From: Arjan van de Ven <arjan@infradead.org>
To: Linus Torvalds <torvalds@osdl.org>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>,
Muli Ben-Yehuda <muli@il.ibm.com>, Ingo Molnar <mingo@elte.hu>,
Thomas Gleixner <tglx@linutronix.de>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>,
Rajesh Shah <rajesh.shah@intel.com>, Andi Kleen <ak@muc.de>,
"Protasevich, Natalie" <Natalie.Protasevich@UNISYS.com>,
"Luck, Tony" <tony.luck@intel.com>, Andrew Morton <akpm@osdl.org>,
Linux-Kernel <linux-kernel@vger.kernel.org>,
Badari Pulavarty <pbadari@gmail.com>
Subject: Re: 2.6.19-rc1 genirq causes either boot hang or "do_IRQ: cannot handle IRQ -1"
Date: Sat, 07 Oct 2006 21:33:05 +0200 [thread overview]
Message-ID: <1160249585.3000.159.camel@laptopd505.fenrus.org> (raw)
In-Reply-To: <Pine.LNX.4.64.0610071154510.3952@g5.osdl.org>
> This is one of those age-old questions: in _theory_ you can do a better
> job in software, but in _practice_ it's just too damn expensive and
> complicated to do a perfect job especially with dynamic decisions, so in
> _practice_ it tends to be better to let hardware make some of the
> decisions.
it seems the right mix at this time is to have the software select the
package, and the hardware pick the core within the package.
Or rather, the software picks which cache domain (and I only count the
largest cache, not L1) and the hardware then has the freedom to do the
right thing inside that. Binding interrupts to a cache domain seems to
be still the right strategy (at least for frequent interrupts like
networking), but to do that right more higher level info is needed than
that the hw has in general. Within the package... it's the opposite
ballgame.
--
if you want to mail me at work (you don't), use arjan (at) linux.intel.com
next prev parent reply other threads:[~2006-10-07 19:33 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-10-05 21:22 Muli Ben-Yehuda
2006-10-06 15:14 ` Eric W. Biederman
2006-10-06 15:50 ` Muli Ben-Yehuda
2006-10-06 16:20 ` Muli Ben-Yehuda
2006-10-06 19:00 ` Andrew Vasquez
2006-10-06 19:42 ` Andrew Morton
2006-10-06 20:02 ` Andrew Vasquez
2006-10-06 20:15 ` Linus Torvalds
2006-10-06 20:18 ` Andrew Morton
2006-10-06 20:42 ` Muli Ben-Yehuda
2006-10-06 20:14 ` Linus Torvalds
2006-10-06 19:53 ` Benjamin LaHaise
2006-10-06 17:47 ` Eric W. Biederman
2006-10-06 20:23 ` Muli Ben-Yehuda
2006-10-06 23:42 ` Eric W. Biederman
2006-10-07 8:03 ` Muli Ben-Yehuda
2006-10-07 16:52 ` Eric W. Biederman
2006-10-07 17:59 ` Muli Ben-Yehuda
2006-10-08 13:39 ` [PATCH 0/3] x86_64 irq fixes Eric W. Biederman
2006-10-08 13:41 ` [PATCH 1/3] i386/x86_64: FIX pci_enable_irq to set dev->irq to the irq number Eric W. Biederman
2006-10-08 13:43 ` [PATCH 2/3] i386/x86_64: Remove global IO_APIC_VECTOR Eric W. Biederman
2006-10-08 13:47 ` [PATCH 3/3] x86_64 irq: Allocate a vector across all cpus for genapic_flat Eric W. Biederman
2006-10-08 19:01 ` Muli Ben-Yehuda
2006-10-08 18:59 ` [PATCH 0/3] x86_64 irq fixes Muli Ben-Yehuda
2006-10-07 19:03 ` 2.6.19-rc1 genirq causes either boot hang or "do_IRQ: cannot handle IRQ -1" Linus Torvalds
2006-10-07 19:33 ` Arjan van de Ven [this message]
2006-10-07 19:57 ` Linus Torvalds
2006-10-09 6:06 ` Eric W. Biederman
2006-10-09 7:40 ` Arjan van de Ven
2006-10-09 14:46 ` Eric W. Biederman
2006-10-09 15:28 ` Protasevich, Natalie
2006-10-09 15:39 ` Arjan van de Ven
2006-10-09 16:02 ` 2.6.19-rc1 genirq causes either boot hang or "do_IRQ: cannothandle " Protasevich, Natalie
2006-10-07 20:24 ` 2.6.19-rc1 genirq causes either boot hang or "do_IRQ: cannot handle " Eric W. Biederman
2006-10-06 16:02 ` Linus Torvalds
2006-10-06 17:22 ` Eric W. Biederman
2006-10-06 18:08 ` Linus Torvalds
2006-10-06 18:48 ` Eric W. Biederman
2006-10-09 5:41 ` [PATCH 1/1] x86_64 irq: Scream but don't die if we receive an unexpected irq Eric W. Biederman
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1160249585.3000.159.camel@laptopd505.fenrus.org \
--to=arjan@infradead.org \
--cc=Natalie.Protasevich@UNISYS.com \
--cc=ak@muc.de \
--cc=akpm@osdl.org \
--cc=benh@kernel.crashing.org \
--cc=ebiederm@xmission.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=muli@il.ibm.com \
--cc=pbadari@gmail.com \
--cc=rajesh.shah@intel.com \
--cc=tglx@linutronix.de \
--cc=tony.luck@intel.com \
--cc=torvalds@osdl.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®