On 04/05/2017 09:36, Fam Zheng wrote:
>> +
>> +    /* If there is a concurrent writer, wait for it.  */
>> +    while (atomic_read(&s->lock) & 1) {
>> +        g_usleep(5);
> What's the difference of g_usleep() from cpu_relax() in
> qemu_co_mutex_lock_unlock?

cpu_relax() did not exist when I wrote this patch. :)  I'll change it.

Paolo

Reply via email to