[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
FreeBSD used a similar method with us.
There is one queue for both realtime signals and standard signals.
For each real time signal, it is pushed back to the end of the queue.
Here is its code, http://bxr.su/FreeBSD/sys/kern/kern_sig.c#351
I need to research how signals are consumed though.
2016-07-17 7:11 GMT-07:00 Charles Cui <charles.cui1984%gmail.com@localhost>:
> I will study FreeBSD logic first and share with you guys.
> 2016-07-17 6:39 GMT-07:00 Christos Zoulas <christos%zoulas.com@localhost>:
>> On Jul 17, 2:02am, charles.cui1984%gmail.com@localhost (Charles Cui) wrote:
>> -- Subject: Re: updates?
>> | Hi Christos,
>> | I considered the questions that you asked.
>> | Here is another patch that improves the previous one in terms of error
>> | checking
>> | I have checked the queue size is the total number instead of per signal.
>> | my patch can pass the tests here,
>> | which can work as an unit test.
>> | In terms of the sig queue delivery order, for real time signals, I need
>> | familiar with the
>> | consuming logics in the netbsd code base. I guess maybe we need to sort
>> | real time signals when enqueuing,
>> | or selecting the correct real time signal when consuming.
>> | My action items,
>> | to familiar with signal consuming logic in netbsd, to make sure real
>> | signals are always consumed based on its signal number (the smaller the
>> | real time signal number, the earlier to be consumed.)
>> Thanks. What does FreeBSD do? Can we have a separate queue for realtime
>> signals and deliver those first?
Main Index |
Thread Index |