Message ID | 1381322418-1349-1-git-send-email-antonio@meshcoding.com |
---|---|
State | Not Applicable, archived |
Headers | show |
From: Antonio Quartulli <antonio@meshcoding.com> Date: Wed, 9 Oct 2013 14:40:02 +0200 > Hello David, > > this is a set of changes intended for net-next/linux-3.13. > > As introduced one year ago (http://article.gmane.org/gmane.org.freifunk.batman/8484) > in the last period we have been busy working on giving our batman-adv > protocol/packet-format all those means to get stable and avoid incompatibilities > in the near future (this is also why we did not send many changes for > linux-3.12..). > > With this patchset we are finally switching to a new compatibility version and > within this development cycle I will send you all those changes aimed to > introduce all the needed mechanism to avoid any further compatibility breakage. > > First of all, in this batch you have the new TVLV code (Type Version Value > Length) which is the first very important milestone towards our goal[1]. > With this mechanism we will be able to improve/change the packet contents layout > (used by the various features) without breaking compatibility with older > protocol versions anymore. > The development of this new component has been started by Spyros Gasteratos > during his Google Summer of Code 2012 and then adjusted, finished and submitted > by Marek Lindner. > > Together with it, you have a set of improvements that we are allowed to apply > thanks to the switching to the new compat version. > These changes are: > - adapt the existing features to make them use the new TVLV mechanism > - move from CRC16 to CRC32c for table correctness check in the TT component > - reorder some of the flags sent over the wire > - remove the __packed attribute from the OGM packet struct. > > Moreover, with this patchset we are totally removing our VISualisation component > that is now implemented in userspace. > > Please pull let me know of any problem. Pulled, thanks.
From: David Miller <davem@davemloft.net> Date: Wed, 09 Oct 2013 13:56:52 -0400 (EDT) > Pulled, thanks. Please don't send me stuff that doesn't build: net/batman-adv/translation-table.c: In function ‘batadv_send_roam_adv’: net/batman-adv/translation-table.c:2210:49: warning: left-hand operand of comma expression has no effect [-Wunused-value] net/batman-adv/translation-table.c:2211:21: warning: left-hand operand of comma expression has no effect [-Wunused-value] net/batman-adv/translation-table.c:2210:6: warning: statement with no effect [-Wunused-value] net/batman-adv/translation-table.c:2211:29: error: expected ‘;’ before ‘)’ token net/batman-adv/translation-table.c:2211:29: error: expected statement before ‘)’ toke That looks like: /* before going on we have to check whether the client has * already roamed to us too many times */ if (!batadv_tt_check_roam_count(bat_priv, client)) goto out; "Sending ROAMING_ADV to %pM (client %pM)\n", orig_node->orig, client); Looks like a merge problem to me.
On Wed, Oct 09, 2013 at 02:53:06PM -0400, David Miller wrote: > From: David Miller <davem@davemloft.net> > Date: Wed, 09 Oct 2013 13:56:52 -0400 (EDT) > > > Pulled, thanks. > > Please don't send me stuff that doesn't build: > > net/batman-adv/translation-table.c: In function ‘batadv_send_roam_adv’: > net/batman-adv/translation-table.c:2210:49: warning: left-hand operand of comma expression has no effect [-Wunused-value] > net/batman-adv/translation-table.c:2211:21: warning: left-hand operand of comma expression has no effect [-Wunused-value] > net/batman-adv/translation-table.c:2210:6: warning: statement with no effect [-Wunused-value] > net/batman-adv/translation-table.c:2211:29: error: expected ‘;’ before ‘)’ token > net/batman-adv/translation-table.c:2211:29: error: expected statement before ‘)’ toke > > That looks like: > > /* before going on we have to check whether the client has > * already roamed to us too many times > */ > if (!batadv_tt_check_roam_count(bat_priv, client)) > goto out; > > "Sending ROAMING_ADV to %pM (client %pM)\n", > orig_node->orig, client); > > Looks like a merge problem to me. Oh damn. Sorry David. I have probably done something wrong during my last rebase on top net-next. I'll fix, double check and resend the pull request. Regards,