From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Tue, 30 Apr 2019 13:44:19 +0300
Subject: net: dsa: bcm_sf2: fix buffer overflow doing set_rxnfc
Git-commit: f949a12fd697479f68d99dc65e9bbab68ee49043
Patch-mainline: v5.1
References: networking-stable-19_05_04
The "fs->location" is a u32 that comes from the user in ethtool_set_rxnfc().
We can't pass unclamped values to test_bit() or it results in an out of
bounds access beyond the end of the bitmap.
Fixes: 7318166cacad ("net: dsa: bcm_sf2: Add support for ethtool::rxnfc")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
---
drivers/net/dsa/bcm_sf2_cfp.c | 6 ++++++
1 file changed, 6 insertions(+)
--- a/drivers/net/dsa/bcm_sf2_cfp.c
+++ b/drivers/net/dsa/bcm_sf2_cfp.c
@@ -130,6 +130,9 @@ static int bcm_sf2_cfp_rule_set(struct d
(fs->m_ext.vlan_etype || fs->m_ext.data[1]))
return -EINVAL;
+ if (fs->location != RX_CLS_LOC_ANY && fs->location >= CFP_NUM_RULES)
+ return -EINVAL;
+
if (fs->location != RX_CLS_LOC_ANY &&
test_bit(fs->location, priv->cfp.used))
return -EBUSY;
@@ -330,6 +333,9 @@ static int bcm_sf2_cfp_rule_del(struct b
int ret;
u32 reg;
+ if (loc >= CFP_NUM_RULES)
+ return -EINVAL;
+
/* Refuse deletion of unused rules, and the default reserved rule */
if (!test_bit(loc, priv->cfp.used) || loc == 0)
return -EINVAL;