kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Gleb Natapov <gleb@redhat.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Avi Kivity <avi@redhat.com>,
	kvm@vger.kernel.org, Jan Kiszka <jan.kiszka@web.de>
Subject: Re: [RFC PATCH 0/2] irq destination caching prototype
Date: Mon, 13 Aug 2012 13:16:08 +0300	[thread overview]
Message-ID: <20120813101608.GS3341@redhat.com> (raw)
In-Reply-To: <20120813101246.GA16147@redhat.com>

On Mon, Aug 13, 2012 at 01:12:46PM +0300, Michael S. Tsirkin wrote:
> On Mon, Aug 13, 2012 at 12:36:41PM +0300, Avi Kivity wrote:
> > On 08/13/2012 12:16 PM, Gleb Natapov wrote:
> > > Here is a quick prototype of what we discussed yesterday. This one
> > > caches only MSI interrupts for now. The obvious problem is that not
> > > all interrupts (namely IPIs and MSIs using KVM_CAP_SIGNAL_MSI) use irq
> > > routing table, so they cannot be cached.
> > 
> > We can have a small rcu-managed hash table to look those up.
> 
> Yes but how small? We probably need at least one entry
> per vcpu, no?
> 
One entry? We will spend more time managing it than injecting interrupts
:) ideally we need entry for each IPI sent and for each potential MSI
from userspace. What happens when hash table is full? We stop caching or
invalidate old entries? If later then cache can go valid->invalid which
may complicate the code.

--
			Gleb.

  reply	other threads:[~2012-08-13 10:16 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-08-13  9:16 [RFC PATCH 0/2] irq destination caching prototype Gleb Natapov
2012-08-13  9:16 ` [RFC PATCH 1/2] Call irq_rt callback under rcu_read_lock() Gleb Natapov
2012-08-13  9:16 ` [RFC PATCH 2/2] Cache msi irq destination Gleb Natapov
2012-08-13  9:32   ` Avi Kivity
2012-08-13  9:34     ` Gleb Natapov
2012-08-13  9:34 ` [RFC PATCH 0/2] irq destination caching prototype Michael S. Tsirkin
2012-08-13  9:36   ` Gleb Natapov
2012-08-13  9:46     ` Michael S. Tsirkin
2012-08-13  9:48       ` Gleb Natapov
2012-08-13  9:36 ` Avi Kivity
2012-08-13 10:12   ` Michael S. Tsirkin
2012-08-13 10:16     ` Gleb Natapov [this message]
2012-08-13 10:21       ` Avi Kivity
2012-08-13 10:24         ` Gleb Natapov
2012-08-13 10:31           ` Avi Kivity
2012-08-13 10:35             ` Gleb Natapov
2012-08-13 10:38             ` Michael S. Tsirkin
2012-08-13 10:58               ` Avi Kivity
2012-08-13 11:01                 ` Gleb Natapov
2012-08-13 11:03                   ` Avi Kivity
2012-08-13 11:12                     ` Gleb Natapov
2012-08-13 11:22                       ` Michael S. Tsirkin
2012-08-13 11:29                         ` Gleb Natapov
2012-08-13 11:43                           ` Gleb Natapov
2012-08-13 12:14                             ` Avi Kivity
2012-08-13 11:30                       ` Avi Kivity
2012-08-13 11:41                         ` Gleb Natapov
2012-08-13 12:13                           ` Avi Kivity
2012-08-13 12:59                           ` Michael S. Tsirkin
2012-08-13 11:19                     ` Michael S. Tsirkin
2012-08-13  9:43 ` Avi Kivity
2012-08-13  9:51   ` Michael S. Tsirkin
2012-08-13  9:53     ` Gleb Natapov
2012-08-13 10:33   ` Gleb Natapov

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=20120813101608.GS3341@redhat.com \
    --to=gleb@redhat.com \
    --cc=avi@redhat.com \
    --cc=jan.kiszka@web.de \
    --cc=kvm@vger.kernel.org \
    --cc=mst@redhat.com \
    /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;
as well as URLs for NNTP newsgroup(s).