![]() |
2009-06-01
, 04:14
|
Posts: 1,258 |
Thanked: 672 times |
Joined on Mar 2009
|
#2
|
![]() |
2009-06-01
, 04:16
|
Posts: 422 |
Thanked: 244 times |
Joined on Feb 2008
|
#3
|
![]() |
2009-06-01
, 05:10
|
Posts: 2,802 |
Thanked: 4,491 times |
Joined on Nov 2007
|
#4
|
What would this mean in terms of a voice call? I am assuming that freemantle doesn't have quite the same gung-ho approach with rti, and will leave a few cycles ready so I can hit "answer" on the incoming call.
CONFIG_GROUP_SCHED=y
CONFIG_FAIR_GROUP_SCHED=y
# CONFIG_RT_GROUP_SCHED is not set
CONFIG_USER_SCHED=y
# CONFIG_CGROUP_SCHED is not set
![]() |
2009-06-01
, 07:50
|
|
Posts: 357 |
Thanked: 115 times |
Joined on Sep 2007
@ Sunny England :)
|
#5
|
Yeah, undoubtedly I/O Wait, but still - a block is a block, so what are the protections in place to prevent this happening when we have something as time critical as a phone call in play?
![]() |
2009-06-01
, 07:55
|
|
Posts: 2,142 |
Thanked: 2,054 times |
Joined on Dec 2006
@ Sicily
|
#6
|
![]() |
2009-06-01
, 08:42
|
Posts: 422 |
Thanked: 244 times |
Joined on Feb 2008
|
#7
|
Application Manager is a good example of this.
What would this mean in terms of a voice call? I am assuming that freemantle doesn't have quite the same gung-ho approach with rti, and will leave a few cycles ready so I can hit "answer" on the incoming call.
Have you beagle players seen anything to suggest the scheduler in freemantle will guarantee cycles for core processes?
I am assuming it is there, but this becomes more imperetive when we are talking about a phone.
Servwise Advanced Hosting - Better, Faster, Smarter
Hum Consulting - IT and Marketing services to sing about