MAVLink app: Do not rate-limit camera trigger feedback

[LorenzMeier]

This is important to ensure no camera trigger messages are being dropped.

[LorenzMeier]

CPU load on master:

 PID COMMAND                   CPU(ms) CPU(%)  USED/STACK PRIO(BASE) STATE 
   0 Idle Task                 1422607 39.042   644/  748   0 (  0)  READY 
   1 hpwork                     137033  4.113   888/ 1780 192 (192)  w:sem 
   2 lpwork                       7088  0.224   640/ 1780  50 ( 50)  READY 
   3 init                        18391  0.000  2008/ 2484 100 (100)  w:sem 
 336 top                            39  0.000  1296/ 1684 100 (100)  RUN   
  95 gps                          8035  0.224   808/ 1476 220 (220)  w:sig 
 103 dataman                       180  0.000   704/ 1180  90 ( 90)  w:sem 
 143 sensors                    167475  5.011  1336/ 1980 250 (250)  w:sem 
 145 commander                   69483  1.944  2896/ 3548 140 (140)  w:sig 
 152 px4io                      105276  3.141   976/ 1380 240 (240)  w:sem 
 165 mavlink_if0                 66952  1.944  1736/ 2436 100 (100)  w:sig 
 166 mavlink_rcv_if0              1236  0.074  1148/ 2100 175 (175)  w:sem 
 178 mavlink_if1                 48296  1.271  1712/ 2388 100 (100)  w:sig 
 179 mavlink_rcv_if1              5004  0.149  1416/ 2100 175 (175)  w:sem 
 236 commander_low_prio            120  0.000   704/ 2996  50 ( 50)  w:sem 
 217 mavlink_if2                308063  9.349  1824/ 2388 100 (100)  w:sig 
 218 mavlink_rcv_if2               511  0.074  1536/ 2100 175 (175)  w:sem 
 234 logger                      17801  0.598  1312/ 3532 250 (250)  w:sem 
 237 log_writer_file                 0  0.000   344/ 1060  60 ( 60)  w:sem 
 285 ekf2                       631821 18.922  5112/ 5780 250 (250)  w:sem 
 292 mc_att_control             154698  4.712  1256/ 1676 250 (250)  w:sem 
 297 mc_pos_control             161517  4.861  1240/ 1876 250 (250)  w:sem 
 308 navigator                    7072  0.224  1224/ 1572 105 (105)  w:sem 

Processes: 23 total, 3 running, 20 sleeping
CPU usage: 56.84% tasks, 4.11% sched, 39.04% idle
DMA Memory: 5120 total, 1536 used 1536 peak
Uptime: 3359.948s total, 1422.608s idle
nsh> free
             total       used       free    largest
Mem:        236864     206384      30480      25680

[LorenzMeier]

With this PR:

 PID COMMAND                   CPU(ms) CPU(%)  USED/STACK PRIO(BASE) STATE 
   0 Idle Task                  270574 37.810   644/  748   0 (  0)  READY 
   1 hpwork                      28150  4.079   816/ 1780 192 (192)  w:sig 
   2 lpwork                       1658  0.199   640/ 1780  50 ( 50)  w:sig 
   3 init                         8239  0.000  2120/ 2484 100 (100)  w:sem 
 328 top                             0  0.000  1208/ 1684 100 (100)  RUN   
  95 gps                          1748  0.298  1056/ 1476 220 (220)  w:sem 
 103 dataman                       181  0.000   704/ 1180  90 ( 90)  w:sem 
 143 sensors                     34035  4.975  1328/ 1980 250 (250)  w:sem 
 145 commander                   13973  1.890  2744/ 3548 140 (140)  w:sig 
 152 px4io                       21663  3.084  1048/ 1380 240 (240)  w:sem 
 165 mavlink_if0                 13805  1.990  1736/ 2436 100 (100)  w:sig 
 166 mavlink_rcv_if0               250  0.000  1336/ 2100 175 (175)  w:sem 
 178 mavlink_if1                 10137  1.592  1712/ 2388 100 (100)  w:sig 
 179 mavlink_rcv_if1               716  0.099  1240/ 2100 175 (175)  w:sem 
 236 log_writer_file                 0  0.000   344/ 1060  60 ( 60)  w:sem 
 217 mavlink_if2                 91961 13.532  1824/ 2388 100 (100)  w:sig 
 218 mavlink_rcv_if2               104  0.000  1536/ 2100 175 (175)  w:sem 
 234 logger                       3756  0.597  1312/ 3532 250 (250)  w:sem 
 318 commander_low_prio             24  0.000   720/ 2996  50 ( 50)  w:sem 
 283 ekf2                       120364 19.004  5112/ 5780 250 (250)  w:sem 
 285 mc_att_control              31419  4.676  1256/ 1676 250 (250)  w:sem 
 292 mc_pos_control              30007  4.975  1240/ 1876 250 (250)  w:sem 
 299 navigator                    1169  0.199  1208/ 1572 105 (105)  w:sem 

Processes: 23 total, 2 running, 21 sleeping
CPU usage: 61.19% tasks, 1.00% sched, 37.81% idle
DMA Memory: 5120 total, 1536 used 1536 peak
Uptime: 684.974s total, 270.575s idle
nsh> free
             total       used       free    largest
Mem:        236864     206368      30496      25888
nsh> 

The main difference are mavlink instances #1 and #2, where #1 consumes 0.4% more CPU load and #2 4.5% more load. This is due to faster scheduling to be able to meet higher rate demands.

[TSC21]

@LorenzMeier can you rebase this?

[LorenzMeier]

This got superseded.