From f2f7731420d56e7164d352d62184408d9570bef1 Mon Sep 17 00:00:00 2001
From: Phil Sutter <psutter@redhat.com>
Date: Fri, 29 May 2020 19:33:22 +0200
Subject: [PATCH] nft: cache: Reset genid when rebuilding cache

This is required in order to avoid a cache rebuild loop if
iptables-nft-restore is called with '--test' parameter and a dump
containing more than a single table.

If non-zero, __nft_build_cache() never updates genid and therefore the
incorrect genid (caused by increment in nft_action()) is never
corrected.

This is a RHEL-only fix, upstream rewrote the whole cache logic which
implicitly resolved this problem.

Fixes: 200bc39965149 ("nft: cache: Fix iptables-save segfault under stress")
Signed-off-by: Phil Sutter <psutter@redhat.com>
---
 iptables/nft-cache.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/iptables/nft-cache.c b/iptables/nft-cache.c
index 07265b7795e4f..bc6e7f7eaebfb 100644
--- a/iptables/nft-cache.c
+++ b/iptables/nft-cache.c
@@ -629,6 +629,7 @@ void nft_rebuild_cache(struct nft_handle *h)
 	if (h->cache_level)
 		__nft_flush_cache(h);
 
+	h->nft_genid = 0;
 	h->cache_level = NFT_CL_NONE;
 	__nft_build_cache(h, level, NULL, NULL, NULL);
 }
-- 
2.27.0