Rich Felker [Thu, 24 Mar 2011 18:18:00 +0000 (14:18 -0400)]
overhaul cancellation to fix resource leaks and dangerous behavior with signals
this commit addresses two issues:
1. a race condition, whereby a cancellation request occurring after a
syscall returned from kernelspace but before the subsequent
CANCELPT_END would cause cancellable resource-allocating syscalls
(like open) to leak resources.
2. signal handlers invoked while the thread was blocked at a
cancellation point behaved as if asynchronous cancellation mode wer in
effect, resulting in potentially dangerous state corruption if a
cancellation request occurs.
the glibc/nptl implementation of threads shares both of these issues.
with this commit, both are fixed. however, cancellation points
encountered in a signal handler will not be acted upon if the signal
was received while the thread was already at a cancellation point.
they will of course be acted upon after the signal handler returns, so
in real-world usage where signal handlers quickly return, it should
not be a problem. it's possible to solve this problem too by having
sigaction() wrap all signal handlers with a function that uses a
pthread_cleanup handler to catch cancellation, patch up the saved
context, and return into the cancellable function that will catch and
act upon the cancellation. however that would be a lot of complexity
for minimal if any benefit...
Rich Felker [Sun, 20 Mar 2011 01:36:10 +0000 (21:36 -0400)]
syscall overhaul part two - unify public and internal syscall interface
with this patch, the syscallN() functions are no longer needed; a
variadic syscall() macro allows syscalls with anywhere from 0 to 6
arguments to be made with a single macro name. also, manually casting
each non-integer argument with (long) is no longer necessary; the
casts are hidden in the macros.
some source files which depended on being able to define the old macro
SYSCALL_RETURNS_ERRNO have been modified to directly use __syscall()
instead of syscall(). references to SYSCALL_SIGSET_SIZE and SYSCALL_LL
have also been changed.
x86_64 has not been tested, and may need a follow-up commit to fix any
minor bugs/oversights.
Rich Felker [Sat, 19 Mar 2011 22:51:42 +0000 (18:51 -0400)]
overhaul syscall interface
this commit shuffles around the location of syscall definitions so
that we can make a syscall() library function with both SYS_* and
__NR_* style syscall names available to user applications, provides
the syscall() library function, and optimizes the code that performs
the actual inline syscalls in the library itself.
previously on i386 when built as PIC (shared library), syscalls were
incurring bus lock (lock prefix) overhead at entry and exit, due to
the way the ebx register was being loaded (xchg instruction with a
memory operand). now the xchg takes place between two registers.
further cleanup to arch/$(ARCH)/syscall.h is planned.
Rich Felker [Sat, 19 Mar 2011 01:53:30 +0000 (21:53 -0400)]
some linux headers useful from user apps.
i'm still not sure whether it's a good idea to include or use any of
these, but i'll add them for now. it may make more sense to just add
official kernel headers to the include path for compiling programs
that need them.
Rich Felker [Fri, 18 Mar 2011 02:55:43 +0000 (22:55 -0400)]
implement wprintf family of functions
this implementation is extremely ugly and inefficient, but it avoids a
good deal of code duplication and bloat. it may be cleaned up later to
eliminate the remaining code duplication and some of the warts, but i
don't really care about its performance.
Rich Felker [Fri, 18 Mar 2011 00:41:37 +0000 (20:41 -0400)]
implement robust mutexes
some of this code should be cleaned up, e.g. using macros for some of
the bit flags, masks, etc. nonetheless, the code is believed to be
working and correct at this point.
Rich Felker [Thu, 17 Mar 2011 17:35:08 +0000 (13:35 -0400)]
avoid function call to pthread_self in mutex unlock
if the mutex was previously locked, we can assume pthread_self was
already called at the time of locking, and thus that the thread
pointer is initialized.
Rich Felker [Thu, 17 Mar 2011 17:17:15 +0000 (13:17 -0400)]
reorder mutex struct fields to make room for pointers (upcoming robust mutexes)
the layout has been chosen so that pointer slots 3 and 4 fit between
the integer slots on 32-bit archs, and come after the integer slots on
64-bit archs.
for some reason these functions are not shaded by the PS/TPS option in
POSIX, so presumably they are mandatory, even though the functionality
they offer is optional. for now, provide them in case any programs
depend on their existence, but disallow any priority except the
default.
Rich Felker [Fri, 11 Mar 2011 02:34:19 +0000 (21:34 -0500)]
fix sem_open and sem_close to obey posix semantics
multiple opens of the same named semaphore must return the same
pointer, and only the last close can unmap it. thus the ugly global
state keeping track of mappings. the maximum number of distinct named
semaphores that can be opened is limited sufficiently small that the
linear searches take trivial time, especially compared to the syscall
overhead of these functions.
Rich Felker [Thu, 10 Mar 2011 23:31:37 +0000 (18:31 -0500)]
optimize pthread termination in the non-detached case
we can avoid blocking signals by simply using a flag to mark that the
thread has exited and prevent it from getting counted in the rsyscall
signal-pingpong. this restores the original pthread create/join
throughput from before the sigprocmask call was added.
Rich Felker [Thu, 10 Mar 2011 14:54:23 +0000 (09:54 -0500)]
fix errno behavior in clock_* functions
these functions are specified inconsistent in whether they're
specified to return an error value, or return -1 and set errno.
hopefully now they all match what POSIX requires.
Rich Felker [Thu, 10 Mar 2011 01:21:23 +0000 (20:21 -0500)]
optimize pthread initialization
the set_tid_address returns the tid (which is also the pid when called
from the initial thread) so there is no need to make a separate
syscall to get pid/tid.
Rich Felker [Thu, 10 Mar 2011 01:07:24 +0000 (20:07 -0500)]
fix race condition in raise - just mask signals
a signal handler could fork after the pid/tid were read, causing the
wrong process to be signalled. i'm not sure if this is supposed to
have UB or not, but raise is async-signal-safe, so it probably is
allowed. the current solution is slightly expensive so this
implementation is likely to be changed in the future.
Rich Felker [Tue, 1 Mar 2011 17:04:36 +0000 (12:04 -0500)]
use -L/...../ -lgcc instead of /...../libgcc.a in musl-gcc wrapper
this should avoid warnings about unused libs when not linking, and
might fix some other obscure issues too. i might replace this approach
with a completely different one soon though.
Rich Felker [Sun, 27 Feb 2011 05:28:59 +0000 (00:28 -0500)]
cleanup utf-8 multibyte code, use visibility if possible
this code was written independently of musl, with support for a the
backwards, nonstandard "31-bit unicode" some libraries/apps might
want. unfortunately the extra code (inside #ifdef) makes the source
harder to read and makes code that should be simple look complex, so
i'm removing it. anyone who wants to use the old code can find it in
the history or from elsewhere.
also, change the visibility of the __fsmu8 state machine table to
hidden, if supported. this should improve performance slightly in
shared-library builds.
Rich Felker [Thu, 24 Feb 2011 21:37:21 +0000 (16:37 -0500)]
various changes in preparation for dynamic linking support
prefer using visibility=hidden for __libc internal data, rather than
an accessor function, if the compiler has visibility.
optimize with -O3 for PIC targets (shared library). without heavy
inlining, reloading the GOT register in small functions kills
performance. 20-30% size increase for a single libc.so is not a big
deal, compared to comparaible size increase in every static binaries.
use -Bsymbolic-functions, not -Bsymbolic. global variables are subject
to COPY relocations, and thus binding their addresses in the library
at link time will cause library functions to read the wrong (original)
copies instead of the copies made in the main program's bss section.
Rich Felker [Thu, 24 Feb 2011 17:34:31 +0000 (12:34 -0500)]
fix backwards conditional in stpncpy
this only made the function unnecessarily slow on systems with
unaligned access, but would of course crash on systems that can't do
unaligned accesses (none of which have ports yet).
Rich Felker [Mon, 21 Feb 2011 03:30:06 +0000 (22:30 -0500)]
use an accessor function for __libc data pointer when compiled as PIC
prior to this change, a large portion of libc was unusable prior to
relocation by the dynamic linker, due to dependence on the global data
in the __libc structure and the need to obtain its address through the
GOT. with this patch, the accessor function __libc_loc is now able to
obtain the address of __libc via PC-relative addressing without using
the GOT. this means the majority of libc functionality is now
accessible right away.
naturally, the above statements all depend on having an architecture
where PC-relative addressing and jumps/calls are feasible, and a
compiler that generates the appropriate code.