If pthread_once() is called with pthread_once_t initialized using
PTREAD_ONCE_INIT, pthread_once does not release pthread_mutex used
internally. This patch fixes that by calling pthread_mutex_destroy()
in the thread which has called init_routine.
Reviewed-by: Corinna Vinschen <corinna@vinschen.de>
Signed-off-by: Takashi Yano <takashi.yano@nifty.ne.jp>
Starting with commit 42faed4128 ("* thread.h (class pthread): Add bool
member canceled."), pthread::testcancel waits infinitely on cancel_event
after it checked if the canceled variable is set. However, this might
introduce a deadlock, if the thread calling pthread_cancel is terminated
after setting canceled to true, but before calling SetEvent on cancel_event.
In fact, it's not at all necessary to wait infinitely. By definition,
the thread is only canceled if cancel_event is set. The canceled
variable is just a helper to speed up code. We can safely assume that
the thread hasn't been canceled yet, if canceled is set, but cancel_event
isn't.
Fixes: 42faed4128 ("* thread.h (class pthread): Add bool member canceled.")
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Take note of schedparam in any pthread_attr_t passed to pthread_create.
postcreate() (racily, after the thread is actually created), sets the
scheduling priority if it's inherited, but precreate() doesn't store any
scheduling priority explicitly set via a non-default attr to
pthread_create, so schedparam.sched_priority has the default value of 0.
(I think this is another long-standing bug exposed by 4b51e4c1. Now we
don't lie about the actual thread priority, it's apparent it's not
really being set in this case.)
Fixes testcase priority2.
Signed-off-by: Jon Turney <jon.turney@dronecode.org.uk>
Currently, _my_tls.tid is set to pthread_null if pthread::self()
is called before pthread::init_mainthread(). As a result, pthread::
init_mainthread() does not set _my_tls.tid appropriately. Due to
this, pthread_join() fails in LDAP environment if the program is
the first program which loads cygwin1.dll.
https://cygwin.com/pipermail/cygwin/2023-June/253792.html
With this patch, _my_tls.tid is re-initialized in pthread::
init_mainthread() if it is pthread_null.
Reported-by: Mümin A. <muminaydin06@gmail.com>
Reviewed-by: Corinna Vinschen <corinna@vinschen.de>
Signed-off-by: Takashi Yano <takashi.yano@nifty.ne.jp>
This way, the sem API is all in the same place, even if the
underlying semaphore class is still in thread.cc.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Rename CygwinCreateThread to create_posix_thread and move
from miscfuncs.cc to create_posix_thread.cc, inbcluding all
related functions. Analogue for the prototypes.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
provide entire internal and external pthread API from inside the
same file.
While I dislike to have another even larger file, this is basically
cleaning up the source and grouping the external API into useful
chunks. Splitting the file cleanly is tricky due to usage of inline
methods is_good_object and verifyable_object_isvalid.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Remove dependency on __sdidinit member of struct _reent to check
object initialization. Like __sdidinit, the __cleanup member of
struct _reent is initialized in the __sinit() function. Checking
initialization against __cleanup serves the same purpose and will
reduce overhead in the __sfp() function in a follow up patch.
Revert mx parameter and mutex lock while operating the list.
Mutex was removed with 94d24160 informing that:
'Use InterlockedCompareExchangePointer to ensure race safeness
without using a mutex.'
But it does not.
Calling pthread_mutex_init and pthread_mutex_destroy from two or
more threads occasionally leads to hang in pthread_mutex_destroy.
To not change the behaviour of other cases where List_insert was called,
List_insert_nolock is added.
POSIX requires that key destructors are called in a loop
for each key with a non-NULL value until all values are
NULL, or until all destructors for non-NULL values
have been called at least PTHREAD_DESTRUCTOR_ITERATIONS
(per POSIX: 4) times.
Cygwinonly called all destructors with non-NULL values
exactly once. This patch fixes Cygwin to follow POSIX.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
So far sig_send's return type is int. The problem with this is
that sig_send returns a sigset_t on __SIGPENDING, and sigset_t
is defined as long type. So the function only returns the lower
32 bit of sigset_t, which is fine on 32 bit, but casts away the
pending RT signals on 64 bit.
Fix this by changing the return type of sig_send to sigset_t, so
as not to narrow down the sigset when returning from handling
__SIGPENDING. Make sure to cast correctly in all invocations
of sig_send.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
This patch set implements the Linux syscalls sched_getaffinity,
sched_setaffinity, pthread_getaffinity_np, and pthread_setaffinity_np.
Linux has a straightforward view of the cpu sets used in affinity masks.
They are simply long (1024-bit) bit masks. This code emulates that view
while internally dealing with Windows' distribution of available CPUs among
processor groups.
After creating a pthread, the stack gets moved to the desired memory
location. While the 32 bit thread wrapper copies the exception handler
information to the new stack (so we have at least *some* exception
handler present), the x86_64 code didn't perform any exception handler
magic. Fix that.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
- Drop hires_[nm]s clocks, rename hires.h to clock.h.
- Implement clk_t class as an extensible clock class in new file clock.cc.
- Introduce get_clock(clock_id) returning a pointer to the clk_t instance
for clock_id. Provide the following methods along the lines of the former
hires classes:
void clk_t::nsecs (struct timespec *);
ULONGLONG clk_t::nsecs ();
LONGLONG clk_t::usecs ();
LONGLONG clk_t::msecs ();
void clk_t::resolution (struct timespec *);
- Add CLOCK_REALTIME_COARSE, CLOCK_MONOTONIC_RAW, CLOCK_MONOTONIC_COARSE
and CLOCK_BOOTTIME clocks.
- Allow clock_nanosleep, pthread_condattr_setclock and timer_create to use
all new clocks (both clocks should be usable with a small tweak, though).
- Bump DLL major version to 2.12.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Updates to misc files to integrate AIO into the Cygwin source tree.
Much of it has to be done when adding any new syscalls. There are
some updates to limits.h for AIO-specific limits. And some doc mods.
pthread_timedjoin_np returns ETIMEDOUT if a thread is still running,
not EBUSY as pthread_tryjoin_np.
Also, clean up initializing timeout in pthread_tryjoin_np.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
- Move pthread_join to thread.cc to have all `join' calls in
the same file (pthread_timedjoin_np needs pthread_convert_abstime
which is static inline in thread.cc)
- Bump API version
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
* Redefine NSPERSEC to NS100PERSEC
* Define NSPERSEC as nanosecs per second
* Define USPERSEC as microsecs per second
* Use above constants throughout where appropriate
* Rename to_us to timespec_to_us and inline
* Rename it_bad to timespec_bad and inline
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
This new function returns the name of the calling thread; works for both
cygthreads and pthreads. All calls to cygthread::name(/*void*/) replaced
by calls to mythreadname(/*void*/).
- pthread_mutex::lock now takes a PLARGE_INTEGER timeout pointer
and uses that in the call to cygwait.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
- Introduce inline helper pthread_convert_abstime. It converts
an absolute timespec to a Windows LARGE_INTEGER timestamp,
depending on the used clock.
- Use this function from pthread_cond_timedwait and semaphore::timedwait
- Merge semaphore::_wait and semaphore::_timedwait into single _wait
method, taking a LARGER_INTEGER timestamp.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
POSIX states as follows about pthread_cond_wait:
If a signal is delivered to a thread waiting for a condition variable,
upon return from the signal handler the thread resumes waiting for the
condition variable as if it was not interrupted, or it returns zero
due to spurious wakeup.
Cygwin so far employs the latter behaviour, while Linux and BSD employ
the former one.
Align Cygwin behaviour to Linux and BSD.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
GDB since commit 24cdb46e [1] can report and use these names.
Add utility function SetThreadName(), which sends a thread name to the
debugger.
Use that:
- to set the default thread name for main thread and newly created pthreads.
- in pthread_setname_np() for user thread names.
- for helper thread names in cygthread::create()
- for helper threads which are created directly with CreateThread.
Note that there can still be anonymous threads, created by system or
injected DLLs.
[1] https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=24cdb46e9f0a694b4fbc11085e094857f08c0419
This patch adds pthread_getname_np and pthread_setname_np.
These were added to glibc in 2.12[1] and are also present in some form on
NetBSD and several UNIXes.
The code is based on NetBSD's implementation with changes to better match
Linux behaviour.
Implementation quirks:
* pthread_setname_np with a NULL pointer segfaults (as linux)
* pthread_setname_np returns ERANGE for names longer than 16 characters (as
linux)
* pthread_getname_np with a NULL pointer returns EFAULT (as linux)
* pthread_getname_np with a buffer length of less than 16 returns ERANGE (as
linux)
* pthread_getname_np truncates the thread name to fit the buffer length.
This guarantees success even when the default thread name is longer than 16
characters, but means there is no way to discover the actual length of the
thread name. (Linux always truncates the thread name to 16 characters)
* Changing program_invocation_short_name changes the default thread name (on
linux, it has no effect on the default thread name)
I'll leave it up to you to decide if any of these matter.
This is implemented via class pthread_attr to make it easier to add
pthread_attr_[gs]etname_np (present in NetBSD and some UNIXes) should it
ever be added to Linux (or we decide we want it anyway).
[1] https://sourceware.org/git/?p=glibc.git;a=blob;f=NEWS
This partially reverts commit 10a30e7 as far as the Cygwin version of
the __getreent function is concerned. Remove _COMPILING_NEWLIB guard
only allowing to use __getreent inline function when building newlib,
since we wan to use it in Cygwin as well.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Signed-off-by: Stefan Assmann <sassmann@redhat.com>
Bump GPLv2+ to GPLv3+ for some files, clarify BSD 2-clause.
Everything else stays under GPLv3+.
New Linking Exception exempts resulting executables from LGPLv3 section 4.
Add CONTRIBUTORS file to keep track of licensing.
Remove 'Copyright Red Hat Inc' comments.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
So far pthread::postcreate() only sets the thread priority at all, only
if the inherit-scheduler attribute is PTHREAD_EXPLICIT_SCHED. This
completely ignores the PTHREAD_INHERIT_SCHED case, since in contrast
to POSIX, a thread does not inherit its priority from the creating
thread, but always starts with THREAD_PRIORITY_NORMAL.
pthread_getschedparam() only returns what's stored in the thread attributes,
not the actual thread priority.
This patch fixes both problems.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
So far the scheduler priority handling is not POSIX compatible.
The priorities use a range of -14 up to +15, which means it's not clear
if the POSIX-required return value of -1 in case of an error is *really*
an error or just the valid priority value -1. Even more confusing, -14
is the *max* value and 15 is the *min* value. Last but not least this
range doesn't match the POSIX requirement of at least 32 priority values.
This patch cleans up scheduler priority handling and moves the valid
priority range to 1 (min) - 32 (max). It also adds a function
sched_get_thread_priority() which will help to make thread priority
more POSIX-like.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
For all pthread init functions, POSIX says
Results are undefined if pthread_FOO_init() is called specifying an
already initialized pthread_FOO object.
So far our pthread init functions tested the incoming object if it's
already an initialized object and, if so, returned EBUSY. That's ok
*iff* the object was already initialized. However, as the example in
https://cygwin.com/ml/cygwin/2016-04/msg00473.html shows, an uninitialized
pthread object could also accidentally look like an initialized object
and then returning EBUSY is not ok.
Consequentially, all those tests are dangerous. Per POSIX, an application
has to know what its doing when calling any of the pthread init functions
anyway, and re-initializing the object is just as well as undefined
behaviour as is returning EBUSY on already initialized objects.
* thread.cc (pthread_attr_init): Drop check for already initialized
object.
(pthread_condattr_init): Ditto.
(pthread_rwlockattr_init): Ditto.
(pthread_mutexattr_init): Ditto.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Hi.
I have noticed that Cygwin's spinlock goes into heavy sleeping code
for each spin. It seems it would be a good idea to actually try to
spin a bit first. There is this 'pause' instruction which let's the
CPU make such busy loops be less busy. Here is a patch to do this.
--
VH
* signal.cc (sigwait): Fix return value to reflect errno in case of
error according to POSIX. Never return EINTR.
* thread.cc (pthread_kill): Return errno if sig_send failed.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
* thread.cc (pthread_getattr_np): Fix memory leak, remove usage of
malloc for small local buffer.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>