From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: Peter Maydell <peter.maydell@linaro.org>
Subject: [Qemu-devel] [PULL 17/17] util/qemu-thread-posix: Fix qemu_thread_atexit* for OSX
Date: Tue, 6 Nov 2018 22:38:03 +0100 [thread overview]
Message-ID: <1541540283-45699-18-git-send-email-pbonzini@redhat.com> (raw)
In-Reply-To: <1541540283-45699-1-git-send-email-pbonzini@redhat.com>
From: Peter Maydell <peter.maydell@linaro.org>
Our current implementation of qemu_thread_atexit* is broken on OSX.
This is because it works by cerating a piece of thread-specific
data with pthread_key_create() and using the destructor function
for that data to run the notifier function passed to it by
the caller of qemu_thread_atexit_add(). The expected use case
is that the caller uses a __thread variable as the notifier,
and uses the callback to clean up information that it is
keeping per-thread in __thread variables.
Unfortunately, on OSX this does not work, because on OSX
a __thread variable may be destroyed (freed) before the
pthread_key_create() destructor runs. (POSIX imposes no
ordering constraint here; the OSX implementation happens
to implement __thread variables in terms of pthread_key_create((),
whereas Linux uses different mechanisms that mean the __thread
variables will still be present when the pthread_key_create()
destructor is run.)
Fix this by switching to a scheme similar to the one qemu-thread-win32
uses for qemu_thread_atexit: keep the thread's notifiers on a
__thread variable, and run the notifiers on calls to
qemu_thread_exit() and on return from the start routine passed
to qemu_thread_start(). We do this with the pthread_cleanup_push()
API.
We take advantage of the qemu_thread_atexit_add() API
permission not to run thread notifiers on process exit to
avoid having to special case the main thread.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20181105135538.28025-3-peter.maydell@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
util/qemu-thread-posix.c | 44 ++++++++++++++++++++------------------------
1 file changed, 20 insertions(+), 24 deletions(-)
diff --git a/util/qemu-thread-posix.c b/util/qemu-thread-posix.c
index dfa66ff..865e476 100644
--- a/util/qemu-thread-posix.c
+++ b/util/qemu-thread-posix.c
@@ -443,42 +443,34 @@ void qemu_event_wait(QemuEvent *ev)
}
}
-static pthread_key_t exit_key;
-
-union NotifierThreadData {
- void *ptr;
- NotifierList list;
-};
-QEMU_BUILD_BUG_ON(sizeof(union NotifierThreadData) != sizeof(void *));
+static __thread NotifierList thread_exit;
+/*
+ * Note that in this implementation you can register a thread-exit
+ * notifier for the main thread, but it will never be called.
+ * This is OK because main thread exit can only happen when the
+ * entire process is exiting, and the API allows notifiers to not
+ * be called on process exit.
+ */
void qemu_thread_atexit_add(Notifier *notifier)
{
- union NotifierThreadData ntd;
- ntd.ptr = pthread_getspecific(exit_key);
- notifier_list_add(&ntd.list, notifier);
- pthread_setspecific(exit_key, ntd.ptr);
+ notifier_list_add(&thread_exit, notifier);
}
void qemu_thread_atexit_remove(Notifier *notifier)
{
- union NotifierThreadData ntd;
- ntd.ptr = pthread_getspecific(exit_key);
notifier_remove(notifier);
- pthread_setspecific(exit_key, ntd.ptr);
-}
-
-static void qemu_thread_atexit_run(void *arg)
-{
- union NotifierThreadData ntd = { .ptr = arg };
- notifier_list_notify(&ntd.list, NULL);
}
-static void __attribute__((constructor)) qemu_thread_atexit_init(void)
+static void qemu_thread_atexit_notify(void *arg)
{
- pthread_key_create(&exit_key, qemu_thread_atexit_run);
+ /*
+ * Called when non-main thread exits (via qemu_thread_exit()
+ * or by returning from its start routine.)
+ */
+ notifier_list_notify(&thread_exit, NULL);
}
-
typedef struct {
void *(*start_routine)(void *);
void *arg;
@@ -490,6 +482,7 @@ static void *qemu_thread_start(void *args)
QemuThreadArgs *qemu_thread_args = args;
void *(*start_routine)(void *) = qemu_thread_args->start_routine;
void *arg = qemu_thread_args->arg;
+ void *r;
#ifdef CONFIG_PTHREAD_SETNAME_NP
/* Attempt to set the threads name; note that this is for debug, so
@@ -501,7 +494,10 @@ static void *qemu_thread_start(void *args)
#endif
g_free(qemu_thread_args->name);
g_free(qemu_thread_args);
- return start_routine(arg);
+ pthread_cleanup_push(qemu_thread_atexit_notify, NULL);
+ r = start_routine(arg);
+ pthread_cleanup_pop(1);
+ return r;
}
void qemu_thread_create(QemuThread *thread, const char *name,
--
1.8.3.1
next prev parent reply other threads:[~2018-11-06 21:48 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-11-06 21:37 [Qemu-devel] [PULL 00/17] Misc patches for QEMU 3.1 hard freeze (?) Paolo Bonzini
2018-11-06 21:37 ` [Qemu-devel] [PULL 01/17] icount: fix deadlock when all cpus are sleeping Paolo Bonzini
2018-11-06 21:37 ` [Qemu-devel] [PULL 02/17] x86: hv_evmcs CPU flag support Paolo Bonzini
2018-11-06 21:37 ` [Qemu-devel] [PULL 03/17] i386: clarify that the Q35 machine type implements a P35 chipset Paolo Bonzini
2018-11-06 21:37 ` [Qemu-devel] [PULL 04/17] ivshmem: fix memory backend leak Paolo Bonzini
2018-11-06 21:37 ` [Qemu-devel] [PULL 05/17] MAINTAINERS: remove or downgrade myself to reviewer from some subsystems Paolo Bonzini
2018-11-06 21:37 ` [Qemu-devel] [PULL 06/17] target/i386: Clear RF on SYSCALL instruction Paolo Bonzini
2018-11-06 21:37 ` [Qemu-devel] [PULL 07/17] memory: learn about non-volatile memory region Paolo Bonzini
2018-11-06 21:37 ` [Qemu-devel] [PULL 08/17] nvdimm: set non-volatile on the " Paolo Bonzini
2018-11-06 21:37 ` [Qemu-devel] [PULL 09/17] memory-mapping: skip non-volatile memory regions in GuestPhysBlockList Paolo Bonzini
2018-11-06 21:37 ` [Qemu-devel] [PULL 10/17] scripts/dump-guest-memory: Synchronize with guest_phys_blocks_region_add Paolo Bonzini
2018-11-06 21:37 ` [Qemu-devel] [PULL 11/17] lsi53c895a: check message length value is valid Paolo Bonzini
2018-11-06 21:37 ` [Qemu-devel] [PULL 12/17] scsi-generic: keep VPD page list sorted Paolo Bonzini
2018-11-06 21:37 ` [Qemu-devel] [PULL 13/17] scsi-generic: avoid out-of-bounds access to VPD page list Paolo Bonzini
2018-11-06 21:38 ` [Qemu-devel] [PULL 14/17] scsi-generic: avoid invalid access to struct when emulating block limits Paolo Bonzini
2018-11-06 21:38 ` [Qemu-devel] [PULL 15/17] scsi-generic: do not do VPD emulation for sense other than ILLEGAL_REQUEST Paolo Bonzini
2018-11-06 21:38 ` [Qemu-devel] [PULL 16/17] include/qemu/thread.h: Document qemu_thread_atexit* API Paolo Bonzini
2018-11-06 21:38 ` Paolo Bonzini [this message]
2018-11-06 23:08 ` [Qemu-devel] [PULL 00/17] Misc patches for QEMU 3.1 hard freeze (?) Peter Maydell
2018-11-08 11:33 ` Peter Maydell
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1541540283-45699-18-git-send-email-pbonzini@redhat.com \
--to=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).