netdev - RE: [PATCH net-next] ethtool: add netlink based get rxfh support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Fri, 4 Nov 2022 19:04:45 +0000
From:   "Mogilappagari, Sudheer" <sudheer.mogilappagari@...el.com>
To:     Andrew Lunn <andrew@...n.ch>
CC:     "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "kuba@...nel.org" <kuba@...nel.org>,
        "corbet@....net" <corbet@....net>,
        "mkubecek@...e.cz" <mkubecek@...e.cz>,
        "Samudrala, Sridhar" <sridhar.samudrala@...el.com>,
        "Nguyen, Anthony L" <anthony.l.nguyen@...el.com>
Subject: RE: [PATCH net-next] ethtool: add netlink based get rxfh support



> -----Original Message-----
> From: Andrew Lunn <andrew@...n.ch>
> Subject: Re: [PATCH net-next] ethtool: add netlink based get rxfh
> support
> 
> > +static int rxfh_prepare_data(const struct ethnl_req_info *req_base,
> > +			     struct ethnl_reply_data *reply_base,
> > +			     struct genl_info *info)
> > +{
> 
> ...
> 
> > +
> > +	ret = ethnl_ops_begin(dev);
> > +	if (ret < 0)
> > +		return ret;
> > +
> > +	/* Some drivers don't handle rss_context */
> > +	if (rxfh->rss_context && !ops->get_rxfh_context)
> > +		return -EOPNOTSUPP;
> 
> You called ethnl_ops_begin(). Just returning here is going to mess up
> rumtime power control, and any driver which expects is
> ethtool_ops->complete() call to be called.
> 
> 	Andrew

Hi Andrew,
Had used other get implementations as reference (which return early 
since _dev_get is not used). However, this patch uses dev_get/put 
due to input parameter. Will fix in v2. 

Got a question wrt rtnl_lock usage. I see lock is acquired for SET
operations and not for GET operations. Is rtnl_lock needed in this
case due to slightly different flow than rest of GET ops?

I apologize regarding lkp error. Had fixed the issue but sent out
old patch by mistake.

Thanks,
Sudheer