Skip to contentSkip to conversation

net: restore ip source validation

PATCH Randolf Pohl 4 participants 7 messages Feb 5, 2010 Proc sysctl

lore
Share & tools
lore
#0 lore raw

[...]

include/linux/sysctl.h enum · line 496

Quoted 10 lines
include/linux/sysctl.h +1 -0
@@ -490,6 +490,7 @@ enum
 	NET_IPV4_CONF_PROMOTE_SECONDARIES=20,
 	NET_IPV4_CONF_ARP_ACCEPT=21,
 	NET_IPV4_CONF_ARP_NOTIFY=22,
+	NET_IPV4_CONF_SRC_VMARK=24,
 	__NET_IPV4_CONF_MAX
 };
 

Must these be contiguous, i.e. 23?

R.

Signature · 2 lines
Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3.5 -
sicherer, schneller und einfacher! http://portal.gmx.net/de/go/atbrowser
#1 lore raw
Quoted 14 lines · Randolf Pohl

[...]

include/linux/sysctl.h enum · line 496

Quoted 10 lines
include/linux/sysctl.h +1 -0
@@ -490,6 +490,7 @@ enum
 	NET_IPV4_CONF_PROMOTE_SECONDARIES=20,
 	NET_IPV4_CONF_ARP_ACCEPT=21,
 	NET_IPV4_CONF_ARP_NOTIFY=22,
+	NET_IPV4_CONF_SRC_VMARK=24,
 	__NET_IPV4_CONF_MAX
 };
 

Must these be contiguous, i.e. 23?

The problem is elsewhere. Here's the fix.

--

kernel/sysctl_check.c +0 -0
index b6e7aae..469193c 100644
@@ -220,6 +220,7 @@ static const struct trans_ctl_table

trans_net_ipv4_conf_vars_table[] = {

        { NET_IPV4_CONF_PROMOTE_SECONDARIES,    "promote_secondaries" },
        { NET_IPV4_CONF_ARP_ACCEPT,             "arp_accept" },
        { NET_IPV4_CONF_ARP_NOTIFY,             "arp_notify" },
+       { NET_IPV4_CONF_SRC_VMARK,              "src_valid_mark" },
        {}

};
---

cheers,
jamal

#2 lore raw
Quoted 8 lines · jamal

The problem is elsewhere. Here's the fix.

--

kernel/sysctl_check.c +0 -0
index b6e7aae..469193c 100644

kernel/sysctl_check.c static const struct trans_ctl_table

Quoted 8 lines
kernel/sysctl_check.c +1 -1
@@ -220,6 +220,7 @@ static const struct trans_ctl_table
 trans_net_ipv4_conf_vars_table[] = {
        { NET_IPV4_CONF_PROMOTE_SECONDARIES,    "promote_secondaries" },
        { NET_IPV4_CONF_ARP_ACCEPT,             "arp_accept" },
        { NET_IPV4_CONF_ARP_NOTIFY,             "arp_notify" },
+       { NET_IPV4_CONF_SRC_VMARK,              "src_valid_mark" },
        {}
 };
---

Well spotted, that fixes it. Your mailer is broken however, it converted tabs into spaces, so I had to add the changed line manually.

Sven

#3 lore raw
Quoted 13 lines · Sven Joachim
Quoted 11 lines · jamal

The problem is elsewhere. Here's the fix.

--

kernel/sysctl_check.c +0 -0
index b6e7aae..469193c 100644
@@ -220,6 +220,7 @@ static const struct trans_ctl_table
 trans_net_ipv4_conf_vars_table[] = {
        { NET_IPV4_CONF_PROMOTE_SECONDARIES,    "promote_secondaries" },
        { NET_IPV4_CONF_ARP_ACCEPT,             "arp_accept" },
Quoted 8 lines

kernel/sysctl_check.c static const struct trans_ctl_table

Quoted 5 lines
kernel/sysctl_check.c +1 -1
        { NET_IPV4_CONF_ARP_NOTIFY,             "arp_notify" },
+       { NET_IPV4_CONF_SRC_VMARK,              "src_valid_mark" },
        {}
 };
---

Well spotted, that fixes it. Your mailer is broken however, it converted tabs into spaces, so I had to add the changed line manually.

Bah. That DEVINET_SYSCTL_ENTRY requires having a binary sysctl assigned, just to use as an index.

Which of course trips over all of the fine checks in sysctl_check.c to keep people from assigning new binary sysctls by accident.

That is the only place in the kernel where he have that problem, I wonder how much work it will be to finish untangling.

Eric

#4 lore raw
Quoted 8 lines · Eric W. Biederman

Bah. That DEVINET_SYSCTL_ENTRY requires having a binary sysctl assigned, just to use as an index.

Which of course trips over all of the fine checks in sysctl_check.c to keep people from assigning new binary sysctls by accident.

That is the only place in the kernel where he have that problem, I wonder how much work it will be to finish untangling.

Isn't that already done in 2.6.33, looking at commit 83ac201b ? Forgive my ignorance, I am a layman.

Sen

#5 lore raw
Quoted 5 lines · Sven Joachim
Quoted 3 lines · Eric W. Biederman

Bah. That DEVINET_SYSCTL_ENTRY requires having a binary sysctl assigned, just to use as an index.

Quoted 8 lines
Quoted 5 lines

Which of course trips over all of the fine checks in sysctl_check.c to keep people from assigning new binary sysctls by accident.

That is the only place in the kernel where he have that problem, I wonder how much work it will be to finish untangling.

Isn't that already done in 2.6.33, looking at commit 83ac201b ? Forgive my ignorance, I am a layman.

In 2.6.33 the enumeration in sysctl.h still serves double duty as an index into a per network device bitmap and as the binary sysctl number. You are correct that the rest of the binary sysctl code is decoupled in 2.6.33.

In 2.6.32 the implementation is also still coupled and that difference is what caused problems for the backport.

I just wince whenever I noticed we have touched sysctl.h.

Eric

#6 lore raw
In reply to Sven Joachim
Quoted 5 lines · Sven Joachim
Quoted 3 lines · Eric W. Biederman

Bah. That DEVINET_SYSCTL_ENTRY requires having a binary sysctl assigned, just to use as an index.

Quoted 8 lines
Quoted 5 lines

Which of course trips over all of the fine checks in sysctl_check.c to keep people from assigning new binary sysctls by accident.

That is the only place in the kernel where he have that problem, I wonder how much work it will be to finish untangling.

Isn't that already done in 2.6.33, looking at commit 83ac201b ? Forgive my ignorance, I am a layman.

In 2.6.33 the enumeration in sysctl.h still serves double duty as an index into a per network device bitmap and as the binary sysctl number. You are correct that the rest of the binary sysctl code is decoupled in 2.6.33.

In 2.6.32 the implementation is also still coupled and that difference is what caused problems for the backport.

I just wince whenever I noticed we have touched sysctl.h.

Eric