mail archive of the barebox mailing list
 help / color / mirror / Atom feed
From: Sascha Hauer <s.hauer@pengutronix.de>
To: Sam Ravnborg <sam@ravnborg.org>
Cc: Barebox List <barebox@lists.infradead.org>
Subject: Re: [PATCH 12/21] net: dhcp: rework
Date: Tue, 28 Nov 2017 08:54:08 +0100	[thread overview]
Message-ID: <20171128075408.esegiqkhcrk4fbvq@pengutronix.de> (raw)
In-Reply-To: <20171125170251.GC25425@ravnborg.org>

On Sat, Nov 25, 2017 at 06:02:51PM +0100, Sam Ravnborg wrote:
> Hi Sasha.
> 
> On Fri, Nov 24, 2017 at 09:12:28AM +0100, Sascha Hauer wrote:
> > The DHCP code is a mess. It is not clear which options are sent to the
> > server and which options are returned from the server. Also environment
> > variables are read from and written to all over the place.
> > 
> > This patch cleans this up. There now is struct dhcp_req_param which is
> > used for options sent to the server and struct dhcp_result which contains
> > the values sent from the server. The values from the server are written
> > to the barebox variables in a single place. Also it's now possible to
> > call the dhcp code without modifying barebox variables at all, storing
> > the result only in the dhcp result struct.
> 
> Browsing the code gave a nice impression that things
> got much simpler.
> Noticed one thing.
> 
> This struct is not the smallest..
> > +
> > +struct dhcp_result {
> > +	IPaddr_t ip;
> > +	IPaddr_t netmask;
> > +	IPaddr_t gateway;
> > +	IPaddr_t nameserver;
> > +	IPaddr_t serverip;
> > +	char *hostname;
> > +	char *domainname;
> > +	char *rootpath;
> > +	char *devicetree;
> > +	char *bootfile;
> > +	char *tftp_server_name;
> > +	uint32_t leasetime;
> >  };
> 
> > +struct dhcp_req_param dhcp_param;
> > +struct dhcp_result *dhcp_result;
> Is this why the struct is xalloc()'er and not a static variable like dhcp_param?
> Maybe the lifetime of the two differs - which may also explain it.

Yes, indeed, it's the lifetime. dhcp_param is static because it's used
only internally during lifetime of the dhcp request. dhcp_result is
returned from the dhcp code, and I don't like to return static data
which potentially gets overwritten in the nect dhcp call.
(IMO It would be even nicer if the dhcp code didn't use static data at
all, maybe next time).

Sascha

-- 
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

  reply	other threads:[~2017-11-28  7:54 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-24  8:12 Networking updates Sascha Hauer
2017-11-24  8:12 ` [PATCH 01/21] driver: Add device_detect_all() function Sascha Hauer
2017-11-24 23:34   ` Sam Ravnborg
2017-11-28  7:58     ` Sascha Hauer
2017-11-28 22:52       ` Sam Ravnborg
2017-11-24  8:12 ` [PATCH 02/21] nvvar: when setting a nvvar to NULL just free the content Sascha Hauer
2017-11-24  8:12 ` [PATCH 03/21] net: Make domainname and nameserver globalvars Sascha Hauer
2017-11-24  8:12 ` [PATCH 04/21] net: Add functions to get/set nameserver and domainname Sascha Hauer
2017-11-24 23:47   ` Sam Ravnborg
2017-11-27 15:18     ` Sascha Hauer
2017-11-24  8:12 ` [PATCH 05/21] net: introduce global.net.server Sascha Hauer
2017-11-24  8:12 ` [PATCH 06/21] net: dhcp: Do not overwrite serverip if it is valid Sascha Hauer
2017-11-25 16:36   ` Sam Ravnborg
2017-11-28  7:42     ` Sascha Hauer
2017-11-24  8:12 ` [PATCH 07/21] net: Use a single gateway Sascha Hauer
2017-11-24  8:12 ` [PATCH 08/21] net: allow udp connections on specified network device Sascha Hauer
2017-11-25 16:41   ` Sam Ravnborg
2017-11-27 14:45     ` Sascha Hauer
2017-11-24  8:12 ` [PATCH 09/21] net: dhcp: Allow to specify " Sascha Hauer
2017-11-25 16:46   ` Sam Ravnborg
2017-11-28  7:50     ` Sascha Hauer
2017-11-24  8:12 ` [PATCH 10/21] net: dhcp: avoid unnecessary casts Sascha Hauer
2017-11-24  8:12 ` [PATCH 11/21] net: dhcp: Coding style fixes Sascha Hauer
2017-11-24  8:12 ` [PATCH 12/21] net: dhcp: rework Sascha Hauer
2017-11-25 17:02   ` Sam Ravnborg
2017-11-28  7:54     ` Sascha Hauer [this message]
2017-11-24  8:12 ` [PATCH 13/21] net: Pick network device based on IP settings Sascha Hauer
2017-11-24  8:12 ` [PATCH 14/21] net: remove "current" network device Sascha Hauer
2017-11-24  8:12 ` [PATCH 15/21] net: ifup: Factor out a eth_discover function Sascha Hauer
2017-11-24  8:12 ` [PATCH 16/21] ifup: Use dhcp C API rather than running command Sascha Hauer
2017-11-24  8:12 ` [PATCH 17/21] net: Provide new way to configure network devices Sascha Hauer
2017-11-24  8:12 ` [PATCH 18/21] net: update network docs Sascha Hauer
2017-11-24  8:12 ` [PATCH 19/21] net: environment: remove ethx setup files Sascha Hauer
2017-11-24  8:12 ` [PATCH 20/21] net: environment: update automounts Sascha Hauer
2017-11-24  8:12 ` [PATCH 21/21] defaultenv: Add README for new network config Sascha Hauer

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=20171128075408.esegiqkhcrk4fbvq@pengutronix.de \
    --to=s.hauer@pengutronix.de \
    --cc=barebox@lists.infradead.org \
    --cc=sam@ravnborg.org \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox