From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pablo Neira Ayuso Subject: [PATCH] genetlink: fix usage of NLM_F_EXCL or NLM_F_REPLACE Date: Mon, 29 Jul 2013 12:30:04 +0200 Message-ID: <1375093804-7534-1-git-send-email-pablo@netfilter.org> Cc: davem@davemloft.net To: netdev@vger.kernel.org Return-path: Received: from mail.us.es ([193.147.175.20]:56780 "EHLO mail.us.es" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751941Ab3G2KaX (ORCPT ); Mon, 29 Jul 2013 06:30:23 -0400 Sender: netdev-owner@vger.kernel.org List-ID: Currently, it is not possible to use neither NLM_F_EXCL nor NLM_F_REPLACE from genetlink. This is due to this checking in genl_family_rcv_msg: if (nlh->nlmsg_flags & NLM_F_DUMP) NLM_F_DUMP is NLM_F_MATCH|NLM_F_ROOT. Thus, if NLM_F_EXCL or NLM_F_REPLACE flag is set, genetlink believes that you're requesting a dump and it calls the .dumpit callback. The solution that I propose is to refine this checking to make it stricter: if ((nlh->nlmsg_flags & NLM_F_DUMP) == NLM_F_DUMP) And given the combination NLM_F_REPLACE and NLM_F_EXCL does not make sense to me, it removes the ambiguity. There was a patch that tried to fix this some time ago (0ab03c2 netlink: test for all flags of the NLM_F_DUMP composite) but it tried to resolve this ambiguity in *all* existing netlink subsystems, not only genetlink. That patch was reverted since it broke iproute2, which is using NLM_F_ROOT to request the dump of the routing cache. Signed-off-by: Pablo Neira Ayuso --- There is still one possibility of breaking user-space: if the application only sets NLM_F_MATCH or NLM_F_ROOT to request a dump, the dump operation will not work anymore. To address this, I have elaborated a list of all existing in-tree subsystems that provide genetlink interfaces that could be affected by git grepping for the "\.dumpit" keyword. Then, I have searched for the user-space code of those genetlink interfaces, to make sure they are using NLM_F_DUMP, this is the result: * nl80211: the iw utility uses NLM_F_DUMP. * openvswitch: version 1.10.0, lib/netlink-socket.c uses NLM_F_DUMP. * nfc: I could just find a nfc-example.git tree: http://code.openbossa.org/?p=nfc/nfc-example.git;a=summary which looks good. * netlabel: netlabel_tools-0.20 looks good. * IPVS: ipvsadm from Simon Horman's git tree looks good. * l2tp: iproute2 code looks good as well. * drdb: drbd-8.4 looks fine, drbd-8.3 does not seem to use the genetlink interface. So it seems recent code always stick to NLM_F_DUMP, which is good. net/netlink/genetlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index 2fd6dbe..145d145 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -571,7 +571,7 @@ static int genl_family_rcv_msg(struct genl_family *family, !capable(CAP_NET_ADMIN)) return -EPERM; - if (nlh->nlmsg_flags & NLM_F_DUMP) { + if ((nlh->nlmsg_flags & NLM_F_DUMP) == NLM_F_DUMP) { struct netlink_dump_control c = { .dump = ops->dumpit, .done = ops->done, -- 1.7.10.4