login
Header Space

 
 

Re: Higer latency with dynamic tick (need for an io-ondemand govenor?)

Score:
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: Woodruff, Richard <r-woodruff2@...>
Cc: Thomas Gleixner <tglx@...>, Ingo Molnar <mingo@...>, <linux-kernel@...>, <linux-pm@...>, <lenb@...>
Date: Sunday, April 20, 2008 - 8:41 am

"Woodruff, Richard" <r-woodruff2@ti.com> writes:


Are you talking about x86? 

On older x86 this effect should have been handled by the C state
algorithm taking the bus master activity register into account (which
should also trigger for interrupts)

But I think the register has been nop'ed on newer platforms 
so indeed we'll need some way to handle this.

-Andi
--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
bug seen with dynticks from CONFIG_HARDIRQS_SW_RESEND, Woodruff, Richard, (Wed May 16, 7:20 pm)
Re: bug seen with dynticks from CONFIG_HARDIRQS_SW_RESEND, Thomas Gleixner, (Thu May 17, 6:10 am)
RE: bug seen with dynticks from CONFIG_HARDIRQS_SW_RESEND, Woodruff, Richard, (Thu May 17, 4:14 pm)
RE: bug seen with dynticks from CONFIG_HARDIRQS_SW_RESEND, Thomas Gleixner, (Thu May 17, 4:38 pm)
RE: bug seen with dynticks from CONFIG_HARDIRQS_SW_RESEND, Woodruff, Richard, (Thu May 17, 6:24 pm)
RE: bug seen with dynticks from CONFIG_HARDIRQS_SW_RESEND, Thomas Gleixner, (Fri May 18, 3:49 am)
Higer latency with dynamic tick (need for an io-ondemand gov..., Woodruff, Richard, (Fri Apr 18, 11:43 am)
Re: Higer latency with dynamic tick (need for an io-ondemand..., Andi Kleen, (Sun Apr 20, 8:41 am)
RE: Higer latency with dynamic tick (need for an io-ondemand..., Woodruff, Richard, (Sun Apr 20, 10:21 am)
RE: Higer latency with dynamic tick (need for an io-ondemand..., Woodruff, Richard, (Sun Apr 20, 10:09 am)
speck-geostationary