net: xfrm: skip policies marked as dead while reinserting policies
Subsystems (3)
PATCH v2 1/1 net: xfrm: skip policies marked as dead while reinserting policies
BUG: KASAN: slab-use-after-free in xfrm_policy_inexact_list_reinsert+0xb6/0x430 Read of size 1 at addr ffff8881051f3bf8 by task ip/668 CPU: 2 PID: 668 Comm: ip Not tainted 6.5.0-rc5-00182-g25aa0bebba72-dirty #64 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13 04/01/2014 Call Trace: <TASK> dump_stack_lvl+0x72/0xa0 print_report+0xd0/0x620 kasan_report+0xb6/0xf0 xfrm_policy_inexact_list_reinsert+0xb6/0x430 xfrm_policy_inexact_insert_node.constprop.0+0x537/0x800 xfrm_policy_inexact_alloc_chain+0x23f/0x320 xfrm_policy_inexact_insert+0x6b/0x590 xfrm_policy_insert+0x3b1/0x480 xfrm_add_policy+0x23c/0x3c0 xfrm_user_rcv_msg+0x2d0/0x510 netlink_rcv_skb+0x10d/0x2d0 xfrm_netlink_rcv+0x49/0x60 netlink_unicast+0x3fe/0x540 netlink_sendmsg+0x528/0x970 sock_sendmsg+0x14a/0x160 ____sys_sendmsg+0x4fc/0x580 ___sys_sendmsg+0xef/0x160 __sys_sendmsg+0xf7/0x1b0 do_syscall_64+0x3f/0x90 entry_SYSCALL_64_after_hwframe+0x73/0xdd
The root cause is:
cpu 0 cpu1xfrm_dump_policyxfrm_policy_walklist_move_tail
xfrm_add_policy ... ... xfrm_policy_inexact_list_reinsert list_for_each_entry_reverse if (!policy->bydst_reinsert) //read non-existent policy
xfrm_dump_policy_donexfrm_policy_walk_donelist_del(&walk->walk.all);
If dump_one_policy() returns err (triggered by netlink socket), xfrm_policy_walk() will move walk initialized by socket to list net->xfrm.policy_all. so this socket becomes visible in the global policy list. The head *walk can be traversed when users add policies with different prefixlen and trigger xfrm_policy node merge.
The issue can also be triggered by policy list traversal while rehashing and flushing policies.
It can be fixed by skip such "policies" with walk.dead set to 1.
---
v2: fix similiar similar while rehashing and flushing policies
--- net/xfrm/xfrm_policy.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-)
net/xfrm/xfrm_policy.c +15 -5
@@ -848,6 +848,9 @@ static void xfrm_policy_inexact_list_reinsert(struct net *net, matched_d =0 ;list_for_each_entry_reverse (policy, &net->xfrm.policy_all, walk.all) { +if (policy->walk.dead) +continue ; +struct hlist_node *newpos =NULL ;bool matches_s, matches_d;
@@ -1253,11 +1256,14 @@ static void xfrm_hash_rebuild(struct work_struct *work)* we start with destructive action. */ list_for_each_entry (policy, &net->xfrm.policy_all, walk.all) { +if (policy->walk.dead) +continue ; +struct xfrm_pol_inexact_bin *bin;u8 dbits, sbits; dir =xfrm_policy_id2dir (policy->index); -if (policy->walk.dead || dir >=XFRM_POLICY_MAX ) +if (dir >=XFRM_POLICY_MAX )continue ;if ((dir &XFRM_POLICY_MASK ) ==XFRM_POLICY_OUT ) {
@@ -1823,9 +1829,11 @@ int xfrm_policy_flush(struct net *net, u8 type, bool task_valid) again:list_for_each_entry (pol, &net->xfrm.policy_all, walk.all) { +if (pol->walk.dead) +continue ; + dir =xfrm_policy_id2dir (pol->index); -if (pol->walk.dead || - dir >=XFRM_POLICY_MAX || +if (dir >=XFRM_POLICY_MAX || pol->type != type)continue ;
@@ -1862,9 +1870,11 @@ int xfrm_dev_policy_flush(struct net *net, struct net_device *dev, again:list_for_each_entry (pol, &net->xfrm.policy_all, walk.all) { +if (pol->walk.dead) +continue ; + dir =xfrm_policy_id2dir (pol->index); -if (pol->walk.dead || - dir >=XFRM_POLICY_MAX || +if (dir >=XFRM_POLICY_MAX || pol->xdo.dev != dev)continue ;
Signature · 1 line
2.25.1
Quoted 19 lines · Dong Chenchen
BUG: KASAN: slab-use-after-free in xfrm_policy_inexact_list_reinsert+0xb6/0x430 Read of size 1 at addr ffff8881051f3bf8 by task ip/668 CPU: 2 PID: 668 Comm: ip Not tainted 6.5.0-rc5-00182-g25aa0bebba72-dirty #64 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13 04/01/2014 Call Trace: <TASK> dump_stack_lvl+0x72/0xa0 print_report+0xd0/0x620 kasan_report+0xb6/0xf0 xfrm_policy_inexact_list_reinsert+0xb6/0x430 xfrm_policy_inexact_insert_node.constprop.0+0x537/0x800 xfrm_policy_inexact_alloc_chain+0x23f/0x320 xfrm_policy_inexact_insert+0x6b/0x590 xfrm_policy_insert+0x3b1/0x480 xfrm_add_policy+0x23c/0x3c0 xfrm_user_rcv_msg+0x2d0/0x510 netlink_rcv_skb+0x10d/0x2d0 xfrm_netlink_rcv+0x49/0x60
Quoted 8 lines
netlink_unicast+0x3fe/0x540 netlink_sendmsg+0x528/0x970 sock_sendmsg+0x14a/0x160 ____sys_sendmsg+0x4fc/0x580 ___sys_sendmsg+0xef/0x160 __sys_sendmsg+0xf7/0x1b0 do_syscall_64+0x3f/0x90 entry_SYSCALL_64_after_hwframe+0x73/0xdd
Thanks for following up.
Quoted 69 lines · Dong Chenchen
BUG: KASAN: slab-use-after-free in xfrm_policy_inexact_list_reinsert+0xb6/0x430 Read of size 1 at addr ffff8881051f3bf8 by task ip/668 CPU: 2 PID: 668 Comm: ip Not tainted 6.5.0-rc5-00182-g25aa0bebba72-dirty #64 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13 04/01/2014 Call Trace: <TASK> dump_stack_lvl+0x72/0xa0 print_report+0xd0/0x620 kasan_report+0xb6/0xf0 xfrm_policy_inexact_list_reinsert+0xb6/0x430 xfrm_policy_inexact_insert_node.constprop.0+0x537/0x800 xfrm_policy_inexact_alloc_chain+0x23f/0x320 xfrm_policy_inexact_insert+0x6b/0x590 xfrm_policy_insert+0x3b1/0x480 xfrm_add_policy+0x23c/0x3c0 xfrm_user_rcv_msg+0x2d0/0x510 netlink_rcv_skb+0x10d/0x2d0 xfrm_netlink_rcv+0x49/0x60 netlink_unicast+0x3fe/0x540 netlink_sendmsg+0x528/0x970 sock_sendmsg+0x14a/0x160 ____sys_sendmsg+0x4fc/0x580 ___sys_sendmsg+0xef/0x160 __sys_sendmsg+0xf7/0x1b0 do_syscall_64+0x3f/0x90 entry_SYSCALL_64_after_hwframe+0x73/0xddThe root cause is:
cpu 0 cpu1
xfrm_dump_policyxfrm_policy_walklist_move_tailxfrm_add_policy ... ... xfrm_policy_inexact_list_reinsert list_for_each_entry_reverse if (!policy->bydst_reinsert) //read non-existent policy
xfrm_dump_policy_donexfrm_policy_walk_donelist_del(&walk->walk.all);If
dump_one_policy()returns err (triggered by netlink socket),xfrm_policy_walk()will move walk initialized by socket to listnet->xfrm.policy_all. so this socket becomes visible in the global policy list. The head *walk can be traversed when users add policies with different prefixlen and triggerxfrm_policynode merge.The issue can also be triggered by policy list traversal while rehashing and flushing policies.
It can be fixed by skip such "policies" with walk.dead set to 1.
Fixes 9cf545ebd591 ("xfrm: policy: store inexact policies in a tree ordered by destination address")Fixes 12a169e7d8f4 ("ipsec: Put dumpers on the dump list")Signed-off-by Dong Chenchen <[email protected]>---
v2: fix similiar similar while rehashing and flushing policies--- net/xfrm/xfrm_policy.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-)net/xfrm/xfrm_policy.c +0
-0
net/xfrm/xfrm_policy.c static void xfrm_policy_inexact_list_reinsert(struct net *net, · line 855
Quoted 8 lines
net/xfrm/xfrm_policy.c +3
-0@@ -848,6 +848,9 @@ static void xfrm_policy_inexact_list_reinsert(struct net *net, matched_d =0 ;list_for_each_entry_reverse (policy, &net->xfrm.policy_all, walk.all) { +if (policy->walk.dead) +continue ; +struct hlist_node *newpos =NULL ;bool matches_s, matches_d;
You can't declare new variables in the middle of execution scope in C.
net/xfrm/xfrm_policy.c static void xfrm_hash_rebuild(struct work_struct *work) · line 1263
Quoted 10 lines
net/xfrm/xfrm_policy.c +3
-0@@ -1253,11 +1256,14 @@ static void xfrm_hash_rebuild(struct work_struct *work)* we start with destructive action. */ list_for_each_entry (policy, &net->xfrm.policy_all, walk.all) { +if (policy->walk.dead) +continue ; +struct xfrm_pol_inexact_bin *bin;u8 dbits, sbits;
Same comment as above.
net/xfrm/xfrm_policy.c static void xfrm_hash_rebuild(struct work_struct *work) · line 1266
Quoted 4 lines
net/xfrm/xfrm_policy.c +1
-1dir =xfrm_policy_id2dir (policy->index); -if (policy->walk.dead || dir >=XFRM_POLICY_MAX ) +if (dir >=XFRM_POLICY_MAX )
This change is unnecessary, previous code was perfectly fine.
Quoted 4 lines
net/xfrm/xfrm_policy.c +0
-0continue ;if ((dir &XFRM_POLICY_MASK ) ==XFRM_POLICY_OUT ) {
net/xfrm/xfrm_policy.c int xfrm_policy_flush(struct net *net, u8 type, bool task_valid) · line 1836
Quoted 10 lines
net/xfrm/xfrm_policy.c +4
-2@@ -1823,9 +1829,11 @@ int xfrm_policy_flush(struct net *net, u8 type, bool task_valid) again:list_for_each_entry (pol, &net->xfrm.policy_all, walk.all) { +if (pol->walk.dead) +continue ; + dir =xfrm_policy_id2dir (pol->index); - if (pol->walk.dead || - dir >= XFRM_POLICY_MAX || + if (dir >= XFRM_POLICY_MAX ||
This change is unnecessary, previous code was perfectly fine.
Quoted 6 lines
net/xfrm/xfrm_policy.c +0
-0pol->type != type)continue ;@@ -1862,9 +1870,11 @@ int xfrm_dev_policy_flush(struct net *net, struct net_device *dev, again:
net/xfrm/xfrm_policy.c int xfrm_dev_policy_flush(struct net *net, struct net_device *dev, · line 1879
Quoted 10 lines
net/xfrm/xfrm_policy.c +4
-2list_for_each_entry (pol, &net->xfrm.policy_all, walk.all) { +if (pol->walk.dead) +continue ; + dir =xfrm_policy_id2dir (pol->index); -if (pol->walk.dead || - dir >=XFRM_POLICY_MAX || +if (dir >=XFRM_POLICY_MAX || pol->xdo.dev != dev)continue ;
Ditto.
Quoted 4 lines
Signature · 2 lines
-- 2.25.1
Quoted 5 lines · Leon Romanovsky
net/xfrm/xfrm_policy.c static void xfrm_hash_rebuild(struct work_struct *work) · line 1266
Quoted 3 lines
net/xfrm/xfrm_policy.c +1
-1dir =xfrm_policy_id2dir (policy->index); -if (policy->walk.dead || dir >=XFRM_POLICY_MAX ) +if (dir >=XFRM_POLICY_MAX )This change is unnecessary, previous code was perfectly fine.
Are you sure? AFAICS walker struct has no 'index' member.
Quoted 8 lines · Florian Westphal
Quoted 5 lines · Leon Romanovsky
net/xfrm/xfrm_policy.c static void xfrm_hash_rebuild(struct work_struct *work) · line 1266
Quoted 3 lines
net/xfrm/xfrm_policy.c +1
-1dir =xfrm_policy_id2dir (policy->index); -if (policy->walk.dead || dir >=XFRM_POLICY_MAX ) +if (dir >=XFRM_POLICY_MAX )This change is unnecessary, previous code was perfectly fine.
Are you sure? AFAICS walker struct has no 'index' member.
But policy has, and we are not interested in validity of it as first check in if (...) will be true for policy->walk.dead.
So it is safe to call to dir = xfrm_policy_id2dir(policy->index) even for dead policy.
Thanks
Herbert Xu
maintainerQuoted 6 lines · Leon Romanovsky
But policy has, and we are not interested in validity of it as first check in if (...) will be true for
policy->walk.dead.So it is safe to call to dir =
xfrm_policy_id2dir(policy->index)even for dead policy.
If you dereference policy->index on a walker object it will read memory before the start of the walker object. That could do anything, perhaps even triggering a page fault.
Cheers,
Signature · 3 lines
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Quoted 11 lines · Herbert Xu
Quoted 6 lines · Leon Romanovsky
But policy has, and we are not interested in validity of it as first check in if (...) will be true for
policy->walk.dead.So it is safe to call to dir =
xfrm_policy_id2dir(policy->index)even for dead policy.If you dereference
policy->indexon a walker object it will read memory before the start of the walker object. That could do anything, perhaps even triggering a page fault.
Where do you see walker object? xfrm_policy_id2dir() is called on policy object, which is defined as "struct xfrm_policy".
Thanks
Quoted 6 lines
Cheers,
Signature · 3 lines
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Quoted 64 lines · Dong Chenchen
Quoted 63 lines · Dong Chenchen
Quoted 63 lines
BUG: KASAN: slab-use-after-free in xfrm_policy_inexact_list_reinsert+0xb6/0x430 Read of size 1 at addr ffff8881051f3bf8 by task ip/668 CPU: 2 PID: 668 Comm: ip Not tainted 6.5.0-rc5-00182-g25aa0bebba72-dirty #64 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13 04/01/2014 Call Trace: <TASK> dump_stack_lvl+0x72/0xa0 print_report+0xd0/0x620 kasan_report+0xb6/0xf0 xfrm_policy_inexact_list_reinsert+0xb6/0x430 xfrm_policy_inexact_insert_node.constprop.0+0x537/0x800 xfrm_policy_inexact_alloc_chain+0x23f/0x320 xfrm_policy_inexact_insert+0x6b/0x590 xfrm_policy_insert+0x3b1/0x480 xfrm_add_policy+0x23c/0x3c0 xfrm_user_rcv_msg+0x2d0/0x510 netlink_rcv_skb+0x10d/0x2d0 xfrm_netlink_rcv+0x49/0x60 netlink_unicast+0x3fe/0x540 netlink_sendmsg+0x528/0x970 sock_sendmsg+0x14a/0x160 ____sys_sendmsg+0x4fc/0x580 ___sys_sendmsg+0xef/0x160 __sys_sendmsg+0xf7/0x1b0 do_syscall_64+0x3f/0x90 entry_SYSCALL_64_after_hwframe+0x73/0xddThe root cause is:
cpu 0 cpu1
xfrm_dump_policyxfrm_policy_walklist_move_tailxfrm_add_policy ... ... xfrm_policy_inexact_list_reinsert list_for_each_entry_reverse if (!policy->bydst_reinsert) //read non-existent policy
xfrm_dump_policy_donexfrm_policy_walk_donelist_del(&walk->walk.all);If
dump_one_policy()returns err (triggered by netlink socket),xfrm_policy_walk()will move walk initialized by socket to listnet->xfrm.policy_all. so this socket becomes visible in the global policy list. The head *walk can be traversed when users add policies with different prefixlen and triggerxfrm_policynode merge.The issue can also be triggered by policy list traversal while rehashing and flushing policies.
It can be fixed by skip such "policies" with walk.dead set to 1.
Fixes 9cf545ebd591 ("xfrm: policy: store inexact policies in a tree ordered by destination address")Fixes 12a169e7d8f4 ("ipsec: Put dumpers on the dump list")Signed-off-by Dong Chenchen <[email protected]>---
v2: fix similiar similar while rehashing and flushing policies--- net/xfrm/xfrm_policy.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-)
<...>
Quoted 19 lines
Quoted 13 lines
net/xfrm/xfrm_policy.c static void xfrm_hash_rebuild(struct work_struct *work) · line 1263
Quoted 9 lines
net/xfrm/xfrm_policy.c +3
-0@@ -1253,11 +1256,14 @@ static void xfrm_hash_rebuild(struct work_struct *work)* we start with destructive action. */ list_for_each_entry (policy, &net->xfrm.policy_all, walk.all) { +if (policy->walk.dead) +continue ; +struct xfrm_pol_inexact_bin *bin;u8 dbits, sbits;Same comment as above.
Quoted 1 line
Quoted 6 lines
net/xfrm/xfrm_policy.c static void xfrm_hash_rebuild(struct work_struct *work) · line 1266
Quoted 3 lines
net/xfrm/xfrm_policy.c +1
-1dir =xfrm_policy_id2dir (policy->index); -if (policy->walk.dead || dir >=XFRM_POLICY_MAX ) +if (dir >=XFRM_POLICY_MAX )This change is unnecessary, previous code was perfectly fine.
Quoted 5 lines
The walker object initialized by
xfrm_policy_walk_init()doesnt have policy.list_for_each_entry()will use the walker offset to calculate policy address. It's nonexistent and different from invalid dead policy. It will read memory that doesnt belong to walker if dereferencepolicy->index.
I think we should protect the memory.
But all operations here are an outcome of "list_for_each_entry(policy, &net->xfrm.policy_all, walk.all)" which stores in policy iterator the pointer to struct xfrm_policy.
How at the same time access to policy->walk.dead is valid while policy->index is not?
Thanks
Quoted 119 lines · Dong Chenchen
Quoted 92 lines · Dong Chenchen
Quoted 64 lines · Dong Chenchen
Quoted 63 lines
Quoted 63 lines
BUG: KASAN: slab-use-after-free in xfrm_policy_inexact_list_reinsert+0xb6/0x430 Read of size 1 at addr ffff8881051f3bf8 by task ip/668 CPU: 2 PID: 668 Comm: ip Not tainted 6.5.0-rc5-00182-g25aa0bebba72-dirty #64 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13 04/01/2014 Call Trace: <TASK> dump_stack_lvl+0x72/0xa0 print_report+0xd0/0x620 kasan_report+0xb6/0xf0 xfrm_policy_inexact_list_reinsert+0xb6/0x430 xfrm_policy_inexact_insert_node.constprop.0+0x537/0x800 xfrm_policy_inexact_alloc_chain+0x23f/0x320 xfrm_policy_inexact_insert+0x6b/0x590 xfrm_policy_insert+0x3b1/0x480 xfrm_add_policy+0x23c/0x3c0 xfrm_user_rcv_msg+0x2d0/0x510 netlink_rcv_skb+0x10d/0x2d0 xfrm_netlink_rcv+0x49/0x60 netlink_unicast+0x3fe/0x540 netlink_sendmsg+0x528/0x970 sock_sendmsg+0x14a/0x160 ____sys_sendmsg+0x4fc/0x580 ___sys_sendmsg+0xef/0x160 __sys_sendmsg+0xf7/0x1b0 do_syscall_64+0x3f/0x90 entry_SYSCALL_64_after_hwframe+0x73/0xddThe root cause is:
cpu 0 cpu1
xfrm_dump_policyxfrm_policy_walklist_move_tailxfrm_add_policy ... ... xfrm_policy_inexact_list_reinsert list_for_each_entry_reverse if (!policy->bydst_reinsert) //read non-existent policy
xfrm_dump_policy_donexfrm_policy_walk_donelist_del(&walk->walk.all);If
dump_one_policy()returns err (triggered by netlink socket),xfrm_policy_walk()will move walk initialized by socket to listnet->xfrm.policy_all. so this socket becomes visible in the global policy list. The head *walk can be traversed when users add policies with different prefixlen and triggerxfrm_policynode merge.The issue can also be triggered by policy list traversal while rehashing and flushing policies.
It can be fixed by skip such "policies" with walk.dead set to 1.
Fixes 9cf545ebd591 ("xfrm: policy: store inexact policies in a tree ordered by destination address")Fixes 12a169e7d8f4 ("ipsec: Put dumpers on the dump list")Signed-off-by Dong Chenchen <[email protected]>---
v2: fix similiar similar while rehashing and flushing policies--- net/xfrm/xfrm_policy.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-)<...>
Quoted 19 lines
Quoted 13 lines
net/xfrm/xfrm_policy.c static void xfrm_hash_rebuild(struct work_struct *work) · line 1263
Quoted 9 lines
net/xfrm/xfrm_policy.c +3
-0@@ -1253,11 +1256,14 @@ static void xfrm_hash_rebuild(struct work_struct *work)* we start with destructive action. */ list_for_each_entry (policy, &net->xfrm.policy_all, walk.all) { +if (policy->walk.dead) +continue ; +struct xfrm_pol_inexact_bin *bin;u8 dbits, sbits;Same comment as above.
Quoted 1 line
Quoted 6 lines
net/xfrm/xfrm_policy.c static void xfrm_hash_rebuild(struct work_struct *work) · line 1266
Quoted 3 lines
net/xfrm/xfrm_policy.c +1
-1dir =xfrm_policy_id2dir (policy->index); -if (policy->walk.dead || dir >=XFRM_POLICY_MAX ) +if (dir >=XFRM_POLICY_MAX )This change is unnecessary, previous code was perfectly fine.
Quoted 5 lines
The walker object initialized by
xfrm_policy_walk_init()doesnt have policy.list_for_each_entry()will use the walker offset to calculate policy address. It's nonexistent and different from invalid dead policy. It will read memory that doesnt belong to walker if dereferencepolicy->index.
I think we should protect the memory.Quoted 8 lines
But all operations here are an outcome of "
list_for_each_entry(policy, &net->xfrm.policy_all, walk.all)" which stores in policy iterator the pointer to structxfrm_policy.How at the same time access to
policy->walk.deadis valid whilepolicy->indexis not?Thanks
1.walker init: its only a list head, no policy
xfrm_dump_policy_startxfrm_policy_walk_init(walk, XFRM_POLICY_TYPE_ANY); INIT_LIST_HEAD(&walk->walk.all); walk->walk.dead = 1;2.add the walk head to
net->xfrm.policy_allxfrm_policy_walklist_for_each_entry_from(x, &net->xfrm.policy_all, all) if (error) { list_move_tail(&walk->walk.all, &x->all); //add the walk to list tail3.traverse the walk list
xfrm_policy_flushlist_for_each_entry(pol, &net->xfrm.policy_all, walk.all)dir = xfrm_policy_id2dir(pol->index);
Quoted 8 lines
it gets policy by &net->xfrm.policy_all-0x130(offset of walk in policy) but when walk is head, we will read others memory by the calculated policy. such as:
walk addr policy addr0xffff0000d7f3b5300xffff0000d7f3b400(non-existent)head walker of
net->xfrm.policy_allcan be skipped bylist_for_each_entry(). but the walker created by socket is located list tail. so we should skip it.
list_for_each_entry_from(x, &net->xfrm.policy_all, all) gives you pointer to "x", you can't access some of its fields and say they exist and other doesn't. Once you can call to "x->...", you can call to "x->index" too.
Thanks
Quoted 42 lines · Dong Chenchen
Quoted 41 lines · Dong Chenchen
Quoted 40 lines
Quoted 6 lines
Quoted 5 lines
The walker object initialized by
xfrm_policy_walk_init()doesnt have policy.list_for_each_entry()will use the walker offset to calculate policy address. It's nonexistent and different from invalid dead policy. It will read memory that doesnt belong to walker if dereferencepolicy->index.
I think we should protect the memory.Quoted 8 lines
But all operations here are an outcome of "
list_for_each_entry(policy, &net->xfrm.policy_all, walk.all)" which stores in policy iterator the pointer to structxfrm_policy.How at the same time access to
policy->walk.deadis valid whilepolicy->indexis not?Thanks
1.walker init: its only a list head, no policy
xfrm_dump_policy_startxfrm_policy_walk_init(walk, XFRM_POLICY_TYPE_ANY); INIT_LIST_HEAD(&walk->walk.all); walk->walk.dead = 1;2.add the walk head to
net->xfrm.policy_allxfrm_policy_walklist_for_each_entry_from(x, &net->xfrm.policy_all, all) if (error) { list_move_tail(&walk->walk.all, &x->all); //add the walk to list tail3.traverse the walk list
xfrm_policy_flushlist_for_each_entry(pol, &net->xfrm.policy_all, walk.all)dir = xfrm_policy_id2dir(pol->index);it gets policy by &net->xfrm.policy_all-0x130(offset of walk in policy) but when walk is head, we will read others memory by the calculated policy. such as:
walk addr policy addr0xffff0000d7f3b5300xffff0000d7f3b400(non-existent)head walker of
net->xfrm.policy_allcan be skipped bylist_for_each_entry(). but the walker created by socket is located list tail. so we should skip it.
Quoted 8 lines
Quoted 6 lines
list_for_each_entry_from(x, &net->xfrm.policy_all, all)gives you pointer to "x", you can't access some of its fields and say they exist and other doesn't. Once you can call to "x->...", you can call to "x->index" too.Thanks
We get a pointer addr not actual variable from
list_for_each_entry_from(), that calculated by walk address dec offset from structxfrm_policy(0x130).
The thing is that you must get valid addr pointer and not some random memory address.
Quoted 9 lines
walk addr:
0xffff0000d7f3b530//allocated by socket, valid -> dec0x130(use macrocontainer_of) policy_addr:0xffff0000d7f3b400 //only a pointer addr -> add0x130
policy->walk:0xffff0000d7f3b530 //its still walker headI think its invalid to read
policy->indexfrom memory that maybe allocated by other user.
This is not how pointers are expected to be used. Once you have pointer to the struct, the expectation is that all fields in that struct are accessible.
Anyway, we discussed this topic a lot.
Thanks
Quoted 4 lines
Thanks!
Dong Chenchen