WITNESS for pthreads

Randall Stewart rrs at lakerest.net
Tue Mar 31 08:11:05 PDT 2009

This was one of the places I was heading (as I wrote privately to  
Daniel ;-D)

I suppose I can share it all i.e. the pthread mutex stuff
will of course work with shared mutexe's but it won't:

a) Build an easy to use semantic for the app to agree on sharing  
memory.. i.e. you
   have left undefined how the process figure out what they are  
sharing. There is
   some value in setting up a easy semantic for app dev's to use.

<i.e. insert the mmap and all the other goo through an additional  

b) What happens when a process exits or hits a core dump while holding  
   of these mutex's? Is this what you are thinking the PROCESS_SHARED  

<i.e. I don't think a process by itself can fully solve this... maybe  
    PROCESS_SHARED could be made to help here>

c) If you build something to do <a> so you have some nice way of naming
   mutex's you can do something similar to our WITNESS option in the
   kernel... this is something the few times I have played in user
   space recently that I have missed... having LOR warnings and such
   can be a useful tool. You can't have this without <a> IMO.

I was am interested in a/b but one of my long term intents is to do  
<c> ;-)


On Mar 31, 2009, at 10:38 AM, John Baldwin wrote:

> On Tuesday 31 March 2009 2:50:27 am Daniel Eischen wrote:
>>> Ok, I have poked around at these... all the mutex attributes  
>>> defined here
>>> do is set the attributes to shared. There does not seem to be any  
>>> standard
>>> naming mechanism.
>> Naming mechanism for what?  Names shouldn't be needed for anything,
>> nor do I think it is desired.
> Off topic: names would be very helpful to port witness to pthreads.   
> The
> thoughts I have had for doing this though would be to add a new _np  
> attribute
> to set the name.  I actually would like to write a 'libwitness' that
> basically overrides the various symbols and provides the name_np  
> attribute
> and implement witness in the shared library on top of whatever  
> pthreads
> library is in use.  This would also allow it to be portable to other  
> OS's.
> (Well, it could break pshared mutexes, but using the pointer-style  
> types, you
> could have the libwitness allocate its own "mutex" structure which has
> a "real" mutex inside of it along with the name and other per-lock  
> data it
> tracks.  It would then forward mutex operations to the real pthreads  
> library
> after performing LOR checks, etc.).
> -- 
> John Baldwin

Randall Stewart
803-317-4952 (cell)

More information about the freebsd-threads mailing list