This is what I am trying to do<div>- I have to 2 processes in the same box.</div><div>- I need to send a buffer of data from one process to another.</div><div>- The other process reads the data &amp; may want to send something back.</div>
<div>- At any time there is only one such buffer in process - i.e. no queue of data to send.</div><div>- the process will send a new buffer only when it gets acknowledgement back from the</div><div>other process that it&#39;s finished reading it.</div>
<div><br></div><div>I am a little unclear as to what I should use here - shared_memory_object or </div><div>managed_shared_memory - it&#39;s not very clear what&#39;s the difference between the two.</div><div><br></div><div>
I tried out a sample from the condition_variable example here &amp; it seems to work.</div><div><a href="http://www.boost.org/doc/libs/1_36_0/doc/html/interprocess/synchronization_mechanisms.html#interprocess.synchronization_mechanisms.conditions">http://www.boost.org/doc/libs/1_36_0/doc/html/interprocess/synchronization_mechanisms.html#interprocess.synchronization_mechanisms.conditions</a></div>
<div><br></div><div><br></div><div>However, in the example the size of the shared memory is based on the size of the trace_queue object.</div><div>In my case, the size needs to be configurable. </div><div><br></div><div>so I am thinking of something like this</div>
<div>- no data array inside the struct.</div><div><br></div><div>struct data</div><div><div>{</div><div><br></div><div>  data(void * p)</div><div>      :  message_in(false), actualdata(p)</div><div>   {}</div><div><br></div>
<div>   //Mutex to protect access to data</div><div>   boost::interprocess::interprocess_mutex      mutex;</div><div>   boost::interprocess::interprocess_condition  cond_empty;</div><div>   boost::interprocess::interprocess_condition  cond_full;</div>
<div>   void * actualdata;</div><div>   bool message_in;</div><div>};</div><div><br></div><div>then</div><div><span class="identifier"><br></span></div><div><span class="identifier">shared_memory_object</span> <span class="identifier">shm</span>
      <span class="special">(</span><span class="identifier">create_only</span><span class="comment"> </span><span class="special">,</span><span class="string">&quot;shared_memory&quot;</span><span class="comment"> </span><span class="special">,</span><span class="identifier">read_write</span> <span class="special">);</span></div>
<div>shm.truncate (sizeof(data) + myconfiguredsize);</div><div><br></div><div>actual data is beyond the struct which holds the condition variable</div><div><br></div><div>i.e. when i write data, i will do </div><div><br></div>
<div><span class="identifier">mapped_region</span> <span class="identifier">region</span>
         <span class="special">(</span><span class="identifier">shm</span> <span class="comment"> </span><span class="special">,</span><span class="identifier">read_write</span> <span class="special">);</span> </div><div>
<span class="keyword">void</span> <span class="special">*</span> <span class="identifier">addr</span>       <span class="special">=</span> <span class="identifier">region</span><span class="special">.</span><span class="identifier">get_address</span><span class="special">();</span></div>
<div><span class="special"><br></span></div><div><span class="special"><br></span></div><div><br></div></div><div>data * d = new (addr) data(addr);</div><div>then</div><div><br></div><div>//Use the condition_variables to protect the write &amp; then memcpy</div>
<div>memcpy(d-&gt;actualdata, thedataiwanttocopy, sizeofthedataiwanttocopy);</div><div><br></div><div><br></div><div>Am I doing this the right way or is there a better way to go about this?</div><div><br></div><div><br></div>