Re: [RFC/PATCH, 2/4] readX_check() performance evaluation II

All of lore.kernel.org
 help / color / mirror / Atom feed

From: Andi Kleen <ak@suse.de>
To: Andi Kleen <ak@suse.de>
Cc: torvalds@osdl.org, willy@debian.org,
	ishii.hironobu@jp.fujitsu.com, linux-kernel@vger.kernel.org,
	linux-ia64@vger.kernel.org
Subject: Re: [RFC/PATCH, 2/4] readX_check() performance evaluation II
Date: Wed, 28 Jan 2004 20:19:59 +0000	[thread overview]
Message-ID: <20040128211959.0d367c30.ak@suse.de> (raw)
In-Reply-To: <20040128211554.0cc890fb.ak@suse.de>

On Wed, 28 Jan 2004 21:15:54 +0100
Andi Kleen <ak@suse.de> wrote:

>
> > 
> > Just set a flag aka "may need checking", and let the check be done by the 
> > actual "read_pcix_error()" code.
> 
> Where would you put the flag? 
> 
> Doing it global may give false errors for the wrong device with async MCEs
> and on SMP.
> 
> For putting it into the pci_dev you need to take logs to walk the list.
> If you delay it to a softirq for safely getting the lock it would be set too late.
> 
> Putting it into a different table indexed by pci index would be also racy 
> with hotplug.

... to follow up myself ...

I suppose moving the pci_dev lists to RCU could make the flag in pci-dev work. But it would be still
a bit tricky with preemptive kernels.

-Andi

WARNING: multiple messages have this Message-ID (diff)

From: Andi Kleen <ak@suse.de>
To: Andi Kleen <ak@suse.de>
Cc: torvalds@osdl.org, willy@debian.org,
	ishii.hironobu@jp.fujitsu.com, linux-kernel@vger.kernel.org,
	linux-ia64@vger.kernel.org
Subject: Re: [RFC/PATCH, 2/4] readX_check() performance evaluation II
Date: Wed, 28 Jan 2004 21:19:59 +0100	[thread overview]
Message-ID: <20040128211959.0d367c30.ak@suse.de> (raw)
In-Reply-To: <20040128211554.0cc890fb.ak@suse.de>

On Wed, 28 Jan 2004 21:15:54 +0100
Andi Kleen <ak@suse.de> wrote:

>
> > 
> > Just set a flag aka "may need checking", and let the check be done by the 
> > actual "read_pcix_error()" code.
> 
> Where would you put the flag? 
> 
> Doing it global may give false errors for the wrong device with async MCEs
> and on SMP.
> 
> For putting it into the pci_dev you need to take logs to walk the list.
> If you delay it to a softirq for safely getting the lock it would be set too late.
> 
> Putting it into a different table indexed by pci index would be also racy 
> with hotplug.

... to follow up myself ...

I suppose moving the pci_dev lists to RCU could make the flag in pci-dev work. But it would be still
a bit tricky with preemptive kernels.

-Andi

next prev parent reply	other threads:[~2004-01-28 20:19 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-01-28  1:54 [RFC/PATCH, 2/4] readX_check() performance evaluation Hironobu Ishii
2004-01-28  1:54 ` Hironobu Ishii
2004-01-28  2:55 ` Linus Torvalds
2004-01-28  2:55   ` Linus Torvalds
2004-01-28  4:44   ` Paul Mackerras
2004-01-28  4:44     ` Paul Mackerras
2004-01-28  8:58   ` Russell King
2004-01-28 16:15     ` Linus Torvalds
2004-01-28 16:15       ` Linus Torvalds
2004-01-28 17:01     ` Grant Grundler
2004-01-28 17:01       ` Grant Grundler
2004-01-28 18:20   ` Matthew Wilcox
2004-01-28 19:19     ` Andi Kleen
2004-01-28 19:19       ` Andi Kleen
2004-01-28 19:33     ` Linus Torvalds
2004-01-28 19:33       ` Linus Torvalds
2004-01-28 19:40       ` Andi Kleen
2004-01-28 19:40         ` Andi Kleen
2004-01-28 20:06         ` Linus Torvalds
2004-01-28 20:06           ` Linus Torvalds
2004-01-28 20:15           ` Andi Kleen
2004-01-28 20:15             ` Andi Kleen
2004-01-28 20:19             ` Andi Kleen [this message]
2004-01-28 20:19               ` [RFC/PATCH, 2/4] readX_check() performance evaluation II Andi Kleen
2004-01-28 20:28             ` [RFC/PATCH, 2/4] readX_check() performance evaluation Linus Torvalds
2004-01-28 20:28               ` Linus Torvalds
2004-01-28 20:30               ` Linus Torvalds
2004-01-28 20:30                 ` Linus Torvalds
2004-01-28 21:09               ` Andi Kleen
2004-01-28 21:09                 ` Andi Kleen
2004-01-28 21:43                 ` Linus Torvalds
2004-01-28 21:43                   ` Linus Torvalds
2004-01-28 21:52                   ` Andi Kleen
2004-01-28 21:52                     ` Andi Kleen
2004-01-28 22:21                     ` Linus Torvalds
2004-01-28 22:21                       ` Linus Torvalds
2004-01-28 22:39                       ` Andi Kleen
2004-01-28 22:39                         ` Andi Kleen
2004-01-28 22:59                         ` Linus Torvalds
2004-01-28 22:59                           ` Linus Torvalds
2004-01-29 12:24                           ` Hironobu Ishii
2004-01-29 12:24                             ` Hironobu Ishii
2004-01-28 22:15                   ` David S. Miller
2004-01-28 22:15                     ` David S. Miller
2004-01-28  3:09 ` Matthew Wilcox
2004-01-28  3:09   ` Matthew Wilcox

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=20040128211959.0d367c30.ak@suse.de \
    --to=ak@suse.de \
    --cc=ishii.hironobu@jp.fujitsu.com \
    --cc=linux-ia64@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=torvalds@osdl.org \
    --cc=willy@debian.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.