[Bug 604753] Re: [eglibc] process shared mutex's fail on armel v7 (thumb)

Ken Werner 604753 at bugs.launchpad.net
Thu Feb 17 14:31:35 UTC 2011


I had a quick look into the apr sources. Their locks/unix/proc_mutex.c can map the proc_mutex functionality to various methods:
 1) Posix sem implemenation (see mutex_posixsem_methods)
 2) SysV sem implementation (see mutex_sysv_methods)
 3) pthread implementation  (see mutex_proc_pthread_methods)
 4) fcntl implementation    (see mutex_fcntl_methods)
 5) flock implementation    (see mutex_flock_methods)

I think only the pthread implemenation (3) is of interest here. The creation of the pthread_mutex happens at:
  locks/unix/proc_mutex.c:proc_mutex_proc_pthread_create
that calls pthread_mutex_init with the following attributes:
  PTHREAD_PROCESS_SHARED, PTHREAD_MUTEX_ROBUST_NP, PTHREAD_PRIO_INHERIT
The apr-phthread-mutex functionality relies on pthread_mutex_lock and pthread_mutex_trylock. The eglibc implemenation for such a mutex calls atomic_compare_and_exchange_val_acq (see also lll_lock/lll_trylock at ports/sysdeps/unix/sysv/linux/arm/nptl/lowlevellock.h) which goes down to __arch_compare_and_exchange_val_32_acq. Using  (e)glibc (>= 2.12.1-0ubuntu11) and GCC (>=4.5) this expands to __sync_val_compare_and_swap (ports/sysdeps/unix/sysv/linux/arm/nptl/bits/atomic.h) which should be safe even on a SMP system.

Some infos regarding the atomic memory operations can be found at:
https://wiki.linaro.org/WorkingGroups/ToolChain/AtomicMemoryOperations#implementation%20details

The next step would be to check whether the testcase fails using a
decent GCC and eglibc (Ubuntu Natty for example).

-- 
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to apr in ubuntu.
https://bugs.launchpad.net/bugs/604753

Title:
  [eglibc] process shared mutex's fail on armel v7 (thumb)



More information about the Ubuntu-server-bugs mailing list