On Wed, Sep 21, 2011 at 7:05 PM, Jack Wang <jack_wang@xxxxxxxxx> wrote:
> Hi James,
>
> Because Dan change device queue depth to 256, and as point by Mark this may
> result in starvation, so I bump the host can queue value to 510.
Hi,
Is this a theoretical or actual concern? Because the mid-layer should
be handling starvation if I understand the mechanics of the
shost->starved_list.
--
Dan
> Best regards!
>
> Jack
>
> From 3647cde88d66760ca336ad127a5b1fdf5c1d28f7 Mon Sep 17 00:00:00 2001
> From: Jack Wang <jack_wang@xxxxxxxxx>
> Date: Thu, 22 Sep 2011 09:50:54 +0800
> Subject: [PATCH] bump host can queue to follow libsas queue depth change
>
> Signed-off-by: Jack Wang <jack_wang@xxxxxxxxx>
> ---
> pm8001_defs.h | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/pm8001_defs.h b/pm8001_defs.h
> index 944afad..704b492 100644
> --- a/pm8001_defs.h
> +++ b/pm8001_defs.h
> @@ -68,7 +68,7 @@ enum port_type {
> #define PM8001_MAX_CCB 512 /* max ccbs supported */
> #define PM8001_MAX_INB_NUM 1
> #define PM8001_MAX_OUTB_NUM 1
> -#define PM8001_CAN_QUEUE 128 /* SCSI Queue depth */
> +#define PM8001_CAN_QUEUE 510 /* SCSI Queue depth */
>
> /* unchangeable hardware details */
> #define PM8001_MAX_PHYS 8 /* max. possible phys */
> --
> 1.7.4.msysgit.0
>
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
