| ??? 11/12/09 14:47 Modified: 11/12/09 14:49 Read: times |
#170733 - individual disa/ena and latencies Responding to: ???'s previous message |
Oliver Sedlacek said:
Note that in my opinion it is not the enable/disable instruction execution time that slows a system but the fact that interrupt latencies are extended that slows a system. Yes, but if the ena/disa execution itself requires atomicity guards (SFR page switching, read-modify-write of a non-bit-addressable SFR (a.k.a. more fun! ;-) )), the latencies might at the end of the day be longer than if only a single global ena/disa would be performed! JW |
| Topic | Author | Date |
| Interrupt fun | 01/01/70 00:00 | |
| at the project in front of you I see ... | 01/01/70 00:00 | |
| It's Atomic | 01/01/70 00:00 | |
| Ring buffer | 01/01/70 00:00 | |
| Diligent fun? | 01/01/70 00:00 | |
individual disa/ena and latencies | 01/01/70 00:00 |