c++ - A faster thread blocks slower thread -

i'm working on point cloud viewer, , design based on 2 thread

• first thread updates point cloud data ( 10 fps)
• second thread d3d renderer render point set screen (about 90 fps)

so code looks this:

std::shared_ptr<pointcloud> pointcloud; critical_section updatelock; void firstthreadproc() {       while(true)       {        /* algorithm processes point cloud, takes time */             entercriticalsection(&updatelock);             pointcloud->update(data,length,...); //also takes time copy , process             leavecriticalsection(&updatelock);       } } /*...*/   std::shared_ptr<d3drenderer> renderer; void secondthreadproc() {     msg msg = { 0 };     while (wm_quit != msg.message)     {         if (peekmessage(&msg, null, 0, 0, pm_remove))         {             translatemessage(&msg);             dispatchmessage(&msg);         }         else         {             entercriticalsection(&updatelock);             renderer->render(pointcloud);             leavecriticalsection(&updatelock);         }     }  } 

i thought second thread way more fast first one, when first 1 entered critical section, second 1 blocked, renderer window should freeze or then. i'm observed right renderer window runs smooth, camera rotate or zoom in/out, good, first thread unstable, fps ranging 10 fps 1 fps.

i'm thinking 2 point cloud buffers, first thread updates second buffer when outsides critical section, swap 2 buffers within critical section. work?

as mentioned in this, critical_section not provide first-in, first-out(fifo) ordering. since second thread way more fast first thread, , whole loop critical section, enter critical section right after leave it. may in critical section , keep first 1 out of it.

my solution put more job of second thread outside critical section, works fine.