LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Andrew Morton <akpm@linux-foundation.org>
To: "Ed L. Cashin" <ecashin@coraid.com>
Cc: linux-kernel@vger.kernel.org, Greg K-H <greg@kroah.com>
Subject: Re: [PATCH 10/12] add module parameter for users who need more outstanding I/O
Date: Mon, 2 Jul 2007 21:41:00 -0700	[thread overview]
Message-ID: <20070702214100.ed81e921.akpm@linux-foundation.org> (raw)
In-Reply-To: <772666addf79d3a3675f5f460228532fcd8f7364.1182883861.git.ecashin@coraid.com>

On Tue, 26 Jun 2007 14:50:12 -0400 "Ed L. Cashin" <ecashin@coraid.com> wrote:

> Add module parameter for users who need more outstanding I/O.
> 

why?

I assume that there is some performance benefit to permitting some more
outstanding IO, but that's just a wild guess.

Assuming that the guess is right, some explanation of the effects of
altering this tunable would be appropriate.


> ---
>  drivers/block/aoe/aoecmd.c |   10 +++++++---
>  1 files changed, 7 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c
> index 8d6540b..9de0024 100644
> --- a/drivers/block/aoe/aoecmd.c
> +++ b/drivers/block/aoe/aoecmd.c
> @@ -17,6 +17,11 @@ static int aoe_deadsecs = 60 * 3;
>  module_param(aoe_deadsecs, int, 0644);
>  MODULE_PARM_DESC(aoe_deadsecs, "After aoe_deadsecs seconds, give up and fail dev.");
>  
> +static int aoe_maxout = 16;
> +module_param(aoe_maxout, int, 0644);
> +MODULE_PARM_DESC(aoe_maxout,
> +	"Only aoe_maxout outstanding packets for every MAC on eX.Y.");
> +
>  static struct sk_buff *
>  new_skb(ulong len)
>  {
> @@ -967,7 +972,6 @@ aoecmd_cfg_rsp(struct sk_buff *skb)
>  	struct aoeif *ifp;
>  	ulong flags, sysminor, aoemajor;
>  	struct sk_buff *sl;
> -	enum { MAXFRAMES = 16 };
>  	u16 n;
>  
>  	h = aoe_hdr(skb);
> @@ -992,8 +996,8 @@ aoecmd_cfg_rsp(struct sk_buff *skb)
>  	}
>  
>  	n = be16_to_cpu(ch->bufcnt);
> -	if (n > MAXFRAMES)	/* keep it reasonable */
> -		n = MAXFRAMES;
> +	if (n > aoe_maxout)	/* keep it reasonable */
> +		n = aoe_maxout;
>  
>  	d = aoedev_by_sysminor_m(sysminor);
>  	if (d == NULL) {
> -- 
> 1.5.2.1
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

  reply	other threads:[~2007-07-03  4:41 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-06-26 18:50 [PATCH 01/12] bring driver version number to 47 Ed L. Cashin
2007-06-26 18:50 ` [PATCH 03/12] mac_addr: avoid 64-bit arch compiler warnings Ed L. Cashin
2007-06-26 18:50 ` [PATCH 02/12] handle multiple network paths to AoE device Ed L. Cashin
2007-07-03  4:29   ` Andrew Morton
2007-07-11 14:46     ` Ed L. Cashin
2007-07-16 22:17     ` [PATCH] stacked ifs (was Re: [PATCH 02/12] handle multiple network paths to AoE device) Ed L. Cashin
2007-07-16 22:31       ` Andrew Morton
2007-07-17  0:01         ` Greg KH
2007-07-18 15:24           ` Jan Engelhardt
2007-06-26 18:50 ` [PATCH 05/12] eliminate goto and improve readability Ed L. Cashin
2007-06-26 18:50 ` [PATCH 07/12] use a dynamic pool of sk_buffs to keep up with fast targets Ed L. Cashin
2007-07-03  4:36   ` Andrew Morton
2007-07-03  4:40     ` David Miller
2007-07-03 18:45       ` Matt Mackall
2007-07-03 19:18         ` Stephen Hemminger
2007-07-06 17:09       ` Ed L. Cashin
2007-06-26 18:50 ` [PATCH 06/12] user can ask driver to forget previously detected devices Ed L. Cashin
2007-06-26 18:50 ` [PATCH 04/12] clean up udev configuration example Ed L. Cashin
2007-06-26 18:50 ` [PATCH 12/12] the aoeminor doesn't need a long format Ed L. Cashin
2007-06-26 19:51   ` Randy Dunlap
2007-06-26 19:59     ` Ed L. Cashin
2007-06-26 18:50 ` [PATCH 10/12] add module parameter for users who need more outstanding I/O Ed L. Cashin
2007-07-03  4:41   ` Andrew Morton [this message]
2007-06-26 18:50 ` [PATCH 11/12] remove extra space in prototypes for consistency Ed L. Cashin
2007-06-26 18:50 ` [PATCH 09/12] remove race between use and initialization of locks Ed L. Cashin
2007-07-03  4:38   ` Andrew Morton
2007-06-26 18:50 ` [PATCH 08/12] only schedule work once Ed L. Cashin
2007-07-03  4:37   ` Andrew Morton

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=20070702214100.ed81e921.akpm@linux-foundation.org \
    --to=akpm@linux-foundation.org \
    --cc=ecashin@coraid.com \
    --cc=greg@kroah.com \
    --cc=linux-kernel@vger.kernel.org \
    --subject='Re: [PATCH 10/12] add module parameter for users who need more outstanding I/O' \
    /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

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).