Databases
StackOverflow2013
Postsdbo
PODoes the WKWYL optimization in make_shared<>() introduce a penalty for some multithreading applications?
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PODoes the WKWYL optimization in make_shared<>() introduce a penalty for some multithreading applications?
    primarykey
    data
    text
    <p>A few days ago I happened to watch <a href="http://channel9.msdn.com/Events/GoingNative/GoingNative-2012/STL11-Magic-Secrets">this very interesting presentation</a> by Stephan T. Lavavej, which mentions the "<em>We Know Where You Live</em>" optimization (sorry for using the acronym in the question title, SO warned me the question might have been closed otherwise), and <a href="http://www.youtube.com/watch?v=L7zSU9HI-6I&amp;feature=gv">this beautiful one</a> by Herb Sutter on machine architecture.</p> <p>Briefly, the "<em>We Know Where You Live</em>" optimization consists in placing the reference counters on the same memory block as the object which <code>make_shared</code> is creating, thus resulting in one single memory allocation rather than two and making <code>shared_ptr</code> more compact.</p> <p>After summing up what I learnt from the two presentations above, however, I started to wonder whether the WKWYL optimization could not <strong>degrade performance</strong> in case <code>shared_ptr</code> is accessed by multiple threads <em>running on different cores</em>.</p> <p>If the reference counters are <em>close</em> to the actual object in memory, in fact, they should be more likely to be fetched into the <strong>same cache line</strong> as the object itself. This in turn, if I got the lesson correctly, would make it more likely that threads will slow down while competing for the same cache line even when they do not need to.</p> <p>Suppose <strong>one</strong> of the threads needs to <strong>update the reference counter</strong> several times (e.g. when copying the <code>shared_ptr</code> around), while <strong>the other ones just need to access the pointed object</strong>: isn't this going to slow down the execution of <strong>all</strong> threads by making them compete for the same cache line? </p> <p><strong>If the refcount lived somewhere else in memory, I would say contention would be less likely to arise</strong>.</p> <p>Does this make a good argument against using <code>make_shared()</code> in similar cases (as long as it implements the WKWYL optimization, of course)? Or is there a fallacy in my reasoning?</p>
    singulars
    1. PO
      singulars
      1. PTAnswer
    1. This table or related slice is empty.
    1. PTQuestion
    1. USsehe
    1. USAndy Prowl
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    2. PO
      singulars
      1. PTAnswer
    3. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. PODoes the WKWYL optimization in make_shared<>() introduce a penalty for some multithreading applications?
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. PODoes the WKWYL optimization in make_shared<>() introduce a penalty for some multithreading applications?
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. PODoes the WKWYL optimization in make_shared<>() introduce a penalty for some multithreading applications?
      1. This table or related slice is empty.
      1. VTUpMod
    1. COCan you summarize what the "WKWYL" optimization is. It is not really a standard expression. In fact, it is so uncommon, this page is already on the front page of a search for it. :)
      singulars
      1. PODoes the WKWYL optimization in make_shared<>() introduce a penalty for some multithreading applications?
      1. USAndrew Tomazos
    2. CO@AndrewTomazosFathomlingCorps: it means "We Know Where You Live". basically, `make_shared()` places the reference counters in the same memory block as the object it creates. see slide 6 of the presentation by Stephan T. Lavavej I linked
      singulars
      1. PODoes the WKWYL optimization in make_shared<>() introduce a penalty for some multithreading applications?
      1. USAndy Prowl
    3. COIs the alternative you are comparing to storing the reference count with the shared object - storing it in a second dynamically allocated block? If so you have avoided the false sharing issue, but have made a much worse double payment for a second dynamic allocation, which is order of magntitudes more expensive than a cache miss.
      singulars
      1. PODoes the WKWYL optimization in make_shared<>() introduce a penalty for some multithreading applications?
      1. USAndrew Tomazos
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload