[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20201201183613.GM2073444@lunn.ch>
Date: Tue, 1 Dec 2020 19:36:13 +0100
From: Andrew Lunn <andrew@...n.ch>
To: Baruch Siach <baruch@...s.co.il>
Cc: Michal Kubecek <mkubecek@...e.cz>, netdev@...r.kernel.org
Subject: Re: [PATCH ethtool] Improve error message when SFP module is missing
On Tue, Dec 01, 2020 at 08:09:21AM +0200, Baruch Siach wrote:
> ETHTOOL_GMODULEINFO request success indicates that SFP cage is present.
> Failure of ETHTOOL_GMODULEEEPROM is most likely because SFP module is
> not plugged in. Add an indication to the user as to what might be the
> reason for the failure.
>
> Signed-off-by: Baruch Siach <baruch@...s.co.il>
> ---
> ethtool.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/ethtool.c b/ethtool.c
> index 1d9067e774af..6d785f830ffa 100644
> --- a/ethtool.c
> +++ b/ethtool.c
> @@ -4856,6 +4856,7 @@ static int do_getmodule(struct cmd_context *ctx)
> err = send_ioctl(ctx, eeprom);
> if (err < 0) {
> perror("Cannot get Module EEPROM data");
> + fprintf(stderr, "SFP module not in cage?\n");
I wonder if this should be limited to ENODEV and EIO? ENOMEM or
EINVAL is probably not going to be fixed by inserting a module.
Andrew
Powered by blists - more mailing lists