_SEARCH and O_EXEC modes defined and required by POSIX 2008.
Rich Felker [Sun, 16 Sep 2012 03:41:07 +0000 (23:41 -0400)]
add crypt_md5 password hash
contributed by nsz
Rich Felker [Sun, 16 Sep 2012 03:17:00 +0000 (23:17 -0400)]
fix stupid bug in updating of O_ACCMODE for O_SEARCH support
this could cause major bugs, and warrants a fix release right away.
Rich Felker [Sat, 15 Sep 2012 07:38:20 +0000 (03:38 -0400)]
0.9.5 release notes
Rich Felker [Sat, 15 Sep 2012 07:19:41 +0000 (03:19 -0400)]
fix syscall asm constraints for arm too
no problems were detected so far, but the constraints seem to have
been invalid just like the mips ones.
Rich Felker [Sat, 15 Sep 2012 07:03:21 +0000 (03:03 -0400)]
revert low rounds-count limits in crypt hashes
it was determined in discussion that these kind of limits are not
sufficient to protect single-threaded servers against denial of
service attacks from maliciously large round counts. the time scales
simply vary too much; many users will want login passwords with rounds
counts on a scale that gives decisecond latency, while highly loaded
webservers will need millisecond latency or shorter.
still some limit is left in place; the idea is not to protect against
attacks, but to avoid the runtime of a single call to crypt being, for
all practical purposes, infinite, so that configuration errors can be
caught and fixed without bringing down whole systems. these limits are
very high, on the order of minute-long runtimes for modest systems.
Rich Felker [Sat, 15 Sep 2012 06:24:12 +0000 (02:24 -0400)]
update mips cancellation-point syscall asm with 7-arg and r25 fixes
these fixes were already made to the normal syscall asm but not the
cancellation point version.
Rich Felker [Sat, 15 Sep 2012 06:22:10 +0000 (02:22 -0400)]
fix buggy constraints in mips inline syscall asm
if same register is used for input/output, the compiler must be told.
otherwise is generates random junk code that clobbers the result. in
pure syscall-wrapper functions, nothing went wrong, but in more
complex functions where register allocation is non-trivial, things
broke badly.
Rich Felker [Sat, 15 Sep 2012 03:52:51 +0000 (23:52 -0400)]
workaround gcc got-register-reload performance problems in malloc
with this patch, the malloc in libc.so built with -Os is nearly the
same speed as the one built with -O3. thus it solves the performance
regression that resulted from removing the forced -O3 when building
libc.so; now libc.so can be both small and fast.
Rich Felker [Sat, 15 Sep 2012 03:38:10 +0000 (23:38 -0400)]
remove forced -O3 from shared library CFLAGS
I originally added -O3 for shared libraries to counteract very bad
behavior by GCC when building PIC code: it insists on reloading the
GOT register in static functions that need it, even if the address of
the function is never leaked from the translation unit and all local
callers of the function have already loaded the GOT register. this
measurably degrades performance in a few key areas like malloc. the
inlining done at -O3 avoids the issue, but that's really not a good
reason for overriding the user's choice of optimization level.
Rich Felker [Fri, 14 Sep 2012 19:32:51 +0000 (15:32 -0400)]
use vfork if possible in posix_spawn
vfork is implemented as the fork syscall (with no atfork handlers run)
on archs where it is not available, so this change does not introduce
any change in behavior or regression for such archs.
Rich Felker [Fri, 14 Sep 2012 01:01:30 +0000 (21:01 -0400)]
strsep is BSD|GNU, not GNU-only; it's originally from BSD
Rich Felker [Fri, 14 Sep 2012 00:56:25 +0000 (20:56 -0400)]
add O_PATH/O_SEARCH support to fcntl.h
I'm not 100% sure that Linux's O_PATH meets the POSIX requirements for
O_SEARCH, but it seems very close if not perfect. and old kernels
ignore it, so O_SEARCH will still work as desired as long as the
caller has read permissions to the directory.
Rich Felker [Tue, 11 Sep 2012 06:23:47 +0000 (02:23 -0400)]
improve mips syscall asm constraints to use immediates, if possible
by using the "ir" constraint (immediate or register) and the carefully
constructed instruction addu $2,$0,%2 which can take either an
immediate or a register for %2, the new inline asm admits maximal
optimization with no register spillage to the stack when the compiler
successfully performs constant propagration, but still works by
allocating a register when the syscall number cannot be recognized as
a constant. in the case of syscalls with 0-3 arguments it barely
matters, but for 4-argument syscalls, using an immediate for the
syscall number avoids creating a stack frame for the syscall wrapper
function.
Rich Felker [Tue, 11 Sep 2012 02:43:22 +0000 (22:43 -0400)]
eliminate assumption that mips syscall restart preserves r25
all past and current kernel versions have done so, but there seems to
be no reason it's necessary and the sentiment from everyone I've asked
has been that we should not rely on it. instead, use r7 (an argument
register) which will necessarily be preserved upon syscall restart.
however this only works for 0-3 argument syscalls, and we have to
resort to the function call for 4-argument syscalls.
Rich Felker [Mon, 10 Sep 2012 23:04:24 +0000 (19:04 -0400)]
asm for memmove on i386 and x86_64
for the sake of simplicity, I've only used rep movsb rather than
breaking up the copy for using rep movsd/q. on all modern cpus, this
seems to be fine, but if there are performance problems, there might
be a need to go back and add support for rep movsd/q.
Rich Felker [Mon, 10 Sep 2012 22:37:27 +0000 (18:37 -0400)]
fix another ppoll issue (missing sigset_t size argument)
Rich Felker [Mon, 10 Sep 2012 22:16:11 +0000 (18:16 -0400)]
reenable word-at-at-time copying in memmove
before restrict was added, memove called memcpy for forward copies and
used a byte-at-a-time loop for reverse copies. this was changed to
avoid invoking UB now that memcpy has an undefined copying order,
making memmove considerably slower.
performance is still rather bad, so I'll be adding asm soon.
Rich Felker [Mon, 10 Sep 2012 22:05:02 +0000 (18:05 -0400)]
fix ppoll with null timeout argument
Rich Felker [Mon, 10 Sep 2012 19:30:52 +0000 (15:30 -0400)]
add LIBCC (compiler runtime) logic and override to configure
this should both fix the issue with ARM needing -lgcc_eh (although
that's really a bug in the libgcc build process that's causing
considerable bloat, which should be fixed) and make it easier to build
musl using clang/llvm in place of gcc. unfortunately I don't know a
good way to detect and support pcc's -lpcc since it's not in pcc's
default library search path...
Rich Felker [Sun, 9 Sep 2012 20:50:20 +0000 (16:50 -0400)]
add setdomainname syscall, fix getdomainname (previously a stub)
Rich Felker [Sun, 9 Sep 2012 20:37:19 +0000 (16:37 -0400)]
mincore syscall wrapper
Rich Felker [Sun, 9 Sep 2012 20:33:47 +0000 (16:33 -0400)]
fix up lfs64 junk for preadv/pwritev
Rich Felker [Sun, 9 Sep 2012 20:29:33 +0000 (16:29 -0400)]
add preadv/pwritev syscall wrappers
Rich Felker [Sun, 9 Sep 2012 20:27:26 +0000 (16:27 -0400)]
fix typo introduced in poll.h
Rich Felker [Sun, 9 Sep 2012 20:09:29 +0000 (16:09 -0400)]
add linux ppoll syscall wrapper
Rich Felker [Sun, 9 Sep 2012 18:58:55 +0000 (14:58 -0400)]
reenable sync_file_range; should no longer break on mips
Rich Felker [Sun, 9 Sep 2012 18:53:06 +0000 (14:53 -0400)]
add 7-arg syscall support for mips
no syscalls actually use that many arguments; the issue is that some
syscalls with 64-bit arguments have them ordered badly so that
breaking them into aligned 32-bit half-arguments wastes slots with
padding, and a 7th slot is needed for the last argument.
Rich Felker [Sun, 9 Sep 2012 05:29:19 +0000 (01:29 -0400)]
inline syscall support for arm
most pure-syscall-wrapper functions compile to the smallest/simplest
code possible (save r7 ; load syscall # ; svc 0 ; restore r7 ; tail
call to __syscall_ret).
Rich Felker [Sun, 9 Sep 2012 05:01:19 +0000 (01:01 -0400)]
inline syscall support for mips
this drastically reduces the size of some functions which are purely
syscall wrappers.
disabled for clang due to known bugs satisfying register constraints.
Rich Felker [Sun, 9 Sep 2012 04:59:30 +0000 (00:59 -0400)]
fix mips syscall_cp_asm code (saved register usage)
Rich Felker [Sun, 9 Sep 2012 04:55:31 +0000 (00:55 -0400)]
fix broken mips syscall asm
this code was using $10 to save the syscall number, but $10 is not
necessarily preserved by the kernel across syscalls. only mattered for
syscalls that got interrupted by a signal and restarted. as far as i
can tell, $25 is preserved by the kernel across syscalls.