...
ssh'd into the systems, and checked the interrupt load rates With the "intcount" command.
...
On vipers, ttyS3 is interrupt 116.
On titans, ttyS3 is interrupt 122
Interrupts/second
DSM | ttyS5-20, IRQ 3 | PC104/GPIO | ttyS2, IRQ 36 | ttyS1, 37 | ttyS3 116/122 | USB | kernel | notes |
|---|---|---|---|---|---|---|---|---|
a1 | 150-900 | 150-900 | 20 | 3.2 | 20 | 1018 | 2.6.35.9-ael1-1-viper Sep 14 10:54:19 MDT 2012 | 1,3 |
a2 | 5.4 | 5.4 | 20 | 3 | 20 | 1019 | 2.6.35.9-ael1-1-viper Sep 14 10:54:19 MDT 2012 | 3 |
a3 | 18.8 | 18.8 | 20 | 2 |
| 13.4 | 2.6.35.9-ael1-2-titan Oct 13 12:45:33 MDT 2012 |
|
a4 |
|
| 20 | 3 | 4 | 13.4 | 2.6.35.9-ael1-2-titan Sep 14 11:29:27 MDT 2012 |
|
a5 | 2300-3900 | 2300-3900 | 20 | 3 | 20 | 1020 | 2.6.35.9-ael1-1-viper Oct 4 13:21:09 MDT 2012 | 1,3 |
a6 | 5.4 | 5.4 | 20 | 3 | 20 | 1044 | 2.6.35.9-ael1-1-viper Sep 14 10:54:19 MDT 2012 | 3 |
a7 |
|
| 20 | 2 | 4 | 15 | 2.6.35.9-ael1-2-titan Sep 14 11:29:27 MDT 2012 |
|
a8 | 1500-2550 | 1500-2550 | 20 | 2 | 20 | 1020 | 2.6.35.9-ael1-1-viper Sep 14 10:54:19 MDT 2012 | 1,3 |
a9 |
|
| 20 | 2 | 20 | 1020 | 2.6.35.9-ael1-1-viper Sep 14 10:54:19 MDT 2012 | 3 |
a10 |
|
| 20 | 13 | 20 | 1020 | 2.6.35.9-ael1-1-viper Sep 14 10:54:19 MDT 2012 | 3 |
a11 |
|
| 20 | 3 | 4 | 18.6 | 2.6.35.9-ael1-2-titan Sep 14 11:29:27 MDT 2012 |
|
a12 |
|
| 20 | 2 | 3 | 15.6 | 2.6.35.9-ael1-2-titan Sep 14 11:29:27 MDT 2012 |
|
a13 |
|
| 20 | 3 | 4 | 12.8 | 2.6.35.9-ael1-2-titan Sep 14 11:29:27 MDT 2012 |
|
a14 |
|
| 20 | 3 | 20 | 1020 | 2.6.35.9-ael1-1-viper Sep 14 10:54:19 MDT 2012 | 3 |
a15 |
|
| 20 | 3 | 4 | 13.2 | 2.6.35.9-ael1-2-titan Sep 14 11:29:27 MDT 2012 |
|
a16 |
|
| 20 | 3 | 4 | 13.4 | 2.6.35.9-ael1-2-titan Sep 14 11:29:27 MDT 2012 |
|
a17 |
|
| 20 | 3 | 4 | 18.2 | 2.6.35.9-ael1-2-titan Sep 14 11:29:27 MDT 2012 |
|
a18 |
|
| 20 | 3 | 4 | 13.8 | 2.6.35.9-ael1-2-titan Sep 14 11:29:27 MDT 2012 |
|
a19 |
|
| 20 | 3 | 4 | 14.6 | 2.6.35.9-ael1-2-titan Sep 14 11:29:27 MDT 2012 |
|
c20 | 52 | 52 | 3 | 2 | 20 | 7.2 | 2.6.35.9-ael1-1-viper Sep 14 10:54:19 MDT 2012 | 4 |
m21 | 350 | 6000-7700 17500 | 3 | 21 | 20 | 14 | 2.6.35.9-ael1-1-viper Oct 3 12:12:46 MDT 2012 | 1,2,4 |
m22 | 64 | 64 | 2 | 3 |
| 3 | 2.6.35.9-ael1-2-titan Oct 2 21:50:26 MDT 2012 | 4 |
Notes:
1: Some vipers have a large pc104 interrupt loadrate. Occurs with both Sep 14 (a1,a8) and the Oct 4 (a5) kernels.
2: Not completely sure why on m21 the IRQ3 load rate is less than the pc104 load. Perhaps that version of the pc104 irq routine exits if pending bits are 0.
3: High USB interrupts are seen on vipers with bluetooth radios interfaced via USB. Not on Titans, even though on a1-19 titans, the bluetooth and flash drives are USB. On a1-19 Vipers, only bluetooth radios are on USB. Assume it is due the different USB interface chips.
4: On c20,m21, m22, USB is used only for flash drives
The big question is why the huge PC104 interrupt load rates on a1, a5, a8 and m21 vipers. Previously traced this to a floating CTS/RTS line, but I don't think that is the current reason. Saw high PC104 interrupt load rate on m21 even with the ribbon cables disconnected from the Emerald headers! Will have to investigate those systems back in the lab.
...