linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Eric Dumazet <edumazet@google.com>,
	Amit Klein <aksecurity@gmail.com>, Willy Tarreau <w@1wt.eu>,
	"David S . Miller" <davem@davemloft.net>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Subject: [PATCH 4.9 50/71] inet: use bigger hash table for IP ID generation
Date: Mon, 28 Jun 2021 10:39:42 -0400	[thread overview]
Message-ID: <20210628144003.34260-51-sashal@kernel.org> (raw)
In-Reply-To: <20210628144003.34260-1-sashal@kernel.org>

From: Eric Dumazet <edumazet@google.com>

commit aa6dd211e4b1dde9d5dc25d699d35f789ae7eeba upstream.

In commit 73f156a6e8c1 ("inetpeer: get rid of ip_id_count")
I used a very small hash table that could be abused
by patient attackers to reveal sensitive information.

Switch to a dynamic sizing, depending on RAM size.

Typical big hosts will now use 128x more storage (2 MB)
to get a similar increase in security and reduction
of hash collisions.

As a bonus, use of alloc_large_system_hash() spreads
allocated memory among all NUMA nodes.

Fixes: 73f156a6e8c1 ("inetpeer: get rid of ip_id_count")
Reported-by: Amit Klein <aksecurity@gmail.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Willy Tarreau <w@1wt.eu>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 net/ipv4/route.c | 42 +++++++++++++++++++++++++++++-------------
 1 file changed, 29 insertions(+), 13 deletions(-)

diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index e9aae4686536..5350e1b61c06 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -70,6 +70,7 @@
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/mm.h>
+#include <linux/bootmem.h>
 #include <linux/string.h>
 #include <linux/socket.h>
 #include <linux/sockios.h>
@@ -463,8 +464,10 @@ static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst,
 	return neigh_create(&arp_tbl, pkey, dev);
 }
 
-#define IP_IDENTS_SZ 2048u
-
+/* Hash tables of size 2048..262144 depending on RAM size.
+ * Each bucket uses 8 bytes.
+ */
+static u32 ip_idents_mask __read_mostly;
 static atomic_t *ip_idents __read_mostly;
 static u32 *ip_tstamps __read_mostly;
 
@@ -474,12 +477,16 @@ static u32 *ip_tstamps __read_mostly;
  */
 u32 ip_idents_reserve(u32 hash, int segs)
 {
-	u32 *p_tstamp = ip_tstamps + hash % IP_IDENTS_SZ;
-	atomic_t *p_id = ip_idents + hash % IP_IDENTS_SZ;
-	u32 old = ACCESS_ONCE(*p_tstamp);
-	u32 now = (u32)jiffies;
+	u32 bucket, old, now = (u32)jiffies;
+	atomic_t *p_id;
+	u32 *p_tstamp;
 	u32 delta = 0;
 
+	bucket = hash & ip_idents_mask;
+	p_tstamp = ip_tstamps + bucket;
+	p_id = ip_idents + bucket;
+	old = ACCESS_ONCE(*p_tstamp);
+
 	if (old != now && cmpxchg(p_tstamp, old, now) == old)
 		delta = prandom_u32_max(now - old);
 
@@ -2936,18 +2943,27 @@ struct ip_rt_acct __percpu *ip_rt_acct __read_mostly;
 
 int __init ip_rt_init(void)
 {
+	void *idents_hash;
 	int rc = 0;
 	int cpu;
 
-	ip_idents = kmalloc(IP_IDENTS_SZ * sizeof(*ip_idents), GFP_KERNEL);
-	if (!ip_idents)
-		panic("IP: failed to allocate ip_idents\n");
+	/* For modern hosts, this will use 2 MB of memory */
+	idents_hash = alloc_large_system_hash("IP idents",
+					      sizeof(*ip_idents) + sizeof(*ip_tstamps),
+					      0,
+					      16, /* one bucket per 64 KB */
+					      0,
+					      NULL,
+					      &ip_idents_mask,
+					      2048,
+					      256*1024);
+
+	ip_idents = idents_hash;
 
-	prandom_bytes(ip_idents, IP_IDENTS_SZ * sizeof(*ip_idents));
+	prandom_bytes(ip_idents, (ip_idents_mask + 1) * sizeof(*ip_idents));
 
-	ip_tstamps = kcalloc(IP_IDENTS_SZ, sizeof(*ip_tstamps), GFP_KERNEL);
-	if (!ip_tstamps)
-		panic("IP: failed to allocate ip_tstamps\n");
+	ip_tstamps = idents_hash + (ip_idents_mask + 1) * sizeof(*ip_idents);
+	memset(ip_tstamps, 0, (ip_idents_mask + 1) * sizeof(*ip_tstamps));
 
 	for_each_possible_cpu(cpu) {
 		struct uncached_list *ul = &per_cpu(rt_uncached_list, cpu);
-- 
2.30.2


  parent reply	other threads:[~2021-06-28 15:21 UTC|newest]

Thread overview: 77+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-28 14:38 [PATCH 4.9 00/71] 4.9.274-rc1 review Sasha Levin
2021-06-28 14:38 ` [PATCH 4.9 01/71] net: ieee802154: fix null deref in parse dev addr Sasha Levin
2021-06-28 14:38 ` [PATCH 4.9 02/71] HID: hid-sensor-hub: Return error for hid_set_field() failure Sasha Levin
2021-06-28 14:38 ` [PATCH 4.9 03/71] HID: Add BUS_VIRTUAL to hid_connect logging Sasha Levin
2021-06-28 14:38 ` [PATCH 4.9 04/71] HID: usbhid: fix info leak in hid_submit_ctrl Sasha Levin
2021-06-28 14:38 ` [PATCH 4.9 05/71] ARM: OMAP2+: Fix build warning when mmc_omap is not built Sasha Levin
2021-06-28 14:38 ` [PATCH 4.9 06/71] HID: gt683r: add missing MODULE_DEVICE_TABLE Sasha Levin
2021-06-28 14:38 ` [PATCH 4.9 07/71] gfs2: Fix use-after-free in gfs2_glock_shrink_scan Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 08/71] scsi: target: core: Fix warning on realtime kernels Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 09/71] ethernet: myri10ge: Fix missing error code in myri10ge_probe() Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 10/71] net: ipconfig: Don't override command-line hostnames or domains Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 11/71] rtnetlink: Fix missing error code in rtnl_bridge_notify() Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 12/71] net/x25: Return the correct errno code Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 13/71] net: " Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 14/71] fib: " Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 15/71] dmaengine: QCOM_HIDMA_MGMT depends on HAS_IOMEM Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 16/71] dmaengine: stedma40: add missing iounmap() on error in d40_probe() Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 17/71] mm: hwpoison: change PageHWPoison behavior on hugetlb pages Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 18/71] batman-adv: Avoid WARN_ON timing related checks Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 19/71] net: ipv4: fix memory leak in netlbl_cipsov4_add_std Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 20/71] net: rds: fix memory leak in rds_recvmsg Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 21/71] udp: fix race between close() and udp_abort() Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 22/71] rtnetlink: Fix regression in bridge VLAN configuration Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 23/71] netfilter: synproxy: Fix out of bounds when parsing TCP options Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 24/71] alx: Fix an error handling path in 'alx_probe()' Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 25/71] net: stmmac: dwmac1000: Fix extended MAC address registers definition Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 26/71] qlcnic: Fix an error handling path in 'qlcnic_probe()' Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 27/71] netxen_nic: Fix an error handling path in 'netxen_nic_probe()' Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 28/71] net: cdc_ncm: switch to eth%d interface naming Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 29/71] net: usb: fix possible use-after-free in smsc75xx_bind Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 30/71] net: ipv4: fix memory leak in ip_mc_add1_src Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 31/71] net/af_unix: fix a data-race in unix_dgram_sendmsg / unix_release_sock Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 32/71] be2net: Fix an error handling path in 'be_probe()' Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 33/71] net: hamradio: fix memory leak in mkiss_close Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 34/71] net: cdc_eem: fix tx fixup skb leak Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 35/71] net: ethernet: fix potential use-after-free in ec_bhf_remove Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 36/71] scsi: core: Put .shost_dev in failure path if host state changes to RUNNING Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 37/71] radeon: use memcpy_to/fromio for UVD fw upload Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 38/71] can: bcm: fix infoleak in struct bcm_msg_head Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 39/71] tracing: Do no increment trace_clock_global() by one Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 40/71] PCI: Mark TI C667X to avoid bus reset Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 41/71] PCI: Mark some NVIDIA GPUs " Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 42/71] ARCv2: save ABI registers across signal handling Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 43/71] dmaengine: pl330: fix wrong usage of spinlock flags in dma_cyclc Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 44/71] net: fec_ptp: add clock rate zero check Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 45/71] can: bcm/raw/isotp: use per module netdevice notifier Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 46/71] usb: dwc3: core: fix kernel panic when do reboot Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 47/71] tracing: Do not stop recording cmdlines when tracing is off Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 48/71] tracing: Do not stop recording comms if the trace file is being read Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 49/71] x86/fpu: Reset state for all signal restore failures Sasha Levin
2021-06-28 14:39 ` Sasha Levin [this message]
2021-06-28 14:39 ` [PATCH 4.9 51/71] i40e: Be much more verbose about what we can and cannot offload Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 52/71] ARM: 9081/1: fix gcc-10 thumb2-kernel regression Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 53/71] Makefile: Move -Wno-unused-but-set-variable out of GCC only block Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 54/71] arm64: perf: Disable PMU while processing counter overflows Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 55/71] Revert "PCI: PM: Do not read power state in pci_enable_device_flags()" Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 56/71] mac80211: remove warning in ieee80211_get_sband() Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 57/71] cfg80211: call cfg80211_leave_ocb when switching away from OCB Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 58/71] mac80211: drop multicast fragments Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 59/71] ping: Check return value of function 'ping_queue_rcv_skb' Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 60/71] inet: annotate date races around sk->sk_txhash Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 61/71] net: caif: fix memory leak in ldisc_open Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 62/71] net/packet: annotate accesses to po->bind Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 63/71] net/packet: annotate accesses to po->ifindex Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 64/71] r8152: Avoid memcpy() over-reading of ETH_SS_STATS Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 65/71] sh_eth: " Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 66/71] r8169: " Sasha Levin
2021-06-28 14:39 ` [PATCH 4.9 67/71] net: qed: Fix memcpy() overflow of qed_dcbx_params() Sasha Levin
2021-06-28 14:40 ` [PATCH 4.9 68/71] net: ll_temac: Avoid ndo_start_xmit returning NETDEV_TX_BUSY Sasha Levin
2021-06-28 14:40 ` [PATCH 4.9 69/71] nilfs2: fix memory leak in nilfs_sysfs_delete_device_group Sasha Levin
2021-06-28 14:40 ` [PATCH 4.9 70/71] i2c: robotfuzz-osif: fix control-request directions Sasha Levin
2021-06-28 14:40 ` [PATCH 4.9 71/71] Linux 4.9.274-rc1 Sasha Levin
2021-06-29 12:50 ` [PATCH 4.9 00/71] 4.9.274-rc1 review Naresh Kamboju
2021-06-29 14:11 ` Guenter Roeck
2021-06-29 22:07   ` Sasha Levin
2021-06-29 18:18 ` Guenter Roeck
2021-06-29 23:43 ` Guenter Roeck

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210628144003.34260-51-sashal@kernel.org \
    --to=sashal@kernel.org \
    --cc=aksecurity@gmail.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=w@1wt.eu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).