[boost] Re: Re: Boost Shmem-Shared memory library first archiveavailable

From: Pavel Vozenilek (pavel_vozenilek_at_[hidden])
Date: 2004-11-24 22:54:02

"Ion Gaztañaga" wrote:

> do you know any windows implementation
> for process-shared condition variables?
Maybe
http://www.garret.ru/~knizhnik/shmem/Readme.htm

> I would love also non-connection oriented message oriented ipc,
> something similar to datagram unix socket or a connection oriented stream
> ipc, like stream unix sockets. As far as I know cygwin a similar uses 
> TCP/IP
> to emulate unix sockets, so I have not seen a pure windows source code
> anywhere.
>
Giallo library (http://giallo.sf.net/ typed from head) is attempt to create
TCP/IP Boost library. It may or may not fit with this request.

[exceptions versus error code]
> When using create only version, if the object is already created, I 
> suppose
> you would like another exception to be thrown, something like
> boost::shmem::already_created. The no throwing versions would return 0 if
> any error is found. Is that right?
>
Copy constructor may throw.

[lean offset_ptr]
> A char pointer could point to any byte in the segment, so (ptrdiff_t -1) 
> would be valid. I can add a new base class for offset_ptr that considers 
> (ptrdiff_t-1) zero, so that the user can use that pointer if he wants.
>
I wanted to write (ptrdiff_t)-1 or 0xFFF...FFF.

>12. The simple algorithm to find fitting memory
>> Maybe if the library has interface to plug in different
>> algorithm (no clue now how it would look like).
>
> I will add another algorith to show that algorithm can be chosen via 
> template parameter.
>
An inspiration for algorithms could be found
on http://www.hoard.org/ (it is complete and complex
framework with lot of allocators).

[fixing buggy STLs vs own containers]
>> Maybe the library could list those offending STLs and instruction how
>> to fix them manually to be useable with shmem. Such a fix should
>> not change behaviour of applications, right?
>
> I will add changes needed for Dinkum STL, but I don't know if publishing 
> template lines and the changes needed can lead to any legal problem.
>
Problems: would be ridiculous IMHO. People may not like to fiddle
with STL but it could be stressed these changes are safe.

>>>>d) support for "transactions": I would like to
>> - transactions with isolation level 1:
>>  - transactions with isolation level 2:
>
> Sounds like it can be a layer above shmem library, another library 
> perhaps.
>
I am not sure:
 - transaction mechanism would likely need to have access
   to very details of shmem on low level (like checking
   whether an object is still there not via exception).
I have feeling shmem should be designed and tested with
hooks to support transaction.

Ideally there would be common framework for transactions
libraries like shmem would implement and obey.

/Pavel