Re: SATA question

From: John-Paul Stewart <jpstewart@xxxxxxxxxxxxxxxx>
Date: Mon, 09 Apr 2007 10:37:05 -0400

AZ Nomad wrote:

On Sun, 08 Apr 2007 15:26:25 -0400, John-Paul Stewart <jpstewart@xxxxxxxxxxxxxxxx> wrote:

The IRQ balancing under discussion is a method for having interrupts handled by the different processors in an SMP system.

[snip]

Then it is irrelevent to this thread.

In message-ID <131bitk83tsv24f@xxxxxxxxxxxxxxxxxx> the original poster wrote in part, "And WTF are all my IRQ on one CPU? It's a Core 2 Duo.... Maybe it's a kernel issue?" Thus IRQ balancing seems relevant to me!
.

References:
- SATA question
  - From: Captain Dondo
- Re: SATA question
  - From: AZ Nomad
- Re: SATA question
  - From: Captain Dondo
- Re: SATA question
  - From: Måns Rullgård
- Re: SATA question
  - From: John-Paul Stewart
- Re: SATA question
  - From: Captain Dondo
- Re: SATA question
  - From: John-Paul Stewart
- Re: SATA question
  - From: Captain Dondo
- Re: SATA question
  - From: John-Paul Stewart
- Re: SATA question
  - From: AZ Nomad
- Re: SATA question
  - From: John-Paul Stewart
- Re: SATA question
  - From: AZ Nomad

Prev by Date: Re: Question regarding persistent mount points for multiple USB drives
Next by Date: Motehrboard comments
Previous by thread: Re: SATA question
Next by thread: Re: SATA question
Index(es):
- Date
- Thread

Relevant Pages

Re: Irq architecture for multi-core network driver.
... the way the default linux kernel configures today's many core ... the above unless you run irqbalance or set smp_affinity by hand. ... How often and how much does irq balancing hurt us. ... open, which cpus those applications run on, and which core the irqs arrive ...
(Linux-Kernel)
[patch 00/75] genirq: Overhaul for 2.6.39
... Further encapsulation of the core state ... Stop setaffinity blindly manipulating affinity mask ... Namespace cleanup ... IRQ_* flag space. ...
(Linux-Kernel)
[SMP BUG] [clockevents: i386 drivers patch] introduces irqbalance-does-not-work-pr
... [clockevents: i386 drivers patch] ... (The irq is not distributed to two Core ... ,most of the irq is distributed to CPU1) ...
(Linux-Kernel)
[patch 3/8] xtensa: variant irq set callbacks
... Allow the core variant code to provide irq enable/disable callbacks. ... +static inline void variant_irq_disable(unsigned int irq) ... +static void xtensa_irq_disable ...
(Linux-Kernel)
Re: two NIC on 2 core system (scheduling problem)
... first is em and second is bge. ... I wonder why system put irq processes almost always to one core. ... In Linux it is possible to use SMP IRQ Affinity to attach IRQs to different CPUs. ... I wonder if there any way to do the same with FreeBSD? ...
(freebsd-net)