[dm-devel] [PATCH RFC] multipath-tools: update "Generic NVMe" options in hwtable

[Martin Wilck]

On Wed, 2022-03-30 at 21:49 +0200, Xose Vazquez Perez wrote:
> Removed values were default.

This is wrong for uid_attribute. You can see it when you run "make
test", which fails with your patch. It's also wrong for
retain_hwhandler, but that entry should indeed be removed, because
keeping it there is misleading at best. hwhandlers have no meaning for
NVMe, and retain_hwhandler is enforced to be true an all kernels since
4.3.

Regards,
Martin

> Check ANA, else fall back to CONST(multibus).
> 
> Cc: Martin Wilck <mwilck at suse.com>
> Cc: Benjamin Marzinski <bmarzins at redhat.com>
> Cc: Christophe Varoqui <christophe.varoqui at opensvc.com>
> Cc: DM-DEVEL ML <dm-devel at redhat.com>
> Signed-off-by: Xose Vazquez Perez <xose.vazquez at gmail.com>
> ---
>  libmultipath/hwtable.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/libmultipath/hwtable.c b/libmultipath/hwtable.c
> index 0e1c0a41..81dca0d5 100644
> --- a/libmultipath/hwtable.c
> +++ b/libmultipath/hwtable.c
> @@ -88,9 +88,9 @@ static struct hwentry default_hw[] = {
>                 /* Generic NVMe */
>                 .vendor        = "NVME",
>                 .product       = ".*",
> -               .uid_attribute = DEFAULT_NVME_UID_ATTRIBUTE,
> -               .checker_name  = NONE,
> -               .retain_hwhandler = RETAIN_HWHANDLER_OFF,
> +               .pgpolicy      = GROUP_BY_PRIO,
> +               .pgfailback    = -FAILBACK_IMMEDIATE,
> +               .no_path_retry = NO_PATH_RETRY_QUEUE,
>         },
>         /*
>          * Apple