From 0977b2383de69dc48e9fa61c5c77878ed08d87fe Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 4 Jan 2019 14:16:59 +0100 Subject: selftests: xfrm: add block rules with adjacent/overlapping subnets The existing script lacks a policy pattern that triggers 'tree node merges' in the kernel. Consider adding policy affecting following subnet: pol1: dst 10.0.0.0/22 pol2: dst 10.0.0.0/23 # adds to existing 10.0.0.0/22 node -> no problems here. But now, lets consider reverse order: pol1: dst 10.0.0.0/24 pol2: dst 10.0.0.0/23 # CANNOT add to existing node When second policy gets added, the kernel must check that the new node ("10.0.0.0/23") doesn't overlap with any existing subnet. Example: dst 10.0.0.0/24 dst 10.0.0.1/24 dst 10.0.0.0/23 When the third policy gets added, the kernel must replace the nodes for the 10.0.0.0/24 and 10.0.0.1/24 policies with a single one and must merge all the subtrees/lists stored in those nodes into the new node. The existing test cases only have overlaps with a single node, so no merging takes place (we can always remove the 'old' node and replace it with the new subnet prefix). Add a few 'block policies' in a pattern that triggers this, with a priority that will make kernel prefer the 'esp' rules. Make sure the 'tunnel ping' tests still pass after they have been added. Signed-off-by: Florian Westphal Signed-off-by: Steffen Klassert --- tools/testing/selftests/net/xfrm_policy.sh | 109 ++++++++++++++++++++++++----- 1 file changed, 91 insertions(+), 18 deletions(-) (limited to 'tools/testing') diff --git a/tools/testing/selftests/net/xfrm_policy.sh b/tools/testing/selftests/net/xfrm_policy.sh index 8db35b99457c..b5a1b565a7e6 100755 --- a/tools/testing/selftests/net/xfrm_policy.sh +++ b/tools/testing/selftests/net/xfrm_policy.sh @@ -46,6 +46,58 @@ do_esp() { ip -net $ns xfrm policy add src $rnet dst $lnet dir fwd tmpl src $remote dst $me proto esp mode tunnel priority 100 action allow } +# add policies with different netmasks, to make sure kernel carries +# the policies contained within new netmask over when search tree is +# re-built. +# peer netns that are supposed to be encapsulated via esp have addresses +# in the 10.0.1.0/24 and 10.0.2.0/24 subnets, respectively. +# +# Adding a policy for '10.0.1.0/23' will make it necessary to +# alter the prefix of 10.0.1.0 subnet. +# In case new prefix overlaps with existing node, the node and all +# policies it carries need to be merged with the existing one(s). +# +# Do that here. +do_overlap() +{ + local ns=$1 + + # adds new nodes to tree (neither network exists yet in policy database). + ip -net $ns xfrm policy add src 10.1.0.0/24 dst 10.0.0.0/24 dir fwd priority 200 action block + + # adds a new node in the 10.0.0.0/24 tree (dst node exists). + ip -net $ns xfrm policy add src 10.2.0.0/24 dst 10.0.0.0/24 dir fwd priority 200 action block + + # adds a 10.2.0.0/24 node, but for different dst. + ip -net $ns xfrm policy add src 10.2.0.0/24 dst 10.0.1.0/24 dir fwd priority 200 action block + + # dst now overlaps with the 10.0.1.0/24 ESP policy in fwd. + # kernel must 'promote' existing one (10.0.0.0/24) to 10.0.0.0/23. + # But 10.0.0.0/23 also includes existing 10.0.1.0/24, so that node + # also has to be merged too, including source-sorted subtrees. + # old: + # 10.0.0.0/24 (node 1 in dst tree of the bin) + # 10.1.0.0/24 (node in src tree of dst node 1) + # 10.2.0.0/24 (node in src tree of dst node 1) + # 10.0.1.0/24 (node 2 in dst tree of the bin) + # 10.0.2.0/24 (node in src tree of dst node 2) + # 10.2.0.0/24 (node in src tree of dst node 2) + # + # The next 'policy add' adds dst '10.0.0.0/23', which means + # that dst node 1 and dst node 2 have to be merged including + # the sub-tree. As no duplicates are allowed, policies in + # the two '10.0.2.0/24' are also merged. + # + # after the 'add', internal search tree should look like this: + # 10.0.0.0/23 (node in dst tree of bin) + # 10.0.2.0/24 (node in src tree of dst node) + # 10.1.0.0/24 (node in src tree of dst node) + # 10.2.0.0/24 (node in src tree of dst node) + # + # 10.0.0.0/24 and 10.0.1.0/24 nodes have been merged as 10.0.0.0/23. + ip -net $ns xfrm policy add src 10.1.0.0/24 dst 10.0.0.0/23 dir fwd priority 200 action block +} + do_esp_policy_get_check() { local ns=$1 local lnet=$2 @@ -160,6 +212,41 @@ check_xfrm() { return $lret } +check_exceptions() +{ + logpostfix="$1" + local lret=0 + + # ping to .254 should be excluded from the tunnel (exception is in place). + check_xfrm 0 254 + if [ $? -ne 0 ]; then + echo "FAIL: expected ping to .254 to fail ($logpostfix)" + lret=1 + else + echo "PASS: ping to .254 bypassed ipsec tunnel ($logpostfix)" + fi + + # ping to .253 should use use ipsec due to direct policy exception. + check_xfrm 1 253 + if [ $? -ne 0 ]; then + echo "FAIL: expected ping to .253 to use ipsec tunnel ($logpostfix)" + lret=1 + else + echo "PASS: direct policy matches ($logpostfix)" + fi + + # ping to .2 should use ipsec. + check_xfrm 1 2 + if [ $? -ne 0 ]; then + echo "FAIL: expected ping to .2 to use ipsec tunnel ($logpostfix)" + lret=1 + else + echo "PASS: policy matches ($logpostfix)" + fi + + return $lret +} + #check for needed privileges if [ "$(id -u)" -ne 0 ];then echo "SKIP: Need root privileges" @@ -270,31 +357,17 @@ do_exception ns4 10.0.3.10 10.0.3.1 10.0.1.253 10.0.1.240/28 do_exception ns3 dead:3::1 dead:3::10 dead:2::fd dead:2:f0::/96 do_exception ns4 dead:3::10 dead:3::1 dead:1::fd dead:1:f0::/96 -# ping to .254 should now be excluded from the tunnel -check_xfrm 0 254 +check_exceptions "exceptions" if [ $? -ne 0 ]; then - echo "FAIL: expected ping to .254 to fail" ret=1 -else - echo "PASS: ping to .254 bypassed ipsec tunnel" fi -# ping to .253 should use use ipsec due to direct policy exception. -check_xfrm 1 253 -if [ $? -ne 0 ]; then - echo "FAIL: expected ping to .253 to use ipsec tunnel" - ret=1 -else - echo "PASS: direct policy matches" -fi +# insert block policies with adjacent/overlapping netmasks +do_overlap ns3 -# ping to .2 should use ipsec. -check_xfrm 1 2 +check_exceptions "exceptions and block policies" if [ $? -ne 0 ]; then - echo "FAIL: expected ping to .2 to use ipsec tunnel" ret=1 -else - echo "PASS: policy matches" fi for i in 1 2 3 4;do ip netns del ns$i;done -- cgit v1.2.3 From fcf86f55f2d4a6b58da5feccd45d9584edc17c5a Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 4 Jan 2019 14:17:04 +0100 Subject: selftests: xfrm: alter htresh to trigger move of policies to hash table ... and back to inexact tree. Repeat ping test after each htresh change: lookup results must not change. Signed-off-by: Florian Westphal Signed-off-by: Steffen Klassert --- tools/testing/selftests/net/xfrm_policy.sh | 44 +++++++++++++++++++++++++++--- 1 file changed, 40 insertions(+), 4 deletions(-) (limited to 'tools/testing') diff --git a/tools/testing/selftests/net/xfrm_policy.sh b/tools/testing/selftests/net/xfrm_policy.sh index b5a1b565a7e6..8ce54600d4d1 100755 --- a/tools/testing/selftests/net/xfrm_policy.sh +++ b/tools/testing/selftests/net/xfrm_policy.sh @@ -28,6 +28,19 @@ KEY_AES=0x0123456789abcdef0123456789012345 SPI1=0x1 SPI2=0x2 +do_esp_policy() { + local ns=$1 + local me=$2 + local remote=$3 + local lnet=$4 + local rnet=$5 + + # to encrypt packets as they go out (includes forwarded packets that need encapsulation) + ip -net $ns xfrm policy add src $lnet dst $rnet dir out tmpl src $me dst $remote proto esp mode tunnel priority 100 action allow + # to fwd decrypted packets after esp processing: + ip -net $ns xfrm policy add src $rnet dst $lnet dir fwd tmpl src $remote dst $me proto esp mode tunnel priority 100 action allow +} + do_esp() { local ns=$1 local me=$2 @@ -40,10 +53,7 @@ do_esp() { ip -net $ns xfrm state add src $remote dst $me proto esp spi $spi_in enc aes $KEY_AES auth sha1 $KEY_SHA mode tunnel sel src $rnet dst $lnet ip -net $ns xfrm state add src $me dst $remote proto esp spi $spi_out enc aes $KEY_AES auth sha1 $KEY_SHA mode tunnel sel src $lnet dst $rnet - # to encrypt packets as they go out (includes forwarded packets that need encapsulation) - ip -net $ns xfrm policy add src $lnet dst $rnet dir out tmpl src $me dst $remote proto esp mode tunnel priority 100 action allow - # to fwd decrypted packets after esp processing: - ip -net $ns xfrm policy add src $rnet dst $lnet dir fwd tmpl src $remote dst $me proto esp mode tunnel priority 100 action allow + do_esp_policy $ns $me $remote $lnet $rnet } # add policies with different netmasks, to make sure kernel carries @@ -370,6 +380,32 @@ if [ $? -ne 0 ]; then ret=1 fi +for n in ns3 ns4;do + ip -net $n xfrm policy set hthresh4 28 24 hthresh6 126 125 + sleep $((RANDOM%5)) +done + +check_exceptions "exceptions and block policies after hresh changes" + +# full flush of policy db, check everything gets freed incl. internal meta data +ip -net ns3 xfrm policy flush + +do_esp_policy ns3 10.0.3.1 10.0.3.10 10.0.1.0/24 10.0.2.0/24 +do_exception ns3 10.0.3.1 10.0.3.10 10.0.2.253 10.0.2.240/28 + +# move inexact policies to hash table +ip -net ns3 xfrm policy set hthresh4 16 16 + +sleep $((RANDOM%5)) +check_exceptions "exceptions and block policies after hthresh change in ns3" + +# restore original hthresh settings -- move policies back to tables +for n in ns3 ns4;do + ip -net $n xfrm policy set hthresh4 32 32 hthresh6 128 128 + sleep $((RANDOM%5)) +done +check_exceptions "exceptions and block policies after hresh change to normal" + for i in 1 2 3 4;do ip netns del ns$i;done exit $ret -- cgit v1.2.3 From 12750abad517a991c4568969bc748db302ab52cd Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 4 Jan 2019 14:17:05 +0100 Subject: xfrm: policy: fix infinite loop when merging src-nodes With very small change to test script we can trigger softlockup due to bogus assignment of 'p' (policy to be examined) on restart. Previously the two to-be-merged nodes had same address/prefixlength pair, so no erase/reinsert was necessary, we only had to append the list from node a to b. If prefix lengths are different, the node has to be deleted and re-inserted into the tree, with the updated prefix length. This was broken; due to bogus update to 'p' this loops forever. Add a 'restart' label and use that instead. While at it, don't perform the unneeded reinserts of the policies that are already sorted into the 'new' node. A previous patch in this series made xfrm_policy_inexact_list_reinsert() use the relative position indicator to sort policies according to age in case priorities are identical. Fixes: 6ac098b2a9d30 ("xfrm: policy: add 2nd-level saddr trees for inexact policies") Signed-off-by: Florian Westphal Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 15 +++++++-------- tools/testing/selftests/net/xfrm_policy.sh | 4 ++-- 2 files changed, 9 insertions(+), 10 deletions(-) (limited to 'tools/testing') diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index e691683223ee..8cfd75b62396 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -886,12 +886,13 @@ static void xfrm_policy_inexact_node_reinsert(struct net *net, struct rb_root *new, u16 family) { - struct rb_node **p, *parent = NULL; struct xfrm_pol_inexact_node *node; + struct rb_node **p, *parent; /* we should not have another subtree here */ WARN_ON_ONCE(!RB_EMPTY_ROOT(&n->root)); - +restart: + parent = NULL; p = &new->rb_node; while (*p) { u8 prefixlen; @@ -911,12 +912,11 @@ static void xfrm_policy_inexact_node_reinsert(struct net *net, } else { struct xfrm_policy *tmp; - hlist_for_each_entry(tmp, &node->hhead, bydst) - tmp->bydst_reinsert = true; - hlist_for_each_entry(tmp, &n->hhead, bydst) + hlist_for_each_entry(tmp, &n->hhead, bydst) { tmp->bydst_reinsert = true; + hlist_del_rcu(&tmp->bydst); + } - INIT_HLIST_HEAD(&node->hhead); xfrm_policy_inexact_list_reinsert(net, node, family); if (node->prefixlen == n->prefixlen) { @@ -928,8 +928,7 @@ static void xfrm_policy_inexact_node_reinsert(struct net *net, kfree_rcu(n, rcu); n = node; n->prefixlen = prefixlen; - *p = new->rb_node; - parent = NULL; + goto restart; } } diff --git a/tools/testing/selftests/net/xfrm_policy.sh b/tools/testing/selftests/net/xfrm_policy.sh index 8ce54600d4d1..71d7fdc513c1 100755 --- a/tools/testing/selftests/net/xfrm_policy.sh +++ b/tools/testing/selftests/net/xfrm_policy.sh @@ -78,8 +78,8 @@ do_overlap() # adds a new node in the 10.0.0.0/24 tree (dst node exists). ip -net $ns xfrm policy add src 10.2.0.0/24 dst 10.0.0.0/24 dir fwd priority 200 action block - # adds a 10.2.0.0/24 node, but for different dst. - ip -net $ns xfrm policy add src 10.2.0.0/24 dst 10.0.1.0/24 dir fwd priority 200 action block + # adds a 10.2.0.0/23 node, but for different dst. + ip -net $ns xfrm policy add src 10.2.0.0/23 dst 10.0.1.0/24 dir fwd priority 200 action block # dst now overlaps with the 10.0.1.0/24 ESP policy in fwd. # kernel must 'promote' existing one (10.0.0.0/24) to 10.0.0.0/23. -- cgit v1.2.3 From e1812933b17be7814f51b6c310c5d1ced7a9a5f5 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Wed, 2 Jan 2019 13:56:57 -0800 Subject: x86/selftests/pkeys: Fork() to check for state being preserved There was a bug where the per-mm pkey state was not being preserved across fork() in the child. fork() is performed in the pkey selftests, but all of the pkey activity is performed in the parent. The child does not perform any actions sensitive to pkey state. To make the test more sensitive to these kinds of bugs, add a fork() where the parent exits, and execution continues in the child. To achieve this let the key exhaustion test not terminate at the first allocation failure and fork after 2*NR_PKEYS loops and continue in the child. Signed-off-by: Dave Hansen Signed-off-by: Thomas Gleixner Cc: bp@alien8.de Cc: hpa@zytor.com Cc: peterz@infradead.org Cc: mpe@ellerman.id.au Cc: will.deacon@arm.com Cc: luto@kernel.org Cc: jroedel@suse.de Cc: stable@vger.kernel.org Cc: Borislav Petkov Cc: "H. Peter Anvin" Cc: Peter Zijlstra Cc: Michael Ellerman Cc: Will Deacon Cc: Andy Lutomirski Cc: Joerg Roedel Link: https://lkml.kernel.org/r/20190102215657.585704B7@viggo.jf.intel.com --- tools/testing/selftests/x86/protection_keys.c | 41 ++++++++++++++++++++------- 1 file changed, 31 insertions(+), 10 deletions(-) (limited to 'tools/testing') diff --git a/tools/testing/selftests/x86/protection_keys.c b/tools/testing/selftests/x86/protection_keys.c index 460b4bdf4c1e..5d546dcdbc80 100644 --- a/tools/testing/selftests/x86/protection_keys.c +++ b/tools/testing/selftests/x86/protection_keys.c @@ -1133,6 +1133,21 @@ void test_pkey_syscalls_bad_args(int *ptr, u16 pkey) pkey_assert(err); } +void become_child(void) +{ + pid_t forkret; + + forkret = fork(); + pkey_assert(forkret >= 0); + dprintf3("[%d] fork() ret: %d\n", getpid(), forkret); + + if (!forkret) { + /* in the child */ + return; + } + exit(0); +} + /* Assumes that all pkeys other than 'pkey' are unallocated */ void test_pkey_alloc_exhaust(int *ptr, u16 pkey) { @@ -1141,7 +1156,7 @@ void test_pkey_alloc_exhaust(int *ptr, u16 pkey) int nr_allocated_pkeys = 0; int i; - for (i = 0; i < NR_PKEYS*2; i++) { + for (i = 0; i < NR_PKEYS*3; i++) { int new_pkey; dprintf1("%s() alloc loop: %d\n", __func__, i); new_pkey = alloc_pkey(); @@ -1152,20 +1167,26 @@ void test_pkey_alloc_exhaust(int *ptr, u16 pkey) if ((new_pkey == -1) && (errno == ENOSPC)) { dprintf2("%s() failed to allocate pkey after %d tries\n", __func__, nr_allocated_pkeys); - break; + } else { + /* + * Ensure the number of successes never + * exceeds the number of keys supported + * in the hardware. + */ + pkey_assert(nr_allocated_pkeys < NR_PKEYS); + allocated_pkeys[nr_allocated_pkeys++] = new_pkey; } - pkey_assert(nr_allocated_pkeys < NR_PKEYS); - allocated_pkeys[nr_allocated_pkeys++] = new_pkey; + + /* + * Make sure that allocation state is properly + * preserved across fork(). + */ + if (i == NR_PKEYS*2) + become_child(); } dprintf3("%s()::%d\n", __func__, __LINE__); - /* - * ensure it did not reach the end of the loop without - * failure: - */ - pkey_assert(i < NR_PKEYS*2); - /* * There are 16 pkeys supported in hardware. Three are * allocated by the time we get here: -- cgit v1.2.3