sh_eth: fix kernel oops in skb_put()

In a low memory situation the following kernel oops occurs:

Unable to handle kernel NULL pointer dereference at virtual address 00000050
pgd = 8490c000
[00000050] *pgdF51e831, *pte000000, *ppte000000
Internal error: Oops: 17 [#1] PREEMPT ARM
Modules linked in:
CPU: 0    Not tainted  (3.4-at16 #9)
PC is at skb_put+0x10/0x98
LR is at sh_eth_poll+0x2c8/0xa10
pc : [<8035f780>]    lr : [<8028bf50>]    psr: 60000113

sp : 84eb1a90 ip : 84eb1ac8 fp : 84eb1ac4
r10: 0000003f r9 : 000005ea r8 : 00000000
r7 : 00000000 r6 : 940453b0 r5 : 00030000 r4 : 9381b180 r3 : 00000000 r2 : 00000000 r1 : 000005ea r0 : 00000000
Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user
Control: 10c53c7d Table: 4248c059 DAC: 00000015 Process klogd (pid: 2046, stack limit = 0x84eb02e8) [...]

This is because netdev_alloc_skb() fails and 'mdp->rx_skbuff[entry]' is left NULL but sh_eth_rx() later uses it without checking. Add such check...

Reported-by Yasushi SHOJI <[email protected]>

Signed-off-by Sergei Shtylyov <[email protected]>

---
This patch is against DaveM's 'net.git' repo.

 drivers/net/ethernet/renesas/sh_eth.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

Index: net/drivers/net/ethernet/renesas/sh_eth.c =================================> --- net.orig/drivers/net/ethernet/renesas/sh_eth.c +++ net/drivers/net/ethernet/renesas/sh_eth.c

(quoted diff) +1 -0

@@ -1481,6 +1481,7 @@ static int sh_eth_rx(struct net_device *
 		if (mdp->cd->shift_rd0)
 			desc_status >>= 16;
 
+		skb = mdp->rx_skbuff[entry];
 		if (desc_status & (RD_RFS1 | RD_RFS2 | RD_RFS3 | RD_RFS4 |
 				   RD_RFS5 | RD_RFS6 | RD_RFS10)) {
 			ndev->stats.rx_errors++;

@@ -1496,12 +1497,11 @@ static int sh_eth_rx(struct net_device *
 				ndev->stats.rx_missed_errors++;
 			if (desc_status & RD_RFS10)
 				ndev->stats.rx_over_errors++;

Quoted 48 lines

In a low memory situation the following kernel oops occurs:

Unable to handle kernel NULL pointer dereference at virtual address 00000050
pgd = 8490c000
[00000050] *pgdF51e831, *pte000000, *ppte000000
Internal error: Oops: 17 [#1] PREEMPT ARM
Modules linked in:
CPU: 0    Not tainted  (3.4-at16 #9)
PC is at skb_put+0x10/0x98
LR is at sh_eth_poll+0x2c8/0xa10
pc : [<8035f780>]    lr : [<8028bf50>]    psr: 60000113

sp : 84eb1a90 ip : 84eb1ac8 fp : 84eb1ac4
r10: 0000003f r9 : 000005ea r8 : 00000000
r7 : 00000000 r6 : 940453b0 r5 : 00030000 r4 : 9381b180 r3 : 00000000 r2 : 00000000 r1 : 000005ea r0 : 00000000
Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user
Control: 10c53c7d Table: 4248c059 DAC: 00000015 Process klogd (pid: 2046, stack limit = 0x84eb02e8) [...]

This is because netdev_alloc_skb() fails and 'mdp->rx_skbuff[entry]' is left NULL but sh_eth_rx() later uses it without checking. Add such check...

Reported-by Yasushi SHOJI <[email protected]>

Signed-off-by Sergei Shtylyov <[email protected]>

---
This patch is against DaveM's 'net.git' repo.

  drivers/net/ethernet/renesas/sh_eth.c |    4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)

Index: net/drivers/net/ethernet/renesas/sh_eth.c =================================>> --- net.orig/drivers/net/ethernet/renesas/sh_eth.c +++ net/drivers/net/ethernet/renesas/sh_eth.c

(quoted diff) +1 -0

@@ -1481,6 +1481,7 @@ static int sh_eth_rx(struct net_device *
  		if (mdp->cd->shift_rd0)
  			desc_status >>= 16;
 
+		skb = mdp->rx_skbuff[entry];
  		if (desc_status & (RD_RFS1 | RD_RFS2 | RD_RFS3 | RD_RFS4 |
  				   RD_RFS5 | RD_RFS6 | RD_RFS10)) {
  			ndev->stats.rx_errors++;

@@ -1496,12 +1497,11 @@ static int sh_eth_rx(struct net_device *
  				ndev->stats.rx_missed_errors++;
  			if (desc_status & RD_RFS10)
  				ndev->stats.rx_over_errors++;

Review static int sh_eth_rx(struct net_device * · line 1507

Quoted 10 lines

(quoted diff) +1 -2

-		} else {
+		} else	if (skb) {
  			if (!mdp->cd->hw_swap)
  				sh_eth_soft_swap(
  					phys_to_virt(ALIGN(rxdesc->addr, 4)),
  					pkt_len + 2);
-			skb = mdp->rx_skbuff[entry];
  			mdp->rx_skbuff[entry] = NULL;
  			if (mdp->cd->rpadir)
  				skb_reserve(skb, NET_IP_ALIGN);

This certainly prevents from a bad access, however, some odd thing is going on. Once I hit a low memory situation with this patch, network thorough-put and response is very bad.

telnet, ping, wget takes loong time.

PING 172.16.2.13 (172.16.2.13) 56(84) bytes of data. 64 bytes from 172.16.2.13: icmp_seq=5 ttld time=0.223 ms 64 bytes from 172.16.2.13: icmp_seq=6 ttld time=0.195 ms 64 bytes from 172.16.2.13: icmp_seq=7 ttld time=0.203 ms 64 bytes from 172.16.2.13: icmp_seq=8 ttld time=0.219 ms 64 bytes from 172.16.2.13: icmp_seq=9 ttld time=0.165 ms 64 bytes from 172.16.2.13: icmp_seq ttld time=0.171 ms 64 bytes from 172.16.2.13: icmp_seq=1 ttld time�23 ms 64 bytes from 172.16.2.13: icmp_seq=2 ttld time€22 ms 64 bytes from 172.16.2.13: icmp_seq=3 ttld timep14 ms 64 bytes from 172.16.2.13: icmp_seq=4 ttld time`06 ms

I'll investigate it.

Shoji-san, can I push this patch to net.git? I doubt that it has
ill effects in itself -- the reason of the slowdown you're seeing should be somewhere else...

In a low memory situation the following kernel oops occurs:

This is because netdev_alloc_skb() fails and 'mdp->rx_skbuff[entry]' is left NULL but sh_eth_rx() later uses it without checking. Add such check...

Reported-by Yasushi SHOJI <[email protected]>

Signed-off-by Sergei Shtylyov <[email protected]>